[Advances in Agronomy 85] Donald L. Sparks - (2005 Elsevier Academic Press)



вход по аккаунту


[Advances in Agronomy 85] Donald L. Sparks - (2005 Elsevier Academic Press)

код для вставкиСкачать
Advisory Board
John S. Boyer
University of Delaware
Paul Bertsch
University of Georgia
Ronald L. Phillips
University of Minnesota
Kate M. Scow
University of California, Davis
Larry P. Wilding
Texas A&M University
Emeritus Advisory Board Members
Kenneth J. Frey
Iowa State University
Eugene J. Kamprath
North Carolina State University
Martin Alexander
Cornell University
Prepared in cooperation with the
American Society of Agronomy Monographs Committee
David D. Baltensperger, Chair
Lisa K. Al-Almoodi
John M. Baker
Kenneth A. Barbarick
David M. Burner
Warren A. Dick
L. Richard Drees
Jeffrey E. Herrick
Bingru Huang
Michel D. Ransom
Craig A. Roberts
David L. Wright
Edited by
Donald L. Sparks
Department of Plant and Soil Sciences
University of Delaware
Newark, Delaware
Elsevier Academic Press
525 B Street, Suite 1900, San Diego, California 92101-4495, USA
84 Theobald’s Road, London WCIX 8RR, UK
This book is printed on acid-free paper.
Copyright ß 2005, Elsevier Inc. All Rights Reserved.
No part of this publication may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopy, recording, or any information
storage and retrieval system, without permission in writing from the Publisher.
The appearance of the code at the bottom of the first page of a chapter in this book
indicates the Publisher’s consent that copies of the chapter may be made for
personal or internal use of specific clients. This consent is given on the condition,
however, that the copier pay the stated per copy fee through the Copyright Clearance
Center, Inc. (www.copyright.com), for copying beyond that permitted by
Sections 107 or 108 of the U.S. Copyright Law. This consent does not
extend to other kinds of copying, such as copying for general distribution,
for advertising or promotional purposes, for creating new collective works, or
for resale. Copy fees for pre-2005 chapters are as shown on the title pages. If no fee
code appears on the title page, the copy fee is the same as for current chapters.
0065-2113/2005 $35.00
Permissions may be sought directly from Elsevier’s Science & Technology Rights
Department in Oxford, UK: phone: (þ44) 1865 843830, fax: (þ44) 1865 853333,
E-mail: permissions@elsevier.com.uk. You may also complete your request on-line
via the Elsevier homepage (http://elsevier.com), by selecting ‘‘Customer Support’’
and then ‘‘Obtaining Permissions.’’
For all information on all Elsevier Academic Press publications
visit our web site at www.books.elsevier.com
ISBN: 0-12-000783-5
05 06 07 08 09 9 8 7 6 5 4 3 2
CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
H. Lin, J. Bouma, L. P. Wilding, J. L. Richardson,
M. Kutı́lek and D. R. Nielsen
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
II. Hydropedology as an Intertwined Branch of Soil Science
and Hydrology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Pedology, Soil Physics, and Hydrology. . . . . . . . . . . . . . . . . . . . .
B. Hydropedology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
III. Fundamentals and Applications of Hydropedology . . . . . . . . . . . . . .
A. Soil Structure and Layering as Indicators of Flow and Transport
Characteristics in Soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Soil Morphology as Signatures of Soil Hydrology . . . . . . . . . . . .
C. Water Movement over the Landscape in Relation to
Soil Cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Hydrology as a Factor of Soil Formation and a Driving Force
of Dynamic Soil System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IV. Future Needs in Advancing Hydropedology . . . . . . . . . . . . . . . . . . .
A. Systems Approaches to Understanding and Communicating
Landscape–Soil–Water Dynamics . . . . . . . . . . . . . . . . . . . . . . . . .
B. From Variability to Pattern and Their Relations to Scale . . . . . .
C. From Pedotransfer Functions to Soil Inference Systems and
Hydropedoinformatics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Education of the Next Generation of Soil Scientists and
Hydrologists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
V. Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
John A. Howard and Elizabeth Hood
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
II. Technology Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Generation of Transgenic Material . . . . . . . . . . . . . . . . . . . . . . . .
B. Protein Expression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
III. Production Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Harvesting/Transport/Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Tissue Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Extraction/Purification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IV. Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. High-Purity Human Health Products . . . . . . . . . . . . . . . . . . . . . .
B. Orally Delivered Products. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Industrial Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
V. Public Acceptance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VI. Conclusions and Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. M. Oliver, C. D. Clegg, P. M. Haygarth and
A. L. Heathwaite
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
II. Pathogens in Livestock Wastes . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Protozoa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Viruses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
III. Detection and Enumeration Techniques . . . . . . . . . . . . . . . . . . . . . .
A. Culture-Based Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Direct Counting Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Molecular Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IV. Transfer from Soil to Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Lateral Surface Pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Matrix Flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Soil Retention Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Bypass Mechanisms in Soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
E. Movement via Growth and Motility. . . . . . . . . . . . . . . . . . . . . . .
F. The Role of Soil Mesofauna. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
V. The Role of Colloids in Facilitating Transfer . . . . . . . . . . . . . . . . . .
VI. Factors Affecting Survival . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Survival in Livestock Wastes . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Survival in Soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Survival in Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VII. Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
L. Wu, M. B. McGechan, C. A. Watson
and J. A. Baddeley
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. The Future Agronomic Challenge . . . . . . . . . . . . . . . . . . . . . . . .
B. Why Model Roots? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Environment/Root Interactions . . . . . . . . . . . . . . . . . . . . . . . . . .
II. Current Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Available Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Selected Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
III. Model Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. General Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Branching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Root Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
E. Data Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IV. Extending the Scope of Current Models . . . . . . . . . . . . . . . . . . . . . .
A. Root Mortality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Interaction with the Environment . . . . . . . . . . . . . . . . . . . . . . . . .
C. Water Uptake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Nutrient Uptake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
E. Photoassimilate Availability and Root Development . . . . . . . . . .
F. Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
V. Structure of an Integrated Model. . . . . . . . . . . . . . . . . . . . . . . . . . . .
VI. Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
R. J. Haynes
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
II. Total Soil Organic Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Attainment of Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Effects of Agricultural Practice . . . . . . . . . . . . . . . . . . . . . . . . . . .
III. Particulate Organic Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Method of Quantification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Nature of the Fraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Amounts Present in Soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Management-Induced Changes . . . . . . . . . . . . . . . . . . . . . . . . . . .
E. Seasonal Fluctuations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
F. Significance to Soil Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IV. Dissolved Organic Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Method of Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Nature of the Fraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Biodegradability of DOM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Adsorbed Organic Matter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
E. Quantities of DOM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
F. Management-Induced Changes . . . . . . . . . . . . . . . . . . . . . . . . . . .
G. Seasonal Fluctuations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
H. Significance to Soil Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
V. Extractable Forms of Organic Matter . . . . . . . . . . . . . . . . . . . . . . . .
A. Hot Water-Extractable Organic Matter . . . . . . . . . . . . . . . . . . . .
B. Dilute Acid-Hydrolyzable C . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Permanganate-Oxidizable C . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VI. Potentially Mineralizable C and N. . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Method of Quantification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Nature of the Fraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Relationship with Other Pools . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Amounts Present in Soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
E. Management-Induced Changes . . . . . . . . . . . . . . . . . . . . . . . . . . .
F. Seasonal Flunctuations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
G. Significance to Soil Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VII. Synthesis and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Significance of Labile Organic Matter Fractions . . . . . . . . . . . . .
B. Practical Value of Labile Organic Matter Fractions . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Yadvinder-Singh, Bijay-Singh and J. Timsina
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Availability of Crop Residues in Rice-Based Cropping Systems . . . .
Management Options for Crop Residues . . . . . . . . . . . . . . . . . . . . . .
Crop Residue Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Kinetics of Crop Residue Decomposition . . . . . . . . . . . . . . . . . . .
B. Factors Affecting Residue Decomposition . . . . . . . . . . . . . . . . . .
C. Fallow Period and Crop Residue Management . . . . . . . . . . . . . .
Crop Residue Management Effects on Nutrient Availability in Soils
A. Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Phosphorus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Potassium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Sulfur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
E. Micronutrients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Effect of Crop Residues on Soil Properties . . . . . . . . . . . . . . . . . . . .
A. Soil Fertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Chemical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Physical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Biological Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
E. Crop Residues for Reclamation of Salt-Affected Soils . . . . . . . . .
Biological Nitrogen Fixation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Phytotoxicity Associated with Crop Residue Incorporation
into the Soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Weed Control and Herbicide Efficiency . . . . . . . . . . . . . . . . . . . . . . .
Emission of Greenhouse Gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Methane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Nitrous Oxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Mitigation Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Agronomic Responses to Crop Residue Management . . . . . . . . . . . .
A. Rice–Wheat Cropping System . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Rice–Rice Cropping System . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Rice–Legume Cropping System . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Other Rice-Based Cropping Systems . . . . . . . . . . . . . . . . . . . . . .
Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Research Needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
G. J. Ash, A. Albiston and E. J. Cother
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
II. Plant Description and Characteristics . . . . . . . . . . . . . . . . . . . . . . . .
III. Plant Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Seed Propagation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Vegetative Propagation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IV. Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Environmental Effects and Heritability . . . . . . . . . . . . . . . . . . . .
B. DNA Markers for Jojoba Clones . . . . . . . . . . . . . . . . . . . . . . . . .
C. Genetics of Wax Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
V. Agronomic Practices and Plant Adaptation . . . . . . . . . . . . . . . . . . . .
A. Yield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Water Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
D. Irrigation and Fertilizer Effects . . . . . . . . . . . . . . . . . . . . . . . . . .
E. Salt Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
F. Soil pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
G. Temperature Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
H. Frost Susceptibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I. Mycorrhizal Status of Jojoba . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VI. Diseases and Insect Pests of Jojoba . . . . . . . . . . . . . . . . . . . . . . . . . .
A. Endophytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B. Insect and Arthropod Pests . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C. Diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VII. Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Numbers in parentheses indicate the pages on which the authors’ contributions begin.
A. Albiston (409), Farrer Centre, School of Agriculture, Charles Sturt University,
Wagga Wagga NSW 2678, Australia
G. J. Ash (409), Farrer Centre, School of Agriculture, Charles Sturt University,
Wagga Wagga NSW 2678, Australia
J. A. Baddeley (181), Crop and Soil Research Group, SAC, Craibstone Estate,
Aberdeen AB21 9YA, United Kingdom
J. Bouma (1), Laboratory of Soil Science and Geology, Wageningen
University and Research Center, 6701 AR Wageningen, The Netherlands
C. D. Clegg (125), Soil Science and Environmental Quality Team, North Wyke
Research Station, Okehampton, Devon EX20 2SB, United Kingdom
E. J. Cother (409), NSW Agriculture, Orange Agricultural Institute, Orange NSW
2800, Australia
P. M. Haygarth (125), Soil Science and Environmental Quality Team, North Wyke
Research Station, Okehampton, Devon EX20 2SB, United Kingdom
R. J. Haynes (221), Discipline of Soil Science, School of Applied Environmental
Sciences, University of Natal, Pietermaritzburg, Scottsville 3209, South Africa
A. L. Heathwaite (125), Centre for Sustainable Water Management, The Lancaster Environmental Centre, Lancaster University, Lancaster LA1 4YQ,
United Kingdom
Elizabeth Hood (91), Arkansas State University, Jonesboro, Arkansas 72403, USA
John A. Howard (91), Applied Biotechnology Institute, College Station, Texas
77845, USA
M. Kutı́lek (1), Faculty of Civil Engineering, Czech Technical University, 16636
Prague, Czech Republic
H. Lin (1), Department of Crop and Soil Sciences, The Pennsylvania State
University, University Park, Pennsylvania 16802, USA
M. B. McGechan (181), Land Economy Research Group, SAC, Bush Estate,
Penicuik EH26 0PH, United Kingdom
D. R. Nielsen (1), Department of Land, Air and Water Resources, University of
California, Davis, California 95616, USA
D. M. Oliver (125), Soil Science and Environmental Quality Team, North Wyke
Research Station, Okehampton, Devon EX20 2SB, United Kingdom, and
Department of Geography, University of Sheffield, Sheffield S10 2TN, United
J. L. Richardson (1), Department of Soil Science, North Dakota State University,
Fargo, North Dakota 58105, USA
Bijay-Singh (269), Department of Soils, Punjab Agricultural University,
Ludhiana 141 004, India
Yadvinder-Singh (269), Department of Soils, Punjab Agricultural University,
Ludhiana 141 004, India
J. Timsina (269), CSIRO Land and Water, Griffith NSW 2680, Australia
C. A. Watson (181), Crop and Soil Research Group, SAC, Craibstone Estate,
Aberdeen AB21 9YA, United Kingdom
L. P. Wilding (1), Department of Soil and Crop Sciences, Texas A & M University,
College Station, Texas 77843, USA
L. Wu (181), Crop and Soil Research Group, SAC, Craibstone Estate, Aberdeen
AB21 9YA, United Kingdom
Volume 85 contains seven excellent and state-of-the-art reviews on topics that
will be of great interest to crop and soil scientists.
Chapter 1 is a comprehensive review on advances in hydropedology. Topics
that are discussed include: hydropedology as an intertwined branch of soil
science and hydrology, fundamentals and applications of hydropedology, and
future needs in advancing hydropedology. Chapter 2 is a timely treatise on
bioindustrial and biopharmaceutical products produced in plants. Key factors
such as transformation, expression, growth, harvest, transport, storage, processing, and purification of the plant material are included. Chapter 3 assesses the
potential for pathogen transfer from grassland soils to surface waters. Topics
include pathogens in livestock waste, detection and enumeration techniques,
transfers from soil to water, the role of colloids in facilitating transfer, and factors
affecting survival. Chapter 4 is an interesting review on plant root-system
architecture models including discussions on model processes and evaluation of
various models with respect to growth, structure, root mortality, and water and
nutrient uptake. Chapter 5 deals with labile organic matter fractions and impacts
on the quality of agricultural soils. Discussions on total particulate dissolved, and
extractable forms of organic matter, as well as potentially mineralizable C and N,
are included. Chapter 6 is a comprehensive review on crop residue management
for nutrient cycling and improving soil productivity in rice-based cropping
systems in the tropics. Topics covered include availability of crop residues in
rice-based cropping systems, management options and crop residue decomposition effects of management on nutrient availability and soil properties, biological
nitrogen fixation, phytotoxicity associated with crop residue incorporation into
soil, and emission of greenhouse gases. The final Chapter deals with agronomic
and management aspects of jojoba, a new crop that is well suited for hot, dry
climates. The seed of jojoba contains a wax that is used in lubricants,
pharmaceuticals and cosmetics.
I appreciate the excellent reviews of the authors.
H. Lin,1 J. Bouma,2 L. P. Wilding,3 J. L. Richardson,4
M. Kutı́lek5 and D. R. Nielsen6
Department of Crop and Soil Sciences, The Pennsylvania State University,
University Park, Pennsylvania 16802, USA
Laboratory of Soil Science and Geology, Wageningen University and
Research Center, 6701 AR Wageningen, The Netherlands
Department of Soil and Crop Sciences, Texas A & M University,
College Station, Texas 77843, USA
Department of Soil Science, North Dakota State University,
Fargo, North Dakota 58105, USA
Faculty of Civil Engineering, Czech Technical University,
16636 Prague, Czech Republic
Department of Land, Air and Water Resources,
University of California, Davis, California 95616, USA
I. Introduction
II. Hydropedology as an Intertwined Branch of Soil Science
and Hydrology
A. Pedology, Soil Physics, and Hydrology
B. Hydropedology
III. Fundamentals and Applications of Hydropedology
A. Soil Structure and Layering as Indicators of Flow and Transport
Characteristics in Soils
B. Soil Morphology as Signatures of Soil Hydrology
C. Water Movement over the Landscape in Relation to Soil Cover
D. Hydrology as a Factor of Soil Formation and a Driving Force
of Dynamic Soil System
IV. Future Needs in Advancing Hydropedology
A. Systems Approaches to Understanding and Communicating
Landscape–Soil–Water Dynamics
B. From Variability to Pattern and Their Relations to Scale
C. From Pedotransfer Functions to Soil Inference Systems and
D. Education of the Next Generation of Soil Scientists and
V. Concluding Remarks
Advances in Agronomy, Volume 85
Copyright 2005, Elsevier Inc. All rights reserved.
0065-2113/05 $35.00
Hydropedology is an intertwined branch of soil science and hydrology
that encompasses multiscale basic and applied research of interactive pedological and hydrological processes and their properties in the unsaturated
zone. The synergistic integration of classical pedology with soil physics,
hydrology, and other related bio- and geosciences into hydropedology suggests a renewed perspective and a more integrated approach to studying
landscape–soil–water dynamics across scales. Pedality, layering of soil horizons, and soil–landscape relationships are three essential characteristics of
soils as occurring on the landscape. Fundamental issues of hydropedology
include (1) soil structure and layering as indicators of flow and transport
characteristics in field soils; (2) soil morphology as signatures of soil
hydrology; (3) water movement over the landscape; and (4) hydrology as a
factor of soil formation and a driving force of dynamic soil system. Hydrology affects and is affected by all of the five natural soil-forming factors
and the four general soil-forming processes. Hence, hydropedology offers
potential opportunities for quantifying soil-forming processes. Future
needs in advancing hydropedology are encapsulated in the philosophy of
‘‘bridging disciplines, scales, data, and education.’’ These include (1) systems
approaches to understanding and communicating landscape–soil–water dynamics; (2) addressing variability using patterns at various scales; (3) enhancing pedotransfer functions and developing soil inference systems and
hydropedoinformatics; and (4) education of the next generation of soil
scientists and hydrologists. Hydropedology calls for adequate attention to
soil morphology (including soil structure) in the field and soil patterns over
the landscape to guide optimal soil physical and hydrological measurements,
field monitoring and experimental designs, and understanding and modeling
of flow and transport in the critical zone. Identification and prediction of
patterns (spatial-temporal organizations) across multiple scales are coming
to the forefront in soil science and hydrology, which offer rich and comprehensive insights regarding variability and the underlying processes. We
suggest various hydropedological approaches to address diverse knowledge
gaps. Given its links to a wide array of environmental, ecological, geological,
agricultural, and natural resource issues of societal importance, hydropedology is emerging as a promising field that could contribute significantly to the
study of the pedosphere, the hydrological cycle, the earth’s critical zone, and
ß 2005 Elsevier Inc.
the earth system.
The U.S. National Research Council (NRC) recently identified integrated
studies of the earth’s critical zone as a compelling research area for the 21st
century (NRC, 2001a). The critical zone, as defined by the NRC (2001a),
extends through the root zone, deep vadose zone, and ground water zone
and includes the land surface and its canopy of vegetation, rivers, lakes, and
shallow seas (Figs. 1 and 2) (See Color Insert). Interactions at this interface
between the solid earth and its fluid envelopes determine the availability of
nearly every life-sustaining resource, hence the term ‘‘critical zone.’’ The
critical zone encompasses the pedosphere—the thin skin of soils on the
earth’s surface that represents a geomembrane across which water and
solutes, as well as energy, gases, solids, and organisms, are actively exchanged with the atmosphere, biosphere, hydrosphere, and lithosphere to
create a life-sustaining environment (Fig. 1).
Soil and water are two critical components of the critical zone. In fact,
soil–water interaction is a key interface between the biotic and abiotic and
thus is a key determinant of the state of the earth system. Water controls a
variety of soil physical, chemical, and biological processes that lead to the
formation of diverse soils that support an array of land uses and biological
communities. On the other hand, soils play a key role in the global hydrological and biogeochemical cycles, contribute to the maintenance of water
quality and ecosystem functions, and act as a living filter and remediation
medium for waste materials. The interactions of soil and water are so
intimate and complex that they cannot be studied in a piecemeal manner,
but rather as a system across spatial and temporal scales. In this spirit,
hydropedology is suggested for synergistic integration of knowledge from
relevant disciplines. Hydropedology is defined here as an intertwined branch
of soil science and hydrology that encompasses multiscale basic and applied
research of interactive pedological and hydrological processes and their
properties in the unsaturated zone. Lin (2003) suggested that hydropedology
functions as a bridge that could address (1) knowledge gaps between pedology, soil physics, hydrology, and other related bio- and geosciences; (2) scale
differences in microscopic, mesoscopic, and macroscopic studies of soil–
water interactions; and (3) data translations from soil survey databases
into soil hydraulic properties. Such a bridging signifies the potential unique
contributions that hydropedology can make to integrated soil and water
sciences. Hydropedology also shifts the focus of geology-rooted classical
pedology—a branch of soil science that integrates and quantifies the
morphology, formation, distribution, and classification of soils as natural
landscape entities—to a hydrology-driven approach with a landscape perspective, reflecting the crucial role of water in many environmental, ecological, geological, agricultural, and natural resource issues of societal
importance. Because interactions between the solid earth and its fluids
control almost every life-sustaining activity, hydropedology holds significant
potential to enhance our understanding of the earth’s critical zone and to
improve the modeling of flow and transport phenomena occurring in the
earth’s surface and subsurface environments.
From a broader perspective, hydropedology plays an increasing role in
interdisciplinary teams and panels formed to address complex environmental research and policy issues (Bouma, 2005). As highlighted in the following
list, seven working models or perceptions of soils may be used to evaluate the
relevancy of hydropedology to the study of the earth’s critical zone—within
each of these models, soil–water interaction plays a critical role.
1. Soil as a natural body: V. V. Dokuchaev’s (1893) theory of soil formation
(i.e., the soil has been formed under the influence of climate, organisms,
relief, parent material, and time) gave birth to modern soil science. Soil
science evolved at the beginning as a branch of geoscience (i.e., soil
viewed as a superficial geological entity and a weathered crust). In the
process of soil formation, water plays an important role, both directly,
through chemical weathering of rock and physical leaching and erosion
processes, and indirectly, through life support of soil biota and vegetation. Soil water is also an active agent in the transformation and translocation of organic and mineral materials and influences the deposition and
resistance of soils to erosion.
2. Soil as a water reservoir and transmitting mantle: Soil is an important
fresh water reservoir and a living filter that impacts water quantity and
quality as well as the overall global hydrological cycle. Without soil,
noncontinuous rainfall or snow could not be transformed into a continuous flow of water to plant roots. Together with ground water, soil also
transforms discontinuous precipitation into continuous discharges recognized as streams and rivers (Kutı́lek and Nielsen, 1994). Transport of
water soluble or suspended materials over the soil surface and through
the soil profile ultimately impacts the quality of surface and ground
waters, the global biogeochemical cycles, and the efficiency and fate of
anthropogenically applied chemicals.
3. Soil as a gas and energy regulating geoderma: Like the skin of human
bodies, the porous soil layer essentially functions as the skin of the earth
(termed ‘‘geoderma’’). It regulates the gas (including water vapor) exchange between the land and the atmosphere, greenhouse gas emissions
from the land, the temperature of the land and plant community, and the
energy balance on the earth’s surface. For example, the phase changes
of water in soil involve storage and release of latent heat that drives
the atmospheric circulation and redistributes both water and heat
globally. Water vapor is the most important of the greenhouse gases,
acting to regulate the earth’s surface temperature by absorbing and
returning to the earth much of the thermal radiation emitted there
(NRC, 1991). The soil and related land uses also play an important role
in regulating global carbon fluxes and sequestration that are related to
global warming.
4. Soil as a component in ecosystems: Soil is the life-giving substance for
vegetation and ecosystems, supporting and regulating the fluxes of air,
water, and nutrients for macro- and microorganisms. For instance, the
water in the uppermost 1–2 m of the earth’s crust (i.e., soil moisture) is
directly linked to the type and functionality of ecosystems and controls
much of the soil–plant–atmosphere continuum. Through its influences on
physical, chemical, and biological processes in the root zone, the soil moisture regime often dictates the pattern of plant distribution over the landscape (Eagleson, 2002). The soil microbial population across the landscape
is also strongly influenced by soil water availability and its distribution,
which in turn impacts microbially mediated biogeochemical reactions such
as nitrification and denitrification (Nielsen et al., 1996; Wagenet, 1998).
5. Soil as a medium for plant growth: Soil is an essential natural resource for
agricultural production and other diverse land uses. Soil–water flow mediates nutrient cycling in agroecosystems, influences crop yield variability,
and determines the need for drainage or irrigation. Precision agriculture,
for instance, requires accurate mapping of soil and moisture variability
across the landscape in order to apply chemicals precisely in the right
location at the right time. Many best management practices in agriculture (e.g., wetland protection or construction, riparian buffer strips to
reduce nonpoint source pollution entering into streams) often require
appropriate understanding of soil–water interactions across the landscape.
6. Soil as a material for engineering: Soil is widely utilized in engineering for
various applications, such as construction material for buildings, roads,
highways, and dams and as a disposal and remediation medium for
wastes of all kinds. It has been reported that the most expensive hazard
in the United States is not earthquakes or flooding, but the one caused by
the enormous structural damages due to soil shrink-swell (NRC, 2001a).
Soil shrink-swell is a phenomenon that is directly caused by soil–water
interactions. Soil mechanical properties as influenced by water content
are also crucial in hillslope stability, landslide or mudflow prevention,
and other structural protections.
7. Soil as an integral part of the environment: Encompassing all of the
preceding models of soil is the overarching environmental arena that is
cross-cutting and multidisciplinary in nature. Soil–water interactions are
significant in a variety of environmental issues of societal importance,
including, for example, sustainable land use planning, watershed management, water quality protection, contaminant fate, waste remediation, and
global environmental change.
Water fluxes into and through soils in the landscape are the essence of life,
and resemble in a way the manner in which blood circulates in a human body
(Bouma, 2005). We could even compare blood pressure with the pressure
potential of water in soil: when it is too high or too low, soil functioning is
clearly hampered. We could therefore speak of a throbbing landscape in which
water enters and leaves, on the timeframe of hours, weeks, years, and centuries.
Once water regimes have been characterized, physical, chemical, and biological
processes can be added, as they strongly depend on the water regime and on
interaction processes with the soil. In fact, the hydrological cycle is viewed as
the integrating process for the fluxes of water, energy, and the chemical elements (NRC, 1991). Richardson et al. (2001) also suggested that the study of
water and its effects on soil is a unifying principle in soil investigations.
The objectives of this chapter are three-fold: (1) to suggest hydropedology
as a promising interdisciplinary area that could contribute significantly
to the integrated studies of the earth’s critical zone; (2) to explore the
fundamental issues of hydropedology and its various applications; and (3)
to propose some needs for the future advancement of hydropedology.
Pedology, soil physics, and hydrology have been identified as the ‘‘cornerstones’’ of hydropedology (Fig. 2), although hydropedology is also linked
to other related bio- and geosciences such as geomorphology, hydrogeology,
ecohydrology, hydroclimatology, and other branches of soil science (Lin,
2003). The three cornerstone disciplines share many common interests in the
interdisciplinary environmental arena, particularly in areas related to water
flow and solute transport through field soils and over the landscape. Although traditionally the three disciplines have had contrasting focuses and
approaches in their investigations, the time is now ripe for synergistic
integration of the three to address complex flow and transport processes in
nature and landscape–soil–water dynamics across scales. Synergies are
expected from integrating pedologists’ expert knowledge of soil–landscape
relationships with soil physicists’ and hydrologists’ mathematical rigor of
flow and transport theory.
1. Different Views of Soil and Scales of Investigation
a. Pedology. Pedologists study soils in their natural settings, so they
view soil as a natural entity and traditionally have focused on field soil
profiles (called pedons) as observed in the landscape. Pedologists describe
soil profiles in situ using observable morphological characteristics and related
landscape features; then they collect bulk soil samples, undisturbed clods,
and sometimes small intact box samples based on soil generic horizons for
laboratory characterizations of various physical, chemical, and micromorphological properties (Fig. 3; See Color Insert). The pedon that pedologists
examine is commonly 1–2 m deep with an area ranging from 1 to 10 m2 (about
1 m wide and several meters long). When an auger or a probe truck is used in
examining field soils (e.g., in soil surveys), the actual volume of soil being
observed is much smaller, but the view of the soil–landscape relationship is
much larger. In the investigations of pedons, pedologists have placed a much
greater emphasis on the vertical relationships of horizons and soil processes
(Fig. 4; See Color Insert) than on the horizontal relationships (Fig. 5),
although the horizontal relationships are what they attempt to delineate in
soil mapping (Buol et al., 2001). Moreover, vertical investigations have been
limited to the upper 2 m beneath the earth’s surface (with an emphasis on the
root zone). In terms of time scale, much of pedological characterizations has
been based on deeper soils on static or static/temporal differentia formed over
a long period of time (up to a geological time scale). Most of the data in soil
survey databases have been collected at a window in time, and there is
precious little soil survey data that are actually of a dynamic basis.
One reason that pedology has received continued attention over the years
from scientists and land users is the success of pedology-based soil survey
interpretations, which have been used extensively. Indicating the relative
limitations for various land uses for any given soil type has been quite useful
for broad land use planning and land evaluation purposes. Of course,
modern problems require more quantitative procedures, as pointed out
elsewhere in this chapter, but it is often forgotten that the alternative for
qualitative or descriptive land use limitations is often nothing at all when
there are no opportunities for extensive measurements. Thus, descriptive
characterizations could still be very valuable, as has been well documented,
and may offer new opportunities as highlighted later in this chapter (e.g.,
Section IV.B).
The emphasis of pedology is now shifting from classification and inventory to understanding and quantifying spatially-temporally variable processes upon which the water cycle and ecosystems depend. While a huge success
in its own right, the exclusive focus on Soil Taxonomy (see illustrations in
Fig. 4) in the pedology community over the last four decades or so has
become an introspective exercise and has resulted in some unintended consequences: (1) Soil Taxonomy does not relate soils to landscapes well; (2) Soil
Taxonomy does not consider dynamic soil properties (such as hydraulic
properties and those affected by short-term land management); (3) Soil
Taxonomy is viewed by many as too complex for non-pedologists; and (4)
soil survey has focused on classifying soils and thus has neglected the
quantification of variability (or specific range of soil properties) within
taxonomic categories and soil map units, thus leading to a common assumption of ‘‘homogeneity’’ within soil taxa and map units by non-pedologists.
However, the purpose of soil surveys is to partition soils and landforms into
stratified subsets that are less variable (Soil Survey Division Staff, 1951,
1993). Quantification of map unit purity for different scales of soil maps is
an area needing improvements in modern soil surveys (Arnold and Wilding,
1991; Lin et al., 2004). An understanding of how soil maps are made, the
map scale involved, and the variability within map units is desirable for
integrating pedology with soil physics and hydrology.
b. Soil Physics. Soil physics deals with the physical properties of the
soil, with an emphasis on the state and transport of matter (especially water)
and energy in the soil. Soil physicists view soil as a porous medium through
which water, solute, gas, and heat move. Traditionally, soil physicists have
emphasized theoretical studies using mathematical models and laboratory
investigations using small samples (often 0.05–0.3 m in diameter and height,
and often consisting of ground-sieved soil materials instead of intact soils).
Small field plots have been used to verify laboratory or theoretical findings
or to understand flow and transport processes at a local scale. For instance,
most of our present understanding of infiltration stems from theoretical
investigations made in the laboratory and on 1 m2 field plots isolated from
many of the factors that are relevant in natural and larger area environments. The contemporary challenge is to extend such small-scale understanding to larger domains. In terms of time scale, classical soil physics
research has typically been in the order of minutes or hours to days, with
few studies lasting months or years.
Emphasis in soil physics over the past four decades has shifted from
laboratory and local scale processes to the field scale and, more recently,
watershed scale transport of water and chemicals (Corwin et al., 1998). The
importance of soil heterogeneity across a field and its impacts on soil
physical and hydrological properties were recognized by Nielsen et al.
(1973) and many others thereafter. However, many of the classical soil
physics theories rely on the assumption of uniform and inert porous materials. Complicating physical factors that occur in the field, such as soil heterogeneity, shrink-swell, soil aggregation, and various macropores, often make
classical soil physical theories too simplistic or invalid (e.g., Bouma, 2005;
Kutı́lek and Nielsen, 1994; Young, 1988). A further inadequacy is that these
Figure 5 Classical soil survey block diagrams showing soil mapping based on soil–landscape
relationships. Also illustrated are two contrasting cases of landscape hydrology in relation to
soil-landscapes. (A) Aerial photo map of a section (6 miles on a side) in Sheboygan Co., WI and
a block diagram showing landscape positions of major soils. (From Hole, 1976.) (B) A block
diagram showing landscape position of a soil catena on a drumlin and adjacent lowlands in
Dodge Co., WI, illustrating water flow direction and dynamic seasonal water table. (From Hole,
1976.) (C) Pattern of soil catena and parent materials in the Clarion-Nicollet-Canisteo association in Kossuth Co., IA. In this soil landscape, water and solutes in the uplands move down
into the Clarion soil profile where they either enter the ground water or move laterally until they
are discharged further down slope as return flow. (Courtesy of L. Steffen.)
theories have mostly ignored the effect of temperature gradients on water
movement. With the reality of field heterogeneous and structured soils, at
least three aspects would warrant a close alliance between soil physicists and
pedologists: (1) quantitative soil structural parameters (including macropores) need to be incorporated into the modeling of various types and scales
of preferential flow and transport; (2) landscape features should be
incorporated into the field and watershed scale soil physical and hydrological models; and (3) scale transfer and input data required for modeling must
be adequately addressed. As Kutı́lek and Nielsen (1994) pointed out, models
of soil hydrology should be related to the reality of a field or watershed by at
least two links: proper characterization of the physical parameters and the
scale of the model. They further stressed that ‘‘without properly taken field
data all our effort is futile.’’
c. Hydrology. Hydrology deals with the hydrological cycle, including
continental water processes at all scales and the global water balance.
Hydrologists traditionally often view soil as a more-or-less homogeneous
layer at the earth’s surface. The evolution of hydrology has been in the
direction of ever-increasing scale, from small catchment to large river basin
to the earth system, and from storm event to seasonal cycle to climatic trend
(NRC, 1991). Historically, the interest in catchment hydrology has been
more related to temporal patterns, in particular, that of streamflow. Because
of the focused interest in streamflow (an integrator of spatial responses),
hydrologists have managed to avoid confronting the challenges of spatial
heterogeneity (Grayson and Blöschl, 2000). A similar history is apparent in
ground water hydrology, where pumping tests have long provided a measure
of integrated aquifer response and thus distracted researchers from the
quantification of aquifer heterogeneity (Anderson, 1997). According to
Grayson and Blöschl (2000), the past few decades have heralded an explosion of interest in spatial variability in hydrology, from the pioneering work
on spatial heterogeneity in runoff producing processes during the 1960s and
early 1970s (e.g., Betson, 1964; Dunne and Black, 1970a,b), through the
development of spatially distributed hydrological models (now often coupled with geographic information system, or GIS) that provide a way to
interpret spatial response (e.g., Abbott and Refsgaard, 1996; Beven, 2002),
to the ever-increasing capabilities of remote-sensing methods that provide
information on state variables of fundamental importance to catchment
hydrology (e.g., Jackson and Le Vine, 1996). In view of the enormous
variability encountered in field soils, deterministic models of flow and transport processes are giving way to spatially stochastic concepts. In the hydrology community, interest now centers on how to describe the random
distribution functions of soil hydraulic properties and the extent of their
spatial correlations at various scales. A conceptual framework defining the
expectation and variance structure of the spatial and temporal heterogeneity
of soil water properties within each soil series or soil map unit from local to
global scales is greatly needed (NRC, 1991).
Soil physical processes play an important role in the hydrological cycle,
because the unsaturated zone is the interface between the atmosphere and
the ground water zone. It is important, therefore, that hydrologists have an
adequate understanding of soil physical processes so that appropriate theory
is used in hydrological models, and that soil physicists continue their endeavors to elucidate the complications that abound in the field and that often
vitiate predictions obtained through unmodified classical theory (e.g., Beven,
1986; Kutı́lek and Nielsen, 1994; Young, 1988). In this regard, pedology
could assist soil physicists and hydrologists in understanding the natural
variability and soil structure in the field. Advances in basic hydrological
research will depend on soil hydrological research, field characterizations of
soil hydraulic properties, and a close connection of mathematical modeling
with adequate field observations. In addition, connecting soil physics, hydrology, and pedology to landscape features would prove to be a meaningful
way to integrate the three. For example, topography provides a strong
clue to both the hydrological regime and pedological characteristics in a
landscape and may help the scaling up or down of observed phenomena.
With the increasing availability of the digital elevation model (DEM), along
with many other geospatial data and sophisticated computer software, the
integration of pedology, soil physics, and hydrology is greatly facilitated.
Debunking Stereotypical Visions and Creating Synergies
Through Integration
To truly ally pedology with soil physics and hydrology, there are some
stereotypical visions that require debunking:
. To many blue-blooded hydrologists and soil physicists, the activities of
pedologists are difficult to judge from a scientific point of view. In their
view, pedologists use ‘‘funny’’ names to describe soils and they make too
many empirical statements about soil behavior that are not necessarily
supported by measurements. On the other hand, pedologists are taken
back by the representations of natural soils in terms of homogeneity and
isotropy that soil physicists and hydrologists often make in their models,
which to pedologists clearly do not reflect real conditions being experienced
in the field.
. Pedology has its roots in soil survey, which considers landscape processes
and soil structure descriptions that have been somewhat neglected in
the period in which soil classification received the most attention. These
two aspects are critical for soil physics and hydrology to improve their
characterization of flow regimes in the field. At the same time, pedologists
can benefit from flow theories in soil physics and hydrology when transforming their qualitative descriptions into quantitative expressions that
are increasingly necessary to respond to the demands from diverse uses of
soil survey information and to provide inputs to environmental policy
The developments of pedology, soil physics, and hydrology are now converging at several fronts, including, for example, (1) facing the field reality,
(2) addressing spatial-temporal variability across scales, (3) understanding
the various processes involved, and (4) using quantitative modeling appropriately. These common grounds lead to the synergies that could be expected
out of integrating the three disciplines, as suggested by recent literatures and
professional activities (e.g., Lin, 2003). For example, combining pedological
and hydrological expertise can be particularly attractive when presenting soils
information to interdisciplinary panels and teams formed to address complex
global environmental issues and policies (Bouma, 2005).
Pedology is a rich discipline that has been overshadowed in the past by
the descriptive and qualitative approaches. But pedology has much to offer
to soil physics and hydrology, and vice versa. For instance, soil mapping
provides the classical foundation for our understanding of soil variation over
the landscape and its underlying causes; soil profile descriptions have been the
major source of information on in situ soil structure and various soil hydromorphological features that are signatures of soil hydrology; soil survey
databases provide a wealth of information that soil physicists and hydrologists could use in their modeling; soil classification offers a hierarchical system
for organizing, modeling, and transferring our knowledge about different soils
around the globe; and soil genesis provides insights regarding soil–landscape
evolution over time. On the other hand, soil hydrology is a major driving force
behind pedogenesis, soil morphology, and soil distribution. It controls a
variety of soil physical, chemical, and biological processes that lead to the
formation of different soils and diverse land uses. Soil moisture regimes play a
critical role in classifying soils, and the spatial-temporal distribution of water
provides clues regarding soil variability and mapping. Furthermore, with
increasing emphasis on human impacts and land management practices, the
rising interest in dynamic soil properties would require more attention to soil
physical and hydraulic properties and their relations to soil taxonomic or map
units. Hence, many knowledge gaps may be closed by integrating classical
pedology, soil physics, and hydrology. Such examples include the following:
. soil structure quantification and modeling its impacts on flow and trans-
port processes;
. preferential flow prediction at different scales and the determination of its
mechanisms and patterns;
. soil hydromorphology quantification and its relations to soil hydrology
(such as water table fluctuation);
. water movement over the landscape and its relations to the soil cover;
. soil–water mapping and soil–landscape modeling across scales;
. soil spatial-temporal variability and the underlying causes, and pattern
identifications and predictions of various soil and hydrological properties
and processes;
. scale bridging from laboratory to field, landscape, region, and globe;
. data bridging through approaches such as pedotransfer functions, including the understanding of the fundamental mechanisms and practical
The promotion of hydropedology as a new interdisciplinary field suggests
a renewed perspective and a more integrated approach to studying landscape–soil–water interactions across spatial and temporal scales. Although
many topics related to hydropedology have been studied considerably in the
past, many unresolved issues remain and future opportunities abound. From
a historic development point of view, some major activities are acknowledged here. Kutı́lek (1966, 1978) recognized the need for combining soil
physical theories with theories of soil genesis. He indicated that we could not
deal with soil hydrology and soil physics without considering the soil properties within the complexity of soil genesis. Later, Kutı́lek and Nielsen (1994)
suggested that the objectives of hydropedological studies are similar to soil
hydrology by a broad spectrum of scales, ranging from the soil pore scale up
to the regional and soil mapping scale. Bouma and Hole (1971) and Bouma
and Anderson (1973) started the investigations of the relationships between
field soil structure and hydraulic conductivity and suggested that soil morphometric analysis has a specific function in improving field estimates of soil
hydraulic conductivity. Thereafter, Bouma published extensively on linking
soil morphology to soil physics (e.g., Bouma, 1984, 1990, 2005). Fritsch and
Fitzpatrick (1994) used a pedo-hydrological method to construct conceptual
landscape–soil–water models that linked soil–landscape features to soil–
water processes, with an emphasis on soil–water flow systems and soil-forming/soil-change processes. Galusky (1997) and Galusky et al. (1998) referenced hydropedology as relating soil morphological indicators to water table
behavior. Rabenhorst et al. (1998), in a collective work done by pedologists in
the 1990s, suggested that soil morphological features could be quantitatively
linked to hydrological or biogeochemical parameters associated with soil
wetness. Richardson and Vepraskas (2001) further demonstrated that soil
morphology is a valuable field tool for evaluating soil hydrology.
In the following sections, we further develop the concept of hydropedology that builds upon the strengths of historic developments as well as
modern scientific and technologic advances.
Back to the Essence of Natural Soils
‘‘What is soil?’’ To answer this seminal question, a good understanding and
appreciation of this gift from nature and its essential characteristics in the
landscape are required. In the context of hydropedology, we suggest that
the following three aspects warrant a close examination—all three point
to the essential differences between natural soils and engineered soil materials
or other types of porous media, and all three are in the domain of pedology.
1. Pedality: Peds are naturally formed soil aggregates (heterogeneous
masses of solid particles bound together) with various strengths, sizes,
and shapes (e.g., strong, very fine granular or weak, coarse prismatic) (see
illustrations in Fig. 3). Ped strength, size, and shape combined are termed
pedality. These are soil structural features routinely described by pedologists in the field. Pedality, along with the interrelated pore space, is the
natural soil ‘‘architecture’’ (soil structure) that is influenced by the five
interacting soil-forming factors at the landscape scale and is governed by
interrelationships between inorganic-organic constituents and physical,
chemical, and biological processes at the meso- or microscopic scales.
A hierarchical organization of soil structure (Fig. 6) seems to be characteristic of most soils, where larger aggregates are often composed of an
agglomeration of smaller aggregates (Soil Survey Division Staff, 1993;
Tisdall and Oades, 1982). Water and chemical movement or retention,
mineral weathering or synthesis, plant root or insect activities, and microorganism habitats all are influenced strongly by such soil architecture.
Indeed, the word ‘‘ped’’ is well reflected in the term ‘‘pedology.’’ This first
essence of natural soils may be summarized as the need to get ‘‘back to
soil structure.’’
2. Layering of soil horizons: Pedons are three-dimensional (3-D) bodies
of soil showing an arrangement of soil horizons that are the results of
soil-forming processes over time. Various kinds and thicknesses of soil
horizons and how they organize in soil profiles reflect long-time pedogenesis and the past and current landscape processes (Figs. 4 and 5). Soil
horizonation is the basis for soil classification and mapping. In the U.S.
Soil Taxonomy, 18 diagnostic surface horizons and 30 diagnostic subsurface horizons have been identified, along with many additional diagnostic
features (Soil Survey Staff, 1999). The fact that natural soils are layered
has three significant implications for soil physics and hydrology: (1) any
Figure 6 Structure at different scales in a silty soil obtained with different instruments. The
fields-of-view differ by about one order of magnitude. Top left: photograph of a vertical profile.
Top middle: X-ray tomography of the A-horizon, resolution 0.5 mm/pixel (pores are dark). Top
right: X-ray micro-tomography, resolution 0.04 mm/pixel (pores are dark). Below, the images
are segmented into the structural units at the corresponding scales. Bottom left: two different
horizons. Bottom middle: dense aggregates (gray) within a loose matrix (white) and a few
macropores (black). Bottom right: pores (gray) within a porous matrix. (Modified from Vogel
and Roth, 2003.)
interface between soil layers slows down water downward movement;
(2) soil layering promotes lateral flow, especially in sloping landscapes
with water-restricting layers underneath; and (3) soil horizons of different
textures and structures often lead to preferential flow. This second essence
of natural soils may be summarized as the need to get ‘‘back to the field.’’
3. Soil–landscape relationships: Landscape is the portion of the land surface
(a population of landforms) that human eyes can comprehend in a single
view (Ruhe, 1969, 1975). Like landscape architecture, the word ‘‘landscape’’ emphasizes a visual aspect, so it is important that we glance
around the surroundings when dealing with soil and water issues in the
field. Landscape encompasses soil, water, vegetation, topography, geomorphology, geology, climate, human activities, and other factors. Landscape evolution has a lot to do with the throughflows of water, chemicals,
and energy (i.e., the hydrological cycle) and the movement of soils,
sediments, and other materials. Soil distribution over the landscape is
closely related to landforms and geomorphological processes (Fig. 5).
Soil–landscape relationships are the foundation for mapping and modeling soils. Depending on which soil-forming factor might dominate in a
given region, there are climosequences, biosequences, toposequences,
lithosequences, or chronosequences of soil distributions over the landscape at a given scale. As human and hydrological impacts on soil
distribution are being increasingly recognized, there are also anthroposequences (i.e., related soils that differ primarily due to the influence of
humans such as land uses) and hydrosequences (similar to the concept of
catena but with a focus on water as the dominant factor, and its scale is
more than hillslope). A systematic understanding of soil–landscape relationships would facilitate the formulation of conceptual and mathematical models of landscape hydrology. This third essence of natural soils
could be summarized as the need to get ‘‘back to the landscape.’’
Catalysts for Promoting Hydropedology
Three factors seem to be the catalysts for promoting hydropedology at
the present time. These factors, however, are not unique to hydropedology;
rather, they reflect the global trends in modern environmental scientific
research and integrated natural resources management:
1. Interdisciplinarity and systems approach: It is well recognized that the
progress of science depends increasingly on an advanced understanding
of the interrelationships among different fields and their components
(AAAS Council, 2001). A number of recent reports of the U.S. National
Research Council have highlighted the significance of integrated soil and
water studies in the context of agriculture (NRC, 1993a), ground water
vulnerability (NRC, 1993b), watershed management (NRC, 1999), earth
sciences (NRC, 2001a), water resources (NRC, 2001b), and environmental
sciences (NRC, 2001c). In addition, watershed approaches to natural
resources management have become a dominant concept at both local
and landscape scales. Watersheds are the logical features in ecosystems
within which to consider the integration of soil, water, landscape, agricultural and forest productivity, and social-economic factors. To address
diverse soil and water issues at various spatial and temporal scales, bridging traditional pedology with soil physics, hydrology, and other related
disciplines is necessary as well as synergistic. This bridging is justified not
only by the interrelationship among these disciplines but also by the
complex nature of the problems.
2. Landscape perspective and multiscale bridging: Over the past decades,
there has been a significant increase in the number of field studies
conducted at the landscape or watershed levels in order to study processes at a scale relevant to the issue at hand (such as nonpoint source
pollution, precision agriculture, sustainable land management, wetland
protection, and watershed management). For instance, the development
of site-specific farming forces researchers to address questions at the
larger scale rather than at the small plot scale that most classical agronomic researchers use (van Kessel and Wendroth, 2001). With landscape
perspective comes the need to address inherent variability in the field and
scale transfer from laboratory or small plot to the larger field and watershed scales, as well as the requirement of meaningful experimental design
and data analysis that take into account the spatial ‘‘scale triplet’’ (i.e.,
spacing, support, and extent) (Blöschl and Sivapalan, 1995) and
corresponding temporal variability (sampling time interval, smoothing
or averaging interval, and length of record) (e.g., Blöschl and Grayson,
2000). The controlling factors by which abiotic and biotic processes occur
along the continuum of a landscape should also be taken into account in
modeling and prediction. Translating information about soil and hydrological properties and processes across scales has emerged as a major
theme in contemporary soil science and hydrology (e.g., Hoosbeek et al.,
1998; Pachepsky et al., 2003).
3. Advancements in geospatial technologies and computer modeling: The era
of information technologies has had a significant impact on modern soil
science and hydrology. Especially relevant is the increasing availability
and wide use of geospatial technologies (including GIS, global positioning systems [GPS], and remote sensing) and simulation modeling. For
example, an integrated system of a geospatial database coupled with
solute transport modeling has been widely sought to address nonpoint
source pollution (e.g., Corwin et al., 1999). Spatially distributed hydrological modeling has become easier to use, and visualization of the results
has greatly improved. However, there has been little change in the
concepts on which the models are based and the ways in which they are
calibrated and used (Beven and Feyen, 2002). A sharp increase in
the quality and quantity of geospatial data, including voluminous
remote-sensing images, coupled with improved data-mining tools and
enhanced database management and distribution systems, will significantly improve our abilities to analyze the vast amount of environmental data
being collected and stored. In addition, new concepts of nonlinear dynamics (such as fractals, chaos, and fuzzy logic) and new tools (such as
spatial-temporal geostatistics, neural networks, and uncertainty analysis)
will help improve the extraction of useful information out of large databases. Such advancements make it feasible to integrate pedology, soil
physics, and hydrology to understand spatially variable and temporally
dynamic processes at various scales.
Domain and Characteristics of Hydropedology
Bearing in mind the aspects discussed previously, we venture to
suggest the domain and characteristics of hydropedology to include the
. Landscape–soil–water systems: Taking a holistic view of the landscape,
with the root in pedology and a focus on water as a driving force,
hydropedology emphasizes the system linkages, the state and pattern of
its component parts, interfacial fluxes, and dynamic changes including
those caused by human activities.
. Soil–water interactions across spatial-temporal scales: Hydropedology
attempts to characterize integrated physical, chemical, and biological
processes of soil–water interactions at all scales, including the transport
of chemicals and energy by the water flow, and the interrelationships
between soil distributions and hydrologic and geomorphic processes.
As illustrated in Fig. 2, hydropedology, in combination with hydrogeology in the vertical dimension, suggests an integrated approach to studying the
interactions of solid earth (soil and rock) and water. Soil investigations
should no longer be limited to the top 2 m of the earth’s surface but extend
well into the deeper vadose zone (including the contact zone with the aquifer
and perhaps the fluxes within that aquifer to the extent that they affect the
dynamic level of the water table). Hydropedology thus requires a concerted
effort to study the soil and underlying material to whatever depth is needed
to meet our scientific needs. Geologists are extending their investigations to
the surface and are including the biosphere and surficical processes, so it is
paramount that soil scientists redirect their efforts to interface with other
geoscientists in making hydropedology an earth, environmental, ecological,
and agronomic science.
In the horizontal perspective (Fig. 2), as a bridge connecting pedology,
soil physics, and hydrology, hydropedology integrates the pedon and
landscape paradigms to link phenomena occuring at the microscopic scale
(e.g., pores and aggregates) to mesoscopic (e.g., pedons and catenas) and
macroscopic (e.g., watersheds, regional, and global) scales. Hydropedology
brings the landscape back to pedology, which has been lost a bit in the soil
classification frenzy.
In terms of time dimension, hydropedology deals with both short- and
long-term characterizations of landscape–soil–water systems, from hourly to
annual and to geological time scales, in order to systematically understand
the role of soils in the hydrological cycle and the role of hydrology in
pedogenesis, soil morphology, soil survey, pedodiversity, and biogeochemical
Four Challenges to be Addressed
There are four overarching themes along the philosophy of ‘‘bridging disciplines, scales, data, and education’’ that hydropedology attempts to address:
1. Good science is not enough; we need useful science as well. While great
strides have been made in soil and water sciences in the last century,
several critical areas still badly need to be further studied, including
scaling issues and human impacts (Hornberger and Boyer, 1995; NRC,
1999). Related to scaling are the complex spatial heterogeneity and temporal dynamics and our limited understanding of landscape–soil–water
processes across scales. On the other hand, it is believed that much of the
science and technology needed to provide the underpinnings necessary for
integrated soil and water management already exists. We have, however,
fallen short in effectively interacting with stakeholders and in translating
our understanding of soil and water systems and the benefits of integrated
management into action (NRC, 1999). Integrated soil and water sciences
in general have yet to develop effective interfaces between what we know
and how we deliver that knowledge. Bouma (2005) suggested a ‘‘joint
learning’’ approach that is essential for such interactive processes.
2. Most current computer models are either ‘‘too good to be real’’ or ‘‘too real
to be good.’’ In the first case, oversimplification undercuts the accuracy or
generality of the results. In the second case, the need for detailed input
data renders the model impractical to apply except in a research setting.
Nevertheless, we recognize that no ideal model exists. Thus, compromises
between the quest for perfection and the complex reality, compounded by
our limited knowledge and/or data, plus natural uncertainty, are facts of
life. Therefore, there is a need for elegant and robust models that can be
based on reliable existing data (NRC, 1999). It is also becoming clearer
that hierarchical approaches may be effective ways to incorporate scales
into models (e.g., Cushman, 1990; Lin and Rathbun, 2003; Vogel and
Roth, 2003).
3. Data rich, information poor. The term ‘‘information’’ here connotes interpretation, synthesis, and utilization of data. The problem is largely due to
data fragmentation, incompleteness, incomparability, or inaccessibility in
spite of past extensive and costly data collections. However, it has been
pointed out that reliable long-term monitoring of data across disciplines
is perhaps the most fundamental in terms of overarching research needs
in the earth’s critical zone (e.g., Hornberger and Boyer, 1995; NRC, 1991,
1999, 2001a; Sposito and Reginato, 1992). On the other hand, advances
in automated sampling and analytical equipment, new remote sensing
and GPS, and computer models tend to change the perceived ‘‘data crisis’’
of the past into a ‘‘data avalanche’’ for the future, burying scientists and
stakeholders alike (Bouma, 1999). Therefore, we need to develop better
database management, analysis, and distribution systems for effective
archiving, comparing, analyzing, visualizing, and modeling of collected
datasets. There is also a need for innovative data mining and knowledge
discovery methods and tools that can intelligently transform database
into useful information.
4. Inspiring classroom teaching is influential; effective public education is
equally critical. New technologies have created many teaching and
learning challenges as well as excitements. The fusion of GIS, remote
sensing, computer modeling, multimedia, and the Internet into the classroom will become more prominent. On the other hand, if we are to put
wise soil and water management practices into action, to enhance the
image of our profession, to increase the necessary funding for basic and
applied research, and to implement collaborative interdisciplinary efforts
in environmental research and policy making, public education is essential. Scientists can and ought to help improve the transfer of knowledge
about soil and water resources to our stakeholders and assist in promoting the public’s awareness and appreciation of the land and water ethics.
Fundamental Scientific Issues of Hydropedology
We believe that the fundamental scientific issues of hydropedology could
be summarized in the following four interrelated areas:
1. soil structure and layering as indicators of flow and transport characteristics in soils;
2. soil morphology as signatures of soil hydrology;
3. water movement over the landscape in relation to soil cover;
4. hydrology as a factor of soil formation and a driving force of dynamic
soil system.
We now elaborate each of these fundamental issues and their related
applications in the following sections.
The term soil structure has been used in U.S. soil surveys, and elsewhere,
to refer to the natural organization of soil particles into individual units
(peds) separated by planes of weakness (Fig. 3). In addition to the shape,
size, and grade of peds, the internal surface features of peds are also described, consisting of (1) coats of a variety of substances unlike the adjacent
soil material and covering part or all of the surfaces, (2) concentration of
material on surfaces caused by the removal of other materials, and (3) stress
formations in which thin layers at the surfaces have undergone reorientation
or packing by stress or shear (Soil Survey Division Staff, 1993). The structural surface features include clay films, clay bridges, sand or silt coats, other
coats, stress surfaces, and slickensides. All of them have significant impacts
on flow and transport processes in field soils. Pores are considered separately
in the U.S. soil survey’s concept of soil structure; however, in Europe,
Canada, Australia, and some other countries, pore-related features (e.g.,
pore-size distribution, connectivity, and turtuosity) are an integral part of
soil structure (e.g., Brewer, 1976; Hodgson, 1997; McKeague et al., 1986).
Thus, the U.S. concept of soil structure is sometimes referred to as ‘‘pedality.’’ Pedality and soil pore space are interrelated, but many soils have
interpedal, intrapedal, and/or transpedal pores that are not necessarily represented by pedality. These pores, formed by biological activities (e.g., root
channels and worm borrows), physical processes (e.g., desiccation cracking
and freezing-thawing), or chemical reactions (e.g., dissolution or binding of
soluble chemicals and organic matter), are critical in determining flow and
transport in field soils. Therefore, in the context of hydropedology, we use
the term ‘‘soil structure’’ to encompass both pedality and pore space. Because
soil structure generally refers to a specific soil horizon, soil layering is thus
treated separately here to reflect the overall organization of a soil profile.
Soil structure and layering must be adequately addressed when measuring, modeling, and interpreting hydrological processes and properties in field
soils. For example, by visualizing flow patterns in soils using dye-staining
techniques, Bouma (1992), Flury et al. (1994), Lin et al. (1996), and many
others have demonstrated for a large variety of soils that structural units are
critical (Fig. 7; See Color Insert). That natural soils are structured to various
degrees at different scales seems to be the rule (Fig. 6), whereas the existence
of a macroscopic homogeneity seems to be the exception (Vogel and Roth,
2003). Indeed, it is the natural structure that reveals the essential difference
between field soils and disturbed soil materials.
1. Soil Structure Formation and Representative Elementary
Volume for Measuring Soil Hydraulic Properties
Pedological processes produce heterogeneity and usually enhance discontinuities inherited from parent materials. One important result of pedogenesis is the formation of various soil structures under the influence of various
soil-forming factors (Figs. 3 and 4). Strength and expression of soil structure
generally increase with time. However, quite disparate processes are responsible for soil structure formation, and each of them may introduce a scale
of its own. Examples for such processes are formation of organo-clay complexes, desiccation cracks, animal burrows, plant root channels, and the
formation of landscape-scale soil structural features due to either topography (such as depressions) or biological and/or chemical processes (such as
calcic pipe or tree pipe). Pedality is usually expressed strongest at the surface
and decreases with depth, resulting in a general trend of increasing ped size
and decreasing ped grade with soil depth. However, tillage and other human
activities could significantly alter soil structure. Changing soil moisture also
changes the expression of soil structure (especially in shrink-swell soils) and
the relative volumes of peds and pores, adding to the complexity of finding
mathematical solutions for modeling water movement in structured soils.
The concept of the representative elementary volume (REV) is well
known (Bear, 1972). However, a long-lasting question remains: What is
the actual size of REV for various soils? We believe that the REV for
measuring soil hydraulic properties should be a function of soil structure:
the larger the peds, the larger the REV. Standard fixed sample volume for
diverse soils could lead to incomparable data. For example, Anderson and
Bouma (1973) showed that any Ksat could be measured in the Bt2 of a
Wisconsin Hapludalf by varying the height of the sample. This was due to
the well-developed blocky structure in that soil horizon, where vertical
continuity of the cracks between the peds decreased as the sample became
longer, resulting in lower Ksat values even though the sample was completely
saturated in all cases. This suggests that soil structure is essential for choosing proper REV for measuring Ksat and other soil hydraulic properties.
Some studies (e.g., Cushman, 1990; Vogel et al., 2002) have suggested a
discrete hierarchy of the REV, where the REV is a local property related to a
given level of soil structural unit (Fig. 8). This is consistent with the hierarchical organization of soil aggregates (Fig. 6) that is characteristic of most
soils (e.g., Oades and Waters, 1991; Tisdall and Oades, 1982).
Preferential Flow in Relation to Soil Structure and Layering
Preferential flow is the process whereby water and dissolved chemicals
move by preferred pathways at an accelerated speed through a fraction of a
porous medium. Preferential flow encompasses macropore flow (also called
bypass flow or short-circuiting), funnel flow, fingering, and others. Vervoort
et al. (1999) suggested that preferential flow might be related to soil structural differences (macropore flow and fractional flow) or textural differences
(fingering flow and funnel flow). Nieber (2000) grouped preferential flow
into macropore flow, gravity-driven unstable flow, heterogeneity-driven
Figure 8 Different concepts of scales and spatial heterogeneity in the unsaturated zone: (A) A conceptual integrated-system model in pedology.
(From Wilding, 2000.) (B) Five quantitative models in hydrology/soil physics: (1) macroscopic homogeneity (thin line), (2) discrete hierarchy (dashed
line), (3) continuous hierarchy (dashed dotted line), (4) classical fractal (thin straight line), and (5) multi-fractal (thick straight lines). (Modified from
Vogel and Roth, 2003.)
flow, oscillatory flow, and depression-focused recharge. Hendrickx and
Flury (2001) recognized the scale dependence of preferential flow and classified preferential flow mechanisms into three scales: pore scale, Darcian scale,
and areal scale, with each scale having a distinct conceptual and physical
basis (Fig. 7 and Table I). The presence of preferential flow in soils leads to
spatial concentration of water flow through unsaturated soil that is not well
described by Darcy’s approach to flow through porous media.
Various pedological features are indicative of possible preferential flow in
field soils, especially if combined with landscape observations (Figs. 4, 7, 9,
and 10) (See Color Insert). For example, these include (1) soil structural
features (such as pedality, coatings, macropores, and slickensides); (2) sloping lamellae (suggesting funnel flow likely to occur); and (3) lithologic
discontinuities (indicating significant changes in particle size distribution or
mineralogy and thus possible finger flow or other types of preferential flow).
A simple field technique was devised by Bouma (1997) to measure bypass
flow as a function of rain intensity and quantity and soil moisture content.
Deriving cracking patterns from theoretical soil swell-shrink characteristics
turned out to be impossible, and very small pores (such as slickenside fissures
shown in Fig. 7B), with a volume that cannot be measured with physical
methods, can conduct large volumes of water (Bouma, 2005; Lin et al.,
1996). A procedure is attractive whereby macropores are first morphologically studied in the field in terms of types, sizes, and vertical continuity,
preferably also functionally characterized by dye-staining. Next, such data
can be fed into models whereby bypass flow is incorporated as a separate
module into existing physical flow models. This represents an effective
combined procedure of hydropedology that avoids purely qualitative
descriptions by soil morphology and not realistic model representations by
soil physics (e.g., Hoogmoed and Bouma, 1980).
Significant progress has been made in the past two decades or so in
understanding preferential flow. However, our ability to predict preferential
flow dynamics, velocity, and pathway is unsatisfactory (Jury, 1999). Quantitative and scalable relationships between preferential flow and soil structure/
texture/layering remain elusive. Although numerous simulation models for
water and chemical movement in soils have been developed, few models have
been tested with adequate field data at multiple scales, especially in combined macropore–micropore systems. Several approaches that have been
taken to incorporate preferential flow into models include (1) the mobile
and immobile water concept, which was the first attempt to deal with
the problem (van Genuchten and Wierenga, 1976), later elaborated into
dual-porosity, dual-permeability, and multiregion approaches (e.g., Ahuja
and Hebson, 1992; Gerke and van Genuchten, 1993; Gwo et al., 1995;
Othmer et al., 1991); (2) the kinematic wave approach, which was used by
German and Beven (1985, 1986) to describe water flow in macropores that
Table I
Three General Scales of Water Flow in Soils and Their Relations to Soil Structure and Preferential Flow
Temporal Conceptual
Seconds Fluid
Hagento days
pR4 DP
Hours to Repremonths
Landscape Days to
(Basic fabric,
flow, film
Pore diameter, Thin section,
Soil columns, soil
Soil profile
profiles, small
flow), funnel
plots (Horizons,
Ksat, TDR,
trans-layer pore
Funnel flow,
P þ I = R þ ET þ Hillslopes,
depressionlandforms (Variability
D þ DWk
focused flow,
within and cross
water table,
pipe flow
soil map units, soilGPR,
landscape structural
soil moisture
Jw ¼
(soil structure)
@H x
physical law
NMR: Nuclear Magnetic Resonance; CT: Computer-assisted Tomography; TDR: Time Domain Reflectometry; GPR: Ground Penetrating Radar.
Q is the volume of water flowing through a cylindrical tube of radius R per unit time, DP is hydrostatic pressure difference across the length L of the
cylindrical tube, and h is water viscosity;
Jw is water flux density (also called specific discharge), K(h) is the unsaturated hydraulic conductivity as a function of soil water matric potential h (in
head unit), H is hydraulic head, and z is vertical distance in soil profile;
P is precipitation (including dew and frost), I is irrigation water, R is surface runoff, ET is evaportranspiration, D is drainage or deep percolation, and
DW is the water storage change in the soil profile.
(Sources: Hendrickx and Flury, 2001; Lin and Rathbun, 2003).
allows flow down the sides of the pores that are not filled with water; (3) the
transfer function model, based on the probability density function of solute
travel time through a soil unit that was suggested by Jury (1982) and Jury
et al. (1986); (4) the morphometric approach, based on soil morphology and
dye-staining that was used by Bouma (1984, 1989, 1990) to model bypass
flow, in which subprocesses were defined, including vertical infiltration at
the soil surface, lateral infiltration from the macropores into the matrix, and
internal catchment from discontinuous macropores; (5) the functional models, generally capacity type approaches, which have been developed to work
with relatively simple preferential flow models that require only a few parameters (e.g., Addiscott, 1977; Corwin et al., 1991); and (6) a single-porosity
model that distinguishes between actual and equilibrium water contents,
which was proposed by Ross and Smetten (2000). Šimunek et al. (2003)
reviewed various models for describing preferential or nonequilibrium flow
and transport in the vadose zone. They stressed the need for intercode
comparison, especially against field data.
Quantification of Soil Structure and Fractal Scaling
Traditionally, soil structure has been evaluated by pedologists in the field
using morphological descriptions or thin sections (Fig. 3), while soil physicists have employed wet and dry sieving, elutriation, and sedimentation to
conduct aggregate analysis, aiming to measure the percentage of water-stable
aggregates in the soil and the extent to which the finer mechanical separates are aggregated into coarser fractions. In the absence of direct quantification, soil structure has also been frequently evaluated by methods that
correlate it to the properties or processes of interest (such as Ksat, water
retention, infiltration rate, and gas diffusion rate). In recent years, noninvasive methods that permit soils to be investigated without undue disturbance
of their natural architecture and that allow 3-D visualization of internal soil
structure and its interactions with water have become increasingly attractive.
These methods include X-ray computing tomography, soft X-ray, nuclear
magnetic resonance, gamma-ray tomography, and others (e.g., Anderson
and Hopmans, 1994; Perret et al., 1999). Image analysis has brought new
opportunities for analyzing soil structure, especially that of the pores, their
sizes, shapes, connectivity, and tortuosity (e.g., Vervoort and Cattle, 2003;
Vogel et al., 2002). Although numerous attempts have been made to find
either statistical relations or deterministic links between soil structural data
and hydraulic properties, a great need still exists to relate in situ soil structure
to field soil hydraulic properties across scales.
Soil structure has been suggested as having fractal characteristics, meaning self-similarity over a range of scales (e.g., Anderson et al., 1998; Bartoli
et al., 1998; Perrier et al., 1999). Yet it is not obvious that soils should exhibit
fractal properties. As Vogel and Roth (2003) pointed out, the self-similarity
of soil structure would suggest some self-similarity in the formation of soil
structure. However, quite different processes are generally involved in soil
structure formation; each of them may introduce a scale of its own. Nevertheless, fractal mathematics (geometrical fractals or probabilistic fractals; cf.
Baveye and Boast, 1998) has been applied to soil particle size, aggregate size,
and pore size, as well as water retention, hydraulic conductivity, preferential
flow, and other soil properties (e.g., Baveye et al., 1999; Pachepsky et al.,
2000). The fundamental equation applying to all fractals is the following
number–size relationship (Mandelbrot, 1982):
NðrÞ ¼ kr
where N(r) is the number of elements of size equal to r (unit length, or
yardstick), k is the number of initiators of unit length, and D is the fractal
dimension. A log-log plot of N(r) vs r yields a straight line (see Fig. 8). In
spite of an impressive body of literature on fractal applications in soil science
(particularly related to soil structure and hydraulic properties), this field of
research still seems in its infancy (Baveye and Boast, 1998).
One way to quantitatively model soil structure is to identify structural
units as form-elements (Vogel and Roth, 2003). For example, at the intermediate scale in Fig. 6 (middle), three structural units could be identified:
‘‘dense aggregates’’ (gray), ‘‘loose matrix’’ (white), and ‘‘macropores’’
(black). Perrier et al. (1999) proposed a generalized approach to modeling
soil structure, called the pore-solid-fractal (PSF) model, which is shown to
exhibit either a fractal or nonfractal pore surface depending on the model
parameters. In the PSF model, the fractal dimension, D, is expressed as
D ¼ d þ logð1
where d is a given Euclidean dimension, P is the proportion of pore phase (like
‘‘macropores’’ in Fig. 6), S is the proportion of solid phase (like ‘‘dense aggregates’’ in Fig. 6), and n is the inverse of the similarity ratio. A third phase,
labeled as fractal (F) phase (like ‘‘loose matrix’’ in Fig. 6, where P þ S þ F ¼
1), is the proportion for the next stage of partitioning that exhibits a selfsimilar manner. With the exception of two special cases corresponding to a
solid mass fractal and a pore mass fractal, the PSF model displays symmetric
power law or fractal pore size and solid size distributions (Perrier et al., 1999).
Soil Layering and Lateral Flow
Soil layering has three significant hydrological implications, as described
in Section II.B.1. Soil layering, especially restrictive horizons, promotes
lateral flow and/or preferential flow that are often poorly represented in
hydrological models. The occurrence of slowly permeable or irregular subsoil horizons or geological formations also could strongly alter flow patterns.
Hence, pedology can have important inputs to improve hydrological models
by taking into account soil layers and landscape features. For example,
numerous water restrictive soil horizons and features have been identified
in Soil Taxonomy, such as fragipan, duripan, ortstein, petrocalcic, petrogypsic, and placic diagnostic horizons, as well as petroferric contact, densic
contact, glacic layer, lithic contact, and paralithic contact (Soil Survey
Staff, 1999). Other subsoil horizons might also gradually develop such that
they increasingly act as an aquitard or aquiclude to downward moving
water, ultimately resulting in water moving laterally within the soil as
subsurface throughflow (Johnson and Hole, 1994). Stratified or dense geological materials (such as glacial till) also often set up a hydrologically
restrictive layer that results in a perched near-surface water table and lateral
water movement.
Another type of lateral flow is caused by hydrophobicity. Dekker et al.
(1984) showed that assumed lateral flow of water on top of a compact spodic
subsurface horizon in the Netherlands did not occur but that lateral movement of water was due to surface runoff originating from hydrophobicity of
the soil surface. Extensive field studies have shown that many soils are
susceptible to hydrophobicity under dry conditions, although land use
history is another important factor (e.g., Dekker and Ritsema, 2003). Soil
survey can provide useful information regarding soil hydrophobicity.
The change in soil hydraulic properties over the boundary of soil layers
is also noteworthy. Jury and Roth (1990) and Hamlen and Kachanoski
(1992) showed that the correlation of hydraulic properties across horizon
boundaries is one of the key impediments to modeling realistically solute
transport at the soil profile scale. Deurer et al. (2003) found that the scaling
factors of measured soil water characteristic functions have a distinct pattern
across the soil horizon boundaries. In layered or stratified parent materials,
water flow and storage in the landscape, and the consequent formation of
wetlands, are influenced by discontinuities in soil hydraulic properties
between layers (Mausbach and Richardson, 1994; Richardson et al., 1992).
Soil morphology is the basis for mapping soils in the landscape, classifying soils into taxonomic categories, and interpreting soil genesis. Soils record
spatial and temporal distribution and circulation of water because actions of
water on soils result in the formation of distinctive morphological features.
Soil morphology reflects both profile hydrology and landscape hydrology by
integrating soil changes over time. Soil horizons, for instance, often develop
in response to water movement (such as leaching or accumulation of certain
materials). A subset of soil morphological characteristics, known as ‘‘hydric
soil indicators’’ (USDA-NRCS, 1998), are directly related to a specific set of
hydrological parameters. Soil morphology is also the testimony of long-term
persistent flow and transport processes occurring in nature, resulting in
visible pedological features such as clay films, tonguing, plinthites, and
diverse soil structures that are hydrologically significant features (Figs. 4,
7, 9, and 10). Soil macro- and micromorphology thus have long been used to
infer soil moisture regimes and hydraulic properties and to provide a basis
for hydrology-related soil genesis and classification (e.g., Bouma, 1992; Lilly
and Lin, 2005).
With growing public interests in wetlands and hydric soils, several works
(e.g., Rabenhorst et al., 1998; Richardson and Vepraskas, 2001; USDANRCS, 1998; Vepraskas and Sprecher, 1997) have underlined the importance of using soil morphology in interpreting soil hydrology. In a historic
review of redox features in relation to soil moisture, Veneman et al. (1998)
indicated that much of the early work was largely qualitative, followed by
efforts to quantify environmental observations with soil morphological
features, leading to current efforts to understand pedogenical processes in
the genesis of seasonally wet soils, and to link soil morphology to quantifiable hydrological or biogeochemical parameters associated with soil wetness
(Rabenhorst et al., 1998).
Redox Features for Identifying Aquic Conditions and Hydric
Soil Indicators
Soil hydromorphology deals with soil morphological features (especially
redoximorphic, or redox, features) caused by water and their relations with
soil hydrology. Redox features (formerly called mottles and low-chroma
colors) are formed by the processes of alternating reduction-oxidation due
to saturation-desaturation and the subsequent translocation or precipitation
of Fe and Mn compounds in the soil (Soil Survey Staff, 1999). Types of
redox features include (see illustrations in Figs. 9 and 10) (1) redox concentrations as accumulations of Fe/Mn oxides (e.g., nodules, concretions,
masses, and pore linings), (2) redox depletions as low-chroma ( 2) features
formed by removal of Fe oxides (including Fe depletions and clay depletions), (3) reduced matrix that changes color upon exposure to air due to
Fe(II) oxidation to Fe(III), and (4) a reaction to an alpha, alpha-dipyridyl
solution if the soil has no visible redox features (Soil Survey Staff, 1999;
Vepraskas, 1992). Redox features, which are usually considered hydric soil
indicators, exclude those hydric soil indicators composed of carbon and
sulfur (USDA-NRCS, 1998).
Hydric soils are defined as soils that formed under conditions of saturation, flooding, or ponding that lasted long enough during the growing season
(repeated periods of more than a few days) to develop anaerobic conditions
in the upper part (usually 0.15–0.3 m) of soil profiles (USDA-NRCS, 1998).
Hydric soils are one of the three requirements (along with hydrophytic
vegetation and wetland hydrology) for identifying jurisdictional wetlands
in the United States. They are identified and delineated in the field using
soil morphological features (i.e., hydric soil indicators). These include a
variety of features that are regional and texture-based, but all are formed
predominantly by the accumulation or loss of Fe, Mn, C, or S compounds
(USDA-NRCS, 1998). While indicators related to Fe/Mn concentrations or
depletions are the most common, other features (e.g., sulfide and various
combinations of carbon accumulations) have been used in specific kinds of
soils that do not develop redox features. There are also so-called problem
soils that seem to be hydric soils but whose morphologies are difficult to
interpret or seem inconsistent with the current landscape, vegetation, or
hydrology (Veneman et al., 1998). These include soils formed in grayishor reddish-colored parent materials, soils with high pH or low organic
matter, Mollisols with thick dark A horizon, Vertisols with shrink-swell,
soils with relict redox features, and disturbed soils such as cultivated soils
and filled areas (Rabenhorst et al., 1998; USDA-NRCS, 1998). Relict redox
features do not reflect contemporary or recent hydrological conditions of
saturation and anaerobiosis; rather, they were likely formed during past
geological wetter climates. Typically, contemporary and recent hydric soil
morphologies have diffuse boundaries, while relict redox features have
abrupt boundaries (USDA-NRCS, 1998).
Certain redox patterns occur as a function of the patterns in which the
ion-carrying water moves through the soil and as a function of the location
of aerated zones in the soil. Characteristic color patterns are thus created by
the reduced Fe and Mn ions removed from a soil if vertical or lateral water
flow occurs, or the oxidized Fe and Mn precipitated in a soil if lack of
sufficient water flux. Consequently, the spatial relationships of redox depletions and redox concentrations may be used to interpret water and air
movement in soils (Vepraskas, 1992). Interpreting directions of water movement from redox patterns is easiest when the features have a consistent
relationship with soil structure including macropores. Vepraskas (1992)
provided four examples that illustrated the basic principles: (1) redox depletions occur around macropores and redox concentrations occur within
matrix (Fig. 9A and B), suggesting that water infiltration along macropores
and reducing condition developed there because of perched saturated layers;
(2) redox concentrations occur around macropores and redox depletions
occur within matrix (Fig. 9C and D), indicating that soil matrix is wet
for periods long enough for reducing condition to be maintained while
macropores become aerated because of faster drainage or plant roots (such
as in rice plant) transport air to macropores when the soil is still flooded;
(3) redox depletions and concentrations have no consistent relationship to
macropores, such as those found in sands or materials with small aggregates
where macropores either are not stable or are relatively small and closely
spaced such that water and air movement into the soil does not follow the
same macropores after each infiltration event; and (4) redox features have a
distribution that combines two of the preceding scenarios, resulting from the
soil profiles where one horizon having one group of features is overlain or
underlain by another horizon containing another group.
Quantification of Soil Hydromorphology
Some recent studies have attempted to quantitatively relate soil morphological features found in soil profiles to quantifiable hydrological parameters
associated with soil wetness (Figs. 10 and 11). These studies, covering a
broad spectrum of geomorphic and climatic conditions across the United
States, were largely associated with the USDA-NRCS Wet Soils Monitoring
Project, which was initiated in 1990 in conjunction with the International
Committee on Aquic Moisture Regime (Rabenhorst et al., 1998). Designed
to collect factual data on the wet properties of various soil climatic regions,
this project amassed field data on water table head, shallow ground water
depth, soil matric potential, soil temperature, redox potential, and presence
of ferrous iron. One facet of the project was to test and comment on hydric
soil indicators and wetland delineations. In many cases, a catena of soils that
provided trends in properties relative to soil wetness was used in the monitoring (e.g., Jenkinson et al., 2002; Reuter and Bell, 2003; Thompson
et al., 1998). Such monitoring efforts provide a foundation for integrating
hydropedology concepts into soil survey programs.
To illustrate, soil morphology is sensitive to long-term, average monthly
water table depths in hydric soils and thus could be used to estimate
statistical (e.g., monthly average) and stochastic (e.g., probabilistic) properties of the monthly water table regime (Galusky et al., 1998). The depth to
gleying has long been used as a crude indicator of the mean position of the
wet season water table (Franzmeier et al., 1983). Soil chroma of 3 to 4 also
has been found to be associated with prolonged periods of water table
saturation (Evans and Franzmeier, 1986; Franzmeier et al., 1983). Similarly,
the depths to redox concretions and depletions have been associated with
water table fluctuations (Vepraskas, 1992). In the United Kingdom, in the
absence of direct measurement, soils can be assigned to one of six soil
wetness classes that describe the height and duration of water logging
based on soil profile features such as the depth to a slowly permeable layer
Figure 11 Quantitative soil morphological indicators for predicting water table behavior:
(A) Regression-estimated and measured monthly water table depths for a costal plain hydric soil
(Aquic Quartzipsamment) in Maryland. (B) Average long-term, monthly water table hydrograph, estimated using the sine function model of Eq. [4] vs actual data. Sine fn 1 is based on
known average March and average annual water-table depths. Sine fn 2 is based on average
March and average annual water table depths that were estimated using soil morphology. (From
Galusky et al., 1998.)
and depth to gleying (Lilly and Matthews, 1994; Lilly et al., 2003). In an
attempt to develop quantitative soil hydromorphological indicators of water
table behavior, Galusky et al. (1998) examined the depth to gleying (d-gley),
depth to soil chroma of 3 to 4 (d-34), and the depths to redox concretions
(d-conc) and depletions (d-depl) in 29 sites in the coastal plain of Maryland.
They found that d-34 correlated the most highly with average monthly water
table levels. This correlation was greatest for the month of March (when
seasonal water table levels in Maryland are generally at their highest) and
decreased during the summer months. The d-gley was also highly correlated with late winter/early spring water table levels. However, they did not
find good correlations between d-conc or d-depl and average monthly water
table levels. Galusky et al. (1998) proposed a first-order autoregressive
model to enhance (extend into the past) existing water table records
(Fig. 11A):
wti ¼ a þ b wti
þ c pri þ d evi ;
where wti is the average monthly water table depth (cm) for month i, wti-1
is the average monthly water table depth (cm) for month i
1, pri is the
cumulative precipitation (cm) for month i, evi is the estimated cumulative
pan evaporation (cm) for month i, and a, b, c, and d are coefficients
estimated from the data. Galusky et al. (1998) also used a sine function to
estimate long-term, average monthly water table hydrograph (Fig. 11B):
est wti ¼ est ann wt þ ampl sin ½2p ðmonthi =12ފ;
where est wti is the average estimated monthly water table for month i, est
ann wt is the average estimated annual water table, ampl is the difference
between the estimated seasonal high water table level and the annual mean,
and 2p/12 is the angular frequency of the function.
While progress has been made, insufficient data exist on the duration
and frequency of high water tables in different soils. There is a need to
determine the dynamics of the water table in benchmark and other important soils so that the duration of saturation and reduction required for
creating aquic conditions may be specified. Currently, aquic conditions as
used in Soil Taxonomy (including endosaturation, episaturation, and anthric
saturation) are not yet quantitatively defined (Soil Survey Staff, 1999).
Similarly, soil drainage classes as used in soil surveys (Soil Survey Division
Staff, 1993) have also been qualitatively determined. There are a great
number of applications for water table data once a significant volume of
quantitative data has been accumulated, particularly if such data are collected with associated landscape features. One of the best uses of such data
would be to more fully develop the relationships between soil profile descriptions and water movement in soil profiles and landscapes. This information
could vastly improve the value of soil surveys and their updates.
Soil Morphology as a Guide to Field Hydrological Measurements
Pedology expertise, especially related to visible soil morphological
features, helps to guide and interpret physical and hydrological measurements in field soils. Morphological descriptions often yield information that
cannot be easily obtained by other methods, such as the shape and strength of
peds, type of macropores, and macropore continuity and connectivity. There
are also examples in which soil morphological data uniquely characterize flow
regimes that would be very difficult to document with classical soil physical or
hydrological techniques, such as redox patterns discussed in Section III.B.1.
Such information can be quite helpful when placing monitoring or measurement equipment in field soils. For example, although many methods are now
available for measuring Ksat (e.g., Dane and Clark, 2002), the effect of sample
volume and pore continuity of interpedal voids on measured Ksat is still
ignored (Bouma, 2005). A study made by Bouma et al. (1989) in a silty
clay soil with glossic features indicated that measurements in the bleached
cracks yielded a Ksat value of 6.9 m/day while values were 0.3 m/day inside
the compact peds (Fig. 12; See Color Insert). Placing samples at random in
this soil led to a very high variability that could not be reduced by applying
statistics, but could be reduced by making a morphological analysis before
samples were taken and by estimating the relative importance of the two
flow regimes in the ped matrix and the glossic features. Thus, studying soil
structure with morphological methods is important when choosing proper
measurement methods and locations for determining Ksat and other soil
hydraulic properties. Various preferential flow patterns as illustrated in
Fig. 7 also suggest that significant variation could result when sampling
soil cores for physical/hydrological measurements in the laboratory.
When working in clay soils, measurement of water table depth has often
presented problems. Levels may fluctuate wildly, particularly after rainfall,
when levels indicated in piezometers and readings of tensiometers may differ
significantly at short distances. Bouma et al. (1980) studied this phenomenon, showing the effects of water flowing along cracks and ped faces, both in
the unsaturated and saturated zone in the soil (Fig. 13). After rainfall, the
water level in the cracks rises very rapidly to subside slowly as water moves
slowly into the unsaturated peds. During that time, the water table level is
not defined. When piezometers intercept these cracks, they show high fluctuations. When they are inside the peds, they are very stable, unless they are
not well sealed on the outside, which is likely to result in crack-flow into the
piezometer, incorrectly suggesting a ‘‘perched’’ water table inside the peds.
This is true when using unlined augerholes for measuring water table levels
in well-structured soils. Then, free water levels are observed at every level to
which an augerhole has been drilled (Fig. 13). When the water moves out of
the cracks after rain, the peds may remain saturated for a while, suggesting
overall saturation of the soil as measured by tensiometry, but the cracks are
already filled with air. Placement of tensiometers with a downward angle
into a vertical profile wall may induce flow along the sides, incorrectly
suggesting saturation, while upward placement avoids this problem. When
tensiometers intercept cracks, they may register temporary ‘‘saturation’’ that
is not measured when the cups occur inside peds (Fig. 13). Understanding
the flow regime, as influenced by macropore patterns, helps explain what
Figure 13 A diagram showing water tables, boreholes (A), piezometers (P), and tensiometers (T) in structured clay soils, illustrating the effects of water flowing along cracks and ped
faces on instrument readings.
otherwise would be highly confusing hydrological measurement results.
Such an understanding can contribute to better instrument design and
measurement protocols. Bouma (1989) summarized this well as ‘‘look first,
then measure.’’
Soil morphology is also helpful in determining water accessibility. Soils
with large compact peds, such as prisms and clods formed by tillage under
adverse conditions, often show concentrations of roots at the ped surface
(Fig. 14; See Color Insert), indicating that the roots were unsuccessful in
penetrating the peds or clods and that preferential flow along roots and ped
surfaces is commonly expected. Field conditions have been observed in
which plants were wilting even though water contents of the root zone were
well above the wilting point. This phenomenon has been attributed to limited
accessibility (Bouma, 2005). Modeling studies, implicitly assuming unlimited accessibility, thus could yield poor results. Droogers et al. (1997) studied
these processes in large undisturbed field samples and defined accessibility as
a function of ped sizes. This, in turn, could be incorporated into existing
simulation models for more realistic prediction of plant growth and crop
Soil Morphological Attributes for Inferring Soil Hydraulic Properties
Soil scientists have been successful in using descriptive morphological
information to make qualitative judgments about a number of soil hydraulic
properties, notably Ksat (e.g., Coen and Wang, 1989; King and Franzmeier,
1981; McKeague et al., 1982; O’Neal, 1949; Soil Survey Division Staff,
1993). Soil morphological data are also suited to grouping soils by their
hydrological functioning. For example, Quisenberry et al. (1993) devised a
descriptive system to classify soils based mainly on water flow pathways and
patterns (uniform flow or different types of preferential flow) using surface
soil texture, subsoil structure, and clay mineralogy. A soil hydrological
classification (termed hydrology of soil types, or HOST) based on soil
morphological attributes has also been developed in the United Kingdom
to predict water movement through soils and substrates (Boorman et al.,
1995; Lilly et al., 1998). The attributes used include the presence or absence
of an organic surface layer, substrate hydrogeology, the depth to a slowly
permeable layer, the depth to gleying, and air capacity values.
Grouping or classifying soils in terms of both soil morphology and
hydraulic properties is also a valuable means of developing simple and
more reliable predictive pedotransfer functions (PTFs) for field soils. For
example, Franzmeier (1991) grouped Ksat of some Indiana soils by soil
classes, called lithomorphic classes, based on origin of parent material,
type of soil horizon, and soil texture. Batjes (1996) used hierarchical pedotransfer rules and functional grouping to predict available water capacity for
the main soil types of the FAO-UNESCO world soil map using soil unit
type, horizon textural class, and organic matter class. Wösten et al. (1990)
demonstrated that class PTFs that use well-defined soil horizons as ‘‘carriers’’ of physical/hydraulic information allow efficient use of soil morphological data because soil horizons can be determined easily and reproducibly
by pedologists (Bouma, 1992). However, as noted by Wösten et al. (1985),
Breeuwsma et al. (1986), and Bouma (1992), pedogenic differences, as
expressed by horizon designations, do not necessarily correspond with
functional differences. Bouma (1992) suggested a more promising threestage protocol that attempts to calculate hydraulic parameters directly
from physical or morphological data. First, good measurements of soil
hydraulic properties that take soil morphological attributes (such as soil
horizon and soil structure described in the field) into account should be
made. Second, measured hydraulic properties should be expressed in terms
of coefficients as defined by, say, the van Genuchten equation and, third,
relate those coefficients by regression analysis to readily available soil properties (such as texture, bulk density, and organic matter content) or more
qualitative groupings of soil horizons.
While qualitative soil morphological attributes have been widely applied,
quantification of such data is generally lacking. Rawls et al. (1993) noted
that a quantitative description of the effects of soil morphological properties
on soil water movement is yet to be established. So far, limited studies have
demonstrated the potential of quantifying soil macromorphology through
field observations or soil micromorphology through thin sections (Lilly and
Lin, 2005). As illustrated by Bouma et al. (1979), Lin et al. (1999b), Vervoort
and Cattle (2003), and Kutı́lek (2004), soil micro- and macromorphometric
data could be used to quantitatively derive soil hydraulic parameters. For
example, in a study conducted by Lin et al. (1999b), field soil hydraulic
conductivity at zero tension (K0) was reasonably predicted from morphometric indices (MI) of soil textural class (MIt), initial moisture state (MIm),
ped grade (MIsg), ped shape (MIst), macropore quantity (MIpq), macropore
size (MIps), and root abundance (MIrq):
K0 ¼
22:4 38:3MIt þ 33:8MIm þ 21:1MIsg þ 47:5MIst
þ 102:4MIpq þ 45:5MIps þ 33:3MIrq :
The utilization of soil structural descriptors in a quantitative fashion, as
used in the preceding example, would be a step forward toward incorporation of soil structure into PTFs and the modeling of flow and transport in
field soils. Quantification of soil morphology would also enhance the understanding of the relationships among different soil morphological features (as
demonstrated by Lin et al., 1999a) and thus permit a better assessment of
soil profile descriptions in relation to water movement in soils and over the
Conceptual models of water movement over the landscape are key aspects
of contaminant transport, watershed management, wetland delineation, and
terrestrial ecosystem functions. Where, when, and how water moves through
various landscapes of different sizes and how water flow impacts soil processes and subsequently soil spatial-temporal patterns need to be better
understood. It has been pointed out that ‘‘how water moves through soil at
the landscape scale’’ is a major research need of the U.S. National Cooperative Soil Survey (NCSS) program (USDA-NRCS, 2001) and that ‘‘a focused
research program to understand soil–water interactions at the landscape
level and at whatever depths needed must be an integral part of national
soil survey in the 21st century’’ (Smith and Hudson, 1999). In 2003, the
NCSS program further recommended that hydropedology be promoted as a
useful framework for modern soil surveys and updates and that whole
landscape hydropedological study be listed as a priority research in the
NCSS program (USDA-NRCS, 2003).
In this regard, traditional 3-D block diagrams used in soil surveys that
show conceptual models of soil–landscape relationships could be useful in
developing conceptual models of water movement over the landscape and in
linking dynamic soil properties to landscape positions. Enhanced 3-D block
diagrams of landscape–soil–water relationships with added information of
water table dynamics, water flow paths, hydric soils, restrictive layers, and
other relevant information could significantly increase the values of soil
survey products. As illustrated in Fig. 5, classical block diagrams of soil–
landscape relationships could be used to indicate landscape hydrology,
illustrating water flow direction and water table dynamics. These block
diagrams could be further linked to watersheds or physiographic regions
to provide valuable conceptual frameworks of water movement over the
landscape in different major land resource areas (MLRAs). MLRAs are
geographically associated land resource units that are geographical areas
(usually several thousand acres in extent) characterized by a particular
pattern of soils, water, climate, and land use (USDA-NRCS, 1997). The
MLRA approach to soil inventory is becoming more and more recognized in
the United States as a focal point for modernizing soil survey information.
Hence, hydropedological study in benchmark areas in various MLRAs
could provide fundamental insights regarding water movement over diverse
landscapes of various scales.
Landscape Hydrology
Landscape perspective is essential in examining interactive hydrological
and pedological processes. Landscape shape controls water flow and ultimately soil distribution over an area. For example, landscapes with numerous depressions, termed ‘‘hummocky landscapes,’’ have landform level and
local ground water flow as well as landscape level surface and subsurface
flow (Lissey 1971; Richardson et al., 2001; Toth, 1963; Winter, 1988).
Hummocky till landscapes display an array of flow patterns in a flownet;
in contrast, smooth landscapes have even and long flow patterns that display
upland recharge and lowland discharge (Fig. 15). Recharge hydrology
removes material from a soil horizon, and water moves to the ground
water, while discharge hydrology adds material to a soil horizon, and
water moves from the ground water (Richardson et al., 2001). Climate also
plays a role in landscape hydrology, as noted by Richardson et al. (1992). In
humid regions, the local highs have a water table that is a subdued replica of
the topography. The higher areas are recharge areas with leached soils, and
depressions are discharge areas often with calcareous soils (Fig. 16A). The
subhumid climate has more variety in the depressions: the higher areas are
distinctly recharged, and the water flow in recharge depression reflects the
episaturation; the flowthrough depression has soils that are calcareous or
unleached within thick A horizons; and the discharge depression may have
saline soils (Fig. 16B). In semiarid regions, the recharge depression becomes
more common (Fig. 16C).
Wetlands are the interface between land and water. The study of
wetland hydrology and wetland soils is thus intimately linked to landscape–soil–water interactions and hence hydropedology. Wetland hydrology
involves the spatial and temporal distribution, circulation, and physiochemical characteristics of surface and subsurface water in the wetland
and its catchment over time and space (Richardson et al., 2001). The edges
of jurisdictional wetlands in the United States are identified by noting the
point at which hydric soils and hydrophytic plants end and the upland
characteristics begin. The hydrological nature of a wetland is the result
of the balance between inflows and outflows of water, the soil and topography in a wetland, and subsurface conditions. Major hydrological inflows include precipitation, flooding rivers, surface flows, ground water,
Figure 15 Flownet representation of (A) a smooth landscape and (B) a hummocky landscape
and their related flow patterns as indicated by arrows. (After Toth, 1963, and Winter, 1988.)
Figure 16 Two-dimensional landscape diagrams showing recharge-discharge hydrology,
equipotential lines, flow lines, and flow direction in hummocky landscapes in different climatic
settings. (From Wysocki et al., 2000.)
and in coastal wetlands, tides (Fig. 17). Most upland areas that have wetland
hydrology occur on landscape positions that receive run-on water from
surrounding landscapes to cause wetness above normal precipitation.
Other upland areas have a seasonal high water table due to high rainfall
and impermeable layers below the soil surface or a ground water table
that seasonally rises close to the surface. Most movement of ground water
is the result of topographic relief, and discharge of ground water at topographically lower elevations results in wetlands or stream flows (Figs. 16
and 17).
Figure 17 A landscape–soil–water system that links soil profile hydrology and hillslope
hydrology. The major processes involved in the landscape hydrologic cycle include: precipitation, infiltration, vegetation interception and return to the atmosphere, surface runoff (overland
flow), subsurface throughflow, upward flow, deep percolation, and ground water flow. One form
of overland flow from a saturated soil is called reflow (or return flow). A concentrated subsurface flow through a chain of connected macropores nearly parallel to the soil surface is called
pipeflow. Also shown is a likely catena of soils along the hillslope.
Hillslope is a fundamental landscape unit and an intermediate scale that
connects point observations to watershed phenomena. Watersheds are comprised of sub-watersheds, which in turn are comprised of multiple hillslopes.
Hillslope processes thus are closely linked to landscape and watershed
hydrology (Fig. 17). In the past decades, hillslope hydrology has received
considerable attention, particularly by hydrologists (e.g., Anderson and Burt,
1990; Kirkby, 1978; Western et al., 1999). As pointed out by Ridolfi et al.
(2003), hillslope hydrology is challenging because a number of processes interact at different scales, significantly contributing to the complexity of the system
that hampers the possibility of a general theory. Some of the most important
issues in hillslope hydrology include the following (Ridolfi et al., 2003):
. horizontal and vertical heterogeneity of soil types and various properties;
. lateral redistribution of water along the hillslope due to the formation of a
saturated zone in the soil and lateral subsurface flow in the vadose zone;
. type and spatial pattern of vegetation along a hillslope and its impacts on
runoff and infiltration;
. different types of climate and precipitation events, which, although possi-
bly spatially uniform over a hillslope, may trigger other mechanisms that
generate spatial dynamics;
infiltration of runoff generated in the uphill part of a slope that occurs at
the rainstorm time scale;
longitudinal hillslope profile and form, and 3-D hillslope geometry and
the presence of spurs and hollows;
geographic position of a hillslope and exposure to the sun and wind,
which may strongly affect evapotranspiration, vegetation distribution,
and soil properties;
boundary conditions, especially at the bottom of a hillslope and the
underlying geological formations;
various land uses and anthropogenic activities.
There have been numerous attempts to relate topographic variability to
soil properties and hillslope hydrology. In the pedology community, for
example, many studies have examined the spatial variations of soil horizon
thickness, particle size distribution, organic carbon, depth to carbonates,
base saturation, depth to redox features, and other soil properties as a
function of hillslope position (e.g., Gerrard, 1981; Kleiss, 1970; Lin et al.,
2004; Mausbach and Richardson, 1994; Moore et al., 1993; Pennock and de
Jong, 1990; Thompson et al., 1998; Walker and Ruhe, 1968). In the hydrology community, the influence of terrain on hillslope hydrology has been
widely investigated (e.g., Anderson and Burt, 1990; Beven, 1997a,b; Kirkby,
1978). A common belief regarding soil moisture distribution over a hillslope
or landscape is that topography becomes increasingly important in wet
periods, but during dry periods soil moisture patterns depend primarily on
soil properties with little effect from topography (e.g., Grayson and Blöschl,
2000). Particularly useful terrain attributes, which are now routinely calculated from a DEM, include topographic wetness index, slope gradient, slope
curvature, specific catchment area, relative elevation, and others. For instance, the topographic wetness index (TWI), also known as the compound
topographic index, is an index of hydrological similarity based on topography and is related to the Horton model and Darcy’s law (Burt and Butcher,
1985; Kirkby, 1975):
TWI ¼ lnða=tanbÞ;
where a is the area draining through a point from upslope (called specific
catchment area), and tanb is the local slope angle. High TWI areas in a
catchment tend to saturate first and therefore indicate potential surface or
subsurface contributing areas. The expansion and contraction of such areas
as a catchment wets and dries is then indicated by the pattern of the TWI
(Beven, 1997a). Based on the TWI, a popular rainfall runoff model, called
TOPMODEL (topography based hydrological model) (Beven and Kirkby,
1979), has been widely used in the hydrology community (Beven, 1997a).
Various studies have attempted to correlate the TWI with actual soil wetness or zones of surface saturation, but the results vary widely (e.g., Sulebak
et al., 2000; Western et al., 1999; Yeh and Eltahir, 1998). There have been
many improvements to Eq. [6], such as (1) incorporating soil transmissivity at
saturation (T0), leading to what is called the soil topographic index,
ln(a/T0 tanb) (Beven, 1986), and (2) considering a as a variable effective
upslope contributing area instead of a fixed value, leading to what is called
the dynamic TOPMODEL (Beven and Freer, 2001). For a more detailed
account of the topographic wetness index and other related aspects of hillslope hydrology, readers are referred to Beven (1997a,b) and Kirkby (1978).
Importance of Geomorphology and Stratigraphy
It is well recognized in the pedology community that geomorphology and
stratigraphy are determinant variables to pedogenesis, soil–landscape patterns,
and soil behavior, particularly at the large-area scale (Wilding, 1994). Geomorphology is the study of the classification, description, nature, origin, and
development of landforms on the earth’s surface, while stratigraphy deals
with rock strata (i.e., soil parent materials in residual soils), especially the
distribution, deposition, and age of sedimentary rocks. At the beginning,
geomorphology was concerned essentially with producing time-dependent
models of landscape evolution. The form of the land was the major focus,
with little mention of process and scant attention to the soil and regolith
materials (Gerrard, 1981). Investigations of drainage basins and storm
hydrographs demonstrated the influence exerted on these phenomena by
the surface covering soil and vegetation. Modern research is increasingly
demonstrating the close dependence of soils, landforms, and geomorphological processes. Geomorphological and pedological processes interact on
hillslopes, especially where the movement of soil and water is considered.
Patterns of landforms are matched, often on a one-to-one correspondence,
by soil patterns (Gerrard, 1981; Wysocki et al., 2000). The characteristic
suite of landforms and soils created by glacial and fluvioglacial deposition is
a classic example. Fluvial and marine processes also produce a characteristic
assemblage of landforms that is paralleled by the soil types.
Similar to soil geomorphology (or pedogeomorphology), hydrogeomorphology is recognized in the hydrology community to address the interrelationships between landforms and processes involving water. Water erosion
and deposition influence the genesis and characteristics of landforms.
Conversely, characteristics of the landform influence surface and subsurface
water movement in the landscape. Water sculpts the landscape through the
processes of runoff, erosion, transport, and deposition, resulting in a treelike network of channels into which the flow becomes concentrated. Networks at the structural basis for interpreting the transport of water and
solutes are best seen at the catchment scale (Horton, 1945; Rodrı́guez-Iturbe
and Rinaldo, 1997; Scheidegger, 1967). However, although empirical
laws describing the 2-D geometry of these networks have existed for about
half a century, there is little quantitative understanding of the dynamics of
channel formation or of the causal relationship between the 3-D network
structure and the precipitation driving the erosion (NRC, 1991). Such an
understanding would reveal fundamental scaling relationships of surface
water hydrology over a broad range of spatial scales (NRC, 1991).
Similarly, water also shapes the vertical soil profile, through the processes
of leaching, translocations, fluctuating water table, shrink-swell, freezethaw, and other processes, that result in various soil structures and a
network of preferential flow pathways. It appears that there is a similarity
between the stream network and the network of water flow pathways in soil
profiles. For example, Deurer et al. (2003) recently suggested a concept of
drainage networks to describe bypass flow pathways in soils at the soil profile
scale. They found that the drainage network in a sandy soil under a coniferous
forest in north Germany closely resembled one of mountainous streams, and
that the fractional area of the entire profile occupied by the network was
found to decrease exponentially with depth. They thought such a network was
related to the law of energy dissipation, which causes a specific tree-like
structure for flow paths in the soil profile, as well as at the catchment scale.
Soil hydrological processes and properties are closely linked to landforms
and parent materials. For instance, a few studies have reported the relation
of soil hydrology to geomorphology and stratigraphy on Wisconsinan-age
till plains that are common in the Midwest (e.g., Evans and Franzmier, 1986;
Jenkinson et al., 2002; Thorp and Gamble, 1972). Jenkinson et al. (2002)
reported that on a dissected till plain underlain with dense till, water was
held up by the low Ksat till. Water thus moved from the interior of the till
plain to the dissected bevel of the plain, where it caused relatively high water
tables in soils that have no redox features.
Soil Variability over the Landscape and within Soil Map Units
The spatial variability of physical soil properties is particularly critical in
hydrology, yet we have relatively poor ways of estimating it at the landscape
and watershed scales. A number of case studies in catchment hydrology have
scrutinized the reliability of soils data and their effect on the representation of
catchment response. For example, Houser et al. (2000) reported that the
addition of spatially variable soil properties based on the Order II soil map
(including clay %, sand %, Ksat, ur, us, and feff, estimated using PTFs from the
literature) produced unrealistic polygon artifacts in the soil moisture patterns
simulated using the TOPMODEL-based Land Atmosphere Transfer Scheme
(Famiglietti and Wood, 1994). In comparison, simulations based on uniform
soil hydraulic properties produced soil moisture patterns that were more
consistent with the observations from airborne push broom microwave
radiometer (Houser et al., 2000). They suggested that it might be possible
to develop a smoothing algorithm that would use the soil polygons to approximate continuously varying, spatially distributed soil parameters. Western
and Grayson (2000) used soil type to spatially distribute hydraulic conductivity measurements in Tarrawarra watershed, assuming uniform conductivity
within each soil type polygon. This produced artificially high soil moisture
values at the interface of the soil types when compared with soil moisture patterns measured by TDR. Vertessy et al. (2000) indicated that the
assumption of uniform conductivity in each of the three land types (differentiated by topography and soil properties) in the La Guenca catchment was
not appropriate as suggested by a large number of soil core Ksat measurements
across the catchment. Instead, they added a random component to the deterministic pattern imposed by land type in their runoff model simulation.
Grayson and Blöschl (2000) claimed that the variability of soil physical
properties within soil types can be as large as or larger than the variability
between soil types. This suggests that caution should be exercised in
distributed hydrological modeling when allocating soil hydraulic properties
on the basis of soil types as indicated by soil maps (using either PTFs or direct
On the other hand, Duffy et al. (1981) demonstrated that when a soil map
was properly used it could help sort out the spatial variability of soil
hydraulic properties. They measured quasi-steady state infiltration rates on
surface soils at 20 locations scattered throughout a 100 ha farm in New
Mexico. If the seven soil series on the farm were ignored, there was basically
no relation between measured and estimated infiltration rates. But when the
infiltration rates were grouped by soil series based on the Order I soil map,
the measured and estimated geometric mean values were highly correlated.
In discussing emerging technologies for scaling field soil–water behavior,
Nielsen et al. (1998) expected that new paradigms for local and regional
scales of homogeneity in pedology and soil classification would emerge, with
soil map units containing spatial and temporal soil–water scale factors.
However, spatial and temporal variability of soil resulting from natural
and man-made processes reduces the certainty, as indicated on existing static
soil maps and in soil survey reports. Although currently available soil maps
and the related databases are often considered as the very best data one can
obtain in environmental and natural resource assessments (Merchant, 1994),
the proper use of existing soil maps and databases is not necessarily
warranted if within map unit variability is not well understood and quantified. For example, in regional ground water vulnerability assessments, the
uses of the soil associations in the State Soil Survey Geographic Database
(STATSGO) (Order IV soil map) are often considered obscure (Merchant,
1994). There are also significant ambiguities with regard to scaling issues of
soil maps (Loague and Green, 1990).
Soil surveys have traditionally overlooked spatial variability within map
units for a variety of reasons, including scale limitations, lack of appropriate
sampling design, and inadequate quantitative data (Lin et al., 2004).
Although acknowledged, variation within soil map units is generally
described qualitatively in vague terms. With the growing use of digital soil
maps and related databases for diverse applications, the variability of soil
taxa and of map units has become more recognized. For example, the nationwide Order II Soil Survey Geographic Database (SSURGO) is believed to be
of little use in site-specific applications if within map unit variability cannot
be quantified. It appears that more detailed Order I soil mapping would be in
great demand for site-specific applications such as precision agriculture and
landscape hydrology. However, virtually every delineation of a map unit in all
soil surveys includes other soil components or miscellaneous areas that are
not identified in the name of a map unit. Many of these components are either
too small to be delineated separately at a given soil survey scale or deliberately
included in delineations of another map unit to avoid excessive detail in the
map or the legend (Soil Survey Division Staff, 1993). These inclusions reduce
the homogeneity or purity of map units and thus require appropriate quantification for use in modeling. Indeed, soil map units are better considered as
landscape units rather than individual soil types (Wysocki et al., 2000). Hall
and Olson (1991) challenged current soil maps: ‘‘Much effort has been
expended on taxonomic classification of soils during the last few years, but
the importance of proper representation of landscape relations within and
between soil mapping units has been virtually ignored. The same mapping
unit is often delineated on convex, concave, and linear slopes. This mapping
results in the inclusion of areas of moisture accumulation, moisture depletion,
and uniform moisture flow within a given mapping unit.’’ While many studies
have suggested the need for a reliable estimate of the proportionate extent of
map unit components within a soil map unit for probabilistic assessment of
soil properties (e.g., Brown and Huddleston, 1991; Foussereau et al., 1993;
Lammers and Johnson, 1991; Lin et al., 2004; Nordt et al., 1991), such
information is still largely lacking. Hence, quantification of map unit purity
for different scales of soil maps is a needed area of improvement in modern
soil surveys (Arnold and Wilding, 1991; Lin et al., 2004). It is encouraging that
the USDA-NRCS has begun such attempts to quantitatively document the
variability of representative mapping units.
We would like to point out that there is a mentality in soil science today
stating that we really do not need to map soils anymore; all we need to do
is to develop variogram functions using geostatistics to estimate spatial
variability from one point to the next. We would challenge this mentality
in dealing with soil spatial-temporal dynamics. First of all, geostatistics
provides powerful interpolative tools after an extensive dataset has been
gathered on a particular soil. However, geostatistics is not a very powerful
extrapolative tool with soil properties, especially to extrapolate from one
tested area to a new area for which the database has not been collected.
This makes geostatistics a rather costly, inefficient method to extrapolate
knowledge from one area to the next. Furthermore, geostatistical functions should be derived from landscape stratified units such as soil types,
slope gradients, geology, land uses, parent materials, and vegetation, and
not indiscriminately across a broad landscape without prior partitioning
of the sources of variability. In this regard, pedological expertise and various
geospatial data such as DEM are helpful in assisting the appropriate
application of geostatistics to landscape analysis.
As the circulatory system is to the body, the fluvial system is to the
landscape. Water is of critical importance to soil morphology, genesis, classification, and mapping. All of the five natural soil-forming factors affect and
are affected by hydrology. The flux factors of soil formation (climate and
vegetation) as well as site factors (topography and parent materials) can be
linked to landscape hydrology, which is further modified by the soil internal
hydrological environment. For instance, climate influences the amount and
timing of soil water availability, and soil moisture in turn influences climate.
The biota growing on and in soils are strongly influenced by water’s presence,
both directly, because organisms require water to live, and indirectly, because
the amount of soil water influences oxygen availability, the temperature
regime, and nutrient transport in soils. Topography frequently directs and
controls the flow of both surface and subsurface water over the landscape.
Parent materials affect the flow of water because they are the sources of the
matrix through which surface water infiltrates and may reflect the materials
through which ground water flows. Time is required for both soil development and change and for water to flow through soils and landscapes. Much
like ‘‘one cannot ignore the role of ground water in performing geologic
work’’ (Domenico and Schwartz, 1998), water in the unsaturated zone
cannot be ignored in soil formation and soil dynamic changes.
Another way to look at the essential role of water in soil formation and
soil dynamic changes is Simonson’s (1959) theory of generalized processes of
soil formation, in which processes and systems linkages were emphasized
over factors. Simonson (1959) suggested four general soil-forming processes:
additions, deletions, transformations, and translocations. All of these processes involve water in significant ways. Water adds material through deposition of eroded sediment and precipitation of dissolved minerals. Water can
also entirely remove soil materials through leaching and erosion. Water
transforms soil material through weathering reactions, and translocates
solid and dissolved materials in mass flow within soil profiles.
Catena Concept and Hydrosequences
Milne’s (1935) catena concept stated that soils along a hillslope are
interconnected. A catena was originally defined as ‘‘a unit of mapping
convenience . . . a grouping of soils which, while they fall wide apart in a
natural system of classification on account of fundamental and morphological differences, are yet linked in their occurrence by conditions of topography and are repeated in the same relationship to each other wherever the
same conditions are met’’ (Milne, 1935). Since then, catenas have been
recognized in a variety of areas under a variety of climatic conditions and
have played an important role in soil and landform studies (Gerrard, 1981).
The wide applications of the catena concept have been complicated by
considerations of parent material variations and climatic differences. The
temporal as well as the spatial aspects of the soils are also important. The
real significance of catenas lies in the recognition of the soil processes and
geomorphic processes—especially those driven by water movement downslope—that are involved in catenary differentiation rather than in the formal
appearance of its product. From Milne’s original example of East African
catena to the latest catenas studied throughout the United States (e.g.,
Jenkinson et al., 2002; Reuter and Bell, 2003), hydrology plays the central
role. Soil profile changes from point to point in accordance with conditions
of drainage and past history of the land surface, and soil differences are
brought about by ‘‘drainage conditions, differential transport of eroded
materials, and leaching, translocation, and redeposition of mobile chemical
constituents’’ (Milne, 1936a,b).
Catenarey soil development occurs in response to the way water moves
through soils and over the landscape (e.g., Hall and Olson, 1991; Lin et al.,
2004; Moore et al., 1993; Thompson et al., 1997). Catenas developed from
similar parent materials, thus are often hydrosequences of related soils with
difference primarily in drainage, especially in depositional landscapes. Soil
hydrosequences have been more quantitatively studied in recent years,
mainly linked to the UDSA-NRCS Wet Soil Monitoring Project (e.g.,
Jenkinson et al., 2002; Reuter and Bell, 2003). As an example, Reuter and
Bell (2003) investigated quantitative relationships between soil hydrology
and morphology in seven landscape positions along a 125 m summit-towetland transect. They installed nested piezometers, observation wells, Pt
electrodes, and thermocouples at multiple depths to monitor water flow
direction, water table, and redox potential on a weekly to biweekly basis
for 4 years, with more intense sampling during spring and fall to capture the
effects of snow melt and plant senescence. Their major findings included the
following: (1) thickness and color of surface horizons in this landscape were
strong indicators of landscape hydrology, especially when redox features
associated with normal water table levels were masked by thick mollic epipedons; (2) profile darkness index (a ratio of A horizon thickness to Munsell
color value and chroma for A horizon) had strong correlation with the
duration of saturation; (3) equal piezometric head with soil depth suggested
a throughflow environment with potential for lateral water movement toward the wetland at the toeslope; and (4) depending on the precipitation,
overland flow, and subsurface flow, the wetland fluctuated between recharge
and discharge hydrology.
Spatial distribution of topographic attributes that characterize water
flow paths also captures spatial variability of soil attributes at the hillslope
scale. Because of the processes that occur along a hillslope, soils may
be quite different in different portions of a landscape, but these processes
and relationships may be similar across a larger area, particularly if geomorphology and stratigraphy are similar. Numerous investigations of catenas
have been completed in order to address this phenomenon (e.g., Evans and
Franzmeier, 1986; Khan and Fenton, 1994; Pennock and de Jong, 1990;
Stolt et al., 1993; Thompson et al., 1998; Veneman and Bodine, 1982). It
follows that the greatest variability in certain soil properties within a physiographic region may occur along a hillslope rather than from one side of
the region to the other (e.g., Lin et al., 2004; Pennock and de Jong, 1990;
Wilding et al., 1994). This warrants a careful investigation and understanding of hillslope soil variability and hydrosequences before making broad
generalizations about soil variability within a physiographic region.
Hydrology/Hydropedology as a Potential Means of Quantifying
Soil-Forming Processes
Jenny’s (1941) soil-forming factorial model states that soil (S) is a function of climate (cl ), organisms (o), topography (r), parent materials ( p), and
time (t). It is expressed as
S ¼ f ðcl; o; r; p; t; . . .Þ;
where the dots indicate additional unspecified factors (such as anthropogenic effects). The factors define the soil in terms of controls on pedogenesis and
soil distribution factors—‘‘an environmental formula’’ that defines the
‘‘state and history of a soil’’ (Jenny, 1941, 1980). Jenny believed that Eq. [7]
could be solved under ideal conditions and that the variables were independent, though he recognized that the factors also may be interrelated (Jenny,
1941, 1980). Interestingly, while the conceptual framework of Eq. [7] has had
a profound impact on pedological research and has been well received by the
soil and earth sciences communities (e.g., Amundson et al., 1994), Jenny
himself stated that ‘‘the fundamental equation of soil formation is of little
value unless it is solved’’ (Jenny, 1941). He further stated that the model
had been presented before (e.g., Hilgard, 1921) but that ‘‘I can solve the
equation. That was the new approach’’ (Jenny, 1980, p. xii). According
to Jenny (1941), the ultimate goal of functional analysis is the formulation
of quantitative laws that permit mathematical treatment. However, no
correlation had been found between controlling factors and soil properties
under field conditions that ‘‘satisfied the requirement of generality and
rigidity of natural laws’’ (Jenny, 1941, p. xii).
Wilding (1994) summarized the difficulties encountered in solving the
Eq. [7], including the following:
. assumption of independence of state factors with no or minimal interac.
factor interchangeability and feedback mechanism;
problems in obtaining partial differentials with nonoverlapping factors;
anthropogenic influences confounding factor variables;
spatial and temporal variability causing high noise in factorial analysis;
difficulty in rigorously reconstructing the time effects on pedogenesis;
lack of time-incremental datasets for developing pedogenic rate changes;
polygenetic pathways of soil genesis and multiple origins of soil properties;
lack of knowledge of precise processes;
lack of suitable database with geographical and geomorphological controls;
difficulty in testing and validating the model.
Since Jenny’s (1941) state-factor model, the progression of pedogenic
model developments has incorporated all, some, or none of the factorial
approach (Wilding, 1994), including the extended state-factor model
(Jenny, 1961, 1980), systems mass-balance process model (Chadwick
et al., 1990; Simonson, 1959), energy flux model (Runge, 1973; Smeck and
Runge, 1971), chemical equilibrium residua model (Chesworth, 1973a,b),
soil–landscape systems model (Hugget, 1975; McSweeney et al., 1994;
Ruhe, 1969), progressive-regressive evolutionary model (Johnson and
Watson-Stegner, 1987; Johnson et al., 1990), coupled reactions-factors-processes model (Ciolkosz et al., 1989), and various simulation models (Bryant
and Olson, 1987; Hoosbeek and Bryant, 1992; Levine and Ciolkosz, 1986;
Minasny and McBratney, 1999, 2001). While most models of soil formation
have been conceptual or qualitative, quantitative and systems models are the
hope of the future (Wilding, 1994). Even for many simulation models offered
today, the factorial approach is used as a control that governs the direction
and magnitude of specific pedogenic processes being simulated.
To quantify the factorial model of Eq. [7], there is much to be done
and many opportunities along the way. Nielsen et al. (1996), in discussing
the opportunity to strengthen soil science from surface soil moisture regimes
derived from increasingly available remote-sensing imagery, pointed out
that reduction of partial differential equations to ordinary differential
equations for describing soil physical, chemical, and biological processes
is an observational and theoretical challenge to both practitioners and
theoreticians. As discussed at the beginning of Section III.D, hydrology
affects and is affected by all of the five natural soil-forming factors and
the four general soil-forming processes. It is believed that pedogenic
processes are mainly driven by the presence and flow of water (e.g., Fritsch
and Fitzpatrick, 1994). Hence, hydrology/hydropedology may offer new
perspectives into solving the factorial Eq. [7]. While we do not yet know
the answers to this challenging task, we speculate that hydrology/hydropedology could potentially offer a more quantitative way to translate the
conceptual model of pedogenic processes into operational mathematical
formulae. As addressed throughout this chapter, it is the goal of hydropedology to be able to quantify interactive pedological and hydrological
processes across scales.
Dynamic Soil Properties
The classical five natural soil-forming factors, plus hydrology and human
activities, contribute to the temporal variation of the soil. Flux factors
(including hydrology) and human activities are more influential in relatively
shorter term dynamic soil properties over site factors. It appears that Jenny
(1941) considered climate as the major pedogenic driving vector acting
through time, with vegetation, topography, and parent material serving as
secondary (Wilding, 1994). The Soil Quality Institute of the USDA-NRCS
considers dynamic soil properties to be those that change with land use
and management (Fig. 18) or natural disturbances and cycles (such as
seasonal and diurnal changes), and that are important for characterizing
soil functions and ecological processes and for predicting soil behavior.
Grossman et al. (2001) also suggested use-dependant properties to be those
Figure 18 Combining dynamic soil properties (or use-dependent) and properties inherent
from natural soil-forming processes (or use-invariant) to form a composite record for soil
interpretations in soil survey databases. The control section used in the Soil Taxonomy is
generally below the dynamic surface soil. The distinction between major soil management types
within the same soil series using the concepts of ‘‘genoform’’ and ‘‘phenoform’’ (Droogers and
Bouma, 1997) separates the morphogenetic properties used in taxonomic units while near
surface temporal properties used in cartographic units that are management driven soil survey
units. (Modified from Grossman et al., 2001.)
soil properties that show change and respond to soil use and management
(such as soil organic matter levels and aggregate stability), and use-invariant
properties to be those soil properties inherent from natural soil-forming
processes that show little change over time and are not affected by soil use
and management (such as mineralogy and particle size distribution).
In the context of hydropedology, hydrology is apparently a major driving
force of dynamic soil systems, including changes in soil types and soil properties. For example, soil type changes could result from erosion, deposition, and
altered hydrological conditions (Ashby, 1987). Fitzpatrick et al. (1992) have
shown that yellow and grey duplex soils (Natraqualfs) in Australia have transformed to saline sulfidic march soils (Sulfaquents) in some sub-catchments
where rising saline water tables have resulted from land clearing. Soil moisture regimes play significant roles in classifying soils in Soil Taxonomy and
other soil classification systems (Soil Survey Staff, 1999; Buol et al., 2001).
In terms of hydrodynamics of soil properties, an example from a Vertisol
is used here to illustrate the importance of hydrology (Fig. 19). This
clay soil was very dry in August, and many extremely coarse (1–3 cm
wide) cracks appeared even in the tillage pan of the Ap2 horizon, thus
in situ measured apparent steady-state infiltration rates were high in both
Ap1 (0–0.1 m) and Ap2 (0.1–0.27 m) horizons. At the end of August, heavy
Figure 19 An illustration of temporal dynamics of apparent steady-state infiltration rates in
the Ap1 and Ap2 horizons of the Ships clay (Chromic Hapludert) at supply potentials of 0 (i0)
and 0.03 m (i 0.03) for a period of about 2 months. The initial gravimetric soil moisture content
(Wi) at each measurement occasion is also indicated. (Modified from Lin et al., 1998.)
rains from a tropical storm occurred for about a week. As a result,
soil moisture content increased significantly, and consequently, soil macroporosity decreased drastically (e.g., many cracks closed). The infiltration
rates measured in early September became very low in both Ap horizons.
After the rains, as the soil dried up under high evaporative demand in
hot summer, surface crusts formed, and intercrust cracks gradually
reappeared at the soil surface. Consequently, infiltration rates in the Ap1
horizon gradually increased. The drying process in the tillage pan was
much slower, and the re-opening of cracks in the Ap2 horizon was not
evident even a month after the rains, thus its infiltration rates remained
fairly low. Besides precipitation, ground water could also have a significant impact on dynamic soil properties. For example, Henry et al. (1985)
investigated the role of ground water discharge as a factor in soil salinization
under Saskatchewan conditions and reported that the salt content in the
soils was linearly proportional to the sodium percentage of the aquifer due to
upward water movement. Salinity over an artesian marine aquifer in the
Glacial Lake Agassiz in North Dakota created a large (about 77,000 ha)
unproductive area surrounding by prime farmland (Doering and Benz, 1972;
J. Richardson, personal communication).
Future needs in advancing hydropedology may be encapsulated in a
philosophy termed ‘‘bridging disciplines, scales, data, and education.’’ This
philosophy is critical in promoting the integration of relevant disciplines and
in educating the next generation of soil scientists and hydrologists.
Soil and water spatial-temporal distribution is both driven by and contributes to the landscape. Thus, a systems approach to understanding and communicating landscape–soil–water dynamics is needed. Such an approach would
facilitate the development of conceptual and mathematical models of landscape
hydrology and pedogenesis. Considering the unfolding research landscape of
the future, Bouma (2005) pointed out the essential role of a systems approach to
solving complex environmental problems. He stated that specialistic input is
not always effective when working in international panels or interdisciplinary
teams. He introduced a joint learning trajectory that can be effective in creating
true cooperation and interchange. Bouma (2005) further suggested that the role
of hydropedology in formulating environmental policies can best be considered
from two points of view: first, ‘‘up to global’’—its role in the international
panels in which policy issues are discussed and negotiations take place. The
issue is not so much that pedologists and hydrologists should be present in such
panels but, rather, that what they have to offer in terms of expertise should be
taken into account when formulating policy options. This requires effective
communication and a critical analysis of our own discipline by taking a broad
view of the issues at hand and attempting to understand the state of mind not
only of colleagues in the natural sciences such as climatology and geology but
also, and more importantly, of economists and political scientists. Second,
‘‘down to local’’—once international treaties have been agreed upon, they
have to be followed by implementation on the national or local level. This
requires input from science, but on a level that recognizes local conditions and
that is prepared to fine-tune its approach accordingly. Work is ideally done by
interdisciplinary teams and a systems approach.
What It Takes to Study Landscape Phenomena
Several aspects that would facilitate holistic studies of landscape-oriented
flow and transport phenomena in natural soils are the following:
1. Soil–landscape mapping is important for capturing patterns of variability,
which can give us a much better handle on the space-time dynamics of flow
systems than point data. For example, Klemes (1986) pointed out that ‘‘It . . .
seems obvious that search for new measurement methods that would yield
areal distributions, or at least reliable areal totals or averages, of hydrologic
variables such as precipitation, evapotranspiration, and soil moisture would
be a much better investment for hydrology than the continuous pursuit of
a perfect massage that would squeeze the nonexistent information out of a
few poor anemic point measurements. . .’’
2. Pattern identification at different scales is a key concept that needs to be
emphasized in landscape studies. Sivapalan (2003), in addressing the
connection between process complexity at hillslope scale and process
simplicity at the watershed scale, suggested that ‘‘One way to achieve this
reconciliation is to focus on common concepts, features, or patterns that
have physical meanings that transcend the range of scales in question,
and which are easily scalable. . . . They will lead to parsimonious models
and, over time, can also assist us in the development of a new theory of
hydrology at the watershed scale by shifting the focus away from smallscale theories at the hillslope (or lower) scales and towards new hydrologic concepts that transcend spatial scales, which are also worthy of
study in their own right.’’
3. Essential to future landscape hydrology is a concerted program of extensive and thorough experimental research on watershed scale dynamics
(e.g., Baveye and Boast, 1999; Grayson and Blöschl, 2000; Hornberger
and Boyer, 1995). This clearly calls for a change of attitude in the scientific
community. The collection and analysis of field data have been undervalued in the present computer modeling frenzy. In the past, many
research publications were devoted to field data collection, analysis, and
interpretation. Indeed, these provided some of the fundamental insights
into pedogenic and catchment processes. Yet today such publications are
very limited (Grayson and Blöschl, 2000; Hornberger and Boyer, 1995).
4. An iterative loop and interaction of ‘‘understanding, sampling, and
modeling’’ is central to enhanced studies of landscape–soil–water
dynamics. As noted by Grayson and Blöschl (2000), we generally begin
with some process understanding, do some sampling to improve that
understanding, and when we have enough understanding to be able to
attempt a conceptualization, we build a model. Hopefully this model
increases our understanding of the processes and, with some more sampling for proper testing, we iteratively refine our modeling and understanding. Because we can rarely sample densely enough to fully capture
the spatial-temporal variability of the system, we must exploit our understanding of dominant processes at different scales, identify patterns that
link point observations to areal phenomena, and use such knowledge to
implement optimal design of sampling.
Hydropedological Approaches to Landscape–Soil–Water Dynamics
Across Scales
Hydropedological approaches call for exploration of the most effective
manner to integrate pedological and hydrological expertise and the use of the
state-of-the-art techniques in mapping, monitoring, and modeling. It is also
important to ‘‘look first, then measure’’ in designing monitoring and measurement protocols based on soil morphology and soil distribution. Integration of
geostatistics with geospatial techniques, coupled with understanding from
pedological and hydrological expertise, would enhance interpolation and extrapolation of point observations to areal coverages. Many previous studies in
soil hydrology have undersampled in space or time, or both. Hydropedological
approaches would require adequate spatial coverage and a sufficiently long
monitoring period to understand the system well. Three hydropedological
approaches to landscape–soil–water studies are suggested here:
1. Mapping, monitoring, and modeling (‘‘3M’’) of landscape–soil–water systems: Much effort by non-pedologists is hampered because soil distribution and processes are not well understood such that site selection for
sampling or monitoring and the design of modeling do not represent
actual distribution and processes. To connect pedon and landscape phenomena, one of the keys lies in the distribution of various soils over the
landscape (i.e., soil patterns). We normally monitor pedons to collect
point data and model landscapes trying to understand areal distributions.
The key connecting the two is the mapping of various soil and other
landscape features. Relatively static properties of soil and landscape
features (such as topography and soil type) may be mapped out to assist
in scaling and modeling of landscape–soil–water dynamics, while more
dynamic properties (such as hydrology and land use) could be monitored
to refine model predictions. The fabric of soil over the landscape could
also help sampling design in both horizontal and vertical directions. For
instance, it is more meaningful for the vertical layout of monitoring
devices to correspond to soil horizons rather than equal depth increments
that ignore soil vertical layering. Mapping also provides a means of
diagnosing the landscape (e.g., identifying some patterns using geospatial
maps, remote-sensing imagery, historical records, or soil–landscape
surveys) before designing experiments and selecting monitoring sites (we
may call this ‘‘map first, then design’’ in landscape studies).
2. Integrating geostatistical and geospatial techniques (‘‘2GS’’) into a
Bayesian hierarchical multiscale modeling framework: Lin and Rathbun
(2003) proposed a Bayesian hierarchical multiscale modeling framework
as an infrastructure for linking soil properties to climatic, topographical,
geological, and vegetative processes, and to bridge data collected at
multiple scales of spatial support. In this framework, the importance of
soil map, DEM, land use, and other geospatial data is emphasized in
enhancing the use of geostatistics and in predicting the spatial-temporal
patterns of soil properties. Enhanced predictions can be achieved through
a combined use of ground-based point observations, GIS-based vector or
raster maps of various scales, and remote-sensing imagery, together with
pedological expertise about soil–landscape distribution. Combining data
collected at different scales of spatial support is achieved by partitioning
the modeling effort into separate process modeling and data modeling
stages. The process modeling stage consists of modeling the joint probability distribution of all variables at the point scale. Given such a model,
the joint distribution of the data collected at different scales of spatial
support may be obtained in the data modeling stage. Under the Bayesian
inferential paradigm, the effects of all sources of variation, including those
attributed to model components and those attributed to the process of
data collection, on the uncertainty of model predictions are readily quantified. Moreover, prior beliefs regarding the spatial distribution of landscape–soil–water systems may be readily incorporated into the Bayesian
hierarchical modeling framework.
3. Strategic spatial modeling and scaling: There are several routes to move
from point scale input at sampling sites to arial coverage of block scale
output using a process-based simulation model (Heuvelink and Pebesma,
1999). The routes depend on the sequence of three separate steps involved:
interpolating, aggregating, and running the model. This issue has been
referred to as the choice between ‘‘calculate first, interpolate later’’ or
‘‘interpolate first, calculate later’’ (Heuvelink and Pebesma, 1999; Stein
et al., 1991). The preferred route suggested by Heuvelink and Pebesma
(1999) is to interpolate point input data first, then run the model at point
locations within a desired block, and last aggregate model outputs spatially for areal coverage, thereby avoiding direct application of the model
at a larger spatial extent. This approach has been used in many coupled
GIS simulation modeling systems to address both spatial and temporal
dimensions (e.g., Clarke et al., 2002; Goodchild et al., 1996).
Sustainable Land Use Planning and Proactive Design
Land use planning provides an excellent basis for joint work of pedologists and hydrologists. Soil and water professions must become more heavily
involved in land use planning across the spectrum of applications of societal
importance. For example, the Dutch ‘‘layer-model’’ planning considers three
‘‘layers’’ (Bouma, 2005). The first one represents the natural dynamics of
land and water; the second one is all the networks of roads, railways, and
waterways; and the third one is human settlements. Ideally, new land use
plans should consider the sequence from one to three, taking into account
first the dynamics of land and water, which is the most difficult to affect or
should not be affected in sensitive areas in which substantial damage could
occur. Next, infrastructure networks have a higher degree of permanence
than settlements, which readily expand and contract. This approach offers
an attractive platform for applied hydropedology, working with other professions as well. The manner in which the natural landscape ‘‘throbs’’ offers
clues as to what can best be done and where it can be done with the lowest
risks and the greatest opportunities.
Most work in pedology and hydrology in the past has been rather reactive
in character. Either questions raised by others were answered or given
conditions were characterized, more often than not representing problems
caused by poor land use (Bouma, 2005). This has, of course, allowed
excellent research, but why not also take an occasional more proactive
approach? Why not take a given soil, consider climate conditions and
landscape features, and design a soil structure that would best satisfy
conflicting demands, for instance, a structure that would allow optimal
rooting, supply a relatively high amount of moisture, be trafficable, and
avoid bypass flow of agrochemicals? Bouma et al. (1999) have made an
attempt to do this. The focus could also be shifted to the landscape scale,
taking into account the ‘‘layer-model’’ mentioned previously. As is, we
dutifully document errors that engineers and architects have made. Why
not proactively design optimal soil and landscape structures based on comparing effects of different flow patterns and deliver these to engineers and
architects with an invitation to realize them in practice?
Pedologists are foremost among those basic soil scientists who help
develop integrated system models to scale up knowledge from small samples
to the global pedosphere (Fig. 8) (Sposito and Reginato, 1992). Pedologists
have studied both the mechanisms and the magnitudes of spatial variability
of soils and landforms as a basis for broad generalizations about soil genesis,
classification, and mapping, while soil physicists and hydrologists have
studied scaling theories such as similitudes and fractals and quantify spatial
variability using methods such as geostatistics and temporal variability using
time series analysis (Lin, 2003). As pointed out by Nielsen and Wendroth
(2003), while there exists a versatile and powerful set of statistical tools for
diagnosing spatially and temporally variable field observations, we have to
explore the cause of variation and improve and expand soil classification
concepts. The efforts made by pedologists, soil physicists, and hydrologists
on soil variability and scaling do not seem to have converged well in the past.
It is to be hoped that hydropedology will generate new opportunities for
such needed synergistic efforts.
There are several possible ways to help de-mystify the mind-boggling
variability of field soils, especially if the synergies are put together jointly
by pedologists, soil physicists, and hydrologists. These include (1) systematic
understanding of soil variability as a function of various space-time factors,
(2) using pattern identification and the concept of pedodiversity, and
(3) organizing soil variability based on hierarchical multiscale frameworks.
Each of these approaches is further discussed in the following sections.
Soil Variability as a Function of Space-Time Factors
Except for some possible scale-invariant soil properties and processes,
different spatial or temporal variations are generally observed depending on
the scale. In a general conceptual framework, the magnitude of soil variability (SV ) is influenced by at least five space-time factors, i.e., spatial extent
(e) or area size, spatial resolution (r) or map scale, spatial location (l ) and
physiographical region, specific soil property or process ( p), and time factor
(t). Conceptually, this may be expressed as:
SV ¼ f ðe; r; l; p; t; . . .Þ;
where the dots indicate additional unspecified factors. Unlike the soilforming factorial model of Eq. [7], Eq. [8] is intended to be a functional
expression rather than a causational relationship. The exact expression of
Eq. [8] is very difficult, if not impossible, to establish, in part because of the
diversity and complexity of the relationship. Nevertheless, Eq. [8] may serve
as a useful conceptual framework for organizing our knowledge about soil
Broadly speaking, it may be expected that as spatial extent (e), spatial
resolution (r), or time scale (t) increase, the magnitude of soil variability
would increase, reaching a possible maximum and then would start to
stabilize or decrease as space or time dimensions continue to increase;
however, the mode and magnitude of such changes would depend on
where the soil is located, in what landscape (i.e., spatial location l ), and
which soil type or specific soil property (i.e., p) is of concern. Numerous
publications have provided evidence that supports the conceptualization of
Eq. [8] (e.g., Burrough, 1993; Heuvelink and Webster, 2001; Lin et al., 2004;
Wilding and Dress, 1983; Wilding et al., 1994). However, there is still a great
need to further the understanding of the complexity, diversity, interactions,
and quantification related to Eq. [8].
For example, the magnitude of soil variability generally increases with
increasing spatial extent from individual pedons to pedons that meet the soil
series concept to mapping units of a given series to all soils within a survey
area. But for some soils, the variability may occur at a limited segment of the
landscape, while for others, maximum variability may occur at long-range
intervals corresponding to mapping units, geomorphic units, or physiographic regions. Spatial variability with increasing spatial extent may be
linear, curvilinear, or other forms depending on soil types and landscape
features. It is apparent that when soil sample size is changed from small
cores to field plots, soil structure becomes more influential; when sample size
is further enlarged from field plots to watersheds, variation in topography,
land use, and other landscape features will have a more significant impact on
soil variability.
Spatial resolution or map scale indicates the level of detail that can be
discerned during the survey process or in cartographic representation of soils
information. As spatial resolution or map scale increases (like ‘‘zoom in’’),
new levels of detail are realized, while general patterns may be lost. On the
other hand, implicit representation of soil variability decreases as data are
aggregated from field-collected information to more generalized soil maps.
Some studies have suggested scale independence or self-similarity of pattern
and form in thematic maps of soil (e.g., Burrough, 1981, 1983a,b, 1993).
Burrough (1989) reviewed the evidence for fractals in soil variation and
concluded that while some properties showed linear log-log variograms over
a range of spatial scales, thus suggesting fractal scaling, many plots displayed
clear breaks of slope that implied a transition from one pattern of variation at
one scale to another pattern at a larger scale. Such departures from the ideal
fractal model led Burrough (1983a) to propose a nested model of stochastic
functions as a basis for understanding multiscale spatial variation. However,
fractal or multifractal models need to be further investigated and elaborated
for potential bridging multiscales of diverse soil and landscape properties and
processes (Crawford et al., 1999; McBratney, 1998).
Not all soil map units at a given scale share the same range of soil
properties. Some parts of the landscape are more variable (such as areas
with steep slopes), and thus a map unit delineating such an area may have a
broader range of characteristics than the one delineated on a more level
portion of the same landscape. However, in addition to topography, the
regional and local scale impacts of climate, hydrology, organisms, parent
materials, and human activities on soil variability can also be significant. For
example, soils on nearly level fluvial flood plains are some of the most
spatially variable soils in a landscape, but if soils have developed from
loess superposed on top of a terrace position, then these soils would be
quite uniform. Wilding et al. (1994) suggested that soil spatial variability
increases with the nature of parent materials in the following order: loess <
till < fluvial deposits < phroclastic and tectonic rocks < drastically
disturbed materials.
Not all soil properties or processes vary in a similar manner. More stable
soil properties such as texture, mineralogy, soil thickness, and color are less
variable than more dynamic properties such as moisture content, infiltration
rate, hydraulic conductivity, redox state, biological activity, and organic
matter content. For soil hydraulic properties, the observed coefficients of
variation are often much higher, commonly over 100% (e.g., Jury, 1986;
Wilding and Drees, 1983). Surface soils have more dynamic changes and
thus tend to have a higher magnitude of variability, while subsoils tend to
have smaller variability. However, such differentiation depends on many
other factors, such as human activities, hydrology, geology, and landforms.
In terms of the time factor, Beckett (1987) pointed out that the temporal
variability of soil nutrient status may equal or exceed spatial variability.
Burrough (1993) suggested that if one soil-forming process dominates for a
long time, it usually leads to a reduction of soil variability. For example,
tropical weathering can give rise to large areas of apparently uniform soils
(e.g., Oxisols), in which variations caused by differences in parent materials
and relief can be reduced by long periods of deep weathering under a tropical
humid climate. Temporal variability of soil physical and hydrological properties has been studied, but dynamic change of soil types (taxonomic units)
used in soil survey and mapping has received little attention (excluding the
change of Soil Taxonomy and taxonomists over time). Although it generally
takes thousands of years to form a natural soil, the growing recognition of
man-made changes to soils has elevated the importance of dynamic soil
changes, especially those related to land use and management.
Pattern Identification at Various Scales
Patterns are everywhere in nature, from the ‘‘blue marble’’ view of the
earth from space, to ‘‘fractal tree-like’’ channels of streams vivid in remotesensing imagery, to dyed pictures of preferential flow observed in field soils.
Pattern, or spatial-temporal organization, offers rich and comprehensive
insights regarding many phenomena in nature. Observation and interpretation of spatial-temporal patterns are thus fundamental to many areas of the
earth sciences such as geology, geomorphology, pedology, and hydrology.
Indeed, soil mapping is based on identifying soil–landscape patterns at
various scales, often depicted in 3-D block diagrams (e.g., Fig. 5) or 2-D
soil maps. We believe that to advance the knowledge base of hydropedology,
and to answer many questions regarding the earth’s critical zone, we need to
explore the information that resides in the myriad of patterns observable in
the pedosphere and the hydrosphere at different space-time scales. We
should emphasize the importance of spatial pattern identification in combination with long-term monitoring in our scientific investigations. For example, as demonstrated in an impressive volume of work complied by Grayson
and Blöschl (2000), there is rich information in spatial patterns that provides
much more stringent tests of hydrological models and much greater insights
into hydrological behavior than traditional methods. They pointed out two
catalysts that brought the issues of patterns to the forefront of hydrologists’
minds (and, we believe, to soil scientists’ minds as well):
1. The readily available DEMs and an array of analysis that is possible with
these data, accelerated by the ever–decreasing cost of computing power
and vastly available geospatial technologies and databases;
2. The rise in environmental awareness of the broader community and its
subsequent impact on the research and management of natural resources.
We now want to know not only the quantity and quality of soil and water
resources, but also from where and when any contaminants come and
where and when best to invest scarce resources to help rectify the problem.
In principle, we now have the tools available to undertake spatialtemporal modeling of environmental response, and the spatially distributed
and temporally dynamic models in combination with attractive color maps
that geospatial technologies generate can seduce even the most skeptical of
politicians and administrators (Grayson et al., 1993). However, while our
ability to generate patterns using computers might be impressive, it is not of
use by itself. What is important is the extent to which these patterns represent reality and to which they provide us with new insights into natural
processes (Grayson and Blöschl, 2000). Observed spatial-temporal patterns
of pedologically and hydrologically important variables are not very common (other than terrain, land use, and in some cases general soil types). To
progress, we will need to make measurements different from those used in
the past, requiring the development of new instruments and approaches. We
will also need to develop more sophisticated methods to analyze spatialtemporal patterns and to correlate measured patterns against model
predictions. Grayson and Blöschl (2000) predicted that testing models by
comparing simulated and observed patterns will eventually become commonplace and will provide a quantum advance in the confidence we could
place on predictions from distributed hydrological models.
Pattern identification may also offer another way of de-mystifying soil
variability. This could be represented either explicitly (e.g., mapped spatial
pattern) or statistically (e.g., geostatistical distribution functions). A number
of recent catchment hydrology field investigations demonstrate how the
understanding and modeling of hydrological processes can be improved by
the use of observed spatial patterns. For example, in the humid climate of
the Tarawarra catchment in Australia, through extensively observed TDR
soil moisture data, spatial-temporal patterns of soil moisture were revealed
(Fig. 20; See Color Insert): in a wet winter, surface and lateral flow was
dominant, producing a topographically organized spatial pattern, while in a
dry summer, there was minimum lateral redistribution and fluxes were
essentially vertical, thus producing a random pattern that was not related
to topography (Grayson et al., 1997; Western et al., 1999). Analysis of
remotely sensed soil moisture patterns in the semi-arid Walnut Gulch watershed in Arizona indicated that, following a rainstorm, these patterns were
organized but this organization faded away after the storm, and the pattern
became random (Houser et al., 2000). The authors suggested that this
change-over was a reflection of the changing control on soil moisture by
rainfall vs soil characteristics during the dry-down process. However, some
spatial patterns of soil moisture are temporally persistent (the notion of
‘‘time stability’’) (Vachaud et al., 1985). Evidence for time stability has
been found by Grayson and Western (1998), Kachanoski and de Jong
(1988), Mohanty and Skaggs (2001), and others. However, time stability of
spatial pattern may be a function of spatial scale and may vary across a
landscape with different soil types, as shown by Kachanoski and de Jong
(1988) and by Zhang and Berndtsson (1991). This implies that soil water
variability needs to be analyzed simultaneously in both space and time. In
this regard, pattern recognition offers special advantages when analyzing
time-dependent properties in space (e.g., Fu, 1982; Zhang and Berndtsson,
Grayson and Blöschl (2000) illustrated the implications of different patterns on hydrological response in a watershed. Figure 21 (See Color Insert)
shows two simulated patterns of soil moisture deficit, each with the same
properties of mean, variance, and correlation length, but one spatially
random and the other ‘‘organized’’ by a wetness index. These two patterns
produce different responses to a given rainfall input: the organized pattern
gives higher and earlier runoff peaks than the random case for small rainfall
events, while the reverse is true for larger precipitation events (Fig. 21).
Four types of spatial patterns may be identified based largely on the
source and nature of data. Such patterns could be obtained through direct
measurements (with or without interpolations), indirect interpretations, or
surrogate correlations (Grayson and Blöschl, 2000):
1. ‘‘Lots of points’’ pattern: When there is a sufficiently dense array of point
measurements, a pattern could be generated by interpreting point data
(Grayson et al., 2002). The array may be based on random, grid, or
various stratified designs. The quality of such an interpolated pattern
depends on the number and distribution of point data, the accuracy of the
original point measurements, and how well the interpolation/extrapolation method reflects the underlying spatial structure of the measured
2. Vector pattern: This pattern includes stream and road networks, geological faults, soil polygons, clay cracking pattern, preferential flow patterns,
and many other line features depicted on a map. This type of pattern can
be easily stored and analyzed in a GIS.
3. Raster pattern: This type of pattern comes largely from remote-sensing
imagery. It could be qualitative binary (e.g., snow or no snow interpreted
from SPOT satellite), multinary (e.g., various land use/land cover interpreted from Landsat satellite), or quantitative (e.g., brightness temperature
obtained from airborne electronically scanned thinned array radiometer,
or ESTAR). Even for a binary pattern, studies have demonstrated that a
wealth of information can be revealed, such as saturated or unsaturated
conditions (Troch et al., 2000), whether runoff occurred or not (Vertessy
et al., 2000), recharge or discharge (Salvucci and Levine, 2000), and with
or without snow cover (Tarboton et al., 2000). This led Grayson and
Blöschl (2000) to call for a change in attitude toward ‘‘non-quantitative’’
data and hydrological model structures. As suggested by Grayson et al.
(2002), remotely sensed data might be used to assist in reducing the
degrees of freedom in distributed hydrological models by providing patterns rather than absolute values of important inputs. Such a philosophy
could also be applied to many pedological and soil survey data.
4. Surrogate pattern: This refers to surrogate data showing correlation to the
pattern of interest but uses data that are much easier to collect in a
spatially distributed fashion (Grayson et al., 2002). For example, we
cannot directly obtain a map of soil hydraulic conductivity or bulk
density from remote sensing, yet it is such parameters that hydrological
models need. As demonstrated by Mattikalli et al. (1998), some of the
remote-sensing instruments (e.g., ESTAR and synthetic aperture radar,
or SAR) can provide information on characteristics related to these
variables, but not the variables themselves. This presents a major challenge for hydrologists and soil scientists, i.e., to build models or pedotransfer functions that are able to exploit the information that is coming
from remote-sensing platforms (Grayson et al., 2002).
Once pattern information is obtained, it can be used in many beneficial
ways, such as (1) designing experimental setup and field data collection
strategy; (2) stratified interpolations/extrapolations of sparse point data;
(3) characterizing and modeling variability such as spatial correlation
and connectivity; (4) refining model structure for enhanced modeling of
landscape–soil–water dynamics; and (5) use in combination with time series
data to provide more realistic space-time simulations of soil moisture and
many other pedological and hydrological phenomena.
Related to pattern identification is yet another positive way of looking at
soil variability, that is, the concept of pedodiversity (or soil diversity), which
is analogous to biodiversity (Ibanez et al., 1995, 1998). There are two
essential components of biodiversity: the variety (or number) of species
(richness) and their spatial distribution or pattern (evenness) (Magurran,
1988). Indices of diversity often attempt to incorporate both components of
diversity into a single figure, or else they tend to neglect one or another
(Ibanez et al., 1995). However, unlike plants, animals, or other organisms,
for which each individual is a discrete entity clearly separated from each
other, soils are a continuum over the landscape, often without a clearcut
distinction between individual soils. Moreover, specific soil properties are
often more of concern to a particular application rather than generic soil
types that are more suited for general land use planning. For example,
precision agriculture is more focused on spatial distribution of soil nutrients
and pH values over a farm field, nonpoint source pollution modeling is more
interested in soil hydraulic properties distribution over the landscape, and
carbon sequestration is more concerned with soil organic matter and inorganic carbon dynamics. Therefore, it is important that pedodiversity
addresses not only the number of soil types, but also specific soil properties
that are often of practical concern. While the concept of pedodiversity is still
in its infancy, we suggest that four components of pedodiversity be differentiated: (1) the number of soil types and their relative abundance within an
area, (2) the variation of specific soil properties, (3) the spatial distribution or
pattern of soil types and various properties, and (4) the temporal dynamics
of soil types and specific properties. The first two components are directly
linked to the development and use of soil maps and the latter two are related
to the identification of soil spatial-temporal patterns.
Hierarchical Frameworks for Bridging Multiscales in
Scale transfer or multiscale bridging remains the heart of many
hydrological and pedological studies (e.g., Baveye and Boast, 1999; Lin,
2003; Sposito, 1998). At present, no single theory that is ideal for spatial
aggregation or upscaling and disaggreagation or downscaling of soils information has emerged. The major contenders seem to be either scaling via a
naturally defined or human-defined hierarchy or through potential continuous hierarchies as suggested by fractal theory (Fig. 8) (e.g., Cushman, 1990;
Lin and Rathbun, 2003; McBratney, 1998; Nielsen and Wendroth, 2003;
Wagenet, 1998; Vogel and Roth, 2003).
Hierarchical frameworks have been conceptualized by soil scientists as a
means for organizing multiple spatial and temporal scales from the soil pore
to the pedosphere (e.g., Hoosbeek and Bryant, 1992; Sposito and Reginato,
1992; Wilding, 2000). Hierarchical complexity has been studied in pedology,
which has long recognized self-organized complexity in the processes of soil
formation, with taxonomic frameworks constructed to summarize that ordering (Buol et al., 2001). If properly constructed, a hierarchy of soil systems
should reflect logical links and quantitative relationships among scales. It
can be argued, however, that the soil scientists’ hierarchy of scales is more an
operational or observational device, based on the ability or feasibility to
measure, rather than fundamental differences in basic processes (Wagenet,
1998). As suggested by Wagenet (1998), an examination of ecological hierarchy theory (Haigh, 1987; O’Neill et al., 1986, 1989) should present some
valuable philosophical and practical concepts pertaining to the translation
of information across scales in soil systems. Hierarchy theory in ecology
defines ‘‘holons,’’ which are nested spatial units characterized by integrated
biological, physical, and chemical processes (Haigh, 1987). In comparison,
soil science uses entities that are less well defined and procedures that are less
Lin and Rathbun (2003) discussed two hierarchical frameworks for bridging multiscales in hydropedology through either a data-driven or a processbased approach (Fig. 22). In the first, the soil mapping hierarchy depicts soil
spatial distribution over landscapes of varying sizes, considering five orders of
soil surveys, spatial aggregations of soil map units, and various applications
of geostatistics. The merger of geostatistics with traditional soil mapping has
led to encouraging new developments of environmental correlation modeling
and landscape-guided soil mapping. In the second, the soil modeling hierarchy deals with soil process models at different scales. While the current
generation of surface and subsurface process models is strongly scale dependent because of process representations, parameter requirements, and
changes of support in model variables, several approaches for scale bridging
Figure 22 Two hierarchical frameworks for bridging multiscale: Hierarchies of (A) soil mapping (for soil distributions) and (B) soil modeling (for soil
processes). SSURGO, STATSGO, and NATSGO are county-, state-, and national-level soil maps, respectively. (From Lin and Rathbun, 2003.)
are available, including upscaling, downscaling, upscaling with downscaling
embedded, strategic cyclical scaling, and strategic spatial scaling (e.g., Lin and
Rathbun, 2003; Mulla and Addiscott, 1999; Root and Schneider, 1995). In
moving beyond the notion of ‘‘trying to model everything,’’ we should be
developing methods to identify dominant processes that control pedological
and hydrological responses in various environments at different scales, and
then develop models to focus on these dominant processes (a notion called the
‘‘dominant processes concept’’) (Grayson and Blöschl, 2000).
Vogel and Roth (2003) discussed different approaches for incorporating
spatial heterogeneity into modeling flow and transport in soils, including
many concepts for the organization of heterogeneities, such as macroscopic
homogeneity, discrete hierarchy, continuous hierarchy, and fractals (Fig. 8).
They further suggested a ‘‘scaleway’’ as a promising tool for predictive
modeling of flow and transport in the subsurface at any scale. This conceptual approach is based on the explicit consideration of spatial structure that
is assumed to be present at any scale of interest, while the microscopic
heterogeneities are replaced by averaged, effective description. The three
ingredients needed in their approach are (1) the structure of the medium,
which must be known, (2) the corresponding effective material properties,
and (3) a process model at the scale of interest. They demonstrated
the scaleway concept for the prediction of a breakthrough curve in an
undisturbed soil column using structural information from two scales.
In view of a common limitation of deterministic approaches to quantify
multiscale dynamics of hydropedological processes, Lin and Rathbun (2003)
also suggested a Bayesian hierarchical multiscale modeling framework, which
has been successfully applied to several environmental applications (e.g.,
Gotway and Young, 2002; Wikle et al., 2001). However, such a framework,
as highlighted in Section IV.A.2, has yet to be applied to soil science and
Pedotransfer functions (PTFs) relate simple soil characteristics often
found in soil surveys to more complex parameters that are needed in modeling
and that are relatively difficult to measure (Bouma, 1989; Bouma and van
Lanen, 1987). The basic idea of PTFs may be generalized to include the
derivation of any needed soil attribute, which is not directly available,
based on available soils data. McBratney et al. (2002) further broadened the
concept of PTF to soil inference systems (SINFERS), in which PTFs form
knowledge rules for inference engines and uncertainty analysis is included in
prediction. Bouma (2005) suggested that SINFERS is a promising approach
that extends the PTF concept beyond the derivation of soil moisture retention
and hydraulic conductivity.
Many publications on the subject of PTFs have appeared in the past two
decades or so, largely centered around the estimation of soil hydraulic properties using basic soils data (e.g., Pachepsky and Rawls, 2005; van Genuchten
et al., 1999; Wösten et al., 2001). This illustrates the significance of the combined approach of hydropedology. The USDA-NRCS and other national and
international organizations are also pursuing PTFs for populating data in soil
survey databases. While various degrees of success have been achieved with
different PTFs (e.g., Pachepsky et al., 1999; Wösten et al., 2001), limitations,
uncertainties, and risks remain. Uncritical application of automatically generated PTFs is likely to produce poor results when used with no feeling for
what may broadly be expected or when the functions were derived from data
obtained for other soils than the ones being characterized (Bouma, 2005).
1. Enhancements of Soil Databases and Pedotransfer Functions
Lin (2003) assessed the current status and future opportunities of PTFs.
He pointed out several areas needing improvements, including the need for
(1) exploring fundamental mechanisms underlying PTFs, (2) linking pedon
data to landscape features, (3) incorporating soil structural information,
(4) considering spatial and temporal scales, and (5) improving practicality
of PTFs to enhance the value of soil survey databases. In the following, we
suggest additional areas for which enhancement of soil databases and PTFs
would be worth pursuing:
1. One critical point made by Lilly and Lin (2005) and Wösten et al. (2001)
is that major progress in PTFs is not to be expected from new statistical
methods, but rather from better data. While empirical, regression, or
functional approaches continue to be used in PTFs, new methods for
developing and using PTFs are increasingly being explored, including
artificial neural networks (e.g., Minasny et al., 1999; Schaap and Bouten,
1996), regression trees (e.g., McKenzie and Jacquier, 1997; Rawls and
Pachepsky, 2002a), and the group method of data handling (e.g.,
Pachepsky and Rawls, 1999). However, the success of any mathematical
or statistical techniques will be heavily dependent on the quality, quantity, comparability, and transferability of the original data stored in the
databases. Without the foundation of reliable and systematic databases,
no matter how sophisticated the techniques used in deriving or using
PTFs, the outputs would be futile and misleading.
2. As alluded to by Lin (2003), it would be more beneficial if flow patterns
(related to flow mechanisms and pathways) could also be determined from
soil survey and related landscape databases. In this regard, classification
or functional grouping of soils based on flow and transport characteristics
would be worthwhile, particularly if linked to soil map units. This could
provide a means of estimating a priori how important preferential flow
phenomenon is in a given soil or location (Jury, 1999).
As illustrated by Moore et al. (1993), Rawls and Pachepsky (2002b),
and others, topography helps the understanding of causation and
correlation of soil properties with landscape positions. Hence, ‘‘toporectifying’’ PTFs through taking into account topographic attributes would
improve landscape-based PTFs.
Most data in traditional soil survey databases are collected during a
specific time window. Recognizing the importance of use-dependent and
other dynamic soil properties, the USDA-NRCS is now considering the
development of a dynamic soil properties database. Such a database,
once developed, would significantly enhance the utility of soil survey
databases and the development of dynamic PTFs. We believe hydropedology is a helpful framework that can provide a bridge connecting
dynamic soil properties and traditional soil survey databases.
Pedology traditionally has focused on natural processes that do not reflect
the effects of short-term soil management. This was done on purpose to
avoid frequently changing classifications of a given soil following different types of soil management. Droogers and Bouma (1997) suggested the
term ‘‘genoform’’ for genetically defined soil series and the term ‘‘phenoform’’ for soil types resulting from a particular form of management in a
given genoform (Fig. 18). Such a distinction between major soil management types within the same soil series facilitates the incorporation of
management effects on soil properties and could potentially enhance
PTFs that involve soil series and land uses as carriers of soil hydraulic
information (e.g., Pulleman et al., 2000; Sonneveld et al., 2002).
Remote-sensing techniques offer significant opportunities for soil scientists to infer the state of soil based on surface-oriented patterns and to
extend these sensor techniques both laterally and vertically to describe the
dynamic 3-D nature of the soil with spatially variable properties across
landscapes (Nielsen et al., 1996). Thus, PTFs utilizing remote-sensing
inputs would be attractive. As alluded to in Section IV.B.2, ‘‘nonquantitative’’ data such as binary or multinary remotely sensed data
could be used to assist in reducing the degrees of freedom in models by
providing patterns rather than absolute values of important inputs.
Soil maps can no longer be static documents. Rather, derivative and
dynamic maps, created for specific purposes or functions, must be generated from original soil maps and tailored to particular applications. Thus,
PTFs, in combination with computer models and geospatial databases,
need to be integrated into expert systems to derive such maps. Until now,
there has been a lack of appropriate means of producing derivative and
dynamic maps such as soil hydraulic properties through space and time.
Soil Reference Systems and Hydropedoinformatics
McBratney et al. (2002) proposed the concept of SINFERS, in which a set
of properly and logically conjoined PTFs serve as the knowledge rules for
inference engines. Such a SINFERS takes known measurements with a given
level of (un)certainty and infers desirable unknown data with minimal inaccuracy allowed in the system. The SINFERS has a source, an organizer, and
a predictor; together they serve as a decision support system for appropriate
use of PTFs with uncertainty estimation. The sources are collections of soil
databases and PTFs, which could also include geospatial data. The organizer arranges and categorizes the PTFs with respect to their required inputs
and soil types from which they were generated. The inference engine is a
collection of ‘‘if-then’’ type of logical rules for selecting appropriate PTFs
with the minimum variance possible. The uncertainty of the prediction is
assessed using Monte Carlo simulations, which can be quantified in terms of
the model uncertainty and input data uncertainty (McBratney et al., 2002).
Sommer et al. (2003) presented an integrated method for soil–landscape
analysis, in which a hierarchical expert system for multidata fusion of
inquires, relief analysis, geophysical measurements (such as EM38), and
remote-sensing data was developed. They further combined the soil-forming
factorial model with the scaleway of Vogel and Roth (2003) to address soil
variability across scales. In their system, soil variability is separated at every
scale into (1) a scale-typical and predictable part and (2) a random part,
which becomes structure at the next lower scale level.
To integrate knowledge, scales, and databases of interactive pedological
and hydrological processes and properties and to streamline information
capture, storage, visualization, modeling, and decision-making, we suggest
hydropedoinformatics. The term is coined from hydroinformatics and pedometrics, both of which have received growing interest in recent years. Hydroinformatics is the study of the flows of knowledge and data related to water
flow and all it transports, together with interactions with both natural and
man-made environments (Abbott, 1991). It is a discipline that has strong
ancestry in computational sciences and artificial intelligence, where GIS and
data mining (artificial neural networks and genetic algorithms specifically)
are the new technologies with probably the widest applicability to the water
industry (Savic and Walters, 1999). Pedometrics is the application of mathematical, statistical, numerical, and artificial intelligence methods to soil
science in general and soil surveys in particular (McBratney, 1986; Webster,
1994). Pedometricians treat soil properties as spatially correlated random
processes and tap the richness of geostatistics and artificial intelligence for
analysis and prediction. However, although pedometrics has helped in elucidating pedogenesis by quantifying relations between individual soil properties and controlling soil-forming factors, solving the full system of
multivariate equations needed to describe soil genesis at different scales
remains one of the biggest challenges for pedometricians (Webster, 1994).
To make hydropedoinformatics meaningful, one critical need is a network
of well-designed and carefully maintained natural laboratories for systematic
field data collections. Soil science and hydrology communities have long
recognized the fundamental need for multiscale, multidisciplinary, and longterm field experiments, including better archiving and sharing of field data
across geographic regions (e.g., NRC, 1999, 2001a). The natural laboratory
concept is the basis for the U.S. National Science Foundation’s Long-Term
Ecological Research Network, established in 1980 for investigating ecological
processes operating over extended periods (months to centuries) at a variety
of spatial scales (from 10 m to continental) (NRC, 2001a). The existing
NCSS program and several other national field experimental networks
(such as the land-grant universities’ experimental stations, the USDAARS’s experimental watersheds, the USGS’s large basin gauging stations,
and the DOE’s waste disposal sites) could serve as good starting points for
exploring such coordinated efforts. The NCSS has provided over 100 years
of soil inventory, measurement, and evaluation, and currently maintains
several national databases (such as SSURGO and STATSGO, official soil
series descriptions, soil characterization laboratory database, soil climate
monitoring network, and wet soil monitoring network). These databases
need to be well coordinated and better utilized in both the development of
PTFs and the construction of integrated hydropedoinformatic systems.
Effective education in the 21st century takes on two new emphases—
integrated multidisciplinary and technology enhanced (e.g., Boyer Commission, 1998; NRC, 1998). It is believed that constructivist-based education must
begin to replace transmission-dominated education if we are to train professionals who can solve interdisciplinary problems. In this respect, integrated
broad-based education and new technologies promise opportunities to infuse
cognitive learning and problem solving into curricula. For example, a wave of
‘‘watershed thinking’’ is spreading across the United States and around the
globe as watershed-based approaches offer a more integrated way to address
natural resources and environmental issues holistically. Accordingly, new
educational programs should provide training for holistic interdisciplinary and
technology-enhanced approaches to soil and water studies. Effective watershed
management also requires the integration of theory, field data, simulation
models, expert judgments, policies and regulations, socio-economic factors,
and ethics in solving practical problems (NRC, 1999).
Interdisciplinary and Integrative Knowledge Base and Skills
The interdisciplinary emphasis of education in the 21st century makes
hydropedology a timely addition to the education of the next generation
of soil scientists and hydrologists. Hydropedology by its very nature is
interdisciplinary. As such, future hydropedologists must have an integrated
knowledge base in pedology, soil physics, and hydrology as well as in other
related bio- and geosciences such as geomorphology, stratigraphy, hydrogeology, hydroclimatology, ecohydrology, landscape ecology, and other
branches of soil science. In a sense, pedology itself is an integrative earth
science. Hence, hydropedology education should cover a broad spectrum of
topics dealing with the earth’s critical zone (Fig. 1). Quantitative hydropedology requires the use of mathematics/(geo)statistics and simulation modeling. Thus, pedometrics including spatial-temporal statistics (e.g., Nielsen
and Wendroth, 2003) should be an integral part of hydropedological education. It is also important that future hydropedologists possess skills in
geospatial and other emerging information technologies, as well as advanced
instrumentation, to enable them to collect, visualize, analyze, and model
spatial-temporal patterns of landscape–soil–water dynamics across scales.
Fundamental Importance of Field Work
Field work is a distinct aspect in geosciences, including hydropedology.
Field work provides the basis for understanding a variety of earth processes
and validating model, laboratory, and remote-sensing results (NRC, 2001a).
In particular, field survey and mapping of landscape–soil–water systems are
fundamental skills for the next generation of hydropedologists. Butler (1980)
pointed out that soil survey is ‘‘one of the basic technologies of soil science.’’
Kutı́lek and Nielsen (1994) suggested that a modeler should work in or at
least supervise the experimental activity in the field. Vice versa, effective field
experimentation requires the theoretical knowledge of the modeler. Hydropedology blends together field work, laboratory experiments, and computer
modeling into an integrated approach to understand landscape–soil–water
dynamics. Sometimes experimental activities take place in the laboratory;
other times they take place in the field, where the comfort of laboratory
control is lost. However, field work (observations, monitoring, experiments,
and verifications) ultimately serves as the initial drive of identifying
real-world problems, formulating theories of processes and events, and
validating models and their predictions. Field work is the foundation of
all, and the logical sequence of hydropedological research is from the field to
laboratory and to modeling, and then back to the field. An adequate understanding of this sequence has a profound impact on how we educate future
generations of hydropedologists.
The critical zone is perhaps the most heterogeneous and complex region
of the earth and the only region of the solid earth readily accessible to direct
observations (NRC, 2001a). Integrated studies of the critical zone require
interdisciplinary and multiscale approaches. Because soil and water are
integral parts of the earth’s critical zone, hydropedology is important to
address interactive pedological and hydrological processes and their properties in the unsaturated zone. Hydropedology represents a paradigm shift in
our basic thinking and approach to ped, pedon, landscape, watershed,
regional, and global scale analysis of soil and water interactions.
The birth of modern soil science started with the recognition of soilforming factors and related processes (Dokuchaev, 1893). The soil as a
natural resource has historically been extensively explored for agricultural
production in order to provide food, feed, fiber, and fuel for the ever-growing
human population. In past decades, awareness of environmental protection
and ecosystem sustainability has driven much of soil science development. In
the 21st century, with the need for integrated approaches to study the earth’s
critical zone, it is paramount that soil science become an integral part of the
earth, environmental, and ecological sciences, in addition to continuing the
services to agricultural communities. Getting back to the root in geosciences,
and hence completing a full circle, provides a more realistic picture of what
soil science can and ought to contribute to our science and society.
In examining the opportunities in the hydrological sciences, a group
of experts under the auspices of the National Research Council claimed
that ‘‘We cannot build the necessary scientific understanding of hydrology
at a global scale from the traditional research and educational programs
that have been designed to serve the pragmatic needs of the engineering
community’’ (NRC, 1991, p. 33). Similarly, we believe that we cannot build
the necessary scientific understanding and appreciation of hydropedology
(and perhaps soil science as a whole) at a global level from the traditional
research and educational infrastructures that have been designed to serve the
pragmatic needs of the agricultural community.
The area of hydrogeosciences has emerged as a compelling discipline
given its links to a broad area of environmental, ecological, geological,
agricultural, and natural resource issues. This area has substantial potential
as an area of tremendous growth. Hydrogeoscientists are encountering a new
intellectual paradigm that emphasizes connections between the hydrosphere
and other components of the earth system (Fig. 1). For example, Entekhabi
et al. (1999) proposed an agenda for land surface hydrology research in the
21st century as hydrological research at the interface between the atmosphere and land surface is undergoing a dramatic change in focus, driven
by new societal priorities, emerging technologies, and better understanding
of the earth system. They also called for the second International Hydrological Decade to open the debate for more comprehensive prioritization of
science and application activities in the hydrological sciences. Another example is the U.S. Department of Energy’s formulation of the National Roadmap
for Vadose Zone Science and Technology, which calls for a national science
program to implement, fund, and coordinate interdisciplinary research into
vadose zone fluid flow and contaminant transport and fate (Stephens et al.,
2002). It is becoming more and more recognized that to understand fully
the distribution of contaminants in the surface and subsurface environments,
one must consider the movement of water and chemicals in the vadose zone,
especially the flow and transport processes occurring in structured soils and
fractured rocks that are of vital concern in nuclear waste disposal and toxic
chemical sites. The third example is the emerging ecohydrology that addresses
the interface between the hydrosphere and the biosphere and that examines the
mutual interaction between the hydrological cycle and ecosystems (Eagleson,
2002; Rodrı́guez-Iturbe, 2000). Soil moisture, a key variable in ecohydrology,
modulates the complex dynamics of the climate–soil–water–vegetation system
and controls the spatial and temporal patterns of vegetation. We believe that
hydropedology is a timely addition to this exciting era of interdisciplinary and
systems approaches to study the pedosphere, the hydrological cycle, the
earth’s critical zone, and the earth system.
H. L. thanks Dr. Donald Sparks for his invitation to contribute this
manuscript to Advances in Agronomy. H. L.’s contribution to this work
was partially supported by a grant from the USDA-CSREES National
Research Initiative (#2002-35102-12547).
Abbott, M. B. (1991). ‘‘Hydroinformatics—Information Technology and the Aquatic Environment’’. Avebury Technical, Aldershot, UK.
Abbott, M. B., and Refsgaard, J. C. (1996). ‘‘Distributed Hydrological Modelling’’. Kluwer,
Dordrecht, The Netherlands.
Addiscott, T. M. (1977). A simple computer model for leaching in structured soils. J. Soil Sci.
28, 554–563.
Ahuja, L. R., and Hebson, C. (1992). ‘‘Root Zone Water Quality Model’’. GPSR Technical
Report No. 2, USDA-ARS, Fort Collins, CO.
American Association for the Advancement of Science (AAAS) Council (2001). Report of the
2001 AAAS Council meeting. Science 292, 2509.
Amundson, R., Harden, J., and Singer, M. (Eds.) (1994). In ‘‘Factors of Soil Formation: A Fiftieth Anniversary Retrospective.’’ Special Publ. #33, Soil Sci. Sco. Am.
Madison, WI.
Anderson, A., McBratney, A. B., and Crawford, J. W. (1998). Applications of fractals to soil
studies. Adv. Agron. 63, 1–76.
Anderson, J. L., and Bouma, J. (1973). Relationships between hydraulic conductivity and
morphometric data of an argillic horizon. Soil Sci. Soc. Am. Proc. 37, 408–413.
Anderson, M. G., and Burt, T. P. (Eds.) (1990). ‘‘Studies in Hillslope Hydrology’’. Wiley, New
Anderson, M. P. (1997). Characterization of geological heterogeneity. In ‘‘Subsurface Flow and
Transport’’ (G. Dagan and S. P. Neuman, Eds.), pp. 23–43. Cambridge Univ. Press,
Cambridge, UK.
Anderson, S. H., and Hopmans, J. W. (Eds.) (1994). ‘‘Tomography of Soil-Water-Root
Processes’’. SSSA Special Pub. #36. Soil Sci. Soc. Am., Inc., Madison, WI.
Arnold, R. W., and Wilding, L. P. (1991). The need to quantify spatial variability. In ‘‘Spatial
Variabilities of Soils and Landforms’’ (M. J. Mausbach and L. P. Wilding, Eds.), pp. 1–8.
SSSA Special Pub. #28. Soil Sci. Soc. Am., Inc., Madison, WI.
Ashby, W. C. (1987). Forests. In ‘‘Restoration Ecology: A Synthetic Approach to Ecological
Research’’ (W. R. Jordan III, M. E. Gilpin, and J. D. Aber, Eds.), pp. 89–109. Cambridge
Univ. Press, Cambridge, UK.
Bartoli, F., Dutartre, P., Gomendy, V., Niquet, S., Dubuit, M., and Vivier, H. (1998). Fractals
and soils structure. In ‘‘Fractals in Soil Science’’ (P. Baveye, J.-Y. Parlange, and B. A.
Stewart, Eds.), Advances in Soil Science, pp. 203–232. CRC Press, Boca Raton, FL.
Batjes, N. H. (1996). Development of a world data set of soil water retention properties using
pedotransfer rules. Geoderma. 71, 31–52.
Baveye, P., and Boast, C. W. (1998). Concepts of ‘‘fractals’’ in soil science: Demixing apples and
oranges. Soil Sci. Soc. Am. J. 62, 1469.
Baveye, P., and Boast, C. W. (1999). Physical scales and spatial predictability of transport
processes in the environment. In ‘‘Assessment of Non-Point Source Pollution in the Vadose
Zone’’ (D. L. Corwin, K. Loague, and Ellsworth, Eds.), pp. 261–280. Geophysical
Monograph 108. American Geophysical Union, Washington, DC.
Baveye, P., Parlange, J.-Y., and Stewart, B. A. (1999).‘‘Fractals in Soil Science’’. Advances in
Soil Science. CRC Press, Boca Raton, FL.
Bear, J. (1972). ‘‘Dynamics of Fluids in Porous Media’’. Elsevier, New York, NY.
Beckett, P. H. T. (1987). Spatial variability of soil K—Status. In ‘‘Proceedings of the 20th
Colloquium’’, pp. 357–380. International Potash Institute, Bern.
Betson, R. P. (1964). What is watershed runoff? J. Geophys. Res. 69, 1541–1552.
Beven, K. (1986). Runoff production and flood frequency in catchment of order n: An alternative approach. In ‘‘Scale Problem in Hydrology’’, (V. K. Gupta, I. Rodriguez-Iturbe, and
E. F. Wood, Eds.), pp. 107–131. Reidel, Dordrecht.
Beven, K. (1997a). TOPMODEL: A critique. Hydro. Process 11, 1069–1085.
Beven, K. (Ed.) (1997b). In ‘‘Distributed Modelling in Hydrology: Application of the TOPMODEL concept’’. Wiley, Chichester.
Beven, K. (2002). Towards an alternative blueprint for a physically based digitally simulated
hydrologic response modeling system. Hydrol. Process 16, 189–206.
Beven, K., and Feyen, J. (2002). The future of distributed modeling. Hydro. Process 16,
Beven, K., and Freer, J. (2001). A dynamic TOPMODEL. Hydro. Process 15, 1993–2011.
Beven, K., and Kirkby, M. J. (1979). A physically based variable contributing area model of
basin hydrology. Hydro. Sci. Bull. 24, 43–69.
Blöschl, G., and Sivapalan, M. (1995). Scale issues in hydrological modelling—A review. Hydro.
Process 9, 251–290.
Blöschl, G., and Grayson, R. (2000). Spatial observations and interpolation. In ‘‘Spatial Pattern
in Catchment Hydrology: Observations and Modeling’’ (R. Grayson and G. Blöschl, Eds.),
pp. 17–50. Cambridge University Press, Cambridge, UK.
Boorman, D. B., Hollis, J. M., and Lilly, A. (1995). ‘‘Hydrology of Soil Types:
A Hydrologically-Based Classification of the Soils of the United Kingdom’’. Institute of
Hydrology Report No. 126, Institute of Hydrology, Wallingford.
Bouma, J. (1984). Using soil morphology to develop measurement methods and simulation
techniques for water movement in heavy clay soils. In ‘‘Proceedings of the ISSS Symposium
on Water and Solute Movement in Heavy Clay Soils’’ (J. Bouma and Raats P. A. C., Eds.),
pp. 298–316. ILRI Pub. 37, Wageningen.
Bouma, J. (1989). Using soil survey data for quantitative land evaluation. Adv. Soil Sci. 9, 177–213.
Bouma, J. (1990). Using morphometric expressions for macropores to improve soil physical
analyses of field soils. Geoderma 46, 3–13.
Bouma, J. (1992). Effects of soil structure, tillage, and aggregation upon soil hydraulic properties. In ‘‘Interacting Processes in Soil Science’’ (R. J. Wagenet, P. Bavege, and B. A.
Stewart, Eds.), pp. 1–36. Lewis Publishers, Boca Raton, FL.
Bouma, J. (1997). Long-term characterization: Monitoring and modelling. In ‘‘Methods for the
Assessment of Soil Degradation’’ (R. Lal, W. H. Blum, C. Valentin, and B. A. Stewart,
Eds.), pp. 339–358. Advances in Soil Science.
Bouma, J. (1999). The strength of the past is the key to the future. In ‘‘Proceedings of the
International Symposium on Soil Resources: Their Inventory, Analysis, and Interpretation
for Use in the 21st Century’’. Minneapolis, MN.
Bouma, J. (2005). Hydropedology as a powerful tool to environmental policy research. Geoderma, in press.
Bouma, J., and Anderson, J. L. (1973). Relationships between soil structure characteristics and
hydraulic conductivity. In ‘‘Field Soil Water Regime’’ (R. R. Bruce, K. Flach, and H. M.
Taylor, Eds.), pp. 77–105. SSSA special publication No. 5. Madison, WI.
Bouma, J., and Hole, F. D. (1971). Soil morphology and hydraulic conductivity of adjacent
virgin and cultivated pedons at two sites: A Typic Argiudoll (silt loam) and a Typic
Eutrochrept (clay). Soil Sci. Soc. Am. Proc. 35, 316–319.
Bouma, J., and van Lanen H. A. J. (1987). Transfer functions and threshold values: From soil
characteristics to land qualities. In ‘‘Proceedings of the ISSS/SSSA Workshop on Quantified Land Evaluation’’ (K. J. Beek et al., Eds.), pp. 106–110. Washington, DC. April
27–May 2, 1986. Int. Inst. for Aerospace Surv. and Earth Sci. Publ. No. 6. ITC Publ.
Enschede, The Netherlands.
Bouma, J., Dekker, L. W., and Haans, J. C. F. M. (1980). Measurement of depth to water table
in a heavy clay soil. Soil Sci. 130, 264–270.
Bouma, J., Droogers, P., and Peters, P. (1999). Defining the ‘‘ideal’’ soil structure in surface soil
of a Typic Fluvaquent in the Netherlands. Soil Sci. Soc. Am. J. 63, 343–348.
Bouma, J., Jongerius, A., and Schoonderbeek, D. (1979). Calculation of saturated hydraulic
conductivity of some pedal clay soils using micromorphological data. Soil Sci. Soc. Am. J.
43, 261–264.
Bouma, J., Jongmans, A. G., Stein, A., and Peek, G. (1989). Characterizing spatially variable
hydraulic properties of a boulder clay deposit. Geoderma 45, 19–31.
Boyer Commission (on Educating Undergraduates in the Research University) (1998). ‘‘Reinventing Undergraduate Education—A Blueprint for America’s Research Universities’’.
Breeuwsma, A., Wösten, J. H. M., Vleeshouwer, J. J., van Slobbe, A. M., and Bouma, J. (1986).
Derivation of land qualities to assess environmental problems from soil survey. Soil Sci.
Soc. Am. J. 50, 186–190.
Brewer, R. (1976). ‘‘Fabric and Mineral Analysis of Soils’’. Robert E. Krieger Publishing
Company, Huntington, NY.
Broderson, W. D. (2003). ‘‘From the Surface Down: An Introduction to Soil Surveys for
Agronomic Use’’. USDA-NRCS, Washington, DC.
Brown, R. B., and Huddleston, J. H. (1991). Presentation of statistical data on map units to the
user. In ‘‘Spatial Variabilities of Soils and Landforms’’ (M. J. Mausbach and L. P. Wilding,
Eds.), pp. 127–147. SSSA Special Pub. #28. Soil Sci. Soc. Am., Inc., Madison, WI.
Bryant, R. B., and Olsen, C. G. (1987). Soil genesis: Opportunities and new directions for
research. In ‘‘Future Developments in Soil Science Research’’ (L. L. Boersma, Ed.),
pp. 301–311. SSSA, Madison, WI.
Buol, S. W., Southard, R. J., Graham, R. C., and McDaniel, P. A. (2001). ‘‘Soil Genesis and
Classification’’, 5th ed. Iowa State University Press, Ames. IA.
Burrough, P. A. (1981). The fractal dimensions of landscapes and other environmental data.
Nature 294, 240–242.
Burrough, P. A. (1983a). Multiscale sources of spatial variation in soil. I. The application of
fractal concepts to nested levels of soil variation. J. Soil Sci. 34, 577–597.
Burrough, P. A. (1983b). Multiscale sources of spatial variation in soil. II. A non-Brownian
fractal model and its application to soil survey. J. Soil Sci. 34, 599–620.
Burrough, P. A. (1989). Fuzzy mathematical methods for soil survey and land evaluation. J. Soil
Sci. 40, 477–492.
Burrough, P. A. (1993). Soil variability: A late 20th century view. Soils Fert. 56, 529–562.
Burt, T. P., and Butcher, D. P. (1985). A basic hillslope runoff model. J. Soil Sci. 36, 469–486.
Butler, B. E. (1980). ‘‘Soil Classification for Soil Survey’’. Clarendon Press, Oxford, UK.
Chadwick, O. A., Brimhall, G. H., and Hendricks, D. M. (1990). From a black to gray box—A
mass balance interpretation of pedogenesis. Geomorphology 3, 369–390.
Chesworth, W. (1973a). The parent rock effect in the genesis of soil. Geoderma 10, 215–225.
Chesworth, W. (1973b). The residua system of chemical weathering: A model for the chemical
breakdown of silicate rocks at the surface of the earth. J. Soil Sci. 24, 69–81.
Ciolkosz, E. J., Waltman, W. J., Simpson, T. W., and Dobos, R. R. (1989). Distribution and
genesis of soils of the northeastern United States. Geomorphology 2, 285–302.
Clarke, K. C., Parks, B. O., and Crane, M. P. (2002). ‘‘Geographic Information Systems and
Environmental Modeling’’. Prentice Hall, NJ.
Coen, G. M., and Wang, C. (1989). Estimating vertical saturated hydraulic conductivity from
soil morphology in Alberta. Can. J. Soil Sci. 69, 1–16.
Corwin, D. L., Loague, K., and Ellsworth, T. R. (1998). Assessing nonpoint source pollution in
the vadose zone. Eos. 79, 219–220.
Corwin, D. L., Loague, K., and Ellsworth, T. R. (1999). Advanced information technologies for
assessing nonpoint source pollution in the vadose zone: Conference overview. J. Environ.
Qual. 28, 357–365.
Corwin, D. L., Waggoner, B. L., and Rhoades, J. D. (1991). A functional model of solute
transport that accounts for bypass. J. Environ. Qual. 20, 647–658.
Crawford, J. W., Baveye, P., Grindrod, P., and Rappoldt, C. (1999). Application of fractals
to soil properties, landscape patterns, and solute transport in porous media. In ‘‘Assessment
of Non-Point Source Pollution in the Vadose Zone’’ (D. Corwin, K. Loague, and T.
Ellsworth, Eds.), pp. 151–164. American Geophysical Union, Washington, DC.
Cushman, J. H. (1990). An introduction to hierarchical porous media. In ‘‘Dynamics of Fluids
in Hierarchical Porous Media’’ (J. H. Cushman, Ed.), pp. 1–6. Academic Press, London.
Dane, J., and Clark, T. (Eds.) (2002). In ‘‘Methods of Soil Analysis: Part 4, Physical Methods’’.
SSSA Book Series No. 5 Soil Sci. Soc. Am., Inc., Madison, WI.
Dekker, L. W., and Ritsema, C. J. (2003). Wetting patterns in water repellent Dutch soils.
In ‘‘Soil Water Repellency’’ (C. J. Ritsema and L. W. Dekker, Eds.), pp. 151–166. Elsevier,
Dekker, L. W., Wosten, J. H. M., and Bouma, J. (1984). Characterizing the soil moisture regime
of a Typic Haplohumod. Geoderma 34, 27–42.
Deurer, M., Green, S. R., Clothier, B. E., Böttcher, J., and Duijnisveld, W. H. M. (2003).
Drainage networks in soils: A concept to describe bypass-flow pathways. J. Hydro. 272,
Doering, E. J., and Benz, L. C. (1972). Pumping an artesian source for water table control.
J. Irrig. Drain. Div. ASCE 98, 275–287.
Dokuchaev, V. V. (1893). ‘‘The Russian Steppes and Study of Soil in Russia, Its Past and
Present’’ (Translated by J. M. Crawford). Ministry of Crown Domains, St. Petersburg,
Domenico, P. A., and Schwartz, F. W. (1998). ‘‘Physical and Chemical Hydrogeology’’, 2nd ed.
Wiley, New York.
Droogers, P., and Bouma, J. (1997). Soil survey input in exploratory modelling of sustainable
land management practices. Soil Sci. Soc. Am. J. 61, 1704–1710.
Droogers, P., van der Meer, F. B. W., and Bouma, J. (1997). Water accessibility to plant roots in
different soil structures occurring in the same soil type. Plant Soil 188, 83–91.
Duffy, C., Wierenga, P. J., and Kselik, R. A. (1981). Variations in infiltration rate based on soil
survey information and field measurements. New Mexico Agric. Exp. Sta. Bull. 680, 1–39.
Dunne, T., and Black, R. D. (1970a). An experimental investigation of runoff production in
permeable soils. Water Resour. Res. 6, 478–490.
Dunne, T., and Black, R. D. (1970b). Partial areal contributions to storm runoff in a small New
England watershed. Water Resour. Res. 6, 1296–1311.
Eagleson, P. S. (2002). ‘‘Ecohydrology: Darwinian Expression of Vegetation Form and
Function’’. Cambridge University Press, Cambridge, UK.
Entekhabi, D., Asrar, G. R., Betts, A. K., Beven, K. J., Bras, R. L., Duffy, C. J., Dunne, T., Koster,
R. D., Lettenmaier, D. P., McLaughlin, D. B., Shuttleworth, W. J., van Genuchten, M. Th.,
Wei, M.-Y., and Wood, E. F. (1999). An agenda for land surface hydrology research and a
call for the second international hydrological decade. Bull. Am. Metero. Soc. 80, 2043–2058.
Evans, C. V., and Franzmeier, D. P. (1986). Saturation, aeration, and color patterns in a
toposequence of soils in North-central Indiana. Soil Sci. Soc. Am. J. 50, 975–980.
Famiglietti, J. S., and Wood, E. F. (1994). Multiscale modeling of spatially variable water and
energy balance processes. Water Resour. Res. 30, 3061–3078.
Fitzpatrick, R. W., Naidu, R., and Self, P. G. (1992). Iron deposits and microorganisms
occurring in saline sulfidic soils with altered soil water regime in the Mt. Lofty Ranges,
South Australia. Catena Suppl. 21, 263–286.
Flury, M., Fluhler, H., Jury, W. A., and Leuenberger, L. (1994). Susceptibility of soils to
preferential flow of water: A field study. Water Resour. Res. 30, 1945–1954.
Foussereau, X., Hornsby, A. G., and Brown, R. B. (1993). Accounting for variability within
map units when linking a pesticide fate model to soil survey. Geoderma 60, 257–276.
Franzmeier, D. P. (1991). Estimation of hydraulic conductivity from effective porosity data for
some Indiana soils. Soil Sci. Soc. Am. J. 55, 1801–1803.
Franzmeier, D. P., Yahner, J. E., Steinhardt, G. C., and Sinclair, Jr., H. R. (1983). Color
patterns and water-table levels in some Indiana soils. Soil Sci. Soc. Am. J. 47, 1196–1202.
Fritsch, E., and Fitzpatrick, R. W. (1994). Interpretation of soil features produced by ancient
and modern processes in degraded landscapes. I. A new method for constructing conceptual soil-water-landscape models. Aust. J. Soil Res. 32, 889–907.
Fu, K. S. (1982). ‘‘Applications of Pattern Recognition’’. CRC Press, Boca Raton, FL.
Galusky, L. P. F., Jr. (1997). Toward the development of quantitative soil morphological
indicators of water table behavior. Ph. D. dissertation. University of Maryland.
Galusky, L. P., Rabenhorst, M. C., and Hill, R. L. (1998). Toward the development of
quantitative soil morphological indicators of water table behavior. In ‘‘Quantifying Soil
Hydromorphology’’ (M. C. Rabenhorst, J. C. Bell, and P. A. McDaniel, Eds.), pp. 77–93.
SSSA Special Pub. #54, Soil Sci. Soc. Am., Inc., Madison, WI.
Gerke, H. H., and van Genuchten, M. Th. (1993). A dual-porosity model for simulating the
preferential movement of water and solutes in structured porous media. Water Resour. Res.
29, 305–319.
German, P. F., and Beven, K. (1985). Kinematic wave approximation to infiltration into soils
with sorbing macropores. Water Resour. Res. 21, 990–996.
German, P. F., and Beven, K. (1986). A distribution function-approach to water-flow in soil
macropores based on kinematic wave theory. Journal of Hydrology 83, 173–183.
Gerrard, A. J. (1981). ‘‘Soils and Landforms—An Integration of Geomorphology and Pedology’’. George Allen & Unwin, London.
Goodchild, M. F., Steyaert, L. T., Parks, B. O., Johnson, C., Maidment, D., Crane, M., and
Glendinning, S. (Eds.) (1996). In ‘‘GIS and Environmental Modeling: Progress and
Research Issues’’. GIS World, Inc., Fort Collins, CO.
Gotway, C. A., and Young, L. J. (2002). Combining incomplete spatial data. J. Am. Stat. Assoc.
97, 632–648.
Grayson, R., and Blöschl, G. (Eds.) (2000). In ‘‘Spatial Pattern in Catchment Hydrology:
Observations and Modeling’’. Cambridge University Press, Cambridge, UK.
Grayson, R. B., and Western, A. W. (1998). Towards areal estimation of soil water content
from point measurements: Time and space stability of mean response. J. Hydro. 207, 68–82.
Grayson, R., Blöschl, G., Barling, R. D., and Moore, I. D. (1993). Process, scale and constraints
to hydrological modeling in GIS. In ‘‘Applications of Geographic Information Systems in
Hydrology and Water Resources’’ (K. Kovar and H. P. Nachtnebel, Eds.), pp. 83–92. Proc.
Vienna Symp. IAHS Pub. #211.
Grayson, R., Blöschl, G., Western, A. W., and McMahon, T. A. (2002). Advances in the use of
observed spatial patterns of catchment hydrological response. Adv. Water Res. 25,
Grayson, R., Western, A. W., Chiew, F. H. S., and Blöschl, G. (1997). Preferred states in spatial
soil moisture patterns: Local and nonlocal controls. Water Resour. Res. 33, 2897–2908.
Grossman, R. B., Harms, D. S., Seybold, C. A., and Herrick, J. E. (2001). Coupling usedependent and use-invariant data for soil quality evaluation in the United States. J. Soil
Water Conserv. 56, 63–68.
Gwo, J. P., Jardine, P. M., Wilson, G. V., and Yeh, G. T. (1995). A multiple-pore-region
concept to modeling mass transfer in subsurface media. J. Hydro. 164, 217–237.
Haigh, M. J. (1987). The holon: Hierarchy theory and landscape research. Catena 10, 181–192.
Hall, G. F., and Olson, C. G. (1991). Predicting variability of soils from landscape models.
In ‘‘Spatial Variabilities of Soils and Landforms’’ (M. J. Mausbach and L. P. Wilding,
Eds.), pp. 9–24. Special Pub. #28. Soil Sci. Soc. Am., Madison, WI.
Hamlen, C. J., and Kachanoski, R. G. (1992). Field solute transport across a soil horizon
boundary. Soil Sci. Soc. Am. J. 56, 1716–1720.
Hendrickx, J. M. H., and Flury, M. (2001). Uniform and preferential flow mechanisms in the
vadose zone. In ‘‘Conceptual Models of Flow and Transport in the Fractured Vadose
Zone’’, pp. 149–187. National Academy Press, Washington, D.C.
Henry, J. L., Bullock, P. R., Hogg, T. J., and Luba, L. D. (1985). Groundwater discharge from
glacial and bedrock aquifers as a soil salinization factor in Saskatchewan. Can. J. Soil Sci.
65, 749–768.
Heuvelink, G. B. M., and Pebesma, E. J. (1999). Spatial aggregation and soil process modeling.
Geoderma. 89, 47–65.
Heuvelink, G. B. M., and Webster, R. (2001). Modelling soil variation: Past, present, and
future. Geoderma. 100, 269–301.
Hilgard, E. W. (1921). ‘‘Soils: Their Formation, Properties, Composition, and Plant Growth in
the Humid and Arid Regions’’. Macmillan Co., London.
Hodgson, J. M. (1997). ‘‘Soil Survey Field Handbook: Describing and Sampling Soil Profiles’’,
3rd ed. Soil Survey Technical Monograph #5, Soil Survey and Land Research Centre,
Silsoe, UK.
Hole, F. D. (1976). Soils of Wisconsin. Wis. Geol. Nat. Hist. Surv. Bull. 87, Soils Ser. 62,
University of Wisconsin Press, Madison, WI.
Hoosbeek, M. R., and Bryant, R. B. (1992). Towards the quantitative modeling of pedogenesis—A review. Geoderma 55, 183–210.
Hoosbeek, M. R., van Es, H. M., and Stein, A. (Eds.) (1998). In Special issue—Modeling spatial
and temporal variability as a function of scale Geoderma 85, 111–254.
Hoogmoed, W. B., and Bouma, J. (1980). A simulation model for predicting infiltration into
cracked clay soil. Soil Sci. Soc. Amer. J. 44, 458–461.
Hornberger, G. M., and Boyer, E. W. (1995). Recent advances in watershed modeling. Rev.
Geophys. 33, 949–957.
Horton, R. E. (1945). Erosional development of streams and their drainage basins: Hydrophysical approach to quantative morphology. Bull. Geol. Soc. Am. 56, 275–370.
Houser, P., Goodrich, D., and Syed, K. (2000). Runoff, precipitation, and soil moisture at
Walnut Gulch. In ‘‘Spatial Pattern in Catchment Hydrology: Observations and Modeling’’
(R. Grayson and G. Bloschl, Eds.), pp. 125–157. Cambridge University Press, Cambridge, UK.
Hugget, R. J. (1975). Soil landscape systems: A model of soil genesis. Geoderma. 13, 1–22.
Ibanez, J. J., De-Alba, S., Bermudez, F. F., and Garcia-Alvarez, A. (1995). Pedodiversity:
Concepts and measures. Catena 24, 215–232.
Ibanez, J. J., De-Alba, S., Lobo, A., and Zucarello, V. (1998). Pedodiversity and global soil
patterns at coarse scales (with discussion). Geoderma 83, 171–192.
Jackson, T. J., and Le Vine, D. E. (1996). Mapping surface soil moisture using an aircraft-based
passive instrument: Algorithm and example. J. Hydro. 184, 85–99.
Jenkinson, B. J., Franzmeier, D. P., and Lynn, W. C. (2002). Soil hydrology on an end moraine
and a dissected till plain in west-central Indiana. Soil Sci. Soc. Am. J. 66, 1367–1376.
Jenny, H. (1941). ‘‘Factors of Soil Formation—A System of Quantitative Pedology’’. McGrawHill, NY.
Jenny, H. (1961). Derivation of state factor equations of soils and ecosystems. Soil Sci. Soc. Am.
J. 25, 385–388.
Jenny, H. (1980). ‘‘The Soil Resource, Origin, and Behavior’’. Springer-Verlag, NY.
Johnson, D. L., and Hole, F. D. (1994). Soil formation theory: A summary of its principal
impacts on geography, geomorphology, soil-geomorphology, quaternary geology, and
paleopedology. In ‘‘Factors of Soil Formation: A Fiftieth Anniversary Retrospective’’ (R.
Amundson, J. Harden, and M. Singer, Eds.), pp. 111–126. Soil Sci. Soc. Am., Special Publ.
#33, Madison, WI.
Johnson, D. L., and Watson-Stegner, D. (1987). Evolution model of pedogenesis. Soil Sci. 143,
Johnson, D. L., Keller, E. A., and Rockwell, T. K. (1990). Dynamic pedogenesis: New views of
some key soil concepts and a model for interpreting quaternary soils. Quat. Res. 33,
Jury, W. A. (1982). Simulation of solute transport using a transfer function mode. Water
Resour. Res. 18, 363–368.
Jury, W. A. (1986). Spatial variability of soil properties. In ‘‘Vadose Zone Modeling of Organic
Pollutants’’ (S. C. Hern and S. M. Melancon, Eds.), pp. 245–269. Lewis Publishers,
Chelsea, MI.
Jury, W. A., Sposito, G., and White, R. E. (1986). A transfer function model of solute
movement through soil. I. Fundamental concepts. Water Resour. Res. 22, 243–247.
Jury, W. A. (1999). Present directions and future research in vadose zone hydrology. In ‘‘Vadose
Zone Hydrology—Cutting across Disciplines’’ (M. B. Parlange and J. W. Hopmans, Eds.),
pp. 433–441. Oxford University Press, NY.
Jury, W. A., and Roth, K. (1990). ‘‘Transfer Function and Solute Movement through Soil:
Theory and Applications’’. Birkhäuser, Basel.
Kachanoski, R. G., and de Jong, E. (1988). Scale dependence and the temporal persistence of
spatial patterns of soil water storage. Water Resour. Res. 24, 85–91.
Khan, F. A., and Fenton, T. E. (1994). Saturated zones and soil morphology in a Mollisol
catena of Central Iowa. Soil Sci. Soc. Am. J. 58, 1457–1464.
King, J. J., and Franzmeier, D. P. (1981). Estimation of saturated hydraulic conductivity from
soil morphological and genetic information. Soil Sci. Soc. Am. J. 45, 1153–1156.
Kirkby, M. J. (1975). Hydrograph modelling strategies.. In ‘‘Process in Physical and Human
Geography’’ (R. Peel, M. Chisholm, and P. Haggett, Eds.), pp. 69–90. Heinemann, London.
Kirkby, M. J. (Ed.) (1978). In ‘‘Hillslope Hydrology’’. Wiley and Sons, New York.
Kleiss, H. J. (1970). Hillslope sedimentation and soil formation in northeastern Iowa. Soil Sci.
Soc. Am. J. 34, 287–290.
Klemes, V. (1986). Dilettantism in hydrology: Transition or destiny? Water Resour. Res. 22,
Kutı́lek, M. (1966). ‘‘Soil Science in Water Management’’ (In Czech: ‘‘Vodohospodarska
Pedologie’’). SNTL, State Publisher of Technical Literature, Prague.
Kutı́lek, M. (1978). ‘‘Soil Science in Water Management’’, 2nd ed. (In Czech: ‘‘Vodohospodarska Pedologie’’). SNTL, State Publisher of Technical Literature, Prague.
Kutı́lek, M. (2004). Soil hydraulic properties as related to soil structure. Soil Tillage Res.,
in press.
Kutı́lek, M., and Nielsen, D. R. (1994). ‘‘Soil Hydrology’’. Catena Verlag, Cremlingen-Destedt,
Lammers, D. A., and Johnson, M. G. (1991). Soil mapping concepts for environmental
assessment. In ‘‘Spatial Variabilities of Soils and Landforms’’ (M. J. Mausbach and L. P.
Wilding, Eds.), pp. 149–160. SSSA Special Pub. #28. Soil Sci. Soc. Am., Inc., Madison, WI.
Levine, E. R., and Ciolkosz, E. J. (1986). A computer simulation model for soil genesis
applications. Soil Sci. Soc. Am. J. 50, 661–667.
Lilly, A. and Lin, H. S. (2005). Using soil morphological attributes and soil structure in
pedotransfer functions. In ‘‘Development of Pedotransfer Functions in Soil Hydrology’’
(Y. Pachepsky and W. Rawls, Eds.), Elsevier. in press.
Lilly, A., and Matthews, K. B. (1994). A soil wetness class map for Scotland: New assessments
of soil and climate data for land evaluation. Geoforum 25, 371–379.
Lilly, A., Ball, B. C., McTaggart, I. P., and Horne, P. L. (2003). Spatial and temporal scaling of
nitrous oxide emissions from the field to the regional scale in Scotland. Nutr. Cycl.
Agroecosyst. 66, 241–257.
Lilly, A., Boorman, D. B., and Hollis, J. M. (1998). The development of a hydrological
classification of UK soils and the inherent scale changes. Nutr. Cycl. Agroecosyst. 50,
Lin, H. S. (2003). Hydropedology: Bridging disciplines, scales, and data. Vadose Zone J. 2, 1–11.
Lin, H. S., and Rathbun, S. (2003). Hierarchical frameworks for multiscale bridging in hydropedology. In ‘‘Scaling Methods in Soil Physics’’ (Y. Pachepsky, D. Radcliffe, and Selim,
Eds.), pp. 353–371. CRC Press, Boca Raton, FL.
Lin, H. S., McInnes, K. J., Wilding, L. P., and Hallmark, C. T. (1996). Effective porosity and
flow rate with infiltration at low tensions into a well-structured subsoil. Trans. ASAE 39,
Lin, H. S., McInnes, K. J., Wilding, L. P., and Hallmark, C. T. (1998). Macroporosity and initial
moisture effects on infiltration rates in Vertisoles and vertic intergrades. Soil Sci. 163, 2–8.
Lin, H. S., McInnes, K. J., Wilding, L. P., and Hallmark, C. T. (1999a). Effects of soil
morphology on hydraulic properties: I. Quantification of soil morphology. Soil Sci. Soc.
Am. J. 63, 948–954.
Lin, H. S., McInnes, K. J., Wilding, L. P., and Hallmark, C. T. (1999b). Effects of soil
morphology on hydraulic properties: II. Hydraulic pedotransfer functions. Soil Sci. Soc.
Am. J. 63, 955–961.
Lin, H. S., Wheeler, D., Bell, J., and Wilding, L. (2004). Assessment of soil spatial variability at
multiple scales. Ecolog. Model. in press.
Lissey, A. (1971). Depression-focused transient groundwater flow patterns in Manitoba. Geol.
Assoc. Can. Spec. Paper 9, 333–341.
Loague, K., and Green, R. E. (1990). Comments on ‘Mapping Pesticide Contamination Potential’ by M. Akram Khan and Tung Liang. Environ. Management 14, 149–150.
Magurran, A. E. (1988). ‘‘Ecological Diversity and Its Measurement’’. Croom Helm, London.
Mandelbrot, B. B. (1982). ‘‘The Fractal Geometry of Nature’’. W. H. Freeman, New York.
Mattikalli, N. M., Engman, E. T., Jackson, T. J., and Ahuja, L. R. (1998). Microwave remote
sensing of temporal variations of brightness temperature and near-surface soil water
content during a watershed-scale field experiment, and its application to the estimation of
soil physical properties. Water Resour. Res. 34, 2289–2299.
Mausbach, M. J., and Richardson, J. L. (1994). Biogeochemistry processes in hydric soil
formation. Curr. Top. Wetland Biogeochem. 1, 68–127.
McBratney, A. B. (1986). ‘‘Introduction to Pedometrics: A Course of Lectures’’. CSIRO
Australia, Division of Soils Technical Memo. 53/1986, Private Bag 2, Glen Osmond, S.A.
McBratney, A. B. (1998). Some considerations on methods for spatially aggregating and
disaggregating soil information. Nutr. Cycl. Agroecosys. 50, 51–62.
McBratney, A. B., Minasny, B., Cattle, S. R., and Vervoort, R. W. (2002). From pedotransfer
functions to soil inference systems. Geoderma 109, 41–73.
McKeague, J. A., Wang, C., and Coen, G. M. (1986). ‘‘Describing and Interpreting
the Macrostructure of Mineral Soils—A Preliminary Report’’. Land Resource Research
Institute, Ottawa, Ontario, Research Branch Agricultural Canada.
McKeague, J. A., Wang, C., and Topp, G. C. (1982). Estimated saturated hydraulic conductivity from soil morphology. Soil Sci. Soc. Am. J. 46, 1239–1244.
McKenzie, N. J., and Jacquier, D. W. (1997). Improving the field estimation of saturated
hydraulic conductivity in soil survey. Aust. J. Soil Res. 35, 803–825.
McSweeney, K., Gessler, P. E., Slater, B., Hammer, R. D., Bell, J., and Petersen, G. W. (1994).
Towards a new framework for modeling the soil-landscape continuum. In ‘‘Factors of Soil
Formation: A Fiftieth Anniversary Retrospective’’ (R. Amundson, J. Harden, and M.
Singer, Eds.), pp. 127–145. Special Publ. #33, Soil Sci. Soc. Am., Madison, WI.
Merchant, J. W. (1994). GIS-based groundwater pollution hazard assessment: A critical review
of the DRASTIC model. Photogram. Eng. Remote Sensing 60, 1117–1127.
Milne, G. (1935). Some suggested units of classification and mapping, particularly for East
African soils. Soil Res. 4, 183–198.
Milne, G. (1936a). Normal erosion as a factor in soil profile development. Nature 138, 548–549.
Milne, G. (1936b). ‘‘A Provisional Soil Map of East Africa’’. East Africa Agric. Res. Stn.,
Amani Memoirs, No. 34. Tanganyika Terr.
Minasny, B., and McBratney, A. B. (1999). A rudimentary mechanistic model for soil production and landscape development. Geoderma 90, 3–21.
Minasny, B., and McBratney, A. B. (2001). A rudimentary mechanistic model for soil formation
and landscape development II. A two-dimensional model incorporating chemical
weathering. Geoderma 103, 161–179.
Minasny, B., McBratney, A. B., and Bristow, K. L. (1999). Comparison of different approaches
to the development of pedotransfer functions for water-retention curves. Geoderma 93,
Mohanty, B. P., and Skaggs, T. H. (2001). Spatio-temporal evoluation and time-stable characteristics of soil moisture within remote sensing footprints with varying soil, slope, and
vegetation. Adv. Water Resour. 24, 1051–1067.
Moore, I. D., Gessler, P. E., Nielsen, G. A., and Peterson, G. A. (1993). Soil attribute prediction
using terrain analysis. Soil Sci. Soc. Am. J. 57, 443–452.
Mulla, D. J., and Addiscott, T. M. (1999). Validation approaches for field-, basin-, and regionalscale water quality models. In ‘‘Assessment of Non-Point Source Pollution in the Vadose
Zone’’ (D. Corwin, K. Loague, and T. Ellsworth, Eds.), pp. 63–78. American Geophysical
Union, Washington, DC.
National Research Council (NRC) (1991). ‘‘Opportunities in the Hydrologic Sciences’’. National Academy Press, Washington, DC.
National Research Council (NRC) (1993a). ‘‘Soil and Water Quality: An Agenda for Agriculture’’. National Academy Press, Washington, DC.
National Research Council (NRC) (1993b). ‘‘Ground Water Vulnerability Assessment—
Contamination Potential Under Conditions of Uncertainty’’. National Academy Press,
Washington, DC.
National Research Council (NRC) (1998). ‘‘Developing a Digital National Library for
Undergraduate Science, Mathematics, Engineering, and Technology Education’’. National
Academy Press, Washington, DC.
National Research Council (NRC) (1999). ‘‘New Strategies for America’s Watersheds’’.
National Academy Press, Washington, DC.
National Research Council (NRC) (2001a). ‘‘Basic Research Opportunities in Earth Science’’.
National Academy Press, Washington, DC.
National Research Council (NRC) (2001b). ‘‘Envisioning the Agenda for Water Resources
Research in the Twenty-First Century’’. National Academy Press, Washington, DC.
National Research Council (NRC) (2001c). ‘‘Grand Challenges in Environmental Sciences’’.
National Academy Press, Washington, DC.
Nieber, J. L. (2000). The relation of preferential flow to water quality, and its theoretical and
experimental quantification. In ‘‘Preferential Flow—Water Movement and Chemical
Transport in the Environment’’, pp. 1–10. Proceedings of the 2nd International Symposium. American Society of Agricultural Engineers, St. Joseph, MI.
Nielsen, D. R., and Wendroth, O. (2003). ‘‘Spatial and Temporal Statistics: Sampling Field Soils
and Their Vegetation’’. Catena Verlag, Cremlingen-Destedt.
Nielsen, D. R., Biggar, J. W., and Erh, K. T. (1973). Spatial variability of field-measured soilwater properties. Hilgardia 42, 215–260.
Nielsen, D. R., Hopmans, J. W., and Reichardt, K. (1998). An emerging technology for scaling
field soil-water behavior. In ‘‘Scale Dependence and Scale Invariance in Hydrology’’ (G.
Sposito, Ed.), pp. 136–166. Cambridge University Press, UK.
Nielsen, D. R., Kutilek, M., and Parlange, M. B. (1996). Surface soil water content regimes:
Opportunities in soil science. J. Hydro. 184, 35–55.
Nordt, L. C., Jacob, J. S., and Wilding, L. P. (1991). Quantifying map unit composition for
quality control in soil survey. In ‘‘Spatial Variabilities of Soils and Landforms’’ (M. J.
Mausbach and L. P. Wilding, Eds.), pp. 183–197. SSSA Special Pub. #28, Soil Sci. Soc.
Am., Inc., Madison, WI.
Oades, J. M., and Waters, A. G. (1991). Aggregate hierarchy in soils. Aust. J. Soil Res. 29,
O’Neal, A. M. (1949). Some characteristics significant in evaluating permeability. Soil Sci. 67,
O’Neill, R. V., DeAngelis, D. L., Waide, J. B., and Allen, T. F. H. (1986). ‘‘A Hierarchical
Concept of Ecosystems’’. Monographs in Population Biology 23. Princeton University
Press, Princeton, NJ.
O’Neill, R. V., Johnson, A. R., and King, A. W. (1989). A hierarchical framework for the
analysis of scale. Landscape Ecol. 3, 193–205.
Othmer, H., Diekkrüger, B., and Kutilek, M. (1991). Bimodal porosity and unsaturated hydraulic conductivity. Soil Sci. 152, 139–150.
Pachepsky, Y., and Rawls, W. J. (1999). Accuracy and reliability of pedotransfer functions as
affected by grouping soils. Soil Sci. Soc. Am. J. 63, 1748–1757.
Pachepsky, Y. A., and Rawls, W. (Eds.) (2005). ‘‘Development of Pedotransfer Functions in
Soil Hydrology’’. Elsevier, Amsterdam, The Netherlands. in press.
Pachepsky, Y. A., Crawford, J. C., and Rawls, W. J. (Eds.) (2000). ‘‘Fractals in Soil Science’’.
Developments in Soil Science, 27. Elsevier, Amsterdam, The Netherlands.
Pachepsky, Y. A., Radcliffe, D. E., and Selim, H. M. (Eds.) (2003). ‘‘Scaling Methods in Soil
Physics’’. CRC Press, Boca Raton, FL.
Pachepsky, Y. A., Rawls, W. J., and Timlin, D. J. (1999). The current status of pedotransfer
functions: Their accuracy, reliability, and utility in field- and regional-scale modeling.
In ‘‘Assessment of Non-point Source Pollution in the Vadose Zone’’ (D. Corwin, K. Loague,
and T. Ellsworth, Eds.), pp. 223–234. American Geophysical Union, Washington, DC.
Pennock, D. J., and de Jong, E. (1990). Regional and caternary variations in properties of
Borolls of southern Saskatchewan, Canada. Soil Sci. Soc. Am. J. 54, 1697–1701.
Perret, J., Prasher, S. O., Kantzas, A., and Langford, C. (1999). Three-dimensional quantification of macropore networks in undisturbed soil cores. Soil Sci. Soc. Am. J. 63, 1530–1543.
Perrier, E., Bird, N., and Rieu, M. (1999). Generalizing the fractal model of soil structure: The
pore-solid fractal approach. Geoderma 88, 137–164.
Pulleman, M. M., Bouma, J., van Essen, E. A., and Meijles, E. W. (2000). Soil organic matter
content as a function of different land use history. Soil Sci. Soc. Amer. J. 64, 689–694.
Quisenberry, V. L., Smith, B. R., Phillips, R. E., Scott, H. D., and Nortcliff, S. (1993). A soil
classification system for describing water and chemical transport. Soil Sci. 156, 306–315.
Rabenhorst, M. C., Bell, J. C., and McDaniel, P. A. (1998). ‘‘Quantifying Soil Hydromorphology’’. SSSA Special Pub. #54, Soil Sci. Soc. Am., Inc., Madison, WI.
Rawls, W. J., and Pachepsky, Y. A. (2002a). Soil consistence and structure as predictors of
water retention. Soil Sci. Soc. Am. J. 66, 1115–1126.
Rawls, W. J., and Pachepsky, Y. A. (2002b). Using field topographic descriptors to estimate soil
water retention. Soil Sci. 167, 423–435.
Rawls, W. J., Ahuja, L. R., Brakensiek, D. L., and Shirmohammadi, A. (1993). Infiltration and
soil water movement. In ‘‘Handbook of Hydrology’’ (D. R. Maidment, Ed.). Chapter 5.
McGraw-Hill Inc., New York.
Reuter, R. J., and Bell, J. C. (2003). Hillslope hydrology and soil morphology for a wetland
basin in south-central Minnesota. Soil Sci. Soc. Am. J. 67, 365–372.
Richardson, J. L., and Vepraskas, M. J. (2001). ‘‘Wetland Soils: Genesis, Hydrology, Landscapes, and Classification’’. LEWIS, CRC Press, Boca Raton, FL.
Richardson, J. L., Arndt, J. L., and Montgomery, J. A. (2001). Hydrology of wetland and
related soils. In ‘‘Wetland Soils: Genesis, Hydrology, Landscapes, and Classification’’ (J. L.
Richardson and M. J. Vepraskas, Eds.), pp. 35–84. LEWIS, CRC Press, Boca Raton, FL.
Richardson, J. L., Wilding, L. P., and Daniels, R. B. (1992). Recharge and discharge of ground
water in aquic conditions illustrated with flownet analysis. Geoderma 53, 65–78.
Ridolfi, L., D’Odorico, P., Porporato, A., and Rodriguez-Iturbe, I. (2003). Stochastic soil
moisture dynamics along a hillslope. J. Hydro. 272, 264–275.
Rodrı́guez-Iturbe, I. (2000). Ecohydrology: A hydrologic perspective of climate-soil-vegetation
dynamics. Water Resour. Res. 36, 3–9.
Rodrı́guez-Iturbe, I., and Rinaldo, A. (1997). ‘‘Fractal River Basins: Chance and SelfOrganization’’. Cambridge Univ. Press, Cambridge, UK.
Root, T. L., and Schneider, S. H. (1995). Ecology and climate: Research strategies and implications. Science 269, 334–341.
Ross, P. J., and Smettem, K. R. (2000). A simple treatment of physical nonequilibrium water
flow in soils. Soil Sci. Soc. Am. J. 64, 1926–1930.
Ruhe, R. V. (1969). ‘‘Quaternary Landscapes of Iowa’’. Iowa State Univ. Press.
Ruhe, R. V. (1975). ‘‘Geomorphology’’. Houghton Mifflin, Boston, MA.
Runge, E. C. A. (1973). Soil development sequences and energy models. Soil Sci. 115, 183–193.
Salvucci, G., and Levine, J. (2000). Groundwater-vadose zone interactions at Trochu.
In ‘‘Spatial Pattern in Catchment Hydrology: Observations and Modeling’’ (R. Grayson
and G. Blöschl, Eds.), pp. 304–328. Cambridge University Press, Cambridge, UK.
Savic, D. A., and Walters, G. A. (1999). Hydroinformatics, data mining and maintenance of UK
water networks. Anti-Corrosion Meth. Materials 46, 415–425.
Schaap, M. G., and Bouten, W. (1996). Modelling water retention curves of sandy soils using
neural networks. Water Resour. Res. 32, 3033–3040.
Scheidegger, A. E. (1967). A stochastic model for drainage patterns into an intramontane
trench. Bull. Int. Assoc. Sci. Hydrol. 12, 15–20.
Simonson, R. W. (1959). Outline of a generalized theory of soil genesis. Soil Sci. Soc. Am. Proc.
23, 152–156.
Šimunek, J., Jarvis, N. J., van Genuchten, M. T., and Gardenas, A. (2003). Review and
comparison of models for describing non-equilibrium and preferential flow and transport
in the vadose zone. J. Hydro. 272, 14–35.
Sivapalan, M. (2003). Process complexity at hillslope scale, process simplicity at the watershed
scale: Is there a connection? Hydro. Proc. 17, 1037–1041.
Smeck, N. E., and Runge, E. C. A. (1971). Phosphorus availability and redistribution in relation
to profile development in an Illinois landscape segment. Soil Sci. Soc. Am. Proc. 35,
Smith, H., and Hudson, B. D. (1999). Soil survey in the twenty-first century. ‘‘Proceedings of the
International Symposium on Soil Resources: Their Inventory, Analysis, and Interpretation
for Use in the 21st Century, June 10–12.’’ Minneapolis, MN.
Soil Survey Division Staff (1951). ‘‘Soil Survey Manual’’. U.S. Department of Agriculture
Handbook No. 18. U.S. Government Printing Office, Washington, DC.
Soil Survey Division Staff (1993). ‘‘Soil Survey Manual’’. U.S. Department of Agriculture
Handbook No. 18. U.S. Government Printing Office, Washington, DC.
Soil Survey Staff (1999). ‘‘Soil Taxonomy—A Basic System of Soil Classification for Making
and Interpreting Soil Surveys’’, 2nd ed. USDA-NRCS Agricultural Handbook No. 436.
U.S. Government Printing Office, Washington, DC.
Sommer, M., Wehrhan, M., Zipprich, M., Weller, U., Zu Castell, W., Ehrich, S., Tandler, B.,
and Selige, T. (2003). Hierarchical data fusion for mapping soil units at field scale.
Geoderma 112, 179–196.
Sonneveld, M. P. W., Bouma, J., and Veldkamp, A. (2002). Refining soil survey information for
a Dutch soil series using land use history. Soil Use Manage 18, 157–163.
Sposito, G. (Ed.) (1998). ‘‘Scale Dependence and Scale Invariance in Hydrology’’. Cambridge
University Press, Cambridge, UK.
Sposito, G., and Reginato, R. J. (Eds.) (1992). ‘‘Opportunities in Basic Soil Science Research’’.
Soil Sci. Soc. Am., Inc., Madison, WI.
Stein, A., Staritsky, I. G., Bouma, J., van Eijnsbergen, A. C., and Bregt, A. K. (1991). Simulation of moisture deficits and areal interpolation by universal cokriging. Water Resour. Res.
27, 1963–1973.
Stephens, D. B., Kowall, S. J., Borns, D., Ellis, D., Everett, L. G., van Genuchten, M. T.,
Graham, M., Parker, F., Weeks, E., and Wilson, J. (2002). Letter to the editor on a national
strategy for vadose zone science and technology. Vadose Zone J. 1, 197–198.
Stolt, M. H., Baker, J. C., and Simpson, T. W. (1993). Soil-landscape relationships in Virginia:
I. Soil variability and parent material uniformity. Soil Sci. Soc. Am. J. 57, 414–421.
Sulebak, J. R., Tallaksen, L. M., and Erichsen, B. (2000). Estimation of areal soil moisture by
use of terrain data. Geographical Anal. 82A, 89–105.
Tarboton, D., Blöschl, G., Cooley, K., Kirnbauer, R., and Luce, C. (2000). Spatial snow cover
processes at Kühtai and Reynolds Creek. In ‘‘Spatial Pattern in Catchment Hydrology:
Observations and Modeling’’ (R. Grayson and G. Blöschl, Eds.), pp. 158–186. Cambridge
University Press, Cambridge, UK.
Thompson, J. A., Bell, J. C., and Butler, C. A. (1997). Quantitative soil-landscape modeling for
estimating the areal extent of hydromorphic soils. Soil Sci. Soc. Am. J. 61, 971–980.
Thompson, J. A., Bell, J. C., and Zanner, C. W. (1998). Hydrology and hydric soil extent within
a Mollisol catena in Southeastern Minnesota. Soil Sci. Soc. Am. J. 62, 1126–1133.
Thorp, J., and Gamble, E. E. (1972). Annual fluctuations of water levels in soils of the Miami
catena, Wayne County, Indiana. Sci. Bull. 5, Earlham College, Richmond, IN.
Tisdall, J. M., and Oades, J. M. (1982). Organic matter and water-stable aggregates in soils.
J. Soil Sci. 33, 141–163.
Toth, J. (1963). A theoretical analysis of groundwater flow in small drainage basins. J. Geophys.
Res. 68, 4197–4213.
Troch, P., Verhoest, N., Gineste, P., Paniconi, C., and Merot, P. (2000). Variable source areas,
soil moisture and active microwave observations at Zwalmbeek and Coet-Dan. In ‘‘Spatial
Pattern in Catchment Hydrology: Observations and Modeling’’ (R. Grayson and G.
Bloschl, Eds.), pp. 187–208. Cambridge University Press, Cambridge, UK.
USDA-NRCS (1997). ‘‘Part 649—Land Resource Regions and Major Land Resource Areas.
National Soil Survey Handbook’’. U S Government Printing Office, Washington, DC.
USDA-NRCS (1998). In ‘‘Field Indicators of Hydric Soils in the United States,’’ Ver. 4.0.
(Hurt, G. W., Whited, P. M., and Pringle, R. F., Eds.) USDA-NRCS, Ft. Worth, TX.
USDA-NRCS (2001). ‘‘National Cooperative Soil Survey Conference Proceedings. Building for
the Future: Science, New Technology and People, June 25–29’’. Fort Collins, CO.
USDA-NRCS (2003). ‘‘National Cooperative Soil Survey Conference Proceedings. Soil Information for a Changing World, June 16–20’’. Plymouth, MA.
Vachaud, G., De Silans Passerat, A., Balabanis, P., and Vauclin, M. (1985). Temporal stability
of spatially measured soil water probability density function. Soil Sci. Soc. Am. J. 49,
van Kessel, K., and Wendroth, O. (2001). Landscape research—Exploring ecosystem processes
and their relationships at different scales in space and time. Soil Tillage Res. 58, 97–98.
van Genuchten, M. T., and Wierenga, P. J. (1976). Mass transfer studies in sorbing porous
media. I. Analytical solutions. Soil. Sci. Soc. Am. J. 40, 473–481.
van Genuchten, M. T., Feike, L., and Wu, L. (Eds.) (1999). ‘‘Proceedings of the International
Workshop on Characterization and Measurement of the Hydraulic Properties of Unsaturated Porous Media’’. University of California, Riverside, CA.
Veneman, P. L. M., and Bodine, S. M. (1982). Chemical and morphological soil characteristics
in a New England drainage-toposequence. Soil Sci. Soc. Am. J. 46, 359–363.
Veneman, P. L. M., Lindbo, D. L., and Spokas, L. A. (1998). Soil moisture and redoximorphic features: A historical perspective. In ‘‘Quantifying Soil Hydromorphology’’
(M. C. Rabenhorst, J. C. Bell, and P. A. McDaniel, Eds.), pp. 1–23. SSSA Special Pub.
#54, Soil Sci. Soc. Am., Inc., Madison, WI.
Vepraskas, M. J. (1992). ‘‘Redoximorphic Features for Identifying Aquic Conditions’’. North
Carolina Agricultural Research Service Technical Bulletin No. 301. NC, USA.
Vepraskas, M. J., and Sprecher, S. W. (Eds.) (1997). ‘‘Aquic Conditions and Hydric Soils: The
Problem Soils’’. SSSA Special Pub. #50, Soil Sci. Soc. Am., Inc, Madison, WI.
Vertessy, R., Elsenbeer, H., Bessard, Y., and Lack, A. (2000). Storm runoff generation at La
Cuenca. In ‘‘Spatial Pattern in Catchment Hydrology: Observations and Modeling’’ (R.
Grayson and G. Bloschl, Eds.), pp. 247–271. Cambridge University Press, Cambridge, UK.
Vervoort, R. W., and Cattle, S. R. (2003). Linking hydraulic conductivity and tortuosity
parameters to pore space geometry and pore size distribution. J. Hydrol. 272, 36–49.
Vervoort, R. W., Radcliffe, D. E., and West, L. T. (1999). Soil structure development and
preferential solute flow. Water Resour. Res. 35, 913–928.
Vogel, H. J., and Roth, K. (2003). Moving through scales of flow and transport in soil. J. Hydro.
272, 95–106.
Vogel, H. J., Cousin, I., and Roth, K. (2002). Quantification of pore structure and gas diffusion
as a function of scale. Eur. J. Soil Sci. 53, 465–473.
Wagenet, R. J. (1998). Scale issues in agroecological research chains. Nutr. Cycl. Agroecosys. 50,
Walker, P. H., and Ruhe, R. V. (1968). Hillslope models and soil formation. II. Closed systems.
Trans 9th Int. Cong. Soil Sci. IV, 561–565.
Webster, R. (1994). The development of pedometrics. Geoderma 62, 1–15.
Western, A. W., and Blöschl, G. (1999). On the spatial scaling of soil moisture. J. Hydro. 217,
Western, A. W., and Grayson, R. (2000). Soil moisture and runoff processes at Tarrawarra.
In ‘‘Spatial Pattern in Catchment Hydrology: Observations and Modeling’’ (R. Grayson
and G. Bloschl, Eds.), pp. 209–246. Cambridge University Press, Cambridge, UK.
Western, A. W., Grayson, R. B., Blöschl, G., Willgoose, G. R., and McMahon, T. A. (1999).
Observed spatial organization of soil moisture and its relation to terrain indices. Water
Resour. Res. 35, 797–810.
Wikle, C. K., Milliff, R. F., Nychka, D., and Berliner, L. M. (2001). Spatiotemporal hierarchical
Bayesian modeling: Tropical ocean winds. J. Am. Stat. Assoc. 96, 382–397.
Wilding, L. P. (1994). Factors of soil formation: Contributions to pedology. In ‘‘Factors of
Soil Formation: A Fiftieth Anniversary Retrospective’’ (R. Amundson, J. Harden, and
M. Singer, Eds.), pp. 15–30. Soil Sci. Soc. Am., Special Publ. #33, Madison, WI.
Wilding, L. P. (2000). Pedology.‘‘Handbook of Soil Science’’ (M. E. Sumner, Ed.-in-Chief),
pp. E1–E4. CRC Press, Boca Raton, FL.
Wilding, L. P., and Drees, L. R. (1983). Spatial variability and pedology. In ‘‘Pedogenesis and
Soil Taxonomy. I. Concepts and Interactions’’ (L. P. Wilding, N. E. Smeck, and G. F. Hall,
Eds.), pp. 83–117. Elsevier, Amsterdam, The Netherlands.
Wilding, L. P., Bouma, J., and Goss, D. (1994). Impact of spatial variability on modeling.
In ‘‘Quantitative Modeling of Soil Forming Processes’’ (R. Bryant and M. R. Hoosbeek,
Eds.), pp. 61–75. Soil Sci. Soc. Am., Special Publ. #39, Madison, WI.
Winter, T. C. (1988). A conceptual framework for assessing cumulative impacts on hydrology of
nontidal wetlands. Environ. Manage 12, 605–620.
Wösten, J. H. M., Bouma, J., and Stoffelsen, G. H. (1985). The use of soil survey data for
regional soil water simulation models. Soil. Sci. Soc. Am. J. 49, 1238–1245.
Wösten, J. H. M., Pachepsky, Y. A., and Rawls, W. J. (2001). Pedotransfer functions: Bridging
the gap between available basic soil data and missing soil hydraulic characteristics.
J. Hydro. 251, 123–150.
Wösten, J. H. M., Schuren, C. H. J. E., Bouma, J., and Stein, A. (1990). Comparing four
methods to generate soil hydraulic functions in terms of their effect on simulated soil water
budgets. Soil Sci. Soc. Am. J. 54, 827–832.
Wysocki, D., Schoeneberger, P., and LaGarry, H. (2000). Geomorphology of soil landscapes.
In ‘‘Handbook of Soil Science’’ (M. E. Sumner, Ed.-in-Chief), pp. E5–E39. CRC Press,
Boca Raton, FL.
Yeh, J.-F., and Eltahir, E. A. B. (1998). Stochastic analysis of the relationship between topography and the spatial distribution of soil moisture. Water Resour. Res. 34, 1251–1263.
Young, E. G. (1988). Soil physics and hydrology. J. Hydro. 100, 411–431.
Zhang, T., and Berndtsson, R. (1991). Analysis of soil water dynamics in time and space by use
of pattern recognition. Water Resour. Res. 27, 1623–1636.
This Page Intentionally Left Blank
John A. Howard1 and Elizabeth Hood2
Applied Biotechnology Institute, College Station, Texas 77845, USA
Arkansas State University, Jonesboro, Arkansas 72403, USA
I. Introduction
II. Technology Options
A. Generation of Transgenic Material
B. Protein Expression
III. Production Options
A. Growth
B. Harvesting/Transport/Storage
C. Tissue Processing
D. Extraction/Purification
IV. Products
A. High-Purity Human Health Products
B. Orally Delivered Products
C. Industrial Enzymes
V. Public Acceptance
VI. Conclusions and Future
Over the past several years there have been many advances in plant
biotechnology that have led to the successful commercialization of agricultural products for crop improvement. Plant biotechnology is now being
considered as a tool to produce non-food products such as biopharmaceuticals and bioindustrial products. This chapter reviews the status of the field
with particular emphasis on different plant systems. Key factors such as
transformation, expression, growth, harvest, transport, storage, processing,
and purification of the plant material are included. The chapter also evaluates the characteristics of different systems and their utility for different types
of products. While no one system stands out as the ideal platform, this
chapter does point to systems that have broader appeal and speculates as
ß 2005 Elsevier Inc.
to future platforms and utilities.
Advances in Agronomy, Volume 85
Copyright 2005, Elsevier Inc. All rights reserved.
0065-2113/05 $35.00
Investigators have been exploring the use of plants as an alternative production system for biologics over the past several years (Daniell et al., 2001; Fischer
and Emans, 2000; Fischer et al., 1999; Giddings, 2001; Hood, 2002; Hood and
Howard, 1999, 2002; Hood and Jilka, 1999; Hood et al., 2002, 2003a; Kusnadi
et al., 1997; Ma et al., 2003). While it seems unlikely that no one production
system could meet all potential needs for the diversity of products, plants do
offer some clear theoretical advantages over other systems. One obvious advantage is that the biologics can be produced free of animal source tissue,
thereby eliminating the fear of transmitting animal pathogens, including prions
responsible for such conditions as ‘‘mad cow disease.’’
Plants also offer the potential for a reduced cost of goods, resulting in both a
lower cost of raw material and the potential for rapid scale-up with limited
facility requirements. Plants offer the least expensive source of biomass. In field
environments, the major inputs come from sunlight, rain, and air. Although
other inputs are added to many commercial crops, they are a relatively minor
expense compared to input requirements for non-plant production systems.
Moreover, animal and microbial production systems require plant material as
the primary carbon source used in converting energy into useful proteins.
Because plants can be rapidly propagated from seeds, they do not have
the limitations found in the raising of large herds, as is the case for transgenic
animal systems. Nor is it necessary to build production facilities to generate
the raw material, as is the case for cell culture systems, for which scale-up
can take years and hundreds of millions of dollars. The shorter scale-up times
and reduced up-front capital expenditures for facilities for plant production
contribute to a lower cost of goods.
There are several additional advantages to plant-based products, particularly
when used for direct delivery. Direct delivery refers to applications in which
the plant material can be used directly in food or feed or as an industrial feedstock
without purification. In some cases, the plant material itself may have value in
the application independent of the recombinant proteins contained in it.
Orally delivered human-health products avoid the cost and safety issues
associated with injectables. This has been studied mostly as it relates to oral
vaccines (Streatfield, 2002; Streatfield and Howard, 2003a,b). With oral delivery, there is no requirement for needles or syringes, and medically trained
personnel are not necessary for administration. The costs of delivery are
reduced, and concerns about contamination stemming from incomplete disposal of needles are removed. The lower cost and convenience should result
in more patients being vaccinated. This is true not only in developing countries, where access to medical staff may be limited, but also in developed
countries, where patient compliance requires booster inoculations.
Plant systems are particularly well suited to yielding large amounts of a desired
product in a relatively small area. For example, a current corn-based vaccine
candidate targeting Enterotoxic Escheria coli (ETEC) can yield over 200,000
doses from a single acre of cultivation. Thus, the world’s supply could be
obtained from a single small farm (Streatfield and Howard, 2003b). Moreover,
because some plant tissues, such as seeds, can store proteins for years without loss
of activity under ambient conditions, a ready supply of material can be manufactured into final form on an as-needed basis. Rapid scale-up, large volumes,
and long term storage are particularly advantageous for industrial enzymes. Low
cost and the ability to use the raw material directly for industrial processes
encourage development in this direction.
These potential advantages have led to a recent increase in interest in
using this technology for the production of new biologics. This chapter is
focused on evaluating the different plant technologies with respect to their
characteristics in producing proteins for plant-made pharmaceuticals and
plant-made industrial products.
The choice of plant to be used depends on a number of factors, including
its cultivation, transformability, growing cost, production and processing of
the target tissue, existence of wild relatives, and degree of outcrossing.
Current systems include corn, soybean, canola, alfalfa, Lemna, tobacco,
and safflower. These crops may be wild plants (e.g., Lemna), domesticated
non-food crops (e.g., tobacco and alfalfa), or food crops (e.g., rice, corn,
soybean, potatos, canola, and safflower).
The type of tissue that is used for protein accumulation is often chosen based
on the type of plant used, or vice versa. For example, Lemna, alfalfa, and tobacco
are leafy crops, thus seed or fruit would not be appropriate for use as production
vehicles. Potatoes are a root crop, and thus the tubers are used. Fleshy fruits such
as tomatoes or bananas can be used for production. However, grain crops such as
corn, soybeans, canola, and safflower have the most stable production vehicle,
the seed. While there are multiple interrelated factors affecting protein accumulation and crop choice, for discussion purposes we have divided the technology
into two parts, (1) generating the transgenic material and (2) protein expression.
Recovery of stably transformed plants exhibiting traits of interest requires
the combination of several technologies: (1) appropriate selectable marker
genes and selection conditions, (2) an efficient culture system that allows
recovery of plants from target tissues, (3) a DNA delivery system, (4) the
choice of a genotype that has agronomic or horticultural relevance, and (5)
joining of these technologies so that transformed, fertile adult plants can be
recovered. Variety is the driver for these technologies because of the plethora
of plant species available as targets for plant production of foreign proteins
(Hood, 2003). For maximum utility and efficiency, however, DNA delivery
systems should be simple, efficient, and preferably inexpensive. This is true
whether the methods are used by scientists in industry or in academic
institutions. Additionally, particularly for industry, the method must be
available for use either because it is in the public domain or because it can
be licensed.
The transfer of foreign DNA into plant cells has been going on for
centuries. The most obvious example of natural DNA transfer to plants
with which we are familiar today is in Agrobacterium tumefaciens, the
perpetrator of crown galls. Until very recently, the result of this DNA
transfer had gone unoticed. However, after some astute observations, the
‘‘tumor-inducing principle’’ was hypothesized in 1958 (Braun, 1958) and
its molecular nature was discovered in the mid-1970s (Chilton et al., 1977).
The stage was set to begin developing DNA transfer technology for crop
Molecular farming requires that DNA for encoding the protein of choice
be introduced at will into the plant of choice. To this end, multiple methods
of human-directed foreign DNA transfer into plant nuclei have been developed over the past 20 years, including Agrobacterium-mediated transformation and several direct gene transfer methods, e.g., microprojectile
bombardment, electroporation, silicon carbide fibers, electrophoresis, and
microinjection (Hood, 1999; Songstad et al., 1995). Microprojectile bombardment has also been used to transform chloroplasts of solanaceous plants
(Daniell et al., 2002; Maliga, 2002), and viral vectors are used to transiently
express genes in plants, primarily tobacco (Lindbo et al., 2001). Of these
methods, the Agrobacterium-mediated nuclear transformation and microprojectile bombardment-mediated nuclear and chloroplast transformation
are the methods most often used today. Viral vectors are being used for
commercial development of therapeutic proteins from tobacco (Grill et al.,
2002; Lindbo et al., 2001). Notable is the lack of recent papers in which the
other direct gene transfer methods are used. However, modifications of some
of these techniques developed in the early 1980s have gained some favor
(Southgate et al., 1998). The operative word here is variety. Crop plants span
many species in many genera, families, and classes, and no ‘‘one size fits all’’
method is appropriate for gene transfer.
Selection of transformed tissues requires the inclusion of genes that allow
identification of the transformed cells. Selectable marker genes come in a
variety of types with quite varied substrates (Hood, 2003). Again, the
operative word is variety, because a large array of methods and markers are
critical to the success of plant biotechnology. Selectable markers can provide
selection pressure on plant tissues, resulting in death of nontransformed
cells, or through the starvation of unwanted cells because selective growth
of transformed cells is supported.
Nature provided the first example of selectively different plant cells—
hormone independent growth in Agrobacterium-incited tumors. The first
engineered plant cell selection system was antibiotic resistance (see Chilton,
2001). In the early 1980s, several labs raced to generate tobacco tissue that
could be selected on kanamycin or G418 (Bevan et al., 1983). Subsequent
experimentation has focused on refinement of these techniques, broadening
of the host range, and broadening of the species amenable to transformation
and selection (reviewed in Wilmink and Dons, 1993).
Selectable markers have recently been deemed undesirable traits because of
their perceived danger to public health due to the potential of allergenicity of
the protein or the potential of resistance gene transfer to gut microorganisms
(antibiotic resistance). Their maintenance in the plant after establishment of the
desired trait is assumed to be unnecessary because these genes have no utility
after transformed plant recovery. However, herbicide resistance genes have
utility because they confer downstream advantages for selection of transformed
plants in the field. These allow growth and recovery of plants prior to establishment of homozygous lines so that early field performance of the transgenic
plants can be assessed. Moreover, the pflp gene (Chen et al., 2000; You et al.,
2003), a selective trait, confers resistance to plant pathogens, also giving it
value. Thus, a discriminating assessment of marker gene value versus risk
should be undertaken before wholesale removal of the trait is endorsed.
Alternative methods have addressed the development of less objectionable selectable markers that have less perceived risk, such as the pflp gene
(Chen et al., 2000; You et al., 2003) and positive selection (Wang et al.,
2000). However, when the risk is determined to outweigh the benefit of
retaining the selective trait gene, the methods being sought for their removal
should be employed. Such methods include (1) co-transformation into unlinked sites then removal through breeding (Komari et al., 1996) and (2)
inclusion of recombination sites to selectively remove genes (reviewed by
Hare and Chua, 2002). A slate of multiple selectable markers can improve
the ability of researchers to maximize recovery of transformed, regenerated
plant materials that are close to the final product line. In the future, it may
also be desirable for non-food crops to use a selectable marker different from
food crops of the same species. This would help to keep the two crops
segregated and easily distinguished in the field.
While there are options for removal of the selectable marker, it is also
possible to express the marker protein in a non-target tissue of the crop.
When seeds are used as a source of recombinant protein, markers may be
designed to be expressed in cell cultures or leaves without presenting a new
protein in the harvested crop. Even when the marker protein may end up in
the harvested crop, the presence of the marker gene or protein may be
inconsequential if the protein product is used in industrial applications or
if it requires purification before being used as a pharmaceutical.
Protein expression is the single most important factor for most recombinant proteins and can dictate the economics of the product as well as
regulatory issues. Expression of the protein depends on many factors, including transcription, translation, targeting, and the ability of the plant to
accumulate the protein.
Native proteins are expressed in all plant tissues and organs. Some basic
‘‘house-keeping’’ proteins are often present in most tissues, e.g., ubiquitin,
and are thus expressed from genes regulated by constitutive promoters. In the
case of foreign protein production of pharmaceuticals and industrial
enzymes, it is desirable to sequester the protein as much as possible into
specific target tissues with the use of tissue-specific promoters that are active
only in limited tissue types. An example is the globulin-1 promoter from maize
(Belanger and Kriz, 1991), which is primarily limited to embryo-specific
expression (Hood et al., 2003; Woodard et al., 2003).
Localization of native protein into subcellular compartments combined
with the tissue-specific expression of their genes allows cells to differentiate
with unique identities and collectively form a eukaryotic organism. In addition to promoter specificity, the subcellular compartments that are noteworthy within that tissue present the array of potential subcellular locations that
are likely to maximize that promoter’s work. For example, targeting to a
glyoxysome in a leaf or root would not be as useful as targeting to this
organelle in a cotyledon that stores oil. An example of the effect of alternative targeting on protein accumulation is shown in Table I. Clearly, cell wall
targeting effected the highest accumulation of protein 1, whereas the vacuole
was the best subcellular location for accumulation of protein 2.
The cellular machinery responsible for targeting proteins is under intense study and has been elegantly reviewed (Kermode, 1996; Pyke, 1999;
Sanderfoot and Raikel, 1999). General pathways for targeting gene products
into specific subcellular compartments are diagrammed in Fig. 1. The rough
endoplasmic reticulum contributes protein to the several compartments that
receive their member proteins from the membrane/secretion pathway. Secretion to the exterior of the plasma membrane is the default pathway, and a
signal sequence that begins this process is generally necessary and sufficient
for a protein to arrive on the cell surface (Vitale and Denecke, 1999).
Table I
Effect of Promoter and Targeting Signal on Protein Accumulation in Maize Seed
Protein location
Protein example
Targeting sequence
Highest T1 seedc
cell walla
cell walla
cell walla
Embryo ERa
SS vacuole
No events
Data from Hood et al. (2003a).
Data from Streatfield et al. (2003).
Percentage given as a percent of total soluble protein.
Figure 1 Targeting pathways for plant proteins.
The plastids, mitochondria, nucleus, and cytoplasm receive their proteins
from cytoplasmically translated messages, and the proteins have specific transit
peptide sequences that allow their import into each organelle (Kermode, 1996).
The plastids are an interesting group of organelles that are derived from a
single pre-organelle, the proplastid (Esau, 1977); they include chloroplasts,
chromoplasts, amyloplasts, and elaioplasts. The import process for chloroplast
proteins encoded by nuclear genes was reviewed by Keegstra and Cline (1999).
The underlying principle of this process is that a transit peptide is necessary
for recognition by a receptor on the surface of the plastid envelope. Some
features of this transit peptide are common to all plastids (de Boer et al.,
1988; Lawrence et al., 1997). However, each member of the plastid group
most likely has features of its import apparatus that limit uptake of proteins
to those specifically required for the unique functions of the plastids (Wan
et al., 1996).
Transit peptides and signal sequences have been used by all players in the
field of molecular farming. In most cases, these signals are a part of the
strategy to achieve maximal accumulation of the target protein. In very few
cases has subcellular in situ localization of the target protein been performed.
In one case, Lt B, the bacterial endotoxin from Escherichia coli, was found in
an unpredicted compartment (Chikwamba et al., 2003).
In order for proteins to accumulate, the protein must be stable to the
particular environment. This includes not only the obvious problem of protease attack, but also more subtle attributes. Specific proteins can have
different stabilities depending on their specific environments. Carbohydrate,
protein, and lipid content, as well as pH and salt, may influence the stability of
the protein. Since these will differ in different plants, tissues, and subcellular
locations, their ability to accumulate will also vary considerably. While this
must be explored empirically today, there is hope that the future may bring
predictability to the fate of proteins based on their specific characteristics
in plants and tissues to achieve the highest overall accumulation.
Production of recombinant proteins refers to the growing, harvesting,
transport, storage, and tissue processing of the crop, as well as the extraction
and purification (Fig. 2). With thousands of species to select from and a wide
variety of possible products, it is highly unlikely that any one system will work
best for each of these steps. Each plant production system has its own distinct
characteristics that may or may not prove advantageous, depending on the
product, so it is important to select the plant system that is best for each
product. Since it is impractical to have thousands of different production
Figure 2
Steps in production with key considerations.
systems, it is preferable to adapt a given system to the needs of the various
products. Fortunately, there are some common features that apply to most
products, enabling a few systems to accommodate most products. These key
features include a potential for low cost of goods, maintenance of protein
integrity, flexibility with regard to time and temperature for harvest, and
maintenance of product safety and environmental safety (Delaney, 2002;
Nikolov and Hammes, 2002). These are discussed in the following sections
as to how they relate to the overall efficiency of the system as well as to the
regulatory aspects.
One of the first decisions to be made when selecting a crop for production
is whether working with a cultivated species would be preferred. Cultivated
species have several advantages, including a higher yield of biomass compared to their wild relatives. An understanding of agronomic practices will
lead to a more reliable supply than that from non-domesticated species, most
of which have not undergone selection for higher yields and pest resistance in
past centuries. These characteristics practiced on domesticated crops have
the net effect of a greatly reduced cost of producing the raw material
compared to competing technologies. Another advantage of cultivated species is that many have been evaluated for antinutritional, allergenic, or toxic
agents. In addition, ample infrastructure for downstream processing and
storage of cultivated crops is established along with the experience of
handling the crop.
Plants only grow well in specific environments due to light, temperature,
and soil conditions. The major commercial crops have been adapted for a
wide range of conditions, which extends their geographic boundaries. In
many cases only one crop per year can be grown in a geographic area.
However, additional production of this same crop can be done in other
geographic locations at different times of the year, extending the seasonal
growing of the crop to year-round production.
Non-cultivated crops have an advantage in that they are unlikely to be
mistaken for food crops and therefore are unlikely to be inadvertently mixed
with the food supply. Unfortunately, they would be more likely than food
crops to outcross with native plants in the environment, which may pose a
larger hazard. Matching the yield of cultivated species with that of wild
species would require decades of research, assuming the crop had the potential. Finally, the impact of these relatively unknown species on product safety
is unknown. Determination of the degree of this impact would undoubtedly
require extensive effort and time.
One consideration for foreign protein production is whether to select a
food crop or a non-food crop. One advantage of food crops is that we know
they are safe for consumption. This is a dramatic advantage in cases where the
final product can include the plant tissue as well as the recombinant protein.
Examples include orally delivered products, such as vaccines, for which the
protein product is not purified from the plant tissue; rather, a formulated
product made from the food crop is orally administered. The use of non-food
crops for orally delivered products would demand that we understand the
potential for toxicants, allergens, or antinutritional agents of these uncharacterized systems. In the case of therapeutics where the protein is purified
from plant tissue there is also an advantage to using food crops. A greater
safety margin can be obtained with known food crops since any host protein
that co-purifies with the recombinant protein would already be part of the
food chain. The only obvious potential advantage for using a non-food crop
is that it may be less likely to inadvertently mix with the food crop.
Another choice to be made is whether to use an open-pollinated or a selfpollinating plant. The advantage of using self-pollinating plants is that there
is a much lower risk that pollen will unintentionally transfer onto other
plants of the same species. Controlled pollen shed of open-pollinated crops
can be used to help alleviate this concern by either physical or genetic means
to prevent outcrossing onto weedy species or related food or feed crops.
For most cultivated crops, self-pollination usually means that the seed
planted by the grower can be saved every year and replanted the following
season. However, it is more difficult to maintain control of the seed and more
likely for a contaminant to accumulate when the same seed source is used to
grow crops season after season. Therefore, the advantage of self-pollinating
crops is partially offset by the potential for amplification of contaminating
plants. This is in contrast to the case of open-pollinated crops that are
produced as hybrids, in which subsequent production is not from plants in
the production fields but from parent seed stocks. Growers do not save seed
because of the poor yield compared to the hybrid seed. Therefore, it is
unlikely that any amplification of contaminating plants will occur.
One additional possibility that exists for plant production systems is that
of plant cell culture systems or hydroponics. These systems have the advantage of being physically contained and avoid the potential disadvantage of
outcrossing with species that are grown in an open environment. Unfortunately, other factors such as higher cost and unknown product safety present
major hurdles for product development.
With the exception of some specialty crops, most crops today are harvested mechanically. Therefore, collecting the plant material itself is usually
not a problem. The problem with harvesting concerns how time sensitive the
plant material is. If fresh fruit is used as the source tissue, it may be critical to
harvest the crop within a narrow time window to avoid degradation of the
crop and/or the protein product. The next related concern is how to move
this tissue into a storage facility without protein degradation. Will the tissue
require refrigeration for any length of time? Fresh fruit and green leafy
material can be a problem in that even if they are harvested at the right
time, they can still degrade during transport or storage. In many cases,
protein degradation can be greatly reduced upon immediate refrigeration.
Some leafy material can be dried; assuming this drying does not degrade the
protein, this option is very appealing. With regard to harvest, seeds are a
preferred choice since they are not time sensitive and do not require special
handling conditions to prevent degradation of the recombinant proteins.
After the plant tissue is harvested and transported to its designated
location, it must be stored for some amount of time before it is processed.
The storage concerns of the host tissue mimic those for transporting the
tissue but may be magnified, since the raw material may be in storage for
years until it is fully processed to its final form. The length of time between
harvesting and processing can vary significantly depending on the plant
tissue. If the protein source is fresh tissue, then the protein will be at risk
for degradation due to the plants’ active metabolic machinery. Fruit crops
are at a disadvantage with regard to storage because of natural degradation
by native enzymes. This limitation can be overcome if the fruit is immediately processed in some form. For example, the fruit can be dried, thus
providing a useful storage system. Alternatively, fruit can be made into a
juice that could be further processed immediately, thereby eliminating the
storage of the host tissue in its native form. This means either that processing
must take place shortly after harvest with no storage or that storage will
require refrigeration. In many cases, both are required.
In contrast, in plant storage organs, e.g., seeds or tubers, the plant part is
in a dormant state with little metabolic activity. In this regard seed tissue has
a distinct advantage because seeds store proteins for years without degrading
proteins. Examples of this phenomenon include the demonstration that
recombinant proteins can remain stable in seed tissue for months to years
(Kusnadi et al., 1998a; Lamphear et al., 2002; Stoger et al., 2000). This
stability may be due to the high concentration of protease inhibitors in
seed, the low water content, or the carbohydrate available to stabilize the
protein. Thus, seeds can be stored under ambient conditions to allow greater
flexibility for processing options.
Tissue processing after the crop is harvested is necessary whether for
direct delivery or for highly purified products. One of the most critical
aspects is the amount of total protein present in the harvested tissue. The
amount of protein as a percentage of total biomass can range from less than
1% to over 40% depending on the plant and tissue source. This feature is
critical because in addition to obtaining relatively high percentages of total
soluble protein for ease in purification, the overall cost of tissue processing
and extraction is directly related to the amount of total biomass. Fresh fruit
is generally at a disadvantage because it is relatively low in overall protein
content (1–3%) and high in water content. In contrast, most seeds are high in
total protein (10–40%) and low in water content (Koehn, 1978). Mechanical
processing of seeds to flour is common and suitable for extraction. In the
case of leaves, maceration can be employed before grinding the tissue. For
fresh fruit or fresh leaf tissue, it may be necessary to extract the product
immediately after processing to avoid protein degradation. In this instance,
tissue processing and extraction need to be considered together. Therefore,
seeds offer protein stability and allow for processing dry tissue and a
relatively low biomass to give an overall advantage prior to extraction.
Tissue can also be processed and separated into fractions enriched with
recombinant proteins by mechanical means. For example, it is possible to
generate a germ or endosperm fraction from grains by using standard
procedures common in the industry today (Watson, 1988). These fractions
can have a much higher protein content than the whole grain due to the
expression technology used. Furthermore, the remaining part of the grain
can be used for other industrial applications. This has been done for maize
for both orally delivered product candidates and purified proteins. In these
cases, the germ fraction contained a 5- to 10-fold enrichment of the recombinant protein on a dry-weight basis compared to the whole grain, due in
part to the type of promoter used (Kusnadi et al., 1998b; Streatfield et al.,
2003). The amount of biomass required for extraction is reduced between
5-fold and 10-fold, thereby reducing the cost of extraction. This leaves 90%
of the grain available for other industrial applications, such as ethanol
production. In this way, the cost of the raw material can be reduced since
it is offset with byproduct credits. Finally, the waste from the process is
greatly reduced, thereby reducing the cost of waste disposal. In this example,
tissue processing is not only a required step, but also an opportunity to
reduce downstream costs.
Extraction is relatively easily done in most cases by simply adding an
aqueous buffer to ground tissue. The pH and salt content can be optimized
on a case-by-case basis to reduce extraction of endogenous proteins and
preferentially solubilize the recombinant protein (Bai and Nikolov, 2001;
Evangelista et al., 1998). Fresh tissue will most likely require refrigeration to
prevent protein degradation. Seeds, however, usually have the advantage of
endogenous protease inhibitors, which allow greater flexibility in extraction
times and temperatures.
Protein can be purified from the aqueous extracts in a manner similar to
extracts from other production systems. For pharmaceutical products, this
will require a cGMP (current Good Manufacturing Practices) facility. Many
control points will be similar to those already used for other systems, but a
few characteristics that are more pertinent to plant-based systems will be
The need to test for animal pathogens in the final product should be
greatly reduced, if not eliminated, when using plant material. Unfortunately,
this concern may be replaced with the need to test or validate protocols to
demonstrate that no pesticides used in growing the crop will be present in the
final product. Often, pesticides are not present on the harvested portion of
the crop. Moreover, the small molecule pesticides will separate easily from
proteins, even if they are present on the harvested material, due to their
vastly different physical properties. Acceptance that the final product is free
of pesticides will most likely require validation before it gains acceptance
from regulatory agencies.
Plants make a number of phenolics, alkaloids, and other secondary
metabolites that can interfere with protein purification. These small molecules sometimes bind to proteins, making purification difficult, or they may
interfere with the applications when the proteins are used directly without
purification. Green leaf material is generally high in these compounds,
although some seeds can also have a high content. Care must be taken either
to select material that is low in these compounds or to validate that there is
no interference with purification or applications.
The advantages and disadvantages for a variety of plant types when
considering all of these characteristics are summarized in Table II. No one
type of plant rises to the top as the clear choice. This suggests the possibility
that one could modify an existing crop specifically for recombinant protein
production. As an example, a cultivated crop could be altered to have a
higher protein content that could translate into higher amounts of recombinant protein. For industrial feedstocks, selecting a major crop that already is
used in industrial applications would be beneficial. For orally delivered
therapeutic proteins or vaccines, a food crop that has GRAS (‘‘generally
recognized as safe’’) status would be best. For protein stability and ease of
transport, a grain would be a good choice.
Combining these characteristics into a single crop would require starting
with a food crop. Because of the perceived danger of intermixing recombinant protein-containing crops with commodity crops, specific preventive
measures can be taken. For example, a colored marker could be used to
differentiate between the industrial grain and the commodity crop. The
public’s acceptance of the crop may change for open-pollinated crops if a
Table II
Characteristics of Plant Systems
Wild species
Clearly distinguishable
from crops
High yields
Experience in growing
Infrastructure and
experience exist
Low yield
Outcross to native plants
Little known about safety
Potential to intermix with crops
used for other purposes
High margin of safety for
human health products
Greater potential to intermix with
food supply
Less potential to intermix
with food supply
Greater potential for toxic,
antinutritional, or allergenic
Fresh tissue
Seed or
dry tissue
Abundant biomass
High protein content
Limited exposure to
Low cost
Infrastructure in place
High cost
Limited knowledge of product
Higher potential to intermix
Clearly distinguished by
Non-transferable genetics
Low cost
Infrastructure and
experience transferable
from commodity crop
Not yet developed
cell cultures
Field grown
Modified food/feed
grain designed
for industrial
male sterile system is used. Alternatively, a self-pollinating crop could be
used. Chemically induced promoters could also be employed such that
expression is only present when an exogenous chemical is applied. By
incorporating these measures, the valued experience and product safety
attributes of food quality grains can be kept and at the same time, the
potential for an unintentional environmental impact can be minimized.
While a general discussion of different plant types provides a framework,
it is also possible to compare actual systems used today. We have summarized the abilities of several different plant species to be host production
systems (Table III). This is a very controversial topic, and a number of
assumptions must be made to even begin to compare them. Therefore, this
Table III
Ratings for Selected Crops as Suitable Protein Production Systems
Cell cultures
should only be used as a general guide. To understand the ratings, the
criteria are listed below as they relate to how the art is practiced today.
. Product safety refers to what we know about the crop as it relates to
antinutritional, allergenic, or toxic agents. An ‘‘A’’ rating means a crop is
a food source with no known problems. A ‘‘B’’ rating means the crop can
be used as a food source; the known toxins and antinutritional or allergenic agents associated with it are not severe. A ‘‘C’’ rating means there
are known toxins, allergens, or antinutritional properties or there is no
information known.
Environmental safety refers to the potential to outcross with weedy species
or food crops and the potential to intermix with the food supply. An ‘‘A’’
rating means there is little outcross potential with weeds or food crops.
A ‘‘B’’ rating is given to those species that, as practiced today, have
reduced this risk to near zero. A ‘‘C’’ rating is given to species for which
there is a significant risk as practiced today.
Cost is defined as the commodity price of the harvested crop, which was
used to provide an index to determine its relative production cost. The
percentage of total protein in the harvested tissue was then used to
calculate the relative amount of recombinant protein that could be present
if expression, based on percent of total protein, was the same in all cases.
An ‘‘A’’ rating was given to the low cost crops, ‘‘B’’ ratings refer to species
that were significantly higher, and ‘‘C’’ ratings were given to species that
were an order of magnitude or more higher than the ‘‘A’’-rated crops.
Laboratory ease: This represents the ease of transforming and targeting
protein accumulation in the harvested tissue. An ‘‘A’’ was given to crops
for which this is routine and can be done in most laboratories. A ‘‘B’’ was
given to crops for which protein accumulation is routine but does require
additional skills and is only practiced in specialized laboratories. A ‘‘C’’
was given to crops for which this process is known to be difficult.
Field knowledge and experience: An ‘‘A’’ rating refers to crops for which
we have ample experience and infrastructure to grow the plant and for
which we have a series of variants that can be useful if specific problems
arise. A ‘‘B’’ rating is given to those for which we have some experience
but the crop experiences are not as well characterized. A ‘‘C’’ refers to
those crops for which we have little experience in growing them.
Harvest/storage/transport: An ‘‘A’’ rating refers to crops that can be
harvested at almost any time, and for which transport and storage is stable
at ambient temperatures for extended times. A ‘‘B’’ rating is for species
that under certain conditions can be stored at cool temperatures but for
which there is flexibility in the harvest date. ‘‘C’’ ratings are for those crops
that easily spoil and that are time-sensitive to harvest, transport, and
storage conditions with or without refrigeration.
. Tissue processing/extraction/purification: An ‘‘A’’ rating is for plant tissue
in which processing can be optimized for extraction and purification. The
tissue would contain low levels of phenolics that can interfere with protein
purification and low protease activity that can degrade the product during
purification. ‘‘B’’ ratings are for those crops that are easily processed but
require a larger biomass to extract the same amount of protein. ‘‘C’’
ratings are for those crops that are difficult to process, are low in protein
content, and have protease activity or phenolics that could interfere with
protein purification.
While each crop has its unique features, Table III illustrates a few key
points. First, no one plant species exists today that is best under all conditions. There has been no attempt to weight each of these categories, because
the importance of each category may be different for different product types.
Therefore, we will consider each of the categories separately and point out
what products are most appropriate for that category and what can be done
to improve a perceived disadvantage.
While some advocate the use of weeds to produce recombinant protein
products, no specific examples have been suggested. Weeds in general score
poorly in all categories. The suggestion of using weeds arises from the fact that
weeds can be easily differentiated from food crops and serve no other useful
purpose. However, they would often be able to outcross with native weeds,
which gives them a lower rating for the environment. Little is known about
how weeds may fair in downstream processes or product safety. It would
require decades of research and cost millions of dollars to fully characterize
these systems. At the end of that time, it may be clear if weeds could be a viable
production system—but the answer may be that they are unacceptable.
Tobacco is one of the favorite plants used in the laboratory; this is
undoubtedly why it was one of the first crops used for plant production
systems. In addition, there is little need for concern about its effect on the
environment or intermixing with food. One disadvantage in using tobacco
for the direct delivery of human health products is its high alkaloid content.
It may also have some antinutritional properties, and it is generally not
palatable to humans. For larger acreage and low-cost products such as
industrial enzymes, tobacco would not be a good choice because of its
relatively high cost of production. This crop fits best with parenteral products that have higher profit margins and high purification requirements
compared to industrial enzymes. In a unique situation, by using a viral vector
system, tobacco can be one of the best systems to deliver small-scale therapeutics in record time. This is being explored for single-chain antibodies and
other small proteins (Grill et al., 2002).
Cell cultures and hydroponics are more suited for containment than all
other plant choices. The physical facility required for growing material is
analogous to microbial or animal cell culture systems, making this one of the
most expensive ways to produce raw material from plants. It is highly
unlikely that this cost disadvantage could be overcome any time soon.
These systems may be some of the best when expressing proteins that are
toxic or deleterious if released.
Sunflower and canola are good choices in most categories. There are some
known allergenic problems associated with sunflower (Zitouni et al., 2000),
which slightly lowers its rating for product safety. This is probably only a
significant factor when producing orally delivered products. Sunflower,
canola, and sorghum have the potential to outcross with weedy species;
therefore, they have the worst rating for safety, which is the ability to
intermix with food and the ability to outcross with weedy species. This safety
issue can be controlled by management practices, depending on how and
where the plants are grown, although they will increase the cost of production slightly and limit the flexibility of areas in which the crops can be grown.
While it may be possible to overcome this limitation, it will be difficult to
convince the public and regulatory agencies that practices are in place such
that these crops pose little risk. The recent entry of safflower as a recombinant protein production vehicle may help to limit some of the unwanted
characteristics and provide an alternative to sunflower.
Bananas and tomatoes may be good choices for orally delivered products
since they are highly palatable, assuming that a dose can be obtained in
convenient amounts of plant tissue. These crops fall short in producing
parenterals because of their low protein content, which impacts purification
costs, and because they are difficult to harvest, transport, store, and extract.
They also do not have the infrastructure or the cost advantages needed to
grow large volume industrial products.
Soybeans are one of the better choices for industrial products, in theory,
based on their high protein content and low cost. They can also be used for
human health products, although they have a slightly lower rating for
product safety because of some known allergens and antinutritional agents
(Kleine-Tebbe et al., 2002; Rihs et al., 1999). This should not be a major
disadvantage for industrial products, however, and its cost advantages and
self-pollination aspects make this crop a good candidate. The major disadvantage of soybeans is that they are very difficult to work with in the
laboratory. This limitation will undoubtedly be reduced with time, and
soybeans may become one of the preferred crops for industrial products in
the future.
Maize has been the most used crop to date for the production of recombinant proteins in plants (Hood et al., 1999). The reasons become apparent
when examining the data in Table III. Maize is well accepted as a safe
product (GRAS) and is widely used in food, feed, and industrial applications
today (Watson, 1988). The production cost of maize is very low, and the
infrastructure can handle large or small acreages for industrial or pharmaceutical products. Storage and transport of seed and protein purification
from flour are very compatible and flexible with current practices without
special handling. In addition, maize has an advantage in that the kernels can
be mechanically separated to yield a germ fraction with enriched protein and
an endosperm fraction with enriched carbohydrate (Watson, 1988). This
allows for easy use of the carbohydrate fraction for industrial applications
such as ethanol production. In this way, not only is the cost of the raw
material reduced but the waste products are handled as well. There are no
known agents in maize that generally interfere with protein purification.
Finally, the grain can be processed with little or no heat inactivation steps
without affecting the nutritional or taste properties, making this a good
candidate for orally delivered products.
The disadvantage of maize is the potential for intermixing with the food
supply. This one feature has brought other crops of lesser strength to the
forefront. Intermixing potential can be handled by management practices,
but the public’s confidence needs to be gained. While skeptics may claim no
food source can ever be used to produce pharmaceutical products, we must
remind ourselves that both eggs and yeast are used to make pharmaceuticals.
Not only is there not a problem with intermixing in these instances, but also
the public has accepted these as distinctive production systems from the food
system. Maize as well as other plant systems must build an infrastructure
dedicated to pharmaceutical protein production, which is as distinct from
food production as edible eggs are from vaccine production.
When should one plant system be chosen over another system, and for
what specific types of products? Table IV summarizes some of the key
characteristics of production as they apply to different types of products.
Table IV
Production Characteristics of Different Product Types
Relative cost
of raw
Relative overall
production cost
High purity
Human health
Oral delivery
The following sections describe progress on these types of products and the
features that are most critical in choosing the best production system.
Most therapeutic proteins were historically derived from animal tissue
because they could only be produced by animals. The advent of recombinant
DNA technology has led to many new hosts for therapeutic protein production
and the design of new therapeutic proteins. The recent explosion of antibodies
as a class of therapeutics, a class that did not exist 20 years ago, has propelled
these proteins to become the major class of new protein therapeutics. Antibodies also represent a challenge for the industry (Chu and Robinson, 2001;
Houdebaine, 2000; Schwartz, 2001). How can these types of products be
produced inexpensively to realize their great potential? Currently, the preferred
production system is animal cell culture because microbial systems are an
inadequate host for making complex mammalian proteins. However, the cost
of production from animal cell culture is quite high, resulting in antibody
products that are too expensive for many applications. This has opened up
the possibility for plants to play a major role for these products because plants
can adequately synthesize these complex proteins and their production costs
are lower than animal cell culture systems (Schillberg et al., 2003).
Another class of proteins that fit the plant technology platform is that of
blood proteins. In many cases, extremely large quantities are needed and at a
very inexpensive price. Proteins such as human serum albumin have been a
target for a low-cost supply for years to increase its applications. In addition,
these proteins can be made in an animal-free system, which is a great concern
when considering the amount of blood required from unknown sources to
obtain the products. Additional proteins, if made in quantity and at a low
cost, may be the basis for an eventual blood replacement. Both hemoglobin
(Merot et al., 2002) and human serum albumin (Farran et al., 2002; Sijmons
et al., 1990) have already been expressed in plants.
Glycoproteins have different glycan structures depending on the host
used for expression. One significant difference is the addition of sialic acid
on some mammalian proteins, which does not occur in other systems. While
this represents a very limited number of proteins, it does represent an
important class of proteins used in the pharmaceutical industry. Sialic acid
allows therapeutic glycoproteins to have a prolonged clearance time in the
blood stream (Rasmussen, 1991). One of the major limitations is that plants
do not usually make sialic acid, although a recent study demonstrated that at
least limited sialylation may be possible (Shah et al., 2003). This option is not
practical for recombinant proteins until plants are better equipped with both
the specific neuraminic acid transferase and the enzymes required to make
sialic acid. While this seems possible, it is not without technical challenges
and will require several years to accomplish. Alternatively, if the therapeutic
proteins are not toxic, it may be possible to use a higher dose to overcome
clearance time and achieve the same physiological results without sialic acid.
While this may add cost to the product, the plant source may still cost less
than alternative systems and easily accommodate the increased cost needed
for a larger dose.
In addition to sialic acid, plant glycoproteins have other subtle differences
from their mammalian counterparts (Faye et al., 1989; Lerouge et al., 1998),
including the addition of xylose and the change in fucose from an alpha(1,6)
to an alpha(1,3) linkage. These carbohydrates have been implicated in the
allergic response as it relates to pollen allergens (Garcia-Casado et al., 1996),
and they may also be responsible for allergic reactions with recombinant
proteins. (van Ree et al., 2000). In several cases, recombinant proteins
expressed in plants contained carbohydrate structures that have been characterized with no apparent effect on activity (Bakker et al., 2001; CabanesMacheteau et al., 1999; Ma et al., 1998; Samyn-Petit, 2001). In a few
examples, these proteins also were examined for their possible role in allergic
response. The minor changes in glycosylation, however, did not appear to be
implicated in an allergenic response (Chargelegue et al., 2000; van Ree et al.,
Commercial high-purity recombinant protein products from plants were
first demonstrated with the diagnostic products avidin (Hood et al., 1997)
and b-glucuronidase (GUS) (Witcher et al., 1998) expressed in corn grain.
The large-scale production of a foreign protein in a recombinant plant has
recently been achieved for bovine trypsin, also expressed in corn grain
(Woodard et al., 2003). These first products represent the prototype for
most high-value therapeutics. In these instances, the raw material is only a
small cost of the final product. What is critical is the concentration of the
protein in the starting material before extraction. As is the case for all types
of products, higher levels of expression and accumulation are the key drivers. Unlike industrial enzymes, however, the price for obtaining the high
concentrations can be easily absorbed, even if this causes a significant loss in
agronomic yield. This is important because there are several variant lines of
crops that could cause an increase in recombinant protein as a percent of dry
weight but that may result in a net decrease in overall agronomic yield. One
example may be breeding recombinant protein genes into opaque mutants of
corn. The yield of opaque corn is significantly lower than normal hybrid
maize. However, the concentration of recombinant protein can be as much
as 3–5 times higher in this grain, allowing significant savings in extraction
costs. This tradeoff is easily acceptable, not only because of the relatively low
cost of raw material compared to the final product but also because of the
relatively low overall acreage requirement.
In general, oral delivery of vaccines or therapeutics is favored because of
the convenience of this method. The economic advantages of oral delivery
when using plant-based production systems that eliminate the need for
purification are also compelling. The key driver is to obtain sufficient levels
of expression to enable oral delivery. The required dose must be in an
amount of plant material that is manageable for consumption at a single
sitting. If the material is not particularly palatable, it could be presented in
pill form. If the material is palatable, a single dose may be in the form of a
wafer or other processed food-like substance. There are several processing
alternatives depending on the available food or feed industry procedures for
the chosen plant material. A few alternatives have been explored, for example, with corn (Watson, 1988). Selecting tissues that contain high levels of
protein as measured by the concentration of the protein per gram of tissue is
advantageous for oral delivery rather than as a percent of total soluble
protein, as is the case for highly purified products. This favors plant tissues
that are low in water content, as opposed to tissues such as fresh fruits.
Preferred plant tissues include seeds, which are low in water content and rich
in protein, and leaf tissue, if it can be dried to reduce water content without
interfering with the product. Preferred crops include dried alfalfa and edible
grains such as corn or rice (Streatfield et al., 2001, 2002, 2003). Potatoes have
been used but may cause some problems with digestion when eaten raw by
humans (Tacket et al., 1998), although recent studies using corn germ
appeared to overcome these mild adverse symptoms (Tacket et al., 2004).
In general, plant tissue that is consumed as food or feed should provide an
extra level of comfort with regard to orally delivered product safety.
Processing costs for orally delivered products are greatly reduced compared to purified proteins, and if it were not for tight quality control and
regulatory standards, the cost would be similar to industrial proteins. Therefore, most plant systems will be economically viable as long as expression is
moderately high. The more critical concern is to demonstrate that during
whatever limited processing is necessary, the protein product is not altered.
Most processing steps for grain products include a high-temperature step that
will most likely inactivate the protein. It has been shown, however, that
antigens can be processed in plant tissues at reduced temperatures, retaining
their native state while at the same time achieving the desired effect (Streatfield
et al., 2002).
Oral delivery can be used with therapeutic proteins, growth hormones, or
vaccines. Because of the large cost advantage in plants, obtaining doses of
1000 times the injectable dose is technically and economically feasible. This
means that even if the oral product requires a 1000-fold higher dose to be
equivalent to an injectable dose, it can be effective and cost the same or less.
Most oral delivery research has focused on vaccines. Immunogenic responses
have been observed with several plant-based vaccine candidates in experimental animal systems. Protection from disease symptoms has been observed
in target animal trials (Lamphear et al., 2002); this has recently included the
possibility of passive immunity for the application to newborns pigs when
boosting the lactogenic immunity in sows (Lamphear et al., 2004). Human
clinical studies have not yet advanced this far and have focused on inducing
an immune response (Kong et al., 2001; Tacket et al., 1998, 2000, 2004).
Presumably, orally consumed proteins are protected by surrounding plant
tissue that allows a sufficient dose to survive the digestive processes of the
stomach and small intestine (Bailey, 2000). This biological encapsulation, or
bioencapsulation, has an effect similar to the use of encapsulating agents
such as liposomes with oral vaccines produced from alternative sources. It
is not yet known if bioencapsulation is true for all plant tissues or if it is
restricted to certain types of tissues. Grains offer the potential for native
protease inhibitors, carbohydrates for protein stability, and a granular matrix that may account for some of the proteins’ slow release and ability to
survive into the gut.
Industrial proteins are used for protein processing and purification, diagnostics, and processes relating to food, feed, and industrial applications.
Most commercial industrial enzymes are currently derived from microbial
sources, either natural or recombinant, primarily because all other systems
are cost prohibitive. However, using plants as a production system is a
developing industry that could be competitive in the next few years (Hood,
2002; Hood and Woodard, 2002).
There are several categories of enzymes, including hydrolases, transferases, oxidation/reduction (redox) enzymes, lyases, isomerases, and ligases.
Only a few are routinely used in industrial applications. Hydrolases such as
amylases and proteases are the most commonly used industrial enzymes.
Other enzyme classes, particularly redox enzymes, will be useful in industry
when they become less expensive and thus available for testing in various
The principal non-food industrial uses for enzymes are starch hydrolysis
(amylases), textile desizing (amylases), leather production (proteases), pulp
processing (xylanases), detergent additives (proteases), and animal feeds,
which represent substantial markets to date. However, a reason for the
limited use of enzymes in certain applications is that no cost-effective source
of the enzyme is available. Transgenic plant systems can meet both the scale
and cost targets for many new applications.
Using an enzyme as a catalyst in a particular process has advantages that
result from the enzyme properties of functioning at ambient temperatures, in
aqueous solution, and usually at near neutral or physiological pH values.
Most enzymes have a very high specificity and function in low concentrations to produce the desired effect. Enzymes can produce a rapid reaction
and have a low level of toxicity. However, these highly active proteins are
generally unstable and require a degree of care and expertise in their preparation and use. In addition, extremes of pH and temperature can limit
activity levels, even if these conditions do not destroy the enzymes. Enzymes
may be inactivated by the presence of various ions, organic molecules, or
solvents, components that are often present in organic reaction systems and
in large-scale industrial processes. Significant changes in the process may be
necessary when switching from chemical to enzymatic catalysts. Alternatively, the enzyme of interest could be engineered using a variety of emerging
technologies to more closely fit into the reaction conditions of the current
target process. Nevertheless, considering the large market opportunity for
industrial enzymes and their environmentally friendly benefits, it is well
worth the effort to generate an efficient production system and make process
or protein changes to accommodate them.
For achieving low-cost industrial enzymes, plants as a production system
have many advantages over current competing technologies. Plants are an
excellent system for cost competition because of their protein expression
potential and minimal production costs. Various plant systems can be used
to produce enzymes, such as tobacco, alfalfa, barley, canola, and corn
(Hood and Woodard, 2002). The first recombinant proteins produced and
sold from transgenic plants were avidin and GUS (Hood et al., 1999;
Witcher et al., 1998). Avidin, the first plant-produced recombinant protein
product to be marketed, was first sold in 1997. Applications of these proteins
include using the GUS protein as a research reagent and using avidin as a
diagnostic reagent and protein purification tool for biotinylated proteins.
Currently, the highest profile plant-based industrial enzyme projects involve enzymes for applications in feed, cleaning agents, processing reagents,
the wood products industry, and biomass conversion. These include mainly
xylanases and cellulases for the textiles and wood products industries, as well
as laccase and trypsin (Bailey et al., 2003; Hood and Woodard, 2002; Hood
et al., 2003b; Woodard et al., 2003). For food applications, enzymes are used
in the conversion of raw materials to form intermediate products that are
more useful in food processing and in food formulations. The treatment of
food products with enzymes makes them more palatable or more stable or
enables the development of some other desirable property. Enzyme production for the food industry is primarily through microbial fermentation. However, major value from edible plant-based protein or enzyme production is
readily apparent. For example, brazzein, a high intensity sweetener, has been
expressed in transgenic maize (Lamphear et al., in press). For food applications, the direct addition of corn meal is possible, allowing for both a
bulking agent and a low-calorie sugar substitute. The cost is kept to a
minimum since there is no added cost in preparing the sweetener, and the
corn meal can be processed using methods similar to conventional methods.
Two potential new areas for enzyme applications are biomass conversion
and the wood products industries. For biomass conversion applications,
enzymes that degrade cell walls will be useful. The resulting products from
those enzymes are monomeric components of walls, primarily 5- and
6-carbon sugars, but also amino acids and lignin products. These will be
substrates for a variety of applications, including fermentation into ethanol
and specialty chemicals. Oxidation/reduction enzymes such as laccase (described previously) and peroxidases (Caramelo et al., 1999; De Jong et al.,
1992; Jensen et al., 1996) will find many new markets as new production
systems are introduced. These enzymes can potentially replace many applications currently using chemicals that are damaging to the environment.
Because of the scale-up potential of the plant system and the low cost of
goods, plants will likely be the system of choice for large-scale enzyme
production for these industries. It is important to explore many plant
systems, enzymes, promoters, and targeting sequences to understand the
factors that affect expression levels, and hence the economics of industrial enzyme production in plants. To date, such experimentation has been
GMO (genetically modified organism) products that enter the food supply are generally not labeled, and thus the public is not given a choice about
consumption of these products. This has led to a controversy about whether
to label GMO products, resulting in a public that is skeptical about their
value and safety. While GMO non-food products can be easily labeled and
are not intended for the food supply, the skepticism and lack of trust have
carried over to them as well. For pharmaceuticals manufactured from any
source, however, the public has a clear choice whether or not to use the drug
product. If the product is effective, they experience the direct benefit. In this
example, the public has been much more accepting. Therefore, there should
not be any major concerns about the plant-produced products themselves.
In fact, the animal-free source, lower cost of goods, and oral delivery should
make the public enthusiastic about these products.
The major public concern is that non-food products produced in plants
will not somehow end up in the food supply. The public usually assumes that
GMOs will be grown with commodity practices that do not limit the intermixing of the plants. In practice, however, these plants are grown with
regulatory standards that are similar to those used for pharmaceutical and
industrial products from non-plant systems. There are new regulations
for these types of products that are similar to practices used in non-plant
systems. It is hoped that in the future the public will have the same confidence
for plants as they now have for eggs or yeast, potential food products that are
currently used for pharmaceutical and industrial protein production.
The regulations and containment practices used for plants, as well as for
other hosts, do not overlook the remote possibility that unintended exposure
may occur, regardless of how insignificant it is. This situation calls for a
safety and risk assessment that would be accepted by the industry, regulatory agencies, special interest groups, and the public. The risk assessment
needs to be science based and could be similar to what is used for other
systems. A system has been proposed for quantification of unintentional
exposure that is based on evaluating the risk that is linked to the hazard
and exposure. Formulas exist for regulated articles, and it has been suggested that these equations be modified for non-food products produced in
plants, allowing a quantitative method of assessing the risk of unintentional
exposure (Howard and Donnelly, 2004).
The requirements for producing non-food products in transgenic plants are
considerably different from those for producing food products. They include
physical isolation, delayed planting times from food crops, agronomic support, dedicated equipment, and frequent monitoring. When taking these
practices into account, the amount of a crop that may inadvertently end up
in the food supply and the associated risk can be calculated. Aprotinin is an
example of a pharmaceutical product that has been produced in plants
(Delaney et al., 2003; Zhong et al., 1999). It has been calculated that even
without any of the required confinement practices, the amount of aprotinin
that could inadvertently end up in the food supply would be well below the
level needed to show an effect (Howard and Donnelly, submitted). This means
that there is no hazard even if the plants were grown and harvested as a
commodity crop. If the required containment practices are used, the calculated
level of risk can be a million times below the non-contained exposure levels.
In addition to potential toxicity, some proteins can be allergenic. Using
the case of aprotinin, we can calculate how much transgenic corn containing
aprotinin used for commercial production must be eaten to induce an
immune response. In this example, one would have to eat 350 tacos or 350
bowls of cereal at one time on three different occasions just to ingest the
minimum dose needed to observe an immune response.
The above calculations seem inconsistent with the fact that the general
public considers these non-food products to be a grave danger. Aprotinin is
one case for which even though the product is intended for non-food
applications, it is already in the food supply from natural sources and no
problems have ever been documented. Other proteins discussed earlier,
trypsin and avidin, are also in the food supply from natural sources, and
we would anticipate similar results. Clearly, the fear of products entering the
food supply is unwarranted for proteins that are already in the food supply
at much higher concentrations than an accidental intermixing would produce. This is not to say that all proteins would have the same risk profile, and
it is this that fuels public fears. Thus, a case-by-case assessment is needed to
allay public concerns if it is applied to all non-food products.
In conclusion, confinement practices can reduce unintentional exposure
for most proposed products to a level that is orders of magnitude below the
slightest concern for food safety. However, safety assessment models need to
be standardized and accepted by the public, regulatory agencies, and special
interest groups. Finally, we need to treat plant-made pharmaceuticals and
plant-made industrial proteins with the same considerations as other pharmaceutical production systems such as eggs or yeast and not as value-added
Plant production technology is still in its infancy. The key to economic
feasibility of products derived from this technology lies first with the expression level of the foreign genes. The best plant expression systems are now
approaching 1% of the dry weight of tissue. This would enable plants to be
the low-cost producer of most proteins compared to other current systems.
However, there is no theoretical reason why plant systems cannot achieve
levels of expression that are at least an order of magnitude higher. Microbes
can produce proteins that comprise as much as 70% of their energy store.
For plants, especially grain with storage proteins, it is reasonable to assume
that the recombinant protein can constitute or replace other proteins that
supply a sink for amino acids. Many advances in technology remain to be
made to take this production system to higher levels.
Direct delivery will be essential for many oral vaccines, and in the future
this process may deliver nutraceuticals and possibly some therapeutics. The
possibility of using products that can be prophylactic, such as specific antibodies, is now approachable using this technology. Feed products easily lend
themselves to direct delivery because growers can control the diet and
supplements of domesticated animals. Direct delivery of industrial products
results in added cost savings for production.
The increase in research efforts for plant-made products will also translate into many more clinical and industrial application trials to demonstrate
efficacy of these products. There will undoubtedly be additional technologies
available that will increase the potency of the orally delivered products.
However, while the data to date are promising, the anticipated increase in
effort will also show that the technology will have limits. These will become
clear as the research is completed.
The next several years should also see detailed regulatory guidelines that
will pave a clear path for later stage clinical trials and manufacturing of
pharmaceutical products. A separate regulatory path for industrial products
should be developed. This will help reduce risks and increase acceptance.
Regulations currently exist for growing non-food crops, but these will need
refinements before large-scale industrial and nutraceutical products can be
The most likely near-term commercialization possibilities for pharmaceuticaltype proteins from plants will be with proteins currently made from animals
or animal cell cultures. These have a much clearer advantage of cost and
safety than products made from microbes. This should set the stage
for additional orally delivered products and industrial products that can
compete with microbial systems for cost. The industry is young, and it is
impossible to even imagine the full range of products at this time.
Bakker, H., Bardor, M., Molthoff, J. W., Gomord, V., Elbers, I., Stevens, L. H., Jordi, W.,
Lommen, A., Faye, L., Lerouge, P., and Bosch, D. (2001). Galactose-extended glycans
of antibodies produced by transgenic plants. Proc. Natl. Acad. Sci. USA 98(5),
Bai, Y., and Nikolov, Z. L. (2001). Effect of processing on the recovery of recombinant
b-glucuronidase (Gus) from transgenic canola. Biotechnol. Prog. 17, 168–174.
Bailey, M. (2000). ‘‘A Model System for Edible Vaccination Using Recombinant Avidin
Produced in Corn Seed’’. Master’s Thesis, Texas A&M University.
Bailey, M. R., Woodard, S. L., Callaway, E., Beifuss, K., Delaney, D., Magallanes-Lundback, M.,
Lane, J., Horn, M. E., Ward, M., Van Gastel, F., Howard, J. A., and Hood, E. E. (2003).
Improved recovery of active recombinant laccase from maize seed. Appl. Microbiol. Biotech.
63(4), 390–397.
Belanger, F. C., and Kriz, A. L. (1991). Molecular-basis for allelic polymorphism of the maize
globulin-1 gene. Genetics 129, 863–872.
Bevan, M. W., Flavell, R. B., and Chilton, M. D. (1983). A chimeric antibiotic resistance gene as
a selectable marker for plant cell transformation. Nature 304, 184–187.
Braun, A. C. (1958). A physiological basis for autonomous growth of the crown gall tumor cell.
Proc. Natl. Acad. Sci. USA 44, 344–349.
Cabanes-Macheteau, M., Fitchette-Laine, A.-C., Loutelier-Bourhis, C., Lange, C., Vien, N. D.,
Ma, J. K. C., Lerouge, P., and Faye, L. (1999). N-Glycosylation of a mouse IgG expressed
ion transgenic tobacco plants. Gycobiology 9(4), 365–372.
Caramelo, L., Martinez, M. J., and Martinez, A. T. (1999). A search for lignolytic peroxidases in
the fungus Pleurotus eryngii involving alpha-keto-gamma thiomethylbutyric acid and lignin
model dimers. Appl. Env. Micro. 65, 916–922.
Chargelegue, D., Vine, N. D., van Dolleweerd, C. J., Drake, P. M., and Ma, J. K. (2000).
A murine monoclonal antibody produced in transgenic plants with plant-specific glycans is
not immunogenic in mice. Transgenic Res. 9, 187–194.
Chen, C.-H., Lin, H.-J., Ger, M.-J., Chow, D., and Feng, T.-Y. (2000). The cloning and
characterization of a hypersensitive response assisting protein that may be associated
with the harpin-mediated hypersensitive response. Plant Mol. Biol. 43, 429–438.
Chikwamba, R. K., Scott, M. P., Mejia, L. B., Mason, H. S., and Wang, K. (2003). Localization
of a bacterial protein in starch granules of transgenic maize kernels. Proc. Natl. Acad. Sci.
USA 100, 11127–11132.
Chilton, M. D. (2001). Agrobacterium. A memoir. Plant Phys. 125, 9–14.
Chilton, M. D., Drummond, M. H., Merlo, D. J., Sciaky, D., Montoya, A. L., Gordon, M. P.,
and Nester, E. W. (1977). Stable incorporation of plasmid DNA into higher plant cells: The
molecular basis of crown gall tumorigenesis. Cell 11, 263–271.
Chu, L., and Robinson, D. K. (2001). Industrial choices for protein production by large-scale
cell culture. Curr. Opin. Biotechnol. 12, 180–187.
Daniell, H., Khan, M. S., and Allison, L. (2002). Milestones in chloroplast genetic engineering:
An environmentally friendly era in biotechnology. Trends Plant Sci. 7, 84–91.
Daniell, H., Streatfield, S. J., and Wycoff, K. (2001). Medical molecular farming: Production of
antibodies, edible vaccines and biopharmaceuticals in plants. Trends Plant Sci. 6, 219–226.
de Boer, D., Cremers, F., Teertstra, R., Smits, L., Hille, J., Smeekens, S., and Weisbeek, P.
(1988). In vivo import of plastocyanin and a fusion protein into developmentally different
plastids of transgenic plants. EMBO J. 7, 2631–2635.
De Jong, E., de Vries, F. P., Field, J. A., van der Zwan, R. P., and de Bont, J. A. M. (1992). Isolation
and screening of basidiomycetes with high peroxidative activity. Mycol. Res. 96, 1098–1104.
Delaney, D. E. (2002). Choice of crop species and development of transgenic product lines.
In ‘‘Plants as Factories for Protein Production’’ (E. E. Hood and J. A. Howard, Eds.),
pp. 139–158. Kluwer Academic Publishers, The Netherlands.
Delaney, D., Jilka, J., Barker, D., Irwin, P., Poage, M., Woodard, S., Horn, M., Vinas, A.,
Beifuss, K., Barker, M., Wiggins, B., Drees, C., Harkey, R., Nikolov, Z., Hood, E., and
Howard, J. (2003). Production of aprotinin in transgenic maize seeds for the pharmaceutical
and cell culture markets. In ‘‘Plant Biotechnology 2002 and Beyond: Proceedings of the 10th
IAPTC&B Congress, Orlando, Florida’’ (I. K. Vasil, Ed.), pp. 393–394. Kluwer Academic
Esau, K. (1977). ‘‘Anatomy of Seed Plants’’. John Wiley and Sons, New York, NY.
Evangelista, R. L., Kusnadi, A. R., Howard, J. A., and Nikolov, Z. L. (1998). Process and
economic evaluation of the extraction and purification of recombinant b-glucuronidase
from transgenic corn. Biotechnol. Prog. 17, 168–174.
Farran, I., Sanchez-Serranno, J. J., Medina, J. F., Prieto, J., and Mingo-Castel, A. M. (2002).
Targeted expression of human serum albumin to potato tubers. Transgenic Res. 11(4),
Faye, L., Johnson, K. D., Sturm, A., and Chrispeels, M. J. (1989). Structure, biosynthesis, and
function of asparagine-linked glycans on plant glycoproteins. Physiologica Plantarum 75,
Fischer, R., and Emans, N. (2000). Molecular farming of pharmaceutical proteins. Transgenic
Res. 9, 279–299.
Fischer, R., Drossard, J., Schillberg, S., and Emans, N. (1999). Towards molecular farming in
the future: Moving from diagnostic protein and antibody production in microbes to plants.
Biotechnol. Appl. Biochem. 30, 101–108.
Garcia-Casado, G., Sanchez-Monge, R., and Chrispeels, M. J. (1996). Role of complex
asparagines-linked glycans in the allergenicity of plant glycoproteins. Glycobiology 6,
Giddings, G. (2001). Transgenic plants as protein factories. Curr. Opin. Biotechnol. 12, 450–454.
Grill, L. K., Lindbo, J., Pogue, G. P., and Turpen, T. H. (2002). Viral vector expression of
foreign proteins in plants. In ‘‘Plants as Factories for Protein Production’’ (E. E. Hood and
J. A. Howard, Eds.), pp. 3–16. Kluwer Academic Publishers, The Netherlands.
Hare, P. D., and Chua, N.-H. (2002). Excision of selectable marker genes from transgenic
plants. Nature Biotechnol. 20, 575–579.
Hood, E. E. (1999). Analysis of plant transformation systems. In ‘‘Application of Transformation Technology in Plant Breeding, Proceedings of the 30th Annual Symposium of the
Korean Breeding Society’’, pp. 33–38. Suwon, Korea, November 18–21, 1999.
Hood, E. E. (2002). From green plants to industrial enzymes. Enzyme Microbial Technol. 30,
Hood, E. E. (2003). Selecting the fruits of your labors. Trends Plant Sci. 8, 357–358.
Hood, E. E., and Howard, J. A. (1999). Protein products from transgenic plants. Agro-Food
Industry Hi-Tech. 3(10), 35–36.
Hood, E. E., and Jilka, J. M. (1999). Plant-based production of xenogenic proteins. Curr. Opin.
Biotechnol. 10, 382–386.
Hood, E. E., and Howard, J. A. (Eds.) 2002. ‘‘Plants as Factories for Protein Production’’.
Kluwer Academic Publishers, Dordrecht, The Netherlands.
Hood, E. E., and Woodard, S. L. (2002). Industrial proteins produced from transgenic plants.
In ‘‘Plants as Factories for Protein Production’’ (E. E. Hood and J. A. Howard, Eds.),
pp. 119–138. Kluwer Academic Publishers, Dordrecht, The Netherlands.
Hood, E. E., Bailey, M. R., Beifuss, K., Horn, M., Magallanes-Lundback, M., Drees, C.,
Delaney, D. E., Clough, R., and Howard, J. A. (2003b). Criteria for high-level expression
of a fungal laccase gene in transgenic maize. Plant Biotechnol. J. 1, 129–140.
Hood, E. E., Horn, M. E., and Howard, J. A. (2003a). Production and application of proteins
from transgenic plants. In ‘‘Plant Biotechnology 2002 and Beyond. Proceedings of the 10th
IAPTC&B Congress, Orlando, Florida’’ (I. K. Vasil, Ed.), pp. 377–382. Kluwer Academic
Publishers, The Netherlands.
Hood, E. E., Kusnadi, A., Nikolov, Z., and Howard, J. A. (1999). Molecular farming of industrial
proteins in transgenic maize. In ‘‘Chemicals via Higher Plant Bioengineering’’ ( Shahidi et al.,
Eds.), pp. 127–147. Kluwer Academic/Plenum Publishers, New York, NY.
Hood, E. E., Witcher, D. R., Maddock, S., Meyer, T., Baszcynski, C., Bailey, M., Flynn, P.,
Register, J., Marshall, L., Bond, D., Kulisek, E., Kusnadi, A., Evangelista, R., Nikolov, Z.,
Wooge, C., Mehigh, R. J., Hernan, R., Kappel, W. K., Ritland, D., Li, C. P., and Howard,
J. A. (1997). Commercial production of avidin from transgenic maize: Characterization of
transformant, production, processing, extraction and purification. Mol. Breeding 3, 291–306.
Hood, E. E., Horn, M. E., and Woodard, S. L. (2002). Monoclonal antibody manufacturing in
transgenic plants-myths and realities. Curr. Opin. Biotechnol. 13, 630–635.
Houdebaine, L. M. (2000). Transgenic animal bioreactors. Transgenic Res. 9, 279–299.
Howard, J. A., and Donnelly, K. C. (2004). A quantitative safety assessment model for nonfood
products produced in agricultural crops. J. Agric. and Environ. Ethics 17(6).
Jensen, L. G., Olsen, O., Kops, O., Wolf, N., Thomsen, K. K., and von Wettstein, D. (1996).
Transgenic barley expressing a protein-engineered, thermostable (1,3-1,4)-b-glucanase
during germination. Proc. Natl. Acad. Sci. USA 93, 3487–3491.
Keegstra, K., and Cline, K. (1999). Protein import and routing systems of chloroplasts. Plant
Cell 11, 557–570.
Kermode, A. (1996). Mechanisms of intracellular protein transport and targeting in plant cells.
Crit. Rev. Plant Sci. 15, 285–423.
Kleine-Tebbe, J., Vogel, L., Crowell, D. N., Haustein, U. F., and Vieths, S. (2002). Sever oral
allergy syndrome and anaphylactic reactions caused bya Bet v 1-related PR-10 protein in
soybean, SAM22. J. Allergy Clin. Immunol. 110(5), 797–804.
Koehn, M. L. (Ed.) 1978. ‘‘Agricultural Statistics, 1978’’. U.S. Department of Agriculture, U.S.
Government Printing Office, Washington, DC.
Komari, T., Hiei, Y., Saito, Y., Murai, N., and Kumashiro, T. (1996). Vectors carrying two
separate T-DNAs for co-transformation of higher plants mediated by Agrobacterium tumefaciens and segregation of transformants free from selection markers. Plant J. 10, 165–174.
Kong, Q., Richter, L., Yang, Y. F., Arntzen, C. J., Mason, H. S., and Thanavala, Y. (2001).
Oral immunization with hepatitis B surface antigen expressed in transgenic plants. Proc.
Natl. Acad. Sci. USA 98, 11539–11544.
Kusnadi, A. R., Evangelista, R. L., Hood, E. E., Howard, J. A., and Nikolov, Z. L.
(1998a). Processing of transgenic corn seed and its effect on the recovery of recombinant b-glucuronidase. Biotechnol. Bioeng. 60(1), 45–52.
Kusnadi, A. R., Hood, E. E., Witcher, D. R., Howard, J. A., and Nikolov, Z. L. (1998b).
Production and purification of two recombinant proteins from transgenic corn. Biotechnol.
Prog. 14, 149–155.
Kusnadi, A. R., Nikolov, Z. L., and Howard, J. A. (1997). Production of recombinant proteins
in transgenic plants: Practical considerations. Biotechnol. Bioeng. 56, 473–484.
Lamphear, B. J., Jilka, J. M., Kesl, L., Welter, M., Howard, J. A., and Streatfield, D. J. (2004).
A corn-based delivery system for animal vaccines: An oral transmissible gastroenteritis in
virus vaccine boosts lactogenic immunity in swine. Vaccine 22, 2420–2424.
Lamphear, B. J., Streatfield, S. J., Jilka, J. M. et al. (2002). Delivery of subunit vaccines in maize
seed. J. Control. Release 85, 169–180.
Lamphear, B. J., Barker, D. K., Brooks, C. A., DeLaney, D. E., Lane, J. R., Beifurs, K., Love,
R., Thompson, K., Mayor, J., Clough, R., Harkey, R., Poage, M., Drees, C., Horn, M.,
Strentfield, S., Nikolov , Z., Woodard, S., Hood, E., Jilka, J., and Howard, J. (in press).
Expression of the Sweet Protein Brazzein in Maize for production of a new commercial
sweetener. Plant Biotechnol. J.
Lawrence, S. D., Kline, K., and Moore, G. A. (1997). Chromoplast development in ripening
tomato fruit: Identification of cDNAs for chromoplast-targeted proteins and characterization of a cDNA encoding a plastid-localized low-molecular-weight heat shock protein.
Plant Mol. Biol. 33(3), 483–492.
Lerouge, P., Cabanes-Macheteau, M., Rayon, C., Fischette-Laine, A. C., Gomord, V., and
Faye, L. (1998). N-glycoprotein biosynthesis in plants: Recent developments and future
trends. Plant Mol. Biol. 38, 31–48.
Lindbo, J. A., Fitzmaurice, W. P., and della-Cioppa, G. (2001). Virus-mediated reprogramming
of gene expression in plants. Curr. Opin. Plant Biol. 4, 181–185.
Ma, J. K.-C., Drake, P. M. W., and Christou, P. (2003). The production of recombinant
pharmaceutical proteins in plants. Nat. Rev. 4, 794–805.
Ma, J. K.-C., Hikmat, B. Y., Wycoff, K., Vine, N. D., Chargelegue, D., Yu, L., Hein, M. B., and
Lehner, T. (1998). Characterization of a recombinant plant monoclonal secretory antibody
and preventive immunotherapy in humans. Nat. Med. 4, 601–606.
Maliga, P. (2002). Engineering the plastid genome of higher plants. Curr. Opin. Plant Biol.
5, 164–172.
Merot, B., Dieryck, W., Philippe, L., Marden, M., Veronique, G., Pagnier, R.-J., Baudino, S.,
and Poyart, C. (2002). Method for producing human hemoglobin proteins using plant cells.
US Patent No. 6,344,600.
Nikolov, Z., and Hammes, D. (2002). Production of recombinant proteins from transgenic
crops. In ‘‘Plants as Factories for Protein Production’’ (E. E. Hood and J. A. Howard,
Eds.), pp. 159–174. Kluwer Academic Publishers, The Netherlands.
Pyke, K. A. (1999). Plastid division and development. Plant Cell 11, 549–556.
Rasmussen, J. R. (1991). Gycosylation of recombinant proteins. In ‘‘Biology of Carbohydrates’’, pp. 179–285. JAI Press Ltd, London.
Rihs, H. P., Chen, Z., Rueff, F., Petersen, A., Rozynek, P., Heimann, H., and Baur, X. (1999).
IgE binding of the recombinant allergen soybean profilin (rGly m 3) is mediated by
conformational epitopes. J. Allergy Clin. Immunol. 104(6), 1293–1301.
Samyn-Petit, B., Gruber, V., Flauhaut, C., Wajda-Dubos, J.-P., Farrer, S., Pons, A.,
Desmaizieres, G., Slomianny, M.-C., Theisen, M., and Delannoy, P. (2001). N-Glycosylation potential of maize: The human lactoferrin used as a model. Glyconjugate J. 18,
Sanderfoot, A. A., and Raikhel, N. V. (1999). The specificity of vesicle trafficking: Coat proteins
and SNAREs. Plant Cell 11, 629–641.
Schillberg, S., Fischer, R., and Emans, N. (2003). Molecular farming of recombinant antibodies
in plants. Cell. Mol. Life Sci. 60, 433–445.
Schwartz, J. R. (2001). Advances in Escherichia coli production of therapeutic proteins. Curr.
Opin. Biotechnol. 12, 195–201.
Shah, M. M., Fujiyama, K., Flynn, C. R., and Joshi, L. (2003). Sialylated endogenous glycoconjugates in plant cells. Nature Biotechnol. 21(12), 1470–1471.
Sijmons, P. C., Dekker, B. M., Schrammeijer, B., Verwoerd, T. C., van den Elzer, P. J., and
Hoekema, A. (1990). Production of correctly processed human serum albumin in transgenic
plants. Biotechnology 8(3), 217–221.
Songstad, D. D., Somers, D. A., and Griesbach, R. J. (1995). Advances in alternative DNA
delivery techniques. Plant Cell Tissue Organ Cult. 40, 1–15.
Southgate, E. M., Davey, M. R., Power, J. B., and Westcott, R. J. (1998). A comparison of methods
for direct gene transfer into maize (Zea mays, L.). In Vitro Cell Biol. Plant 34, 218–224.
Stoger, E., Vaquero, C., Torres, E., Sack, M., Nicholson, L., Drossard, J., Williams, S., Keen, D.,
Perrin, Y., and Christou, P. (2000). Cereal crops as viable production and storage systems for
pharmaceutical scFv antibodies. Plant Mol. Biol. 42, 583–590.
Streatfield, S. J., Jilka, J. M., Hood, E. E., Turner, D. D., Bailey, M. R., Mayor, J. M.,
Woodard, S. L., Beifuss, K. K., Horn, M. E., Delaney, D. D., Tizard, I. R., and Howard,
J. A. (2001). Plant-based vaccines: Unique advantages. Vaccine 19, 2742–2748.
Streatfield, S. J., Mayor, J. M., Barker, D. K., Brooks, C., Lamphear, B. J., Woodard, S. L.,
Beifuss, K. K., Vicuna, D. V., Massey, L. A., Horn, M. E., Delaney, D. D., Nikolov, Z. L.,
Hood, E. E., Jilka, J. M., and Howard, J. A. (2002). 2000 Congress Symposium on
Molecular Farming: Development of an edible subunit vaccine in corn against enterotoxigenic strains of Escherichia coli. In Vitro Cell. Dev. Biol. Plant 38, 11–17.
Streatfield, S. (2002). The greening of vaccine technology. Curr. Drug Disc. November, 15–18.
Streatfield, S. J., Lane, J. R., Brooks, C. A. et al. (2003). Corn as a production system for human
and animal vaccines. Vaccine 21, 812–815.
Streatfield, S. J., and Howard, J. A. (2003a). Plant-based vaccines. Int. J. Parasitol. 33, 479–493.
Streatfield, S. J., and Howard, J. A. (2003b). Plant production systems for vaccines. Expert Rev.
Vaccines 2(6), 763–775.
Tacket, C. O., Mason, H. S., Losonsky, G., Clements, J. D., Levine, M. M., and Arntzen, C. J.
(1998). Immunogenicity in humans of a recombinant bacterial antigen delivered in a
transgenic potato. Nat. Med. 4, 607–609.
Tacket, C. O., Mason, H. S., Losonsky, G., Este, M. K., Levine, M. M., and Arntzen, C. J.
(2000). Human immune responses to a novel Norwalk virus vaccine delivered in transgenic
potatoes. J. Infect. Dis. 182, 302–305.
Tacket, C. O., Pasetti, M., Edelman, R., Clements, J. D., Howard, J. A., and Streatfield, S.
(2004). Immunogenicity of recombinant LT-B delivered orally to humans in transgenic
corn. Vaccine 22, 4385–4389.
Twyman, R. M., Stoger, E., Schillberg, S., Christou, P., and Fischer, R. (2003). Molecular
farming in plants: Host systems and expression technology. Trends Biotechnol. 21, 570–578.
van Ree, R., Cabanes-Macheteau, M., Akkerdaas, J., Milazzo, J.-P., Loutelier-Bourhis, C.,
Rayon, C., Villalba, M., Koppelmans, S., Aalbers, R., Rodriguez, R., Faye, L., and
Lerouge, P. (2000). b(1,2)-Xylose and a(1,3)-fucose residues have a strong contribution
in IgE binding to plant glycoallergens. J Biol. Chem. 75(15), 11451–11458.
van Ree, R., van der Wal, J.-W., Pen, J., and Aalberse, R. C. (1994). Expression of a- amylase
from Bacillus licheniformis in transgenic tobacco introduces IgE-binding N-glycans.
In ‘‘Specificity of Grass Pollen Allergens’’, pp. 155–142. Doctorate Dissertation, Universite
it van Amsterdam.
Vitale, A., and Denecke, J. (1999). The endoplasmic reticulum—Gateway of the secretory
pathway. Plant Cell 11, 615–628.
Wan, J. X., Blakeley, S. D., Dennis, D. T., and Ko, K. (1996). Transit peptides play a major role
in the preferential import of proteins into leucoplasts and chloroplasts. J. Biol. Chem.
271(49), 31227–31233.
Wang, A. S., Evans, R. A., Altendorf, P. R., Hanten, J. A., Doyle, M. C., and Rosichan, J. L.
(2000). A mannose selection system for production of fertile transgenic maize plants from
protoplasts. Plant Cell Rep. 19, 654–660.
Watson, S. A. (1988). Corn marketing, processing and utilization. In ‘‘Corn and Corn Improvement’’ (G. F. Sprague and J. W. Dudley, Eds.), 3rd ed., pp. 885–942. No. 18, American
Society of Agronomy, Inc., Crop Science Society of America, Inc., Soil Science Society of
America, Inc., Madison, WI.
Wilmink, A., and Dons, J. J. M. (1993). Selective agents and marker genes for use in transformation of monocotyledonous plants. Plant Mol. Biol. Rptr. 11, 165–185.
Witcher, D. R., Hood, E. E., Peterson, D., Bailey, M., Bond, D., Kusnadi, A., Evangelista, R.,
Nikolov, Z., Wooge, C., Mehigh, R., Kappel, W., Register, J., and Howard, J. A. (1998).
Commercial production of B-glucuronidase (GUS): A model system for the production of
proteins in plants. Mol. Breeding 4, 301–312.
Woodard, S. L., Mayor, J. M., Bailey, M. R., Barker, D. K., Love, R. T., Lane, J. R., Delaney,
D. D., McComas-Wagner, J. M., Mallubhotla, H. D., Hood, E. E., Dangott, L. J., Tichy,
S. E., and Howard, J. A. (2003). Maize-derived bovine trypsin: Characterization of the first
large-scale, commercial protein product from transgenic plants. Biotechnol. Appl. Biochem.
38, 123–130.
You, S. J., Liau, C. H., Huang, H. E., Feng, T. Y., Prasad, V., Hsiao, H. H., Lu, J. C., and
Chan, M. T. (2003). Sweet pepper ferredoxin-like protein (pflp) gene as a novel selection
marker for orchid transformation. Planta 217, 60–65.
Zhong, G.-Y., Peterson, D., Delaney, D. E., Bailey, M., Witcher, D. R., Register, J. C. III,
Bond, D., Lin, C.-P., Marshall, L., Kulisek, E., Ritland, D., Meyer, T., Hood, E. E., and
Howard, J. A. (1999). Commercial production of aprotinin in transgenic maize seeds. Mol.
Breeding 5, 345–356.
Zitouni, N., Errahali, Y., Metche, M., Kanny, G., Moneret-Vautrin, D. A., Nicolas, J. P., and
Fremont, S. (2000). Influence of refining steps on trace allergenic protein content in
sunflower oil. J. Allergy Clin. Immunol. 106(5), 962–967.
D. M. Oliver,1,2 C. D. Clegg,1 P. M. Haygarth1 and A. L. Heathwaite3
Soil Science and Environmental Quality Team, North Wyke Research Station,
Okehampton, Devon EX20 2SB, United Kingdom
Department of Geography, University of Sheffield, Sheffield S10 2TN,
United Kingdom
Centre for Sustainable Water Management,
The Lancaster Environmental Centre, Lancaster University,
Lancaster LA1 4YQ, United Kingdom
I. Introduction
II. Pathogens in Livestock Wastes
A. Bacteria
B. Protozoa
C. Viruses
III. Detection and Enumeration Techniques
A. Culture-Based Methods
B. Direct Counting Approaches
C. Molecular Methods
IV. Transfer from Soil to Water
A. Lateral Surface Pathways
B. Matrix Flows
C. Soil Retention EVects
D. Bypass Mechanisms in Soil
E. Movement via Growth and Motility
F. The Role of Soil Mesofauna
V. The Role of Colloids in Facilitating Transfer
VI. Factors AVecting Survival
A. Survival in Livestock Wastes
B. Survival in Soil
C. Survival in Water
VII. Concluding Remarks
Contamination of surface waters with pathogenic micro-organisms is an
area of growing importance in the context of diVuse agricultural pollution.
Hydrological pathways linking farmed land to receiving waters may operate
Advances in Agronomy, Volume 85
Copyright 2005, Elsevier Inc. All rights reserved.
0065-2113/05 $35.00
as vectors of disease transmission. RunoV from grassland systems may be
particularly important. In this chapter, we synthesize and evaluate recent and
contextual studies relating to the issue. The chapter is necessarily wide ranging
and interdisciplinary but we have focused largely on the hydrological, soilbased, and microbiological perspectives. The potential for pathogen presence
in livestock wastes is demonstrated through prevalence studies, and
subsequent loading of grasslands with contaminated wastes generates a potential surface store of pathogens. These microbes may then be transferred to
the wider environment when source and transport drivers are combined in, for
example, precipitation events. The delivery of contaminated agricultural
drainage waters into first order streams may impact the quality and ecological
balance of watercourses if the micro-organisms of concern are still viable. This
chapter evaluates both die-oV and transfer processes operating from source
through to the end point receptors in surface waters. Gaps in knowledge are
identified and appear to be due to the contribution of heterogeneity and
hydrological complexity of agricultural catchments and the complications of
ß 2005, Elsevier Inc.
prevalence data derived via a range of methodologies.
The spread of human pathogens throughout populations is a major health
concern (Sharma et al., 2003; Theron and Cloete, 2002), drawing attention
from scientific and government bodies in the interest of public welfare (Ketley,
1997; Miettinen et al., 2001). Surface water can function as a vector of disease
transmission and so human health may be compromised as a result of contaminated receiving waters being used as sources of potable water, for recreational purposes, and to irrigate food crops. Linked to health issues is absence
from work due to resulting gastrointestinal illness, which is estimated to cost
the UK economy over £1 billion per annum (Jones, 1999) and the United
States approximately $20 billion in lost productivity each year (Gerba, 1996).
Calculations using U.S. medical data suggest that Escherichia coli 0157 alone
may cost the United Kingdom around £30 M annually in healthcare (Jones,
1999), and water-borne micro-organisms are now responsible for approximately one-quarter of hospital patients throughout the world (Gerba, 1996).
Coupled with this problem is the potential impact on commercial enterprises
associated with contaminated recreational waters and implications associated
with European legislation, such as the EU Bathing Waters Directive.
One industry thought to play a significant role in the dissemination of
pathogens through the environment is agriculture, in particular, grassland
livestock farming, because of the carriage of such micro-organisms within
infected animals (Nicholson et al., 2000). Livestock wastes, if used eYciently
and managed eVectively, provide a host of benefits to both farmers and the
environment. Recycling livestock waste to land curbs excessive fecal waste
accumulation in farm storage systems (Unwin et al., 1986), builds soil quality,
and returns valuable nutrients back to grassland systems (Hooda et al., 2000).
However, there are a number of more serious pollutant problems linked with
‘‘washoV’’ of surface applied wastes that may impact the quality and ecological balance of watercourses. Such problems are increasingly prominent within modern farming, in which ‘‘recycling’’ of farm wastes has more and more
become an adopted ‘‘disposal’’ strategy (Chadwick and Chen, 2002). It is this
routine procedure of disposal alongside the unavoidable grazing seasons that
threatens contamination of drainage waters with enteric pathogens such as,
for example, E. coli 0157, Salmonella spp, Campylobacter spp, and Cryptosporidium parvum. Consequently, there is a potential linkage between agricultural practices and pathogen dissemination through the environment,
driven by hydrological events. While much eVort has been exerted in studying
the food chain links of pathogen movement and epidemiology (Locking et al.,
2001), these hydrological drivers of pathogen transfer in agricultural systems
are an alternative route of disease transmission that warrant further examination to highlight the potential pathways for contamination other than
those via the food chain. As a result, the farm environment has come under
increased scrutiny as a vector for disease transmission (Aitken, 2003; Jones,
2001; Mawdsley et al., 1995; Stanley and Jones, 2003; Trevena et al., 1999).
This Chapter draws on the available literature to identify the current state
of our knowledge with regard to the introduction of micro-organisms to
grassland systems via farm waste products. A framework of key research
needs, with regard to both transfer and survival, is presented in Fig. 1. This is
an area of emerging importance in the context of diVuse pollution associated
with disease-causing micro-organisms, and what is certain is that this is a
major issue in grassland ecology and catchment water quality.
The rumen and digestive tract of agricultural livestock is host to a rich
diversity of microflora and therefore can also act as a reservoir for pathogenic micro-organisms (Rasmussen et al., 1993). A consequence of this is
that micro-organisms such as E. coli 0157, Salmonella spp., Campylobacter
jejuni, Listeria monocytogenes, C. parvum, and Giardia intestinalis may contaminate fecal deposits and livestock wastes that are excreted and applied
onto grasslands, respectively. As a result, it is inevitable that the soil system
will, at least periodically, harbour fecally derived microbes both at the soil
surface and within the network pore structure. In addition to pathogenic
bacteria and protozoa, viruses such as Rotavirus have also been recorded in
livestock wastes (Lund and Nissen, 1983); however, few zoonotic viruses
Figure 1 Framework of key research needs.
infect cattle (Pell, 1997), so this chapter centers on bacterial and protozoan
pathogens. The fecally derived micro-organisms of concern within grassland
agriculture are detailed in the following.
1. Escherichia coli
E. coli are short, Gram-negative, enteric bacteria that are common inhabitants of the mammalian gut. While most strains of E. coli are harmless,
a few strains are potentially pathogenic, such as strains 0157, 0111, 026,
0103, and 0145. Verotoxin-producing E. coli 0157 is considered to be the
most severe strain and was first identified as a major disease-causing microorganism in 1982, symptomatically ranging from mild diarrhea through to,
in extreme cases, life-threatening illnesses.
E. coli 0157 has been identified in many cattle worldwide, with the
incidence of detection varying between studies. For example, up to 16% of
cattle in UK (Chapman et al., 1997) and Finnish (Lahti et al., 2003) herds
and 1.9–5% of U.S. cattle (Sargeant et al., 2000; Zhao et al., 1995) were
reportedly infected, and in the southern hemisphere, the incidence of E. coli
infection is as low as 0.2% (Hallaran and Sumner, 2001). However, work by
the U.S. Department of Agriculture (USDA) (1997) has identified the sporadic nature and seasonal diVerences in carriage of E. coli 0157 in cattle
because of the varying incidence among herds that can range up to 100% if
studies repeatedly sample through time.
While much information has been gleaned from studies of E. coli 0157
food infections, there is still much to be understood about routes of transmission though soil and water. The current UK annual infection rate in the
human population is estimated at approximately 2 cases per 100,000
(Jones, 1999), and is low in comparison with Salmonella and Campylobacter
infections. However, the rapid rise in the number of cases and complications associated with this pathogen have focused increased attention on the
severity of risk associated with E. coli 0157.
Salmonella spp.
Salmonella spp. are Gram-negative facultative aerobic rods closely related
to E. coli and Shigella spp. Salmonella spp. are commonly found in the gut of
animals, and many strains are pathogenic to humans. Cells of Salmonella
spp. are capable of prolonged survival in unfavorable conditions outside
their host (Winfield and Groisman, 2003). Although the main reservoir of
Salmonella spp. is generally considered to be poultry, cattle also harbor this
pathogen (Huston et al., 2002; McEvoy et al., 2003), with reported shedding
rates of up to 109 cells per day in feces of chronically aVected cattle
(McGuirk and Peek, 2003). The reported numbers of cattle harboring
Salmonella spp. indicate some variability, with infection rates generally
between 1% (Zibilske and Weaver, 1978) and 10% (Heinonen-Tanski et al.,
1998) of cattle.
The increased amount of research associated with Salmonella spp. in
poultry has led to an improved understanding and management regarding
the control of this pathogen that has consequently resulted in a decrease in
the number of reported cases of salmonellosis in the United Kingdom
(Mawdsley et al., 1995). However, this bacterium still causes many problems
throughout the world and is estimated to cost the U.S. economy up to $3
billion each year (ERS/USDA, 2003).
Campylobacter spp.
Campylobacter spp. are microaerophilic, Gram-negative curved rods and
have recently emerged as a major human gastrointestinal pathogen (Ketley,
1997) that is now recognized as one of the most common causes of gastroenteritis. For example, in the United States, Campylobacter spp. are responsible for the greatest number of food-borne diseases, with almost 2 million
cases of infection per year (Madigan et al., 2003). Campylobacters are
common to the intestinal tract of humans and animals, and the strain
associated with most reported human infections, C. jejuni, causes greater
than 90% of Campylobacter enteritis in the United Kingdom (Stanley and
Jones, 2003).
An investigation recently carried out in Denmark determined that 23%
of all animals tested within 24 Danish dairy herds were positive for thermophilic Campylobacters (Nielsen, 2002). In U.S. dairy herds, C. jejuni
was identified in 37.7% of fecal samples of animals taken across a number
of states (Wesley et al., 2000), and 89.4% of British beef cattle at slaughter
were reportedly infected with Campylobacter (Stanley et al., 1998). Some
caution should be taken with the interpretation of prevalence results relating to Campylobacter infection, as some discrepancies may exist in
data because of problems with sample collection and location. For example, the prevalence of Campylobacter spp. within 25 Californian beef
herds was 5% for rectal samples; however, the prevalence in fecal
samples was only 0.5% (Hoar et al., 1999), suggesting inactivation of
cells post-defecation.
Listeria monocytogenes
L. monocytogenes is a Gram-negative facultative anaerobic rod that
is commonly found in soil and water and is emerging as an important
food-borne pathogen (Bassler et al., 1995). Although L. monocytogenes is
more often associated with dairy product contamination, this pathogen
has been identified in cattle feces (Pell, 1997) and so may present a potential
risk through environmental routes of transmission. Jones (1999) quotes
public health laboratory data that state that approximately 130 cases
of food poisoning in the United Kingdom during 1997 were related to
Listeria spp.
Cryptosporidium parvum
C. parvum is a zoonotic parasite that infects the gastrointestinal tract
of warm-blooded animals, including humans, causing the disease Cryptosporidiosis. Cattle feces are a primary source of Cryptosporidium, with
estimated maximum oocyst shedding by Californian beef cattle ranging
between 3 103 to 2.3 105 oocysts per cow per day (Atwill et al., 2003;
Hoar et al., 2000). The prevalence of Cryptosporidium infection among cattle
within German and Canadian herds may range from 36% (Joachim et al.,
2003) to 100% (O’Handley et al., 1999).
Giardia spp.
Giardia spp. are a single-celled parasite that can cause the disease Giardiasis through the consumption of fecally contaminated water; it is common in
domestic and wild animals (Olson et al., 1997). The prevalence of Giardia
among livestock can diVer, with 38% and 29% incidence rates reported for
sheep and cattle, respectively, and younger animals identified as having a
greater incidence of Giardia infection (Olson et al., 1997). Giardia and Cryptosporidium combined were responsible for one-third of drinking waterassociated disease outbreaks during 1993 to 1994 (Kramer et al., 1996),
which highlights the importance of the protozoan pathogen community.
Viruses are obligate intracellular parasites composed of genetic elements
and those of an enteric origin that are shed by grazing animals. They may be
of concern in the soil and aquatic habitat, where they may be able to persist
extracellularly and remain viable. Enteric viruses in polluted waters are not
well documented, although the few studies that have addressed this issue
have identified some, such as bovine enterovirus, within livestock polluted
sources (Lund and Nissen, 1983). Although viral pathogens sourced from
livestock wastes represent a much reduced health risk to humans than
bacterial and protozoan pathogens (Nicholson et al., 2000), the real risk
remains unclear. Viruses in soils and water sources pose a greater risk to
human health when human sewage sludge is applied to land. Routes of
transfer of viral particles derived from livestock wastes will therefore be
useful to demonstrate the transfer pathways available to human viruses
dissipated from surface applied sewage sludge. Such viruses may include
Rotavirus, Calicivirus, and Enteroviruses such as Poliovirus and Coxsackievirus. In this chapter we do not consider viruses further (see review by
Jin and Flury [2002] for a discussion of virus fate and transport in porous
While issues of pathogen carriage rates have been discussed, caution
should be taken when assessing prevalence data within livestock and herds.
Enumeration techniques in any discipline are only as sensitive as their limit of
detection, and we should bear in mind that when animals within a herd test
negative with respect to a pathogen of concern, this result may potentially be
a false negative result, because the target pathogen is actually present in
numbers below the detection level of the technique. This is important because
the potential to infect other animals in the herd through the subsequent
contamination of pasture and drinking water will exist. This demonstrates
the advantages of prevalence studies that monitor over long periods.
Determining the presence of specific micro-organisms in environmental samples requires reliable enumeration techniques. The techniques currently employed can be divided into methods that rely upon the culturing
of viable micro-organisms, direct counting approaches, and those that utilize
molecular techniques. These are described briefly here; more comprehensive
reviews of detection methods can be found in Rompre et al. (2002) and
APHA (1998).
The most commonly used approach to assess bacterial cell numbers in soil
and water samples is standard plate-counting techniques through the enumeration of colony forming units (cfu’s), the most probable number (MPN)
method, and membrane filtration (MF) techniques. These approaches rely
on the ability of the target population in a sample to grow on solid agar or in
a liquid culture media. Table I summarizes the advantages and limitations
associated with the classic methods of bacterial cell detection.
1. Plate Counts
The plate count method provides a quick, easy, and inexpensive method
of bacterial enumeration. Soil and water samples are spread plated onto
specific solid agar media and usually are incubated for 24 h at temperatures
Table I
Summary of the Range of Classical Detection Techniques Available for Bacterial Micro-organisms
Plate count/Indicator
Massa et al. (2001)
A reliable alternative to MPN
and MF method. Method of
choice in polluted waters due to
its economic advantages in terms
of space, time, and materials.
Considerable imprecision
is inherent in plate counts,
therefore common practice
to replicate delivery to culture plates.
Competition from antagonistic
organisms. Poor detection of slowgrowing or VBNC micro-organisms.
Unresolved problems associated with
identifying most appropriate
agar medium.
Advantageous with soil samples
as avoids both clogging of
membrane pores and colony
spread around soil particles.
Works well with turbid and
colored waters.
Particularly useful for low
concentrations of organisms.
Hedges (2002)
Rompre et al. (2002)
Stoddard et al. (1998)
Stoddard et al. (1998)
Rompre et al. (2002)
Shipe and Cameron (1954)
(continued )
Table I (Continued)
Greater recovery than MF method.
MF/Indicator organisms
Massa et al. (2001)
Bissonnette et al. (1977)
Volumes filtered in cases where cell
counts are likely to be low.
Greater eYciency than MPN in
Produces more reproducable
Massa et al. (1989)
Brodsky and
Schiemann (1975)
Van Poucke and
Nelis (2000)
EVorts to further reduce detection
time have seen two-step
procedures involving
fluorescence labelling of colonies
on membrane filters.
Lower sensitivity in recovering coliforms
when compared to MPN.
Unreliable in turbid waters as a result of
colloidal clogging of membrane pores.
Suspended sediments cause spreading
Braswell and
Hoadley (1974)
Jacobs et al. (1986)
Maxey (1970)
Massa et al. (2001)
Competition from antagonistic organisms.
Poor detection of slow growing or VBNC
Unresolved problems associated with
identifying most appropriate medium.
Demanding in terms of time (48–96 h) and
Bissonnette et al. (1977)
Rompre et al. (2002)
MPN, most probable number; MF, membrane filtration; VBNC, viable but nonculturable.
Rompre et al. (2002)
Green et al. (1975)
Lin (1976)
Tobin and Dutka (1977)
Fleisher and
McFadden (1980)
Sharpe and Michaud (1975)
Competition from antagonistic organisms.
Poor detection of slow growing or VBNC
micro-organisms. Recovery performance
diVers between various commercial brands
of membrane filters.
Multiplicative error associated with
the conversion of colony numbers
per 100 ml of sample when
diVerent sample filter
volumes are used. Unresolved
problems associated with
identifying most appropriate
agar medium.
Randomness of their distribution
on the surface of the
filter may complicate counts.
in the range of 35–44 8C, depending on media used, before enumeration.
However, the interpretation of viable count data should be considered
carefully, as recovery of all target organisms may not occur because of the
presence of viable but non-culturable (VBNC) bacteria in samples. In this
VBNC state, organisms are still metabolically active but are no longer able
to grow and divide on conventional media and therefore are unable to
produce colonies (Colwell et al., 1985; Rollins and Colwell, 1986), leading
to an underestimate of the true count. This viable technique is generally
reliable and is widely employed in both the public and private sectors as a
basic routine method for bacterial enumeration.
Most Probable Number
The MPN method is a statistical approach involving the serial dilution of
samples to determine the highest dilution yielding growth. Assessment of the
number of positive samples is then made to derive the MPN of bacteria in
samples through reference to probability tables or computer programs
(Briones and Reichardt, 1999). This method is useful in situations of low
cell densities in samples (less than one viable organism per milliliter), as the
plate count method lacks statistical robustness at such low concentrations
(Herbert, 1990). In addition, Rompre et al. (2002) note that this inexpensive
method proves useful in circumstances in which turbid or colored water
complicates the MF method; however, estimates of bacterial density are
known to vary over a 10-fold range for identical samples (Massa et al.,
2001), highlighting potential limitations of the method.
Membrane Filtration
The MF technique is a widely used method incorporating media-based
specificity to enumerate micro-organisms (APHA, 1998). The method
involves the use of a membrane filter (generally 0.45 mm in diameter) that
is capable of capturing bacteria from liquids. A sample is filtered, and the
filter pad is then transferred to a selective growth media prior to incubation.
Herbert (1990) reported that a major advantage of MF over conventional
plating methods is the speed with which samples can be processed.
More importantly, however, this technique allows large sample volumes
to be filtered in cases where cell numbers are likely to be low. At present,
the MF technique is the most widely adopted method for routine enumeration of coliforms (Rompre et al., 2002). However, as with other
culture-based methods, its ability to recover stressed or injured coliforms is
still not clear.
In conjunction with plate counts and MF, specific enzymatic activity
may be incorporated to enhance their sensitivity. Media containing chromogenic or fluorogenic substrates for the enzymes b-D-galactosidase
and b-D-glucuronidase enable the detection of coliforms and E. coli, respectively, and are being used more and more frequently (Fricker and Fricker,
1996). There are a large number of studies that detail specific substrate
incorporation into media for the analysis of environmental samples during
the culturing stage (Brenner et al., 1993; Byamukama et al., 2000; Clark
et al., 1991; Geissler et al., 2000; George et al., 2001; Rice et al., 1990; Sueiro
et al., 2001). Color and fluorescence production result from cleavage
by specific enzymes of chromogenic and fluorogenic substrates (e.g., indoxyl-b-D-glucuronide, 5-bromo-4-chloro-3-indolyl-b-D-glucuronide, and
4-methylumbelliferyl-b-D-galactopyranoside). For example, E. coli has
been detected in river water through the hydrolysis of the fluorogenic substrate 4-methylumbelliferyl-b-D-glucuronide (George et al., 2001). The
addition of such substrates to both solid and liquid cultivation media not
only improves the sensitivity of culture-based methods but also provides
results more easily and rapidly (Geissler et al., 2000; Sueiro et al., 2001).
It is important to acknowledge the availability of direct counting methods
in addition to those already discussed. Direct microscopy is an accepted and
common methodology for enumerating micro-organisms within soil suspensions, but it is not used routinely for the identification bacteria of fecal
origin. However, standard procedures for detecting protozoan pathogens
in surface waters involve filtration of large volumes of water, followed by
elution with a detergent solution, concentration via centrifugation, and
separation of the protozoa using immunomagnetic techniques (Straub and
Chandler, 2003). For enumeration and identification, the oocysts can then
be stained and visualized using microscopy.
In addition to conventional detection methods involving selective culturing, biochemical methods, and direct counting approaches, there are also
more recently developed molecular techniques, such as DNA hybridizations
and polymerase chain reaction (PCR) detection that can identify specific
micro-organisms through targeting specific gene sequences. Although molecular techniques are generally not used in routine sampling regimes, they
are employed to confirm the identity of isolates for epidemiological studies.
Recent advances in hybridization technology using microarrays may allow
the rapid screening of environmental samples for the presence of specific
pathogens in the near future (Call et al., 2003; Rompre et al., 2002). Currently, PCR amplification of partial or full sequences of genes allows the
detection of low numbers of organisms in samples. For example, E. coli 0157
has been detected in soil at concentrations as low as 10 cfu g 1 (Campbell
et al., 2001), highlighting the usefulness of PCR amplification in detecting
low numbers of pathogens in environmental samples. Through targeting nucleic acid sequences, the need to rely on the expression of specific
physiological and biochemical traits associated with other approaches is
eliminated, thus PCR may be valuable in the identification of those microorganisms that are in a state of nonculturability. Successful application
of PCR has been demonstrated in a number of studies, for example, in the
detection of the listeriolysin O gene, allowing a rapid alternative to standard
techniques in the detection of L. monocytogenes (Bessesen et al., 1990). The
quantification of C. parvum oocysts in municipal water samples has been
investigated using quantitative PCR amplification of a partial gene sequence
for oocyst cell wall production (Chung et al., 1999). The approach resulted
in the enumeration of oocysts within one order of magnitude and demonstrates the diYculties in accurate enumeration of specific micro-organisms in
environmental samples using quantitative PCR-based approaches (Rompre
et al., 2002).
Fecal loading of grasslands provides the potential for transfer of microorganisms, introduced via livestock wastes, to watercourses. If this surface
store of microbes is coupled with hydrological drivers such as storm events,
the risk of transfer is increased. However, such transfer is subject to a series
of spatial and temporal controls (Fig. 2) that dictate this potential for
transfer. There are a variety of available transfer routes through which
potential pathogens may be transported from soil to receiving waters, but
the factors that control the transfer of microbes through soils are not well
understood (Hornberger et al., 1992). The transport mechanisms of microorganisms within soils can be divided into physical, geochemical, and
biological processes (Tim et al., 1988). The physical processes include advection, whereby potential pathogens are carried in bulk water and move
according to the water velocity, and dispersion, which involves the spreading
of micro-organisms as they move along the water path. The geochemical
processes act to delay microbial transfer through the soil matrix and consist
of filtration, sorption, and sedimentation mechanisms. Finally, biological
Figure 2 Controls governing potential pathogen transfers from agricultural settings.
processes, such as growth and chemotactic responses, may influence pathogen transfer through the soil habitat as demonstrated with other introduced
micro-organisms (e.g., Reynolds et al., 1989).
Pathogenic micro-organisms and, similarly, non-pathogens may enter
surface water via overland flow pathways, by subsurface transfer routes in
highly permeable soils, or through artificial field drainage. The transfer from
grassland soils to surface waters is largely facilitated by hydrology and,
conceptually, microbial transfer can be associated with two tiers of hydrological energy. Slow flow microbial transfers operating between storms are
thought to be associated with the steady percolation of precipitation inputs
through the soil profile. This contrasts with overland flows resulting from
high-energy precipitation events, which enable the physical movement of
soils, manures, and potential pathogens into streams, creating a more rapid
and direct transfer route. Hence, hydrological events can aVect water quality
within catchments, and this is emphasized during dry periods and summer
grazing seasons when heavy rainfall events are capable of increasing stream
fecal bacteria levels by 100-fold (Rodgers et al., 2003).
Figure 3 shows a conceptual model of microbial transmission through
the agricultural environment. It highlights two important components:
(1) the routes of transfer available from source to receptor and (2) the
continum of micro-organism die-oV from source to receptor. Figure 4 illustrates naturally occurring flow pathways associated with the soil system
and shows transfer modes operating within them—as both free and attached
microbial consortia. However, as noted by Camper et al. (1993), it is not
only porous medium hydrodynamics that govern microbial contaminant
transport. Bacterial characteristics such as size and motility and properties
of the soil itself such as surface conditions and particle dimensions all
interact to determine microbial fluxes. In addition, the direct voiding of
excreta into farm streams and ditches by livestock provides a potential
transfer of fecally derived micro-organisms independent of energy-driven
flow mechanisms (Crabill et al., 1999).
The generation of surface runoV provides a potential vehicle for the rapid
translocation of entrained soil, waste, and biological colloids (see Fig. 4).
The physical force applied to the soil surface, resulting from the kinetic
energy associated with the overland flow pathway and impacting raindrops,
can disrupt the upper soil layer. This may dislodge soil particles along
with sorbed microbes, free attached micro-organisms into the overlying
water, and physically break down and transfer fecal matter. Microbial
Figure 3 Conceptual model of micro-organism transmission from surface applied fecal
waste to surface waters.
contaminants transferred within this above-ground lateral flow then eVectively escape the filtering eVects of the subsurface equivalent lateral transfer
The role of rainfall and the resulting flow signatures in dictating microbial transfer from soil to water has been described by Abu-Ashour and
Lee (2000), Fenlon et al. (2000), Cook and Baker (2001), and Vinten et al.
(2002). Abu-Ashour and Lee (2000) suggest that precipitation events are
a major factor dictating both vertical and horizontal movements of bacteria in soil. Having investigated E. coli transport on sloping soil surfaces via
surface runoV, these authors confirmed the importance of this process as a
mechanism of bacterial transport on soil surfaces.
Figure 4 Natural transfer pathways available to fecally derived micro-organisms applied to
soil surfaces.
The significance of the overland flow pathway in facilitating microbial
movement has been acknowledged with respect to methods of slurry application in a comparison of surface applied versus incorporation techniques
(Quinton et al., 2003). Incorporation of slurries into soil under laboratory
conditions indicated that a reduced number of fecal bacteria might be
transported from the soil system. While this suggests that under field conditions slurry incorporation would be a preferred method of application to
protect water courses, Quinton et al. (2003) warned of the potential discrepancies that may be observed as a result of such a scaling up, largely
resulting from desiccation of the surface applied slurry, and hence the
reduced threat given such conditions. Pathogen transport processes
operating during overland flow have recently been summarized in a suite
of proposed transport scenarios (Tyrell and Quinton, 2003). These involved
the discussion of incorporation of free microbes into overland flow, mobilization of soil or waste particles into overland flow, carrying attached
microbes, and detachment of microbes from soil surfaces arising from
shearing forces of raindrop or flow action. There is a clear need to quantify
the eYciency of overland flow in facilitating the wash-in of fecal material
from pasture to stream (Nagels et al., 2002). However, while wash-in of fecal
matter has long been recognized as a consequence of overland flows generated within the contributing areas of a catchment (McDonald and Kay,
1981), within large and complex watersheds the bacteriological quality of a
stream is the resultant eVect of a variety of indistinguishable sources, and so
determination of the loading capability of a particular transfer route is
In assessing cell movement and deposition in overland flows, consideration should be given to processes such as splash and flow detachment.
Work by Vinten et al. (2002) drew attention to the importance of these
hydrological processes at the soil surface and highlights the promotion of
microbial transfer via the mobilization of slurry colloids following energy
transfer from impacting raindrop momentum. Similarly, the energy associated with heavy rain is able to erode the soil and release considerably high
numbers of fecally derived pathogens to runoV waters from grasslands
(Heinonen-Tanski and Uusi-Kamppa, 2001). If antecedent soil conditions
are conducive to generating overland flow pathways and heavy rains occur
shortly after slurry application, there is great potential for significant runoV
of fecal micro-organisms following their entrainment into the surface flow.
This is complementary to studies by Abu Ashour and Lee (2000), who
concluded that cells may be carried in surface runoV following detachment
from soil particles or alternatively may experience transfer, remaining in the
sorbed phase. However, surface runoV does not always contribute heavily to
the bacterial loading of receiving waters. Following slurry application at a
field experiment in Scotland, losses of E. coli in surface runoV were only
0.003% of cells present after 75 mm of rain in the 49 days post-wastespreading (Vinten et al., 2002). This does not solely reflect the eYciency of
the transfer pathway and is likely to incorporate a die-oV of introduced cells.
So although overland flows may provide an eYcient microbial export route,
their impact may be only short lived or of reduced eVect at times.
Using a novel soil tilting table apparatus, Mawdsley et al. (1996)
simulated the horizontal surface transfer of the protozoan pathogen
C. parvum under controlled laboratory conditions. Following the addition
of contaminated livestock waste to soil blocks with a slope of 7.5%, the
movement of this micro-organism in runoV was detected for a minimum of
21 days, and in one case was still being laterally exported after 70 days.
Throughout the duration of the experiment, a significantly higher oocyst
concentration was found in leachate as opposed to the runoV across the soil
surface, a finding similar to that of Vinten et al. (2002) with respect to E. coli.
However, considering the rapidity of the overland flow pathway, possible
pollution from occasional runoV events, even on soils where leaching predominates, must still be considered (Mawdsley et al., 1996). Furthermore,
Mawdsley et al. (1996) postulated that on true impermeable heavy clay soil,
a greater proportion of oocysts would be lost in surface runoV. However, we
have no proof that overland flow, once started, actually delivers materials to
streams and primary water systems. It may instead provide a pulsing mechanism of transfer or be interrupted by buVers before having an impact on
receiving waters.
Much research has focused on the vertical transfer of pathogens in soil
leachate and the similarities with colloid filtration theory. Vertical displacement through the soil profile of these micro-organisms has been demonstrated in a variety of soil column experiments (Aislabie et al., 2001; Brush
et al., 1999; Gagliardi and Karns, 2000; Warnemuende and Kanwar, 2002;
Wollum and Cassel, 1978). The moisture content of the soil determines
bacterial movement as the continuous water films permit bacterial transfer
because the microbial population of the soil is limited to the aqueous phase
and the solid–liquid interface. It has been proposed that appreciable bacterial movement in soil can only occur if there are enough water-filled pores of
the diameter required to enable a continuous pathway (Bowen and Rovira,
1999). A lack of movement at moisture tensions below saturation has been
shown, and thus with increasing hydraulic conductivity there is a rise in
bacterial transport (Rahe et al., 1978).
When water drains from pores, microbial movement through the soil
structure depends on sieving eVects imposed by pore openings. Gagliardi
and Karns (2000) concluded that if soil pores avoid becoming clogged, E.
coli 0157 is able to travel below the soil surface layers for periods in excess of
2 months following the initial application. Manure that has remained on the
soil surface for extended periods prior to a rain event is still capable of
delivering E. coli, and potentially other cells, by transfer through the profile
with the onset of hydrological inputs. With successive rain events, more cells
are transferred in the resulting leachate, but at levels much reduced in
comparison with counts recorded in the first leaching event (Saini et al.,
There is a general increase in bacterial movement within saturated soils
compared with drier soils; however, the occurrence of percolating water
increases the potential for translocation of micro-organisms through the
soil matrix. Culley and Phillips (1982) were able to demonstrate that fecal
streptococci (FS) were capable of movement through soil profiles into field
drains located at a depth of 75 cm, provided water was present to facilitate
this downward transfer. The importance of percolating water cannot be
stressed enough, and in combination with the action of higher organisms,
it represents one of the most important microbial transfer mechanisms in
soils (Hekman et al., 1995).
Fenlon et al. (2000) demonstrated the delivery of E. coli to drains to
reach a cumulative maximum of 11% of applied cells following rainfall
events occurring for the 7 days after slurry application. Their study
suggested that, following the initial substantial pulse of cells to the
drains immediately after slurry spreading, as much as 80–90% of total
E. coli was retained within the soil matrix. Fenlon et al. (2000) reached
the conclusion that, provided a large enough hydrological event occurred close to the timing of slurry application, bacteria can be flushed
from the soil matrix in considerable numbers. Their work also emphasised the importance of storing farm wastes until the soil conditions
were suitable for their application given the ability of the matrix to
retain such a large proportion of added cells.
Soil type and condition also contribute to the determination of the
route of cell movement from soil to water. Bacterial travel times are
more rapid in coarser textured soils with larger pore spaces as opposed
to finer textured soils (Huysman and Verstraete, 1993; Tan et al., 1992).
Gannon et al. (1991) stressed that all bacterial species are filtered out to some
extent by the soil matrix, with bacterial transport strongly correlated with
cell size, and Cook and Baker (2001) showed the soil matrix, within lysimeters, to be eVective in retaining micro-organisms added to soil via a
carrier substrate.
The translocation of pathogens down soil profiles parallels that of other
microbial populations and relies on pore size openings, the soil matrix
system, and water characteristics. Figure 5 shows the range of diameters
for diVerent pathogen classes and highlights some specific examples of cell/
oocyst/particle sizes. The soil sieving eVects experienced by free-flowing
microbes include straining, sedimentation, and sorption mechanisms.
Figure 5
Pathogen cell/oocyst/particle dimensions relative to soil particle fractions.
1. Straining
There is evidence that pore clogging by bacteria and protozoa may restrict
microbial transfer by physically blocking a pore entrance and may even alter
soil hydraulic conductivity at a localized scale (Thullner et al., 2002). The
extent to which this process of bioclogging occurs is a function of the particle
size of the porous medium and the diameter of the microbial consortia that
transfer through the system. Thus, micro-organisms suspended in flow are
eVectively strained by the soil matrix and accumulate on soil particles when
pore openings are too small to permit their passage. The hydraulic conductivity of the soil is then reduced not only by the accumulating cells, but also
as a function of the subsequent excretion of extracellular polymers. The
immobile regions that may exist in the filter matrix in the form of ineVective
micropores (Kim and Corapcioglu, 2002) may trap microbes in these dead
end pores and can result in a trapping of potential pathogens, from where
they may act as a significant reservoir of contamination. It is unclear what
risk trapped microbes pose.
The concept of filtering capabilities associated with the soil matrix is
crucial within the management strategy of buVer strips. It may be considered
that sorption and filtration processes are more eVective in protecting surface
waters than relying on organism die-oV (Aislabie et al., 2001). The role of
buVer strips in removing pathogens, such as C. parvum, from carrying water
was evaluated by Atwill et al. (2002). BuVers constructed with silty clay or
loam or at lower bulk densities were most successful at filtering the oocysts,
in contrast to sandy loams and higher bulk density soils. However, when
water is in excess of field capacity, it has been suggested that the main flow
of water occurs through pores larger than all colloid sizes except for protozoa. This in turn implies that straining may only be eVective for protozoan
pathogens, and even so, under extremely wet conditions protozoa will still be
mobile (McGechan, 2002). Brush et al. (1999) asserted that future research
still needs to focus on diVerentiating the relative importance of the various
removal mechanisms that operate and on oocyst–medium interactions.
The natural filtering of microbes as they dissipate through the soil has
been likened to that associated with colloidal movements with both, in
contrast to dissolved constituents, accommodating a low diVusivity. As a
result, the conceptual use of models formulated for colloid transfer has been
investigated with respect to microbiological transport, providing theoretical
frameworks for modeling bacterial movement (Johnson et al., 1995). However, filtration theory calculations of the distances traveled by bacteria
underestimate true translocation lengths within the soil (Simoni et al.,
1998), with increased travel distances resulting from heterogeneity in the
adhesion properties within the bacterial populations. Furthermore, Harter
et al. (2000) noted that the presence of dead end pores and angular particles
arranged to form planar contact potentially enhance localized straining
unaccounted for by filtration theory. Bearing in mind this size exclusion
principle, Camper et al. (1993) addressed the physiological condition of
bacterial cells and proposed that, given adverse conditions, cells may become
starved, experience a reduction in their diameter, and then successfully
infiltrate the previously inaccessible pore entrance. Similarly, Macleod et al.
(1988) demonstrated that starvation influenced cell penetration rates and
clogging through porous media. Biological factors such as motility, growth,
and physiological stress introduce uncertainty in terms of the extent that
colloid filtration principles can successfully interpret microbial movements,
and consequently, colloid filtration theory should not be used as a rigid
predictive model, but rather as a tool for interpreting microbial buVering
(McDowell-Boyer et al., 1986).
2. Sorption
Soil type may influence microbial transfer because of diVerences in
sorption properties relating to the associated colloidal material of the soil
(Schijven et al., 2002). The major soil components aVecting sorption of
bacteria are clay and organic matter (Aislabie et al., 2001). E. coli sorption
within diVerent soil types has enabled verification that soils accommodating higher clay content sorb greater numbers of bacteria due to a greater
specific surface area (Ling et al., 2002). Sorption of C. parvum onto suspended particles in soil water has also been observed. Hydrophobicity and
zeta potential were highlighted to exert a significant influence in the adhesion mechanisms of the protozoan oocysts (Drozd and Schwartzbrod, 1996).
The authors stressed that sorption of this pathogen to soil surfaces cannot
be attributed to a single factor and that there is a complex suite of forces
interacting to govern microbial retention. This has also been shown in the
study of Nielsen et al. (2001), who concluded that bacterial sorption is a
function of electrostatic forces, Van der Waal interactions, extracellular
polysaccharides, and cell hydrophobicity.
3. Sedimentation
Sedimentation of microbes in pore water may also play a role in restricting microbial transfer through soil; however, it is of little significance for the
smaller virus particles, which tend to be naturally buoyant and thus are
unlikely to settle. Similarly, motile bacteria by their very nature are less likely
to undergo sedimentation. Instead, it is larger micro-organisms in excess of
5 mm that are considered most likely to undergo the sedimentation processes
(Yao et al., 1971).
The significance of large pores and voids in facilitating the movement of
water and colloidal material through the soil has long been acknowledged
(Allaire-Leung et al., 2000; Beven and Germann, 1982; Buttle and Leigh, 1997;
Ehlers, 1975; Jacobsen et al., 1997; McLeod et al., 1998; Williams et al., 2000).
A number of studies have focused attention on macropore flow as a vector of
microbial transfer. This work concluded that macropore flow is a major mechanism for microbial transport within soil (Abu-Ashour et al., 1998; Fontes
et al., 1991; Gannon et al., 1991; Harvey, 1997; Mawdsley et al., 1995;
Smith et al., 1985). Such preferential flow pathways serve as routes of relatively
rapid water flow and allow cells, among other colloids and contaminants, to
successfully bypass the sieving and constraining architecture of the soil matrix
(see Fig. 4). Although macropores often make up only a small volume of the soil
body, they serve as important routes for both the lateral and vertical transfer of
cells entrained in the carrying water. Macropores may be formed naturally or
through soil fauna activity, plant root presence, or soil shrinkage. The
interconnected pore domain carved through the action of earthworms demonstrates the eVectiveness macropores can have in promoting micro-organism
transfer (Joergensen et al., 1998). In their absence, the distribution of microbes
may be mostly confined to the uppermost zones of the soil profile. Soil column
experiments such as those used by Huysman and Verstraete (1993) detail
significant micro-organism movement through these larger pores following
the reduction of the retardation component imposed by the more tortuous
and constricting micropores.
Provided that the input of water to a soil system is suYcient to initiate
water flow in larger pores, suspended bacteria may move rapidly through the
profiles of well structured soils (Smith et al., 1985). Smith et al. (1985)
concluded that macropore-facilitated transfer of cells is frequent and stated
that any macroporous soil has the potential to rapidly transport cells to the
depth to which these pores extend given suYcient hydrological delivery.
Macropores may play a vital role in governing microbial movement, especially in respect to long-distance biological contaminant transport, provided
that conditions that initiate water film development, and consequently flow,
are met (Fontes et al., 1991).
Not all bacteria and protozoa introduced into soil depend on advection and
dispersion mechanisms for transfer through interconnected pores and fissures.
Instead, distribution within porous media can result as a direct consequence
of micro-organism growth and motility. Self movement of microbes
occurs through the action of flagellar rotation. Reynolds et al. (1989) demonstrated transfer rates approaching 0.5 cm h 1 within packed sand cores, and
McCaulou et al. (1995) suggested that motile bacteria were possibly able to
detach from a solid surface under their own locomotive power. This suggests
that bacterial and protozoan travel times calculated in the absence of motility
may incorporate a degree of error (McCaulou et al., 1995).
In addition, the motility of one particular micro-organism may influence
that of another. Brown et al. (2002) draw attention to the role of free living
protozoa, which may prey on bacterial cells and consequently act as vehicles
for bacterial transmission. Therefore, this intracellular location not only
provides a potential protective niche for ingested cells but also may facilitate
their transfer through the soil. However, with microbial motility promoting
relatively short travel distances, it may be that such active bacterial movement holds importance only within micro-environments and is of little
significance in relation to widespread movement at the field scale, where
hydrological flows exert a greater influence.
Mesofaunal activity within grassland soils may contribute to the dispersal
of microbes, and hence potential pathogens, either directly, through the
attachment of micro-organisms to larger soil inhabitants, or indirectly,
through the creation of larger pore networks within the soil matrix. Earthworms account for the largest proportion of biomass attributed to soil
animals under pasture (Lee, 1985) and so contribute toward important eVects
on the physical structure of soil through burrow formation. Earthwormworked soils can promote up to a 50% reduction in the surface runoV of
slurries applied to grasslands, most likely as a result of the worms increasing
soil porosity and regenerating connections between the soil surface and
drainage cracks in deeper horizons (Scholefield, personal communication).
Opperman et al. (1987) concluded that Eisenia foetida aided the translocation of coliform bacteria, derived from cattle slurry, to a depth of 17.5 cm,
whereas in their absence the bacteria were restricted to the upper zone of the
soil. This complements a study that reported significant vertical FC transfer
through earthworm burrows to the subsoil following land application of
slurry (Joergensen et al., 1998). Thorpe et al. (1996) found that earthworms
promoted bacterial transport to a depth of 30 cm in the soil, though they
pointed out that burial of inoculated litter rather than an increase in macropore flow, due to the earthworm channels, was a more important mechanism
for cell transfer.
In clay soils, earthworm tunnels play a less important role because of their
inherent instability resulting from soil swelling and shrinking processes
(Joergensen and Seitz, 1998) and the lower number of earthworms in clay
soils in general. Interestingly, Joergensen et al. (1998) noted that the drilosphere soil (that which lines earthworm burrows), with its increased potential to hold water and sorb suspended cells, was more densely populated with
fecal bacteria compared with the bulk soil.
The physical alteration of soil structure is not the only means by which
indigenous soil organisms aid transfer of introduced microbial populations.
Gammack et al. (1992) suggested that earthworms, along with mites and
millipedes, act as vehicles for bacterial transport through direct attachment
of cells to such mesofauna. They also note a mechanism of transfer associated with the ingestion of organic material and its subsequent movement
through the gut, which they propose enhances both horizontal and lateral
movement, though this is probably minimal.
Colloids may play an important role in assisting the successful transfer of
introduced micro-organisms from soil to receiving waters. Slurries, manures,
and excreta contain vast numbers of particulate and colloidal materials spanning a range of diameters, with many already being host to sorbed microorganisms (McGechan and Lewis, 2002). Large voids facilitate relatively easy
colloidal movement, and so any microbe that is associated with a migrating
colloid may not only improve its survival chances but also emerge from the soil
ahead of the wetting front. However, in their excellent review, Kretzschmar
et al. (1999) noted that prediction of the importance of colloid-facilitated
contaminant transport is complicated by the at present poor understanding
of processes relating to colloid release, transport, and deposition. Their review
assessed colloidal transfer in depth and provided the framework from which
concepts of biological attachment to suspended colloids may be extrapolated.
There are two particularly important properties associated with colloids
that enable them to function as important contaminant carriers. The first is
that colloids have a very large specific surface area, in excess of 10 m2 g 1
(Kretzschmar et al., 1999). Second, these colloids remain stable in suspension for significant periods, and if bacteria attach to the large available
surface area, their dispersal through the soil may be aided considerably.
Those colloids that remain more in the center stream of the flow path
are likely to remain uncaptured, and thus migrate along these faster, more
permeable flow paths. Consequently, those micro-organisms that form
a microbe–colloid complex are likely to experience the same transfer or
buVering processes of their nonbiological colloid counterparts.
A number of factors contribute to the initial sorption of cells to suspended
material in soil water. The preferential sorption of specific microbial cells
cannot be explained simply by referring to a single force eVect (see Section
IV.C.2) and is related also to contact time between cell and colloid, colloid size,
and conditions around the soil particle such as wettability and surface texture
(Drozd and Schwartzbrod, 1996). In addition, interactions between bacteria
may result, and therefore attachment to a colloid does not necessarily occur in
a monolayer. Attached bacteria will detach and be deposited back into
the flowing suspension as a critical flow velocity is approached. This introduces
the concept of reversible and irreversible attachment, which is discussed briefly
here but is described further by Palmateer et al. (1993). Permanent attachment
involves the anchoring of the micro-organism to the particle with which
it is interacting. This therefore suggests a direct mode of contact between
microbe and particle. This contact is provided through the association of the
microbes pili, fimbriae, or flagella with the colloid in a ‘‘cementing’’ attachment
(Palmateer et al., 1993). Reversible attachment does not involve a true physical
association; it is a sorption producing a concentrating eVect at the particle
surface. Given suYcient turbulence, however, the resulting shear forces are able
to desorb the microbe relatively easily.
Recently, Dai and Boll (2003) concluded that C. parvum and Giardia
lamblia oocysts fail to attach to natural soil particles and that, in instances
of overland flow, oocysts would travel independently of the particle load
within turbid plumes. The authors suggested that this has relevance to
management practices because the way to deal with freely suspended
micro-organisms would be to minimize overland flow, whereas if these
microbes had been particulate-attached, the emphasis would have been on
targeting sediment transport controls. In contrast to this protozoan example, there are quantitative investigations into bacterial accumulation upon
suspended particles exported from agricultural fields (Palmateer et al., 1993).
Palmateer et al. (1993) stated that the particulate load within agricultural
drainage can accommodate cells at levels as high as 103 to 105 per mm2 as a
function of the sorption process. As a consequence of particulate attachment, they also demonstrated that the transport of fecally derived E. coli in
this sorbed phase may travel kilometers within agricultural drains.
Soil colloids and colloid-facilitated transport of contaminants may
potentially play a significant role in assisting potential pathogen transfer.
Although the apparent capability of colloids to promote contaminant
transfer has been acknowledged, it must be emphasized that the actual
importance is far from understood (Mills and Saiers, 1993), and very few
studies have been published that detail direct evidence for colloid-facilitated
transport of contaminants (Kretzschmar et al., 1999).
Having demonstrated the potential for pathogen transfer in grassland
environments, it becomes evident that the soil system must, in part, act as
a temporary store for introduced micro-organisms as well as providing the
physical routes of connection between source application and the end point
receptor. One of the challenges in grassland research is to understand how
soil systems accommodate introduced micro-organisms and influence their
persistence or, alternatively, how micro-organisms adapt to the sudden
dramatic change in environmental surroundings. The remainder of this
chapter deals with this aspect of pathogen presence in agricultural settings.
The survival and epidemiology of pathogens is a key issue when considering the potential for contamination of the surrounding land and water
and, as illustrated in Figure 2, is an important temporal component exerting
influence over the risk associated with transfer. A logarithmic first order
exponential decay equation is often assumed through statistical analysis of
bacterial die-oV. The decline of a bacterial population with time can be
described mathematically as
Mt ¼ M0 e
where Mt is microbial concentration at time t, M0 is initial microbial concentration, k is the first order rate coeYcient for the net mortality rate for
organisms/day, and t is time in days.
This equation contributes to the population decline illustrated in Fig. 6.
Such an equation worked satisfactorily in the study of Stoddard et al. (1998),
although it was occasionally complicated through bacterial regrowth. It is
possible to extend first order kinetics to model bacterial die-oV rates
and include terms for the antagonistic eVects of biotic and abiotic factors
(Wilkinson et al., 1995). A variety of other bacterial decay equations were
considered by DeGuise et al. (1999) in their discussion of foundations for
modelling bacterial contamination.
Excretal fecal waste is deposited directly onto grasslands by grazing
livestock. A crust may form on deposited excreta within 2 days (Thelin
and GiVord, 1983), producing a favorable location for bacterial survival
Figure 6
Factors contributing to survival of micro-organisms introduced to soils.
within the deposit. The crusting limits interactions with the soil and the
atmosphere, providing a microclimate and protective niche for microbial
survival until the next rain event provides a means of transfer. The crusting
process can also protect micro-organisms from intense sunlight and heat for
at least one summer, implying that fecal indicator bacteria may persist for
over 1 year in crusted bovine feces (Buckhouse and GiVord, 1976). This
suggests that fecal deposits may provide a long-term continuous source of
microbial pollution to surrounding areas. Bacteria on the surface of recently
deposited feces are the first to be aVected by UV in sunlight (Scottish Executive and Food Standards Agency Scotland, 2001). Wang et al. (1996) showed
that E. coli 0157 survived up to 7, 8, and 10 weeks at 37 8C, 22 8C, and 5 8C,
respectively, demonstrating that the pathogen can persist for lengthy periods
within the protective niche of the fecal deposit, though temperature and water
activity were noted as being influential on E. coli 0157 persistence. There have
been complementary reports suggesting that E. coli 0157 may survive for
extended periods even under very dry conditions (Jiang et al., 2002), and
Fukushima et al. (1999) observed low levels of pathogenic E. coli surviving in
bovine feces for over 4 months. Fukushima et al. (1999) also found that
pathogenic strains 0111 and 026 survived for periods of up to 18 weeks. The
protozoan Giardia survives for much shorter periods, remaining infective for
only 1 week in cattle feces (Olson et al., 1999). Fluctuations in salinity within
the fecal deposit have been documented as exerting a control on Cryptosporidium survival within excreta (Bradford and Schijven, 2002). A fecal deposit
may experience a change in salinity through exposure to either rain (decrease
in salinity) or urine (increase in salinity). An increase in the salinity of the fecal
waste reduced the electrostatic interactions between charged particles,
and Bradford and Schijven (2002) speculated that induced changes in salinity
influenced the release of micro-organisms from protective sites and so
interfered with survival curves of the associated microbes.
The relatively uniform mix of excreta and urine produced by housed
livestock, collected in a liquid form, is termed slurry (Chadwick and Chen,
2002). The pathogen content of slurry is a function of dilution, dry matter
content, temperature, storage time, animal source, and pH, among other
factors (Mitscherlich and Marth, 1984). Slurries accommodate a more
uniform microbial contamination than their solid manure counterparts as
a result of the greater mobility of micro-organisms in this liquid material
(Chadwick and Chen, 2002). McGee et al. (2001) have reported E. coli 0157
persistence in stored cattle slurry in excess of 3 months. However, despite its
survival in slurry, McGee et al. (2001) warned that it may not represent a
predominant source of transmission in agricultural settings because of substantial decline in numbers observed. The long survival times in their study
may be explained by the large inoculation of bacteria received by the slurry,
which exceeded the numbers in feces. A greater persistence of cells was seen
in slurry of a higher dry matter content, complementing findings in which
E. coli 0157 inoculated into slurry reaches undetectable levels over five times
quicker than when inoculated into cattle feces (Maule, 2000). Earlier work of
Kovacs and Tamasi (1979) detailed survival times of E. coli and Salmonella
spp. in slurry, with maximum survival durations of 1 week and 28 weeks,
respectively. Surprisingly, Salmonella survived seven times longer at 20 8C
than at 4 8C, where it persisted for only 4 weeks; the authors claimed it to be a
potential function of the predominance of the Salmonella component in the
samples. However, it may be that the lower temperatures induced a VBNC
state in the Salmonella spp. and avoided detection, thus leading to underestimates of Salmonella cell counts at low temperatures. A 30–50% reduction
in the number of viable C. parvum oocysts within slurry has been recorded
for temperatures of 4 8C after 100 days, demonstrating a healthy persistence
of the parasite within these liquid wastes (Warnes and Keevil, 2003).
Solid Manure (Farmyard Manure)
DiVerences in the survival of pathogens in diVerent manures have been
reported. For example, within ovine manure, Kudva et al. (1998) reported
the persistence of E. coli 0157 for a period of 21 months, but that if the
manure was aerated, the survival times for 0157 could be reduced to 4
months. In bovine manure, survival was not as lengthy, with aerated bovine
manure allowing E. coli 0157 to persist for only 47 days. Within aerated
manures, a drying eVect resulting from the mixing was suggested as
the causal agent of bacterial decline. Kudva et al. (1998) also observed a
lower survival rate of the bacteria within manure under laboratory conditions compared with manure exposed to the environment. The proposed
explanation for this was, in part, that physical dimensions of environmental
manure piles provided micro-niches that were not reproducible under the
conditions of laboratory experiments. As for micro-organism survival in
slurry, it was concluded that wastes of a higher solid content prolonged
survival (Kudva et al., 1998).
Governing factors of bacterial and protozoan (both indigenous
and introduced) survival in soil are well documented (e.g., Acea et al.,
1988; England et al., 1993; GriYths and Young, 1994; Heijnen and Vanveen,
1991; Jamieson et al., 2002). Variables such as temperature, moisture content, soil type, pH, sunlight, and presence of indigenous micro-organisms,
nutrients, and organic matter all exert an influence on survival within the soil
(Cools et al., 2001; Reddy et al., 1981; Van Donsel et al., 1967). As
an important first step in developing an understanding of pathogen persistence in agricultural systems, it is essential to address the biotic and abiotic
factors that govern the life cycle of any introduced micro-organism in the
soil. Table II summarizes the influential controls of resulting micro-organism die-oV curves, which can be extended to encompass pathogenic strains.
What complicates the evaluation of die-oV within agricultural settings is the
dissemination of micro-organisms from the source material into the soil.
Once in the soil, the exact nature of die-oV coeYcients at the field scale are
diYcult to determine through the combined eVect of both true organism
decline rates and their dilution into the soil system. Table III provides a
summary of selected pathogen die-oV rates in a variety of environmental
After the introduction of micro-organisms to soil, either through direct
deposition of livestock feces or via a carrier substrate such as manure or
slurry, most bacteria and protozoa have diYculty surviving. Common inhabitants of the gastrointestinal tract are not adapted to survive in soils; their
preferred habitat facilitates optimal growth in warm (37 8C), moist, and
highly nutritious conditions such as those found in the mammalian digestive
tract. While there are many factors that may individually aVect microbial
survival in soils, it is likely that many interactions of biotic and abiotic eVects
combine to provide a detrimental environment for pathogen survival (Fig. 6),
and thus it becomes diYcult to determine the extent of individual factor
influences. The following section briefly discusses some factors that aVect the
survival of microbes, including pathogens, entering the soil habitat.
1. Nutrient Availability
The availability of nutrients in the soil habitat exerts an influence on the
persistence of microbes. Readily utilizable organic carbon (C) is often a
limiting growth factor, and so a lack of available nutrients may induce a
starvation response within bacteria and may have an impact on their survival.
Consequently, the capability of cells to survive starvation may influence their
persistence in their new surrounding (Acea et al., 1988) as they undergo both a
physiological and morphological change under nutrient limitation. Such
population changes may be a result of cells failure to lower their metabolic
rate to satisfy the low usable organic C conditions (Klein and Casida, 1967).
High soil organic matter content may support survival and potentially
even promote regrowth of fecal microbes (Gerba et al., 1975). The increased
organic matter not only acts to increase nutrient retention and provide a
C source, but also proves beneficial through improved moisture retention.
Nutrient availability presents itself as an important control on microbial
survival, but it must be remembered that coliforms and gut bacteria prefer
simple C rather than complex organic matter. Soil does not contain a
large amount of simple sugars but rather more complex C sources, again
highlighting the unfavorable nature of the foreign soil environment.
Nutrient availability
Soil moisture status
Soil temperature
Water temperature
Soil type
Influence on micro-organism survival
Cells may become sublethally stressed and physiologically injured in the absence
of nutrients in water and soil
Increased moisture content favors micro-organism survival. However, in some
instances increased water content may be unfavorable, allowing protozoa to
transfer and prey on bacteria and enabling larger pores to fill which oVer cells
less protection against predators
Increasing soil temperature leads to a decline in micro-organism numbers.
Persistence is greater in winter months. Freezing conditions can prove
Survival of introduced micro-organisms is promoted through lower water
temperatures. Cell die-oV is accelerated during summer, though this may
be a combined eVect of increased sunlight inactivation
Imposes structural and textural influences. Clay soils promote greatest survival.
Sandy soils have a poor water holding capacity which limits survival. The
structural network creates a habitable pore space. Soil type influences
cellular adsorption
Example references
Klein and Casida (1967)
Entry et al. (2000)
Mubiru et al. (2000)
Postma et al. (1989)
Reddy et al. (1981)
Cools et al. (2001)
Davenport et al. (1976)
Stoddard et al. (1998)
McGee et al. (2002)
Thomas et al. (1999b)
Wang and Doyle (1998)
England et al. (1993)
Fenlon et al. (2000)
Maule (1999)
Stotzky and Rem (1966)
Young and Ritz (2000)
Table II
Influence of Environmental Factors on the Survival of Micro-organisms Introduced into Soil and Water
Predation in soils
UV exposure
Organic matter
Micro-organism species
Stream bed sediment
Organic matter increases nutrient retention, provides a carbon source and
improves moisture retention—all of which are beneficial to micro-organism
survival and growth
The variety of interacting physical, biological and chemical factors will aVect
micro-organisms depending on the susceptibility of the species
Survival is greater in stream bed sediments in comparison with the overlying
water column. They provide protection against predation and UV
inactivation and act as a source of nutrients
Acea et al. (1988)
Brown et al. (2002)
Tappeser et al. (1998)
Artz and Killham (2002)
Korhonen and Martikainen (1991)
Van Donsel et al. (1967)
Barcina et al. (1989)
Mofidi et al. (2002)
Morita et al. (2002)
Sinton et al. (2002)
Gerba et al. (1975)
Mitscherlich and Marth (1984)
Burton et al. (1987)
Craig et al. (2002)
Davies et al. (1995)
Van Donsel and Geldreich (1971)
Predation in waters
Increased survival in sterile soils. Most predation of bacterial cells occurs via
grazing protozoa. However, survival may be increased with an
intracellular location within protozoal trophozoites
Removal of protoaoan communities by filtration prolongs bacterial survival
in aquatic environments. Cells associated with particles are less susceptible to
A near neutral pH supports micro-organism survival
Can cause inactivation of micro-organisms through the formation of lesions
in DNA and via photo-oxidation mechanisms. Therefore survival is reduced
at the soil surface and in less turbid waters
Table III
Survival of Fecally Derived Micro-organisms within DiVerent Media
Fecal coliforms
Temperature (8C) Survival (days)
Stoddard et al. (1998)
50% population reduction after
manure application
Liquid manure
Nonsterile river
Nonsterile soil
Sandy soil
Loam soil
Clay soil
Drinking trough
Drinking trough
water þ feces
Bovine feces
Aerated bovine Environmental
Ovine manure
Media used ¼ EMB agar
Stoddard et al. (1998)
Kovacs & Tamasi (1979)
Bogosian et al. (1996)
Decline to 1 log10 CFU
Inoculation level of 106 CFU ml
Fenlon et al. (2000)
McGee et al. (2002)
Feces added to trough
Inoculation level of 105 CFU g
Aerated by mixing
Wang et al. (1996)
Kudva et al. (1998)
Fecal streptococci
Escherichia coli
Eschericia coli K12
Escherichia coli 0157
Liquid manure
Untreated water
Filtered water
Silty clay loam
Silt loam
Loamy sand
Cattle feces
Cattle feces
Liquid manure
Aerated by mixing
Fukushima et al. (1999)
Kovacs & Tamasi (1979)
Korhonen & Martikainen (1991)
Jenkins et al. (2002)
Olson et al. (1999)
Olson et al. (1999)
Kovacs and Tamasi (1979)
0.2 micron filter
Days to reach
99% inactivation
Escherichia coli 0157,
011 & 026
Campylobacter jejuni
Cryptosporidium parvum
Glardia cysts
Aerated ovine
Bovine feces
2. Soil Moisture Status
The moisture content of the soil habitat may be an important stress
factor; a number of studies have reported that microbial survival is dependant upon soil moisture content (Entry et al., 2000; Jenkins et al., 2002;
Mubiru et al., 2000; Postma et al., 1989; Reddy et al., 1981). Inextricably
linked with moisture content is the oxygen status of the soil, with slightly
anaerobic conditions favoring the persistence of microbes. In general, the
rate of microbial die-oV increases with a decrease in soil moisture; as an
example, the die-oV rate of E. coli 0157 within diVerent soil types has been
determined to be dominated by diVerences in soil water availability (Mubiru
et al., 2000). Entry et al. (2000) concluded that FC experienced prolonged
survival when accompanied by an increase in moisture within grass buVer
strips. However, increasing the water content of a soil does not always prove
beneficial to an introduced bacterial population. Under conditions of excessive moisture, a considerable dilution of usable organic C may result, creating unfavorable conditions for E. coli survival (Klein and Casida, 1967) if
the cells are no longer associated with fecal wastes. However, Jenkins et al.
(2002) examined the role of water potential on the inactivation kinetics of
C. parvum oocysts and concluded that soil water potential was less important than soil type and temperature eVects on oocyst inactivation.
Soil Temperature
The temperature of British soils is, on average, 15 8C (Cools et al., 2001);
however, micro-organisms that are potentially pathogenic to humans have
an optimal growth temperature of 37 8C. Interestingly, once in the soil
environment, pathogen survival rates can vary inversely, with temperatures
below 15 8C (Davenport et al., 1976). Cools et al. (2001) found that a lower
incubation temperature combined with a higher soil moisture content prolonged survival, with numbers of E. coli reaching the detection limit at day 80
under 100% field capacity. The survival of E. coli was greater at 5 8C than at
25 8C and the higher soil temperature resulted in limits of detection being
reached as early as 26 days after inoculation. Quantitatively, die-oV rate may
be defined as doubling with a 10 8C increase in temperature, in the range of
5–30 8C (Reddy et al., 1981). Faust (1982) also showed that FC bacteria
can remain viable for lengthy periods, provided soil temperatures are relatively low, but if high temperatures were combined with other unfavorable
conditions, die-oV increased. The study of Van Donsel et al. (1967) investigated FC survival throughout a year and determined a quicker rate of die-oV
during summer. Freezing conditions will often reduce survival, and this will
take greater eVect during times of high soil moisture levels (Stoddard et al.,
1998). High temperatures in combination with low moisture levels have been
noted as the conditions most detrimental to Salmonella typhimurium survival
(Zibilske and Weaver, 1978). Jenkins et al. (2002) concluded that soil temperature was an important factor governing C. parvum survival, observing,
at temperatures between 30 and 40 8C, that the oocyst inactivation rate was
directly related to increasing temperature, complementing the earlier results
of Fayer et al. (1998).
Soil Type
There has been much work addressing soil structural and textural influences on general micro-organism survival (e.g., England et al., 1993; GriYths
and Young, 1994; Heijnen and Vanveen, 1991; Postma and Vanveen, 1990;
Rogasik et al., 1999; Stotzky and Rem, 1966; Young and Ritz, 1998). Most
of these studies have concentrated on nonpathogenic organisms, although it
is possible to apply the same principles to pathogenic micro-organisms
introduced to soil via animal wastes.
Soil texture determines the water content holding capacities of soils and
therefore aVects microbial survival through reasons discussed earlier. In
addition, soils that are poor at retaining water will accommodate a greater
number of discontinuities in water films, thus restricting the movement of
grazing protozoa and improving the survival potential of prey cells
(Heijnen and Vanveen, 1991). Soil texture, through the amalgamation of
soil aggregates and organic matter, also acts as a provider of microhabitats
that may aVect the survival of micro-organisms in the soil. The presence of
clay in the soil can enhance the retention of micro-organisms, including
pathogens, and increase the provision of protective niches. England et al.
(1993) claimed that both clay type and content are important in determining
microbial persistence.
Soil type also dictates the soil structure and physical makeup of the soil
pore network. The habitable pore space (Young and Ritz, 2000) that arises
through the given soil structure means that organisms of diVerent diameters
may only inhabit pores to which they can gain physical access. The influx of
cells into a particular size pore is also a function of the water status of the
soil, but those cells that inhabit smaller pores become less susceptible to
predation by larger micro-organisms, which cannot access the narrow pore
networks (Young and Ritz, 2000).
Survival times noted in the literature vary according to soil type and the
micro-organism under investigation. Maule (1999) observed E. coli 0157
survival in laboratory-based soil and grass microcosms of over 130 days
under continual illumination. Fenlon et al. (2000) reported survival of the
same pathogenic strain of E. coli in loam and clay soils to exceed, in some
instances, 20 weeks, and the study of Ogden et al. (2002) noted 0157
persistence of approximately 105 days in a loamy sand.
Tappeser et al. (1998) proposed that predation through protozoa is one of
the most important factors controlling inoculated bacterial populations.
However, its importance with regard to fecally derived micro-organisms is
unclear. England et al. (1993) suggested that the size selective feeding behavior of protozoa can assist in microbe survival. For example, within a soil of
increased clay content there exists the potential for cells to increase their
mass and volume via sorption to clay colloids, resulting in a clay–cell
complex that may avoid protozoan ingestion because of the increased size
of the cells. More recently, it has been noted that the role of protozoa in
terms of acting as bacterial reservoirs has received little attention (Brown
et al., 2002). A number of bacterial pathogens are able to survive and
replicate in protozoa, many of which are common in soils and water.
Brown et al. (2002) proposed that the stressful soil conditions imposed
upon E. coli 0157 and other pathogens can be reduced through the protective
niche provided by an intracellular location within protozoan trophozoites.
6. Soil pH
Reddy et al. (1981) concluded that a soil pH of between 6 and 7 oVered
optimal conditions for bacterial survival. Van Donsel et al. (1967) examined
the eVect of low pH on bacterial survival and showed that for a very acidic
peat soil, organism survival was reduced to 0.1% in less than 10 days.
Increasing the pH of the same soil type extended survival times, although
multiplication was prevented. Once pHs of 5.6 to 6.3 were approached,
E. coli were able to multiply to a very high level and remained in the
environment for as long as 110 days. This reduced survival capacity of
fecally derived bacteria at lower pHs has been confirmed by Gerba et al.
(1975) and more recently by Sjogren (1994).
In surface waters, the survival of fecally derived bacteria is a function of
their ability to endure physical, chemical, and biological stresses associated
with the aquatic habitat. Microbial persistence in water is critical, as it
determines the potential for detrimental eVects ‘‘downstream’’ of fecal
sources and is a major route of dispersal. Introduced cells may become
physiologically debilitated through exposure to environmental stresses,
and natural waters have long been noted as unfavorable environments
for introduced micro-organisms (Artz and Killham, 2002; Korhonen and
Martikainen, 1991; Scarce, 1964). Within agricultural settings, pathogens
may persist in ditch and drainage waters, drinking trough water, and
stream and river waters, potentially in a VBNC state as an adopted survival
strategy (Nilsson et al., 1991; Rollins and Colwell, 1986; Roszak and
Colwell, 1985, 1987; Thomas et al., 1999a). Listed in the following sections
are a variety of factors that may exert an influence on potential pathogen
survival in water.
1. Nutrient Availability
A number of studies have documented the role of nutrient availability in
modulating micro-organism survival rates. The multiplication rate of bacterial cells is determined by the level of utilizable nutrients made available to
them (Lechevallier et al., 1991). Survival times of E. coli are limited in
comparison with the native microbial community of aquatic systems
(Jones, 1999), suggesting that enteric bacteria have diYculty competing
with natural microflora for the low concentration of available nutrients
(Burton et al., 1987). It is also important to consider the nutrient availability
associated with suspended particles (both soil and waste derived) in surface
waters. Maki and Hicks (2002) stated that suspended sediments impose a
variety of influences on bacterial survival, and it is claimed that nutrient
levels can be 10 to 100 times higher on suspended particle surfaces as
opposed to the surrounding aquatic environment (Paerl, 1975). Hence,
available nutrients associated with suspended sediments assist bacterial
growth of those cells that attach to particles in comparison with cells
suspended freely in water (Crump et al., 1998). Subsequently, through the
utilization of nutrients bound to suspended particles, there is an increased
potential for cells to remain viable and avoid physiological stresses associated with starvation (Maki and Hicks, 2002).
Water Temperature
Increasing water temperatures often reduces the survival of microorganisms in aquatic systems. Campylobacter spp. survival in water has
been observed, with low temperatures (5 8C), along with nutrient availability, favoring survival, suggesting that water systems may act as a significant
reservoir for Campylobacter infection (Thomas et al., 1999b). Other studies
agree that low temperatures encourage micro-organism survival (Rice et al.,
1992), although disagreement concerning exactly how long particular
microbes persist in the environment is evident (e.g., Personne et al., 1998;
Wang and Doyle, 1998). The survival of E. coli 0157 alongside nonpathogenic strains is likewise temperature dependent, hence the ability of pathogenic E. coli serotypes to persist in aquatic environments in combination
with low infective doses reinforces the public health concern associated with
this bacteria (Jones et al., 2002). Cell concentrations generally decline
through summer months, probably reflecting the influence of the rising
water temperatures. However, there is a simultaneous increase in UV radiation distributed to surface waters, and so the eVects on micro-organism
survival of individual environmental influences cannot be easily defined.
Olson et al. (1999) suggested that C. parvum can remain viable at temperatures of 4 8C, and in waters of 25 8C oocysts may remain infectious for up
to 12 weeks (Fayer et al., 1998). The literature demonstrates that microorganisms of fecal origin may be suYciently robust to endure aquatic
environments for extended periods and thus harbor great potential for the
spread of disease through water as a vector.
As well as natural waters, water troughs on farms are an important
reservoir of pathogenic micro-organisms (LeJeune et al., 2001; Rice and
Johnson, 2000; Shere et al., 1998). The study of McGee et al. (2002) concluded that E. coli 0157 survival within trough water located in the field
could last as long as 24 days at temperatures varying between 2 8C and
15 8C. Survival was promoted if the water troughs were stored in a shed
rather than being left exposed in field conditions. Trough water kept in
the laboratory at a constant 15 8C enabled detection of E. coli 0157 cells 31
days after the start of the experiment, highlighting the improved survival
characteristics associated with more constant temperatures.
UV Radiation
The bactericidal eVects of sunlight through UV-B may result in photobiological DNA damage (Sinton et al., 2002). Gameson and Gould (1975)
suggested that exposure to solar radiation is the most important factor
regarding bacterial decline in waters, and more recently its importance
with regard to pathogen persistence in aquatic environments has received
attention (e.g., Mofidi et al., 2002; Morita et al., 2002; Sinton et al., 2002). In
a comparison of E. coli survival in illuminated and nonilluminated systems,
Barcina et al. (1989) determined light to be a decisive regulatory factor
governing cellular metabolic activity and concluded that those cells exposed
to visible light were progressing through defined stages of dormancy. However, the importance of UV eVects must be put into context, as within
agricultural settings they have a much reduced consequence on microbial
survival within turbid waters. Likewise, the significance of UV eVects is of
little relevance with respect to subsurface hydrological pathways, and so the
impact exerted by UV on microbial population decline is more a process
operating further down the chain of contamination, once potential pathogens enter more clear waters, in a much diluted concentration, downstream
of where drainage waters meet streams.
Biotic factors also exert an influence upon the survival of foreign microorganisms entering aquatic environments. There can be a great deal of
antagonistic activity associated with indigenous microbes through predation
and competition, and the importance of predation in regulating the survival
of all micro-organisms within surface waters is well documented (e.g., Flint,
1987; Korhonen and Martikainen, 1991). Simple experiments comparing
survival rates in filtered and unfiltered waters demonstrate the impact of
removing protozoan populations from samples (Artz and Killham, 2002;
Korhonen and Martikainen, 1991). Not only does this suggest that, through
filtration, removal of the protozoan community from the bacterial population minimizes predation, but it also implies that bacterial cells are able to
obtain available nutrients much more readily through reduced competition.
In the absence of other micro-organisms, E. coli has been shown to survive
for periods in excess of 260 days at temperatures ranging between 4 and
25 8C (Flint, 1987).
Survival in Stream Bed Sediments
Stream sediments may support microbial survival (Davies et al., 1995;
Van Donsel and Geldreich, 1971). Sorption to, and the subsequent sedimentation of, suspended particles provides increased protection to microbes by
limiting interactions with biotic and abiotic antagonistic factors such as
predation and sunlight or by increasing nutrient availability (Craig et al.,
2002). The significance of this microbial store was highlighted by ObiriDanso and Jones (2000), and by Craig et al. (2002) and Shiaris et al.
(1987), in which the numbers of FC bacteria in sediment were reported
be 1,000 and 10,000 times greater, respectively, than those found in the
overlying water column. This substantiates the earlier work of Grimes
(1975), in which channel dredging increased FC counts downstream. The
rise in bacterial concentrations was attributed to the disturbance and resuspension of the bed sediments and the FC bound to those sediments.
Medema et al. (1998) related the sedimentation of cells to Stoke’s law,
whereby the settling velocity is a function of particle size, water viscosity,
and the diVerence in density of the particle and water. Subsequently, those
cells that attach to suspended sediment upon entering a watercourse are
much more likely to settle with the bed sediment in contrast to freely
suspended cells in the water column (Wilkinson et al., 1995). This sediment
reservoir is capable of harboring potentially pathogenic bacteria for periods
amounting to several months because the favorable conditions and their
accumulation in the upper layers of the sediment allows for potential resuspension during times of high turbulence within the water body (Burton et al.,
1987). Equally, recreational use of surface waters may disrupt the bed
sediments and give rise to temporary health hazards associated with surface
This chapter has identified the two major components to consider in
relation to the emergence of potentially pathogenic micro-organisms in
grassland environments—survival and transfer. Ultimately, characteristic
survival curves must be combined with the dynamics of hydrology to appreciate the real extent of risk in terms of pathogen transmission to the wider
public. The literature available to the scientific community at present lacks
the bridging of these two fundamental components, though recent studies
tend to group these factors together.
Awareness of the environment as a reservoir for enteric micro-organisms
and the potential routes available to them has revealed a number of gaps in
our knowledge. In particular, the identification of hydrological connectivity
from surface applied sources to the aquatic receptor needs to be investigated.
The heterogeneity and hydrological complexity of agricultural catchments
means that transfer routes can vary their relative contaminant contribution
loads both spatially and temporally. Thus, extrapolation of our understanding of vertical and lateral flux processes observed at smaller scales cannot be
readily applied to the catchment scale. We must prioritize the transfer routes
of greatest significance in relation to where the maximum risk of hydrological connection between fecal sources and surface waters exists and then act
to reduce their potential threat. Hydrology is highlighted as the key component in governing pathogen emergence in receiving waters. Though this may
sound obvious, what remains to be fully catalogued, as with other diVuse
agricultural pollutants such as N and P, is a more comprehensive understanding of the role of hydrology. In addition to this, the concept of energy
exchanges at the soil surface and energies operating within the soil domain
itself deserves further investigation in order to explore associations with flow
velocities, colloid mobilization thresholds, and the potential particle association of these micro-organisms. The key risks all point to mobilization of
surface applied wastes. To protect watercourses, we need to manage the
environment and hydrology to work for us in curtailing pollution of our own
water and develop improved communication between scientists and farmers.
The other area of confusion is the lack of transferability of results relating
to cattle and herd pathogen prevalence, which stems from the large and diverse
range of microbiological methods available as detection tools. Whether it is
the choice of media to culture bacteria or the molecular technique adopted
to detect protozoan oocysts, diVerent laboratories across the world use an
array of methods that inevitably complicate the comparison of prevalence
data from country to country and between regions. However, at the same time
our ability to detect pathogens in the environment has increased markedly
through development of alternative molecular approaches that now function
as incredibly important tools in the arena of environmental detection.
While this chapter addresses grassland farming, it must be remembered
that arable farming may also act as a vector of disease transmission. In
particular, organic arable farming needs to be addressed in terms of its
relative risk contribution to the wider population through the direct consumption of contaminated food. The protection of surface waters can only
be achieved through development and continuous evolution of recommendations, regulatory guidelines, and legislation once the processes and apparent governing scenarios of pollutant transfer and delivery are understood.
The health risks associated with water-borne disease may be kept to a
minimum with eVective control of surface water quality, which may be
promoted through an improved understanding of sources, distribution,
delivery, and survival of pathogens in agricultural settings.
Abu-Ashour, J., Joy, D. M., Lee, H., Whiteley, H. R., and Zelin, S. (1998). Movement of
bacteria in unsaturated soil columns with macropores. Trans. ASAE 41, 1043–1050.
Abu-Ashour, J., and Lee, H. (2000). Transport of bacteria on sloping soil surfaces by runoV.
Environ. Toxicol. 15, 149–153.
Acea, M. J., Moore, C. R., and Alexander, M. (1988). Survival and growth of bacteria
introduced into soil. Soil Biol. Biochem. 20, 509–515.
Aislabie, J., Smith, J. J., Fraser, R., and McLeod, M. (2001). Leaching of bacterial indicators of
faecal contamination through four New Zealand soils. Aust. J. Soil Res. 39, 1397–1406.
Aitken, M. N. (2003). Impact of agricultural practices and river catchment characteristics on
river and bathing water quality. Water Sci. Technol. 48, 217–224.
Allaire-Leung, S., Gupta, S. C., and Moncrief, J. F. (2000). Water and solute movement in soil
as influenced by macropore characteristics—1. Macropore continuity. J. Contamin. Hydrol.
41, 283–301.
American Public Health Association (1998). ‘‘Standard Methods for the Examination of Water
and Wastewater’’, 20th ed., APHA, Washington.
Artz, R. R. E., and Killham, K. (2002). Survival of Escherichia coli O157:H7 in private drinking
water wells: Influences of protozoan grazing and elevated copper concentrations. FEMS
Microbiol. Lett. 216, 117–122.
Atwill, E. R., Hoar, B., Pereira, M. D. C., Tate, K. W., Rulofson, F., and Nader, G. (2003).
Improved quantitative estimates of low environmental loading and sporadic periparturient
shedding of Cryptosporidium parvum in adult beef cattle. Appl. Environ. Microbiol. 69,
Atwill, E. R., Hou, L. L., Karle, B. A., Harter, T., Tate, K. W., and Dahlgren, R. A. (2002).
Transport of Cryptosporidium parvum oocysts through vegetated buVer strips and
estimated filtration eYciency. Appl. Environ. Microbiol. 68, 5517–5527.
Barcina, I., Gonzalez, J. M., Iriberri, J., and Egea, L. (1989). EVect of visible-light on progressive
dormancy of Escherichia coli cells during the survival process in natural fresh-water. Appl.
Environ. Microbiol. 55, 246–251.
Bassler, H. A., Flood, S. J. A., Livak, K. J., Marmaro, J., Knorr, R., and Batt, C. A. (1995). Use
of a fluorogenic probe in a PCR-based assay for the detection of Listeria monocytogenes.
Appl. Environ. Microbiol. 61, 3724–3728.
Bessesen, M. T., Luo, Q., Rotbart, H. A., Blaser, M. J., and Ellison, R. T. (1990). Detection of
Listeria monocytogenes by using the polymerase chain-reaction. Appl. Environ. Microbiol.
56, 2930–2932.
Beven, K., and Germann, P. (1982). Macropores and water-flow in soils. Water Resour. Res. 18,
Bissonnette, G. K., Jezeski, J. J., McFetters, G. A., and Stuart, D. G. (1977). Evaluation of
recovery methods to detect coliforms in water. Appl. Environ. Microbiol. 33, 590–595.
Bogosian, G., Sammons, L. E., Morris, P. J. L., Oneil, J. P., Heitkamp, M. A., and Weber, D. B.
(1996). Death of the Escherichia coli K-12 strain W3110 in soil and water. Appl. Environ.
Microbiol. 62, 4114–4120.
Bowen, G. D., and Rovira, A. D. (1999). The rhizosphere and its management to improve plant
growth. Adv. Agron. 66, 1–102.
Bradford, S. A., and Schijven, J. (2002). Release of Cryptosporidium and Giardia from dairy calf
manure: Impact of solution salinity. Environ. Sci. Technol. 36, 3916–3923.
Braswell, J. R., and Hoadley, A. W. (1974). Recovery of Escherichia coli from chlorinated
secondary sewage. Appl. Environ. Microbiol. 28, 328–329.
Brenner, K. P., Rankin, C. C., Roybal, Y. R., Stelma, G. N., Scarpino, P. V., and Dufour, A. P.
(1993). New medium for the simultaneous detection of total coliforms and Escherichia coli
in water. Appl. Environ. Microbiol. 59, 3534–3544.
Briones, A. M., and Reichardt, W. (1999). Estimating microbial population counts by ‘most
probable number’ using Microsoft Excel (R). J. Microbiolog. Meth. 35, 157–161.
Brodsky, M. H., and Schiemann, D. A. (1975). Influence of coliform source on evaluation of
membrane filters. Appl. Microbiol. 30, 727–730.
Brown, M. W. R., Smith, A. W., Barker, J., Humphrey, T. J., and Dixon, B. (2002). E. coli 0157
persistence in the environment. Microbiology 148, 1–2.
Brush, C. F., Ghiorse, W. C., Anguish, L. J., Parlange, J. Y., and Grimes, H. G. (1999).
Transport of Cryptosporidium parvum oocysts through saturated columns. J. Environ.
Qual. 28, 809–815.
Buckhouse, J. C., and GiVord, G. F. (1976). Water quality implications of cattle grazing on a
semiarid watershed in southeastern Utah. J. Range Manage. 29, 109–113.
Burton, G. A., Gunnison, D., and Lanza, G. R. (1987). Survival of pathogenic bacteria in
various fresh-water sediments. Appl. Environ. Microbiol. 53, 633–638.
Buttle, J. M., and Leigh, D. G. (1997). The influence of artificial macropores on water and solute
transport in laboratory soil columns. J. Hydrol. 191, 290–313.
Byamukama, D., Kansiime, F., Mach, R. L., and Farnleitner, A. H. (2000). Determination of
Escherichia coli contamination with chromocult coliform agar showed a high level of
discrimination eYciency for diVering fecal pollution levels in tropical waters of Kampala,
Uganda. Appl. Environ. Microbiol. 66, 864–868.
Call, D. R., Borucki, M. K., and Loge, F. J. (2003). Detection of bacterial pathogens in
environmental samples using DNA microarrays. J. Microbiolog. Meth. 53, 235–243.
Campbell, G. R., Prosser, J., Glover, A., and Killham, K. (2001). Detection of Escherichia coli
0157:H7 in soil and water using multiplex PCR. J. Appl. Microbiol. 91, 1004–1010.
Camper, A. K., Hayes, J. T., Sturman, P. J., Jones, W. L., and Cunningham, A. B. (1993).
EVects of motility and adsorption rate coeYcient on transport of bacteria through
saturated porous-media. Appl. Environ. Microbiol. 59, 3455–3462.
Chadwick, D. R., and Chen, S. (2002). Manures. In ‘‘Agriculture, Hydrology and Water
Quality’’ (P. M. Haygarth and S. C. Jarvis, Eds.), pp. 57–82. CAB International,
Wallingford and New York.
Chapman, P. A., Siddons, C. A., Cerdan-Malo, A. T., and Harkin, M. A. (1997). A 1-year study of
Escherichia coli 0157 in cattle, sheep, pigs and poultry. Epidemiol. Infect. 119, 245–250.
Chung, E., Aldom, J. E., Carreno, R. A., Chagla, A. H., Kostrzynska, M., Lee, H., Palmateer,
G., Trevors, J. T., Unger, S., Xu, R., and De Grandis, S. A. (1999). PCR-based
quantitation of Cryptosporidium parvum in municipal water samples. J. Microbiolog.
Meth. 38, 119–130.
Clark, D. L., Milner, B. B., Stewart, M. H., Wolfe, R. L., and Olson, B. H. (1991). Comparativestudy of commercial 4-methylumbelliferyl-beta-d-glucuronide preparations with the standard methods membrane filtration fecal coliform test for the detection of Escherichia coli in
water samples. Appl. Environ. Microbiol. 57, 1528–1534.
Colwell, R. R., Brayton, P. R., Grimes, D. J., Roszak, D. B., Huq, S. A., and Palmer, L. M.
(1985). Viable but non-culturable Vibrio cholerae and related pathogens in the environment—implications for release of genetically engineered microorganisms. Bio-Technology
3, 817–820.
Cook, M. J., and Baker, J. L. (2001). Bacteria and nutrient transport to tile lines shortly after
application of large volumes of liquid swine manure. Trans. ASAE 44, 495–503.
Cools, D., Merckx, R., Vlassak, K., and Verhaegen, J. (2001). Survival of E. coli and
Enterococcus spp. derived from pig slurry in soils of diVerent texture. Appl. Soil Ecol. 17,
Crabill, C., Donald, R., Snelling, J., Foust, R., and Southam, G. (1999). The impact of sediment
fecal coliform reservoirs on seasonal water quality in Oak Creek, Arizona. Water Res. 33,
Craig, D. L., Fallowfield, H. J., and Cromar, N. J. (2002). Comparison of decay rates of faecal
indicator organisms in recreational coastal water and sediment. Water Sci. Tech. Water
Supply 2, 131–138.
Crump, B. C., Baross, J. A., and Simenstad, C. A. (1998). Dominance of particle-attached
bacteria in the Columbia River estuary, USA. Aquatic Microb. Ecol. 14, 7–18.
Culley, J. L. B., and Phillips, P. A. (1982). Bacteriological quality of surface and subsurface
runoV from manured sandy clay loam soil. J. Environ. Qual. 11, 155–158.
Dai, X., and Boll, J. (2003). Evaluation of attachment of Cryptosporidium parvum and Giardia
lamblia to soil particles. J. Environ. Qual. 32, 296–304.
Davenport, C. V., Sparrow, E. B., and Gorden, R. C. (1976). Fecal indicator bacteria
persistence under natural conditions in an ice covered river. Appl. Environ. Microbiol. 32,
Davies, C. M., Long, J. A. H., Donald, M., and Ashbolt, N. J. (1995). Survival of fecal
microorganisms in marine and fresh-water sediments. Appl. Environ. Microbiol. 61,
DeGuise, K. A., Wolfe, M. L., and Dillaha III, T. A. (1999). ‘‘A Foundation for Modeling
Bacterial Contamination of Surface Waters due to Nonpoint Sources’’. ASAE Annual
International Meeting. Toronto, Canada.
Drozd, C., and Schwartzbrod, J. (1996). Hydrophobic and electrostatic cell surface properties of
Cryptosporidium parvum. Appl. Environ. Microbiol. 62, 1227–1232.
Economic Research Service/U.S. Department of Agriculture (ERS/USDA) (2003). ‘‘Economics
of Foodborne Disease: Salmonella’’. http://www.ers.usda.gov/briefing/FoodborneDisease/
Ehlers, W. (1975). Observations on earthworm channels and infiltration on tilled and untilled
loess soil. Soil Sci. 119, 242–249.
England, L. S., Lee, H., and Trevors, J. T. (1993). Bacterial survival in soil—eVect of clays and
protozoa. Soil Biol. Biochem. 25, 525–531.
Entry, J. A., Hubbard, R. K., Thies, J. E., and Fuhrmann, J. J. (2000). The influence of
vegetation in riparian filterstrips on coliform bacteria: II. Survival in soils. J. Environ.
Qual. 29, 1215–1224.
Faust, M. A. (1982). Relationship between land-use practices and fecal bacteria in soils.
J. Environ. Qual. 11, 141–146.
Fayer, R., Trout, J. M., and Jenkins, M. C. (1998). Infectivity of Cryptosporidium parvum
oocysts stored in water at environmental temperatures. J. Parasitol. 84, 1165–1169.
Fenlon, D. R., Ogden, I. D., Vinten, A., and Svoboda, I. (2000). The fate of Escherichia coli
and E. coli O157 in cattle slurry after application to land. J. Appl. Microbiol. 88, 149S–156S.
Fleisher, J. M., and McFadden, R. T. (1980). Obtaining precise estimates in coliform
enumeration. Water Res. 14, 477–483.
Flint, K. P. (1987). The long-term survival of Escherichia coli in river water. J. Appl. Bacteriol.
63, 261–270.
Fontes, D. E., Mills, A. L., Hornberger, G. M., and Herman, J. S. (1991). Physical and chemical
factors influencing transport of microorganisms through porous-media. Appl. Environ.
Microbiol. 57, 2473–2481.
Fricker, E. J., and Fricker, C. R. (1996). Use of two presence/absence systems for the detection
of E. coli and coliforms from water. Water Res. 30, 2226–2228.
Fukushima, H., Hoshina, K., and Gomyoda, M. (1999). Long-term survival of Shiga toxinproducing Escherichia coli O26, O113, and O157 in bovine feces. Appl. Environ. Microbiol.
65, 5177–5181.
Gagliardi, J. V., and Karns, J. S. (2000). Leaching of Escherichia coli 0157:H7 in diverse soils under
various agricultural management practices. Appl. Environ. Microbiol. 66, 877–883.
Gameson, A., and Gould, D. (1975). EVects of solar radiation on the mortality of some
terrestrial bacteria in seawater. In ‘‘Discharge of Sewage from Sea Outfalls’’ (A. Gameson,
Ed.), pp. 209–217. Pergamon Press, London.
Gammack, B. M., Paterson, E., Kemp, J. S., Cresser, M. S., and Killham, K. (1992). Factors
aVecting movement of micro-organisms in soils. In ‘‘Soil Biochemistry’’ (G. Stotzky and
L. M. Bolla, Eds.), 7, pp. 263–305. Marcel Dekker, Inc, New York.
Gannon, J. T., Mingelgrin, U., Alexander, M., and Wagenet, R. J. (1991). Bacterial transport
through homogeneous soil. Soil Biol. Biochem. 23, 1155–1160.
Geissler, K., Manafi, M., Amoros, I., and Alonso, J. L. (2000). Quantitative determination of
total coliforms and Escherichia coli in marine waters with chromogenic and fluorogenic
media. J. Appl. Microbiol. 88, 280–285.
George, I., Petit, M., Theate, C., and Servais, P. (2001). Use of rapid enzymatic assays to study
the distribution of faecal coliforms in the Seine river (France). Water Sci. Technol. 43, 77–80.
Gerba, C. P. (1996). Pathogens in the environment. In ‘‘Pollution Science’’ (I. L. Pepper,
C. P. Gerba, and M. L. Brusseau, Eds.), pp. 279–294. Academic Press, New York.
Gerba, C. P., Wallis, C., and Melnick, J. L. (1975). Fate of wastewater bacteria and viruses in
soil. J. Irrig. Drain. Div. 101, 157–174.
Green, B. L., Clausen, E., and Litsky, W. (1975). Comparison of the new millipore HC
with conventional membrane filters for the enumeration of fecal coliform bacteria. Appl.
Microbiol. 30, 697–699.
GriYths, B. S., and Young, I. M. (1994). The eVects of soil-structure on protozoa in a clay-loam
soil. Eur. J. Soil Sci. 45, 285–292.
Grimes, D. J. (1975). Release of sediment bound fecal coliforms by dredging. Appl. Microbiol.
29, 109–111.
Hallaran, G., and Sumner, J. (2001). Prevalence of E. coli O157 in dairy cows presented for
slaughter in Victoria. Aust. Vet. J. 79, 707–708.
Harter, T., Wagner, S., and Atwill, E. R. (2000). Colloid transport and filtration of Cryptosporidium parvum in sandy soils and aquifer sediments. Environ. Sci. Technol. 34, 62–70.
Harvey, R. W. (1997). Microorganisms as tracers in groundwater injection and recovery
experiments: A review. FEMS Microbiol. Rev. 20, 461–472.
Hedges, A. J. (2002). Estimating the precision of serial dilutions and viable bacterial counts. Intl.
J. Food Microbiol. 76, 207–214.
Heijnen, C. E., and Vanveen, J. A. (1991). A determination of protective microhabitats for
bacteria introduced into soil. FEMS Microbiol. Ecol. 85, 73–80.
Heinonen-Tanski, H., Niskanen, E. M., Salmela, P., and Lanki, E. (1998). Salmonella in animal
slurry can be destroyed by aeration at low temperatures. J. Appl. Microbiol. 85, 277–281.
Heinonen-Tanski, H., and Uusi-Kamppa, J. (2001). RunoV of faecal microorganisms and
nutrients from perennial grass ley after application of slurry and mineral fertiliser. Water
Sci. Technol. 43, 143–146.
Hekman, W. E., Heijnen, C. E., Burgers, S., Vanveen, J. A., and Vanelsas, J. D. (1995).
Transport of bacterial inoculants through intact cores of 2 diVerent soils as aVected by
water percolation and the presence of wheat plants. FEMS Microbiol. Ecol. 16, 143–157.
Herbert, R. A. (1990). Methods for enumerating microorganisms and determining biomass in
natural environments. Meth. Microbiol. 22, 1–39.
Hoar, B. R., Atwill, E. R., Elmi, C., Utterback, W. W., and Edmondson, A. J. (1999).
Comparison of fecal samples collected per rectum and oV the ground for estimation of
environmental contamination attributable to beef cattle. Am. J. Vet. Res. 60, 1352–1356.
Hoar, B. R., Atwill, E. R., and Farver, T. B. (2000). Estimating maximum possible environmental loading amounts of Cryptosporidium parvum attributable to adult beef cattle. Quant.
Microbiol. 2, 21–36.
Hooda, P. S., Edwards, A. C., Anderson, H. A., and Miller, A. (2000). A review of water quality
concerns in livestock farming areas. Sci. Total Environ. 250, 143–167.
Hornberger, G. M., Mills, A. L., and Herman, J. S. (1992). Bacterial transport in porousmedia—evaluation of a model using laboratory observations. Water Resour. Res. 28,
Huston, C. L., Wittum, T. E., and Love, B. C. (2002). Persistent fecal Salmonella shedding in
five dairy herds. J. Am. Vet. Med. Assoc. 220, 650–655.
Huysman, F., and Verstraete, W. (1993). Water-facilitated transport of bacteria in unsaturated
soil columns—influence of cell-surface hydrophobicity and soil properties. Soil Biol.
Biochem. 25, 83–90.
Jacobs, N. J., Zeigler, W. L., Reed, F. C., Stukel, T. A., and Rice, E. W. (1986). Comparison
of membrane filter, multiple-fermentation-tube, and presence-absence techniques for
detecting total coliforms in small community water systems. Appl. Environ. Microbiol. 51,
Jacobsen, O. H., Moldrup, P., Larsen, C., Konnerup, L., and Petersen, L. W. (1997). Particle
transport in macropores of undisturbed soil columns. J. Hydrol. 196, 185–203.
Jamieson, R. C., Gordon, R. J., Sharples, K. E., Stratton, G. W., and Madani, A. (2002).
Movement and persistence of faecal bacteria in agricultural soils and subsurface drainage
water: A review. Can. Biosys. Eng. 44, 1.1–1.9.
Jenkins, M. B., Bowman, D. D., Fogarty, E. A., and Ghiorse, W. C. (2002). Cryptosporidium
parvum oocyst inactivation in three soil types at various temperatures and water potentials.
Soil Biol. Biochem. 34, 1101–1109.
Jiang, X., Morgan, J., and Doyle, M. P. (2002). Fate of Escherichia coli 0157:H7 in manure
amended soil. Appl. Environ. Microbiol. 68, 2605–2609.
Jin, Y., and Flury, M. (2002). Fate and transport of viruses in porous media. Adv. Agron. 77,
Joachim, A., Krull, T., Schwarzkopf, J., and Daugschies, A. (2003). Prevalence and control of
bovine cryptosporidiosis in German dairy herds. Vet. Parasitol. 112, 277–288.
Joergensen, R. G., Kuntzel, H., Scheu, S., and Seitz, D. (1998). Movement of faecal indicator
organisms in earthworm channels under a loamy arable and grassland soil. Appl. Soil Ecol.
8, 1–10.
Joergensen, R. G., and Seitz, D. (1998). Depth transfer of fecal indicator organisms in an arable
clayey soil. Zeitschrift Pflanzenernahrung Bodenkunde 161, 309–314.
Johnson, W. P., Blue, K. A., Logan, B. E., and Arnold, R. G. (1995). Modeling bacterial
detachment during transport through porous-media as a residence-time-dependent process.
Water Resour. Res. 31, 2649–2658.
Jones, D. L. (1999). Potential health risks associated with the persistence of Escherichia coli
O157 in agricultural environments. Soil Use Manage. 15, 76–83.
Jones, D. L., Campbell, G., and Kaspar, C. W. (2002). Human enteric pathogens.
In ‘‘Agriculture, Hydrology and Water Quality’’ (P. M. Haygarth and S. C. Jarvis, Eds.),
pp. 133–153. CAB International, Wallingford and New York.
Jones, K. (2001). Campylobacters in water, sewage and the environment. J. Appl. Microbiol. 90,
Ketley, J. M. (1997). Pathogenesis of enteric infection by Campylobacter. Microbiology–UK 143,
Kim, S. B., and Corapcioglu, M. Y. (2002). Contaminant transport in dual-porosity media with
dissolved organic matter and bacteria present as mobile colloids. J. Contamin. Hydrol. 59,
Klein, D. A., and Casida, L. E. (1967). Escherichia coli die-out from natural soil as related to
nutrient availability and the indigenous microflora. Can. J. Microbiol. 13, 1461–1470.
Korhonen, L. K., and Martikainen, P. J. (1991). Survival of Escherichia coli and Campylobacter
jejuni in untreated and filtered lake water. J. Appl. Bacteriol. 71, 379–382.
Kovacs, F., and Tamasi, G. (1979). Survival times of bacteria in liquid manure. Acta Vet.
Academ. Hung. 27, 47–54.
Kramer, M. H., Herwaldt, B. L., Craun, G. F., Calderon, R. L., and Juranek, D. D. (1996).
Waterborne disease: 1993 and 1994. J. Am. Water Works Assoc. 88, 66–80.
Kretzschmar, R., Borkovec, M., Grolimund, D., and Elimelech, M. (1999). Mobile subsurface
colloids and their role in contaminant transport. Adv. Agron. 66, 121–193.
Kudva, I. T., Blanch, K., and Hovde, C. J. (1998). Analysis of Escherichia coli 0157:H7 survival
in ovine or bovine manure and manure slurry. Appl. Environ. Microbiol. 64, 3166–3174.
Lahti, E., Ruoho, I., Rantala, L., Hanninen, M. L., and Honkanen-Buzalski, T. (2003).
Longitudinal study of Escherichia coli O157 in a cattle finishing unit. Appl. Environ.
Microbiol. 69, 554–561.
Lechevallier, M. W., Schulz, W., and Lee, R. G. (1991). Bacterial nutrients in drinking-water.
Appl. Environ. Microbiol. 57, 857–862.
Lee, K. E. (1985). ‘‘Earthworms, Their Ecology and Relationships with Soils and Land Use’’.
Academic Press, New York.
LeJeune, J. T., Besser, T. E., Merrill, N. L., Rice, D. H., and Hancock, D. D. (2001). Livestock
drinking water microbiology and the factors influencing the quality of drinking water
oVered to cattle. J. Dairy Sci. 84, 1856–1862.
Lin, S. D. (1976). Evaluation of millipore HA and HC membrane filters for the enumeration of
indicator bacteria. Appl. Environ. Microbiol. 32, 300–302.
Ling, T. Y., Achberger, E. C., Drapcho, C. M., and Bengtson, R. L. (2002). Quantifying
adsorption of an indicator bacteria in a soil-water system. Trans. ASAE 45, 669–674.
Locking, M. E., O’Brien, S. J., Reilly, W. J., Wright, E. M., Campbell, D. M., Coia, J. E., Browning,
L. M., and Ramsay, C. N. (2001). Risk factors for sporadic cases of Escherichia coli O157
infection: The importance of contact with animal excreta. Epidemiol. Infect. 127, 215–220.
Lund, E., and Nissen, B. (1983). The survival of enteroviruses in aerated and unaerated cattle
and pig slurry. Agricult. Wastes 7, 221–233.
Macleod, F. A., Lappin-Scott, H. M., and Costerton, J. W. (1988). Plugging of a model rock
system by using starved bacteria. Appl. Environ. Microbiol. 54, 1365–1372.
Madigan, M. T., Martinko, J. M., and Parker, J. (2003). ‘‘Brock Biology of Microorganisms’’,
10th ed. Prentice Hall, New Jersey.
Maki, R. P., and Hicks, R. E. (2002). Salmonella typhimurium survival and viability is unaltered
by suspended particles in freshwater. J. Environ. Qual. 31, 1702–1709.
Massa, S., Brocchi, G. F., Peri, G., Altieri, C., and Mammina, C. (2001). Evaluation of recovery
methods to detect faecal streptococci in polluted waters. Lett. Appl. Microbiol. 32, 298–302.
Massa, S., Grandi, M., Poda, G., Cesarondi, D., Cangamella, F., and Trovatelli, L. D. (1989).
Coliform detection from river waters: Comparison between mPN and MF techniques.
Water Air Soil Pollution 43, 135–145.
Maule, A. (1999). Environmental aspects of E. coli 0157. Intl. Food Hygiene 9, 21–23.
Maule, A. (2000). Survival of verocytotoxigenic Escherichia coli 0157 in soil, water and on
surfaces. J. Appl. Microbiol. Symp. Suppl. 88, 71S–78S.
Mawdsley, J. L., Bardgett, R. D., Merry, R. J., Pain, B. F., and Theodorou, M. K. (1995).
Pathogens in livestock waste, their potential for movement through soil and environmental
pollution. Appl. Soil Ecol. 2, 1–15.
Mawdsley, J. L., Brooks, A. E., Merry, R. J., and Pain, B. F. (1996). Use of a novel soil tilting
table apparatus to demonstrate the horizontal and vertical movement of the protozoan
pathogen Cryptosporidium parvum in soil. Biol. Fertil. Soils 23, 215–220.
Maxey, R. B. (1970). Non-lethal injury and limitations of recovery of coliform organisms on
selective media. J. Milk Food Technol. 33, 445–448.
McCaulou, D. R., Bales, R. C., and Arnold, R. G. (1995). EVect of temperature-controlled
motility on transport of bacteria and microspheres through saturated sediment. Water
Resour. Res. 31, 271–280.
McDonald, A., and Kay, D. (1981). Enteric bacterial concentrations in reservoir feeder
streams—baseflow characteristics and response to hydrograph events. Water Res. 15,
McDowell-Boyer, L. M., Hunt, J. R., and Sitar, N. (1986). Particle transport through porous
media. Water Resour. Res. 22, 1901–1921.
McEvoy, J. M., Doherty, A. M., Sheridan, J. J., Blair, I. S., and McDowell, D. A. (2003). The
prevalence of Salmonella spp. in bovine faecal, rumen and carcass samples at a commercial
abattoir. J. Appl. Microbiol. 94, 693–700.
McGechan, M. B. (2002). Transport of particulate and colloid-sorbed contaminants through
soil, part 2: Trapping processes and soil pore geometry. Biosys. Eng. 83, 387–395.
McGechan, M. B., and Lewis, D. R. (2002). Transport of particulate and colloid-sorbed
contaminants through soil, part 1: General principles. Biosys. Eng. 83, 255–273.
McGee, P., Bolton, D. J., Sheridan, J. J., Earley, B., Kelly, G., and Leonard, N. (2002). Survival
of Escherichia coli O157:H7 in farm water: Its role as a vector in the transmission of the
organism within herds. J. Appl. Microbiol. 93, 706–713.
McGee, P., Bolton, D. J., Sheridan, J. J., Earley, B., and Leonard, N. (2001). The survival of
Escherichia coli O157: H7 in slurry from cattle fed diVerent diets. Lett. Appl. Microbiol. 32,
McGuirk, S. M., and Peek, S. (2003). Salmonellosis in cattle: A review. American Association of
Bovine Practitioners 36th Annual Conference, September 15–17, 2003. Columbus, OH.
McLeod, M., Schipper, L. A., and Taylor, M. D. (1998). Preferential flow in a well drained and a
poorly drained soil under diVerent overhead irrigation regimes. Soil Use Manage. 14, 96–100.
Medema, G. J., Schets, F. M., Teunis, P. F. M., and Havelaar, A. H. (1998). Sedimentation of
free and attached Cryptosporidium oocysts and Giardia cysts in water. Appl. Environ.
Microbiol. 64, 4460–4466.
Miettinen, I. T., Zacheus, O., von BonsdorV, C. H., and Vartiainen, T. (2001). Waterborne
epidemics in Finland in 1998–1999. Water Sci. Technol. 43, 67–71.
Mills, A. L., and Saiers, J. E. (1993). Particle-associated transport of pollutants in subsurface
environments. In ‘‘Particulate Matter and Aquatic Contaminants’’ (S. S. Rao, Ed.),
pp. 105–126. CRC Press, Inc, Florida.
Mitscherlich, E., and Marth, E. H. (1984). ‘‘Microbial Survival in the Environment’’. SpringerVerlag, Berlin.
Mofidi, A. A., Meyer, E. A., Wallis, P. M., Chou, C. I., Meyer, B. P., Ramalingam, S., and
CoVey, B. M. (2002). The eVect of UV light on the inactivation of Giardia lamblia and
Giardia muris cysts as determined by animal infectivity assay (P-2951-01). Water Res. 36,
Morita, S., Namikoshi, A., Hirata, T., Oguma, K., Katayama, H., Ohgaki, S., Motoyama, N.,
and Fujiwara, M. (2002). EYcacy of UV irradiation in inactivating Cryptosporidium
parvum oocysts. Appl. Environ. Microbiol. 68, 5387–5393.
Mubiru, D. N., Coyne, M. S., and Grove, J. H. (2000). Mortality of Escherichia coli O157:H7 in
two soils with diVerent physical and chemical properties. J. Environ. Qual. 29, 1821–1825.
Nagels, J. W., Davies-Colley, R. J., Donnison, A. M., and Muirhead, R. W. (2002). Faecal
contamination over flood events in a pastoral agricultural stream in New Zealand. Water
Sci. Technol. 45, 45–52.
Nicholson, F. A., Hutchinson, M. L., Smith, K. A., Keevil, C. W., Chambers, B. J., and Moore,
A. (2000). A study on farm manure applications to agricultural land and an assessment of
the risks of pathogen transfer into the food chain. A report to The Ministry of Agriculture
Fisheries and Food, UK.
Nielsen, E. M. (2002). Occurrence and strain diversity of thermophilic Campylobacters in cattle
of diVerent age groups in dairy herds. Lett. Appl. Microbiol. 35, 85–89.
Nielsen, J. L., Mikkelsen, L. H., and Nielsen, P. H. (2001). In Situ detection of cell surface
hydrophobicity of probe-defined bacteria in activated sludge. Water Sci. Technol. 43, 97–103.
Nilsson, L., Oliver, D., and Kjelleberg, S. (1991). Resuscitation of Vibrio vulnificus from the
viable but non-culturable state. J. Bacteriol. 173, 5054–5059.
Obiri-Danso, K., and Jones, K. (2000). Intertidal sediments as reservoirs for hippurate negative
Campylobacters, Salmonellae and faecal indicators in three EU recognised bathing waters
in North West England. Water Res. 34, 519–527.
Ogden, I. D., Hepburn, N. F., MacRae, M., Strachan, N. J. C., Fenlon, D. R., Rusbridge, S. M.,
and Pennington, T. H. (2002). Long-term survival of Escherichia coli O157 on pasture
following an outbreak associated with sheep at a scout camp. Lett. Appl. Microbiol. 34,
O’Handley, R. M., Cockwill, C., McAllister, T. A., Jelinski, M., Morck, D. W., and Olson, M. E.
(1999). Duration of naturally acquired giardiosis and cryptosporidiosis in dairy calves and
their association with diarrhea. J. Am. Vet. Med. Assoc. 214, 391–396.
Olson, M. E., Goh, J., Phillips, M., Guselle, N., and McAllister, T. A. (1999). Giardia cyst and
Cryptosporidium oocyst survival in water, soil, and cattle feces. J. Environ. Qual. 28,
Olson, M. E., Thorlakson, C. L., Deselliers, L., Morck, D. W., and McAllister, T. A. (1997).
Giardia and Cryptosporidium in Canadian farm animals. Vet. Parasitol. 68, 375–381.
Opperman, M. H., McBain, L., and Wood, M. (1987). Movement of cattle slurry through soil
by Eisenia foetida (Savigny). Soil Biol. Biochem. 19, 741–745.
Paerl, H. W. (1975). Microbial attachment to particles in marine and freshwater ecosystems.
Microb. Ecol. 2, 73–83.
Palmateer, G., McLean, D. E., Kutas, W. L., and Meissner, S. M. (1993). Suspended
particulate/bacterial interaction in agricultural drains. In ‘‘Particulate Matter and Aquatic
Contaminants’’ (S. S. Rao, Ed.), pp. 1–40. CRC Press, Inc., Florida.
Pell, A. N. (1997). Manure and microbes: Public and animal health problem? J. Dairy Sci. 80,
Personne, J. C., Poty, F., Vaute, L., and Drogue, C. (1998). Survival, transport and dissemination of Escherichia coli and enterococci in a fissured environment. Study of a flood in a
karstic aquifer. J. Appl. Microbiol. 84, 431–438.
Postma, J., and Vanveen, J. A. (1990). Habitable pore-space and survival of Rhizobium
leguminosarum biovar trifolii introduced into soil. Microb. Ecol. 19, 149–161.
Postma, J., Vanveen, J. A., and Walter, S. (1989). Influence of diVerent initial soil-moisture
contents on the distribution and population-dynamics of introduced Rhizobium
leguminosarum biovar trifolii. Soil Biol. Biochem. 21, 437–442.
Quinton, J. N., Tyrrel, S. F., and Ramos, M. C. (2003). The eVect of incorporating slurries on
the transport of faecal coliforms in overland flow. Soil Use Manage. 19, 185–186.
Rahe, T. M., Hagedorn, C., McCoy, E. L., and Kling, G. F. (1978). Transport of antibioticresistant Escherichia coli through Western Oregon hillslope soils under conditions of
saturated flow. J. Environ. Qual. 7, 487–494.
Rasmussen, M. A., Cray, W. C., Casey, T. A., and Whipp, S. C. (1993). Rumen contents as a
reservoir of enterohemorrhagic Escherichia coli. FEMS Microbiol. Lett. 114, 79–84.
Reddy, K. R., Khaleel, R., and Overcash, M. R. (1981). Behavior and transport of microbial
pathogens and indicator organisms in soils treated with organic wastes. J. Environ. Qual.
10, 255–266.
Reynolds, P. J., Sharma, P., Jenneman, G. E., and McInerney, M. J. (1989). Mechanisms of
microbial movement in subsurface materials. Appl. Environ. Microbiol. 55, 2280–2286.
Rice, E. W., Allen, M. J., and Edberg, S. C. (1990). EYcacy of beta-glucuronidase assay
for identification of Escherichia coli by the defined-substrate technology. Appl. Environ.
Microbiol. 56, 1203–1205.
Rice, E. W., and Johnson, C. H. (2000). Short communication: Survival of Escherichia coli
O157:H7 in dairy cattle drinking water. J. Dairy Sci. 83, 2021–2023.
Rice, E. W., Johnson, C. H., Wild, D. K., and Reasoner, D. J. (1992). Survival of Escherichia coli
O157:H7 in drinking-water associated with a waterborne disease outbreak of hemorrhagic
colitis. Lett. Appl. Microbiol. 15, 38–40.
Rodgers, P., Soulsby, C., Hunter, C., and Petry, J. (2003). Spatial and temporal bacterial quality
of a lowland agricultural stream in northeast Scotland. Sci. Total Environ. 314–316, 289–302.
Rogasik, H., Crawford, J. W., Wendroth, O., Young, I. M., Joschko, M., and Ritz, K. (1999).
Discrimination of soil phases by dual energy x-ray tomography. Soil Sci. Soc. Am. J. 63,
Rollins, D. M., and Colwell, R. R. (1986). Viable but non-culturable stage of Campylobacter
jejuni and its role in survival in the natural aquatic environment. Appl. Environ. Microbiol.
52, 531–538.
Rompre, A., Servais, P., Baudart, J., de-Roubin, M. R., and Laurent, P. (2002). Detection and
enumeration of coliforms in drinking water: Current methods and emerging approaches.
J. Microbiolog. Meth. 49, 31–54.
Roszak, D. B., and Colwell, R. R. (1987). Survival strategies of bacteria in the naturalenvironment. Microbiolog. Rev. 51, 365–379.
Saini, R., Halverson, L. J., and Lorimor, J. C. (2003). Rainfall timing and frequency influence on
leaching of Escherichia coli RS2G through soil following manure application. J. Environ.
Qual. 32, 1865–1872.
Sargeant, J. M., Gillespie, J. R., Oberst, R. D., Phebus, R. K., Hyatt, D. R., Bohra, L. K., and
Galland, J. C. (2000). Results of a longitudinal study of the prevalence of Escherichia coli
O157 : H7 on cow-calf farms. Am. J. Vet. Res. 61, 1375–1379.
Scarce, L. E., Rubenstein, S. H., and Megregian, S. (1964). Survival of indicator bacteria
in receiving waters under various conditions. Great Lakes Res. Div. Univ. Mich. 11,
Schijven, J. F., Hassanizadeh, S. M., and de Bruin, H. A. M. (2002). Column experiments to
study nonlinear removal of bacteriophages by passage through saturated dune sand.
J. Contamin. Hydrol. 58, 243–259.
Scottish Executice and Food Standards Agency Scotland (2001). ‘‘Joint FSA/SE Task Force on
E. coli 0157—Final Report’’. http://www.foodstandards.gov.uk/scotland/ecoli.htm.
Sharma, S., Sachdeva, P., and Virdi, J. S. (2003). Emerging water-borne pathogens. Appl.
Microbiol. Biotechnol. 61, 424–428.
Sharpe, A. N., and Michaud, G. L. (1975). Enumeration of high numbers of bacteria using
hydrophobic grid-membrane filters. Appl. Microbiol. 30, 519–524.
Shere, J. A., Bartlett, K. J., and Kaspar, C. W. (1998). Longitudinal study of Escherichia coli
O157 : H7 dissemination on four dairy farms in Wisconsin. Appl. Environ. Microbiol. 64,
Shiaris, M. P., Rex, A. C., Pettibone, G. W., Keay, K., McManus, P., Rex, M. A., Ebersole, J.,
and Gallagher, E. (1987). Distribution of indicator bacteria and Vibrio-parahaemolyticus in
sewage-polluted intertidal sediments. Appl. Environ. Microbiol. 53, 1756–1761.
Shipe, E. L., and Cameron, G. M. (1954). A comparison of the membrane filter with the most
probable number method for coliform determinations from several waters. Appl. Microbiol.
2, 85–88.
Simoni, S. F., Harms, H., Bosma, T. N. P., and Zehnder, A. J. B. (1998). Population
heterogeneity aVects transport of bacteria through sand columns at low flow rates. Environ.
Sci. Technol. 32, 2100–2105.
Sinton, L. W., Hall, C. H., Lynch, P. A., and Davies-Colley, R. J. (2002). Sunlight inactivation
of fecal indicator bacteria and bacteriophages from waste stabilization pond eZuent in
fresh and saline waters. Appl. Environ. Microbiol. 68, 1122–1131.
Sjogren, R. E. (1994). Prolonged survival of an environmental Escherichia coli in laboratory soil
microcosms. Water Air Soil Pollution 75, 389–403.
Smith, M. S., Thomas, G. W., White, R. E., and Ritonga, D. (1985). Transport of Escherichia
coli through intact and disturbed soil columns. J. Environ. Qual. 14, 87–91.
Stanley, K. N., and Jones, K. (2003). Cattle and sheep farms as reservoirs of Campylobacter.
J. Appl. Microbiol. 94, 104S–113S.
Stanley, K. N., Wallace, J. S., Currie, J. E., Diggle, P. J., and Jones, K. (1998). The seasonal
variation of thermophilic Campylobacters in beef cattle, dairy cattle and calves. J. Appl.
Microbiol. 85, 472–480.
Stoddard, C. S., Coyne, M. S., and Grove, J. H. (1998). Fecal bacteria survival and infiltration
through a shallow agricultural soil: Timing and tillage eVects. J. Environ. Qual. 27, 1516–1523.
Stotzky, G., and Rem, L. T. (1966). Influence of clay minerals on micro-organisms I: Montmorillonite and kaolinite on bacteria. Can. J. Microbiol. 12, 547–563.
Straub, T. M., and Chandler, D. P. (2003). Towards a unified system for detecting waterborne
pathogens. J. Microbiolog. Meth. 53, 185–197.
Sueiro, R. A., Araujo, M., Santos, C. J., Gomez, M. J., and Garrido, M. J. (2001). Evaluation of
Coli-ID and MUG Plus media for recovering Escherichia coli and other coliform bacteria
from groundwater samples. Water Sci. Technol. 43, 213–216.
Tan, Y., Bond, W. J., and M., G. D. (1992). Transport of bacteria during unsteady unsaturated
soil water flow. Soil Sci. Soc. Am. J. 56, 1331–1341.
Tappeser, B., Jager, M., and Eckelkamp, C. (1998). ‘‘Survival, Persistence, Transfer—An
Update on Current Knowledge on GMOs and the Fate of Their Recombinant DNA’’.
Thelin, R., and GiVord, G. F. (1983). Faecal coliform release patterns from faecal material of
cattle. J. Environ. Qual. 12, 57–63.
Theron, J., and Cloete, T. E. (2002). Emerging waterborne infections: Contributing factors,
agents and detection tools. Crit. Rev. Microbiol. 28, 1–26.
Thomas, C., Gibson, H., Hill, D. J., and Mabey, M. (1999a). Campylobacter epidemiology: An
aquatic perspective. J. Appl. Microbiol. 85, 168S–177S.
Thomas, C., Hill, D. J., and Mabey, M. (1999b). Evaluation of the eVect of temperature and
nutrients on the survival of Campylobacter spp. in water microcosms. J. Appl. Microbiol.
86, 1024–1032.
Thorpe, I. S., Prosser, J. I., Glover, L. A., and Killham, K. (1996). The role of earthworm
Lumbricus terrestuis in the transport of bacterial inocula through soil. Soil Biol. Biochem.
23, 132–139.
Thullner, M., Mauclaire, L., Scrhroth, M. H., Kinzelbach, W., and Zeyer, J. (2002). Interaction
between water flow and spatial distribution of microbial growth in a two-dimensional flow
field in saturated porous media. J. Contamin. Hydrol. 58, 169–189.
Tim, U. S., Mostaghimi, S., and Dillaha III, T. A. (1988). ‘‘Modelling the Movement and
Persistence of Bacteria and Viruses in Porous Media’’. ASAE Meeting Presentation, Paper
No. 88–2627.
Tobin, R. S., and Dutka, B. J. (1977). Comparison of the surface structure, metal binding, and
fecal coliform recoveries of nine membrane filters. Appl. Environ. Microbiol. 34, 69–79.
Trevena, W. B., Willshaw, G. A., Cheasty, T., Domingue, G., and Wray, C. (1999). Zoonotic
transmission of vero cytotoxin producing Escherichia coli 0157 infection from farm animals
to humans within Cornwall and West Devon, UK. Communic. Dis. Public Health 2, 263–268.
Tyrell, S. F., and Quinton, J. N. (2003). Overland flow transport of pathogens from agricultural
land receiving faecal wastes. J. Appl. Microbiol. 94, 87S–93S.
Unwin, R. J., Pain, B. F., and Whinham, W. N. (1986). The eVect of rate and time of application
of nitrogen in cow slurry on grass cut for silage. Agricult. Wastes 15, 253–268.
USDA (1997). ‘‘An Update: Escherichia coli 0157:H7 in Humans and Cattle’’. Centers for
Epidemiology and Animal Health. Fort Collins, Colorado.
Van Donsel, D. J., and Geldreich, E. E. (1971). Relationships of Salmonella to fecal coliforms in
bottom sediments. Water Res. 5, 1079–1087.
Van Donsel, D. J., Geldreich, E. E., and Clarke, N. A. (1967). Seasonal variations in survival of
indicator bacteria in soil and their contribution to storm-water pollution. Appl. Microbiol.
15, 1362–1370.
Van Poucke, S. O., and Nelis, H. J. (2000). Rapid detection of fluorescent and chemiluminescent
total coliforms and Escherichia coli on membrane filters. J. Microbiolog. Meth. 42, 233–244.
Vinten, A. J. A., Lewis, D. R., Fenlon, D. R., Leach, K. A., Howard, R., Svoboda, I., and
Ogden, I. (2002). Fate of Escherichia coli and Escherichia coli O157 in soils and drainage
water following cattle slurry application at 3 sites in southern Scotland. Soil Use Manage.
18, 223–231.
Wang, G. D., and Doyle, M. P. (1998). Survival of enterohemorrhagic Escherichia coli O157:H7
in water. J. Food Protect. 61, 662–667.
Wang, G. D., Zhao, T., and Doyle, M. P. (1996). Fate of enterohemorrhagic Escherichia coli
O157:H7 in bovine feces. Appl. Environ. Microbiol. 62, 2567–2570.
Warnemuende, E. A., and Kanwar, R. S. (2002). EVects of swine manure application on
bacterial quality of leachate from intact soil columns. Transact. ASAE 45, 1849–1857.
Warnes, S., and Keevil, C. W. (2003). Survival of Cryptosporidium parvum in faecal wastes and
salad crops. In ‘‘Cryptosporidium parvum in Food and Water’’ (G. DuVy, Ed.), pp. 15–24.
Teagasc, Dublin.
Wesley, I. V., Wells, S. J., Harmon, K. M., Green, A., Schroeder-Tucker, L., Glover, M., and
Siddique, I. (2000). Fecal shedding of Campylobacter and Arcobacter spp. in dairy cattle.
Appl. Environ. Microbiol. 66, 1994–2000.
Wilkinson, J., Jenkins, A., Wyer, M., and Kay, D. (1995). Modeling fecal-coliform dynamics in
streams and rivers. Water Res. 29, 847–855.
Williams, A., Scholefield, D., Dowd, J., Holden, N., and Deeks, L. (2000). Investigating
preferential flow in a large intact soil block under pasture. Soil Use Manage. 16, 264–269.
Winfield, M. D., and Groisman, E. A. (2003). Role of nonhost environments in the lifestyles of
Salmonella and Escherichia coli. Appl. Environ. Microbiol. 69, 3687–3694.
Wollum, II, A. G., and Cassel, D. K. (1978). Transport of microorganisms in sand columns. Soil
Sci. Soc. Am. J. 42, 72–76.
Yao, K. M., Habibian, M. T., and O’Melia, C. R. (1971). Water and waste water filtration:
Concepts and applications. Environ. Sci. Technol. 5, 1105–1112.
Young, I. M., and Ritz, K. (1998). Can there be a contemporary ecological dimension to soil
biology without a habitat? Discussion. Soil Biol. Biochem. 30, 1229–1232.
Young, I. M., and Ritz, K. (2000). Tillage, habitat space and function of soil microbes. Soil
Tillage Res. 53, 201–213.
Zhao, T., Doyle, M. P., Shere, J., and Garber, L. (1995). Prevalence of enterohemorrhagic
Escherichia coli 0157-H7 in a survey of dairy herds. Appl. Environ. Microbiol. 61,
Zibilske, L. M., and Weaver, R. W. (1978). EVect of environmental factors on survival of
Salmonella typhimurium in soil. J. Environ. Qual. 7, 593–597.
L. Wu,1 M. B. McGechan,2 C. A. Watson1 and J. A. Baddeley1
Crop and Soil Research Group, SAC, Craibstone Estate,
Aberdeen AB21 9YA, United Kingdom
Land Economy Research Group, SAC, Bush Estate,
Penicuik EH26 0PH, United Kingdom
I. Introduction
A. The Future Agronomic Challenge
B. Why Model Roots?
C. Environment/Root Interactions
II. Current Work
A. Available Models
B. Selected Models
III. Model Processes
A. General Description
B. Branching
C. Growth
D. Root Architecture
E. Data Structure
IV. Extending the Scope of Current Models
A. Root Mortality
B. Interaction with the Environment
C. Water Uptake
D. Nutrient Uptake
E. Photoassimilate Availability and Root Development
F. Management
V. Structure of an Integrated Model
VI. Concluding Remarks
Improving our understanding of the relationships between soil conditions
and plant growth, both above and below ground, will contribute to the
development of cropping systems that are less reliant on mineral fertilizers
for crop nutrition. Although many models predicting the flows of nutrients
between plants and soil have been developed, few of these deal in detail with
Advances in Agronomy, Volume 85
Copyright 2005, Elsevier Inc. All rights reserved.
0065-2113/05 $35.00
root architecture and dynamics. In this chapter, we review seven widely cited
models of root architecture and development in terms of their ability to
improve predictions of plant and soil nutrient flows. We have examined
processes related to root system architecture and development, compared
mathematical expressions and parameters used in the selected models, and
summarized common processes and parameters for simulating root systems.
This outcome should benefit researchers and model developers, preventing
the need to spend limited resources on repeating the same process.
Detailed conclusions include the fact that both inter-branching distance
and insertion angle are essential parameters for representing root architecture. Additionally, in a three-dimensional model an extra parameter, radial
angle, should be used for determining the location of a branch relative to the
root from which it originated. Root growth is simulated by elongation rate
and elongation direction, with root component diameter also represented in
some models. Almost all the three-dimensional models reviewed calculate the
current direction of newly formed root segments using the previous direction
of tip extension together with an angle related to geotropism.
This review was carried out as the first stage in a research program on
integrating root growth models with soil nutrient cycling models. For this
purpose, the review suggests that, in order to optimize practical applications of these models in cropping systems, there is a need to integrate a
number of additional processes, including root longevity and mortality,
environmental responses, and eVects of management such as tillage or the
pesticide application regime. The form of root mortality relevant to nutrient cycling in soil is that due to natural senescence of root components.
This diVers from catastrophic death of roots due to attack by pathogenic
fungi, which has been considered in one existing root model. To achieve
the required objectives, there is also a need to strengthen the integration of
above-ground plant component dynamics with root system development,
particularly in relation to breeding new crop varieties for sustainable
ß 2005 Elsevier Inc.
agricultural systems.
Agricultural sustainability is concerned with production of agricultural
products over a long period of time. Sustainability ensures that agricultural
systems can be operated in such a way that output quantity and quality can
be maintained year by year, without degradation of the environment. Global
population is anticipated to increase in the decades ahead, although with a
declining rate of growth (UN, 2001). A major agricultural challenge will be
to achieve a significant increase in agricultural productivity on currently
available land, in order to meet the food requirements for this population,
but at the same time conserving natural resources. Recent productivity rises
have occurred in part due to increase in use of chemicals, new technologies,
and mechanization. For example, statistical data for 1961–2001 from the
United Nations Food and Agriculture Organization (http://apps.fao.org)
show a significant positive relationship between global cereal crop production and nitrogen fertilizer consumption. Although these changes have
achieved many positive results in biological, economic, and social terms,
there have also been significant costs (Bruinsma, 2003; Norse, 2003; Schaller,
1993). Prominent among these are degradation of soil structure (Jordahl and
Karlen, 1993; Pimentel et al., 1995), contamination of water with physical,
chemical, and biological pollutants (Carpenter et al., 1998; Foster et al.,
1986; Goulding, 2000; Peterson et al., 2001; Smith et al., 1999; Vitousek
et al., 1997), and gaseous emissions to the atmosphere (Bobbink et al., 1998;
Galloway, 1995; Jenkinson, 2001). It is an ongoing challenge to develop and
demonstrate management practices that increase the sustainability of agricultural systems. In Europe, following the Mid-Term Review of the Common Agricultural Policy, it is likely that future agriculture will bifurcate into
high-throughput systems producing major food commodities, and more
extensive systems primarily focused on the provision of environmental
goods. Management practices and the use of agrochemicals will diVer widely
under these two scenarios. The second option includes organic production
(currently 4% of agricultural land in the United Kingdom [Soil Association,
Accurate assessment of the long-term consequences of environmental
perturbation is very important for the sustainable use of agricultural
resources. Long-term agroecosystem experiments at various locations
worldwide have attempted to evaluate biological, biogeochemical, and environmental changes associated with agricultural systems over time periods
of 100 years or more (Steiner, 1995), e.g., the Broadbalk Experiment at
the Rothamsted Experimental Station, UK (van Bergen et al., 1997), the
Sanborn Field experiment at the University of Missouri, USA (Upchurch
et al., 1985), and the Old Rotation Study at Auburn University, USA
(Mitchell and Entry, 1998). Although providing valuable data, these sites
cannot represent all environmental, economic, or biological conditions occurring throughout the world. The augmentation of such site-specific, empirical information by process-based knowledge oVers the potential for
predicting the sustainability of agricultural systems in a wider range of
environments. This includes the diagnosis of problems, prescription of alternative ways of improving agroecosystem performance, and quantification of
sustainability in terms of productivity, stability over time (the constancy
of production over time), resiliency to disturbance, and equitability of
benefits derived from an agroecosystem.
An eVective sustainable agriculture system strives to develop a farming
strategy that optimizes management practices such that crop productivity is
maintained while adverse environmental impacts are reduced. Field evaluations of nutrient management practices are cumbersome, error prone,
complicated, and expensive. It is often more eYcient to study agroecosystem performance using computer models than to experiment with the
system itself. Models can simulate and evaluate a range of nutrient management scenarios and aid in evaluation of alternative chemical (in terms of
application rate, source, and timing) and field management practices, representing an eYcient and cost-eVective alternative to field evaluations. With the
development of information technology and computer hardware, increasing
numbers of simulation and other mathematical models have been developed
(Grant, 1997; Johnsson et al., 1987; Parton et al., 1987; Williams and Renard,
1985) and applied to the analysis of productivity and sustainability of complex
agricultural systems (Jones et al., 1991).
Root systems are central to the acquisition of water and nutrients by
plants (Fitter et al., 1991) but are also a major pathway for the input of
carbon and nutrients to soil (Persson, 1978; Ruess et al., 1996; Vogt, 1991).
Roots have often been described as the ‘‘invisible’’ or ‘‘hidden’’ parts of a
plant (Waisel et al., 1996; Weaver, 1926). Despite their obvious importance,
much less is known about the dynamics of live roots than about aboveground plant organs. This is because roots grow in soil from which they
cannot be extricated or readily observed without destroying both individual
roots and the overall architecture of the root system.
Plant root systems are complex structures that exist in a spatial and
temporal mosaic of resource availability. Attempts have been made to
describe the spatial deployment of root systems through the characterization
of root architectural parameters by various schemes (Fitter, 1982, 1986;
Hackett, 1968). Although valuable, these are spatial descriptions only of
roots spread out in two dimensions and at one point in time. In reality, three
dynamic root system processes, namely production, extension (growth), and
mortality, contribute to the appearance, transformation, and disappearance
of root systems in both space and time.
Due to the inaccessibility of root systems, special techniques are required
to investigate the standing stock, distribution, and turnover. Traditional
descriptions of root architecture and characteristics have been made by
destructive techniques, such as monolith washing, soil coring, or trench
wall methods (Bledsoe et al., 1999; Böhm, 1979; Milchunas et al., 1992).
These methods obtain a snapshot of root systems at a specific time. Root
systems, however, are plastic and interact dynamically with physical, chemical, and biological factors in the soil that vary in time and space. If a
destructive sampling method is used, it is diYcult to track root system developments such as root growth rate and direction, mortality, and branching.
Furthermore, it is unable to distinguish a static, unchanging root system from
one that is highly dynamic with production balanced by disappearance. Nondestructive, observational techniques using rhizotrons or minirhizotrons
provide a nondestructive, in situ method of monitoring the production,
growth, and mortality of individual roots over time (Bland and Dugas, 1988;
Tierney and Fahey, 2001; van Noordwijk et al., 1985; Watson et al., 2000).
Understanding root demography is central to the study of plant growth
and development, carbon and nutrient cycling, and water movement within
the plant/soil system. The complexity of both biotic and abiotic interactions,
combined with stochastic changes in root architecture, makes it diYcult to
understand below-ground dynamics on the basis of experimentation alone.
Goss and Watson (2003) highlighted the need to refine cropping system
models to take account of root dynamics. Models can be used to predict
root system architecture in various plant species and to investigate the
appropriateness of hypotheses employed in a model by comparing simulated
and observed root system morphologies. They can also be used to simulate
synchronized development of a root system, and (in conjunction with uptake
models) to simulate soil water and/or nutrient uptake behavior. Attempts
have been made to simulate the branched, hierarchical nature of plant root
systems and plant–environment interactions mathematically (Clausnitzer
and Hopmans, 1994; Diggle, 1988a; Dunbabin et al., 2002; Fitter et al.,
1991; Hackett and Rose, 1972; Lungley, 1973; Lynch et al., 1997; OzierLafontaine et al., 1999; Pagès et al., 1989). Following an earlier review of the
physiological processes of root development by Coutts (1987), Pagès (1999)
reviewed some of the underlying principles of processes that must be
incorporated into models of root development and architecture. Pagès
(1999) concluded that the common basis for the models is that they are
phenomenological models that translate and combine morphogenetic rules.
Some of the models also consider the eVects of environmental factors on
assimilation rate. Experimentation can help refine models, and then
simulated results can be used to test and improve the hypotheses upon
which the models depend, and in turn add to our understanding of plant
processes and functions as well as suggesting knowledge gaps that require
further experimentation.
Field and laboratory experiments have demonstrated the complex interactions between the production, growth, and mortality of individual roots
and local environmental factors surrounding the plant. Abiotic factors such
as soil temperature (Forbes et al., 1997; Gavito et al., 2001; Vincent and
Gregory, 1989; Watson et al., 2000), soil water content (Brissette and
Chambers, 1992; Chiatante et al., 1999; Derner et al., 2001; Huck et al.,
1983; Joslin et al., 2000), and mechanical resistance from the soil (Bengough
and Mullins, 1990; Bingham and Bengough, 2003; Laboski et al., 1998;
Misra and Gibbons, 1996) have an important role in root development.
Huang et al. (1991) concluded from their experiment with wheat seedlings
(Triticum aestivum L.) that the length and dry matter production of both
seminal and crown roots increased gradually to a maximum as the temperature increased to 25 8C, but then declined as it rose to 30 8C. Biotic factors,
including arbuscular and ecto-mycorrhizal colonization and infection by
fungi, some of which are pathogenic (Forbes et al., 1996; Hooker et al.,
1992; Niemi et al., 2002), also aVect root architecture, functions, and mortality. Carbon (C) and nitrogen (N) status in plant organs and soil (Bingham
et al., 1997; Boukcim et al., 2001) and previous root system architecture also
influence root development (Pulgarin et al., 1988).
The proportion of CO2 in the atmosphere is currently rising to unprecedented levels, and this will have direct eVects on plants, including their
root systems. Elevated atmospheric CO2 may increase root biomass and root
length density (Derner et al., 2001; Fitter et al., 1997; Newton et al., 1994),
elongation rate (Drennan and Nobel, 1996), and the level of root exudation
(Norby et al., 1987), as well as stimulating the branching process (Berntson
and Woodward, 1992). King et al. (1997) investigated morphology and
tissue quality of seedling root systems of Pinus taeda and Pinus ponderoa
aVected by varying levels of CO2, temperature, and nitrogen supply. They
found a large increase in root length under elevated CO2, with increasing
temperature and nitrogen supply giving further increases in root length.
A root system, considered as a collection of sources and sinks, is
simulated as a submodel in many larger models that describe either crop–
environment relationships or matter transfer (water, carbon, or nutrients) in
the soil–plant–atmosphere continuum (SPAC). In such cases, root biomass,
as a proportion of plant biomass, root density distribution (in space), and
root length density distribution, are all used to control water and/or mineral
uptake. A secondary assumption also needs to be made: that the spatial
distribution of the roots is homogeneous in the soil layer and the uptake
is similar among all roots. Root distribution is assessed in terms of the
penetration and proliferation of roots down to the penetrated depth, with
consideration of the eVects of one or more environmental factors on root
growth (Asseng et al., 1997). There are some models that describe temporal
development of root distribution as a diVusion process (Acock and
Pachepsky, 1996; de Willigen et al., 2002; Gerwtiz and Page, 1974). For
example, Acock and Pachepsky (1996) developed a two-dimensional convective-diVusive root system model in which the proliferation and growth of
roots in all directions are considered to result from a diVusion-like gradient,
whereas the convection-like propagation of roots downward is perceived to
be caused by geotropism. Mmolawa and Or (2000) reviewed some expressions for these parameters as applied to root-zone solute dynamics under
drip irrigation. Because root growth diVers in terms of direction, spacing,
elongation rate, and functional activity, such assumptions represent an
oversimplification (Rengel, 1993). These models are not discussed here, as
they ignore root architecture.
With the development of computer hardware and software, various root
system architecture models have been developed over the last three decades.
Pioneering work in the simulation of root systems was carried out by Lungley
(1973). Some models purely simulate root static structure (Henderson et al.,
1983), root system growth and development in two dimensions (Lungley,
1973; Porter et al., 1986; Rose, 1983), or root architecture in three dimensions
(Bernston, 1994; Diggle, 1988a; Fitter et al., 1991; Pagès et al., 1989). Other
models involving the root system relate to water uptake (Clausnitzer and
Hopmans, 1994; Doussan et al., 1998; Tsutsumi et al., 2002), nutrient uptake
(Grant and Robertson, 1997), and uptake-dependent growth (King et al.,
2003; Somma et al., 1998).
There are diVerent approaches to the description of root systems in the
models. The most common is topology of the branching process (Acock and
Pachepsky, 1996; Clausnitzer and Hopmans, 1994; Diggle, 1988a; Fitter
et al., 1991; Hackett and Rose, 1972; Lungley, 1973; Lynch et al., 1997;
Pagès et al., 1989). Roots are classified according to branching order, and
each order has its own characteristics in terms of growth rate, life span,
and branching ability. Fractal geometry has also been used in connection
with root architecture simulation (Ozier-Lafontaine et al., 1999; Shibusawa,
1994). In this method, the network of a root system is described as being selfsimilar or following scale-invariant branching rules. This is achieved by
deducing properties of the entire root system from basic rules governing
individual bifurcations and the geometry of each segment or branch.
A stochastic (as opposed to deterministic) approach has also been practiced
for the description of root system architecture and development (Jourdan
and Rey, 1997). Stochastic processes (e.g., automata, probability, and
graphic models) have been used to simulate the topology of branched
structures and root development (growth, mortality, and branching).
Because biological hypotheses are not quantified, such a model is purely
The current chapter builds on an earlier review (Pagès, 1999) to present an
in-depth, systematic review of individual models and their equations. At this
level of detail, it becomes clear that published models often use diVerent
terminology to describe processes of root growth and development, and a
number of diVerent mathematical expressions have been used to describe the
same process. It has been found to be helpful, when considering further
development of such models, to formalize the process descriptions and to
distinguish those that are essential to the main processes of the models. In this
chapter, seven existing models are discussed and compared. These models
have been developed in various diVerent areas of the world, cover a range of
diVerent plant species, and have been frequently cited in the literature. Although some of the reviewed models have submodels to simulate water flow,
nutrient transport, and carbohydrate allocation to various plant components,
we limit the analysis here to processes directly associated with the root system.
The seven models reviewed here are listed under abridged headings (either
the model name or the organization at which the model was developed); they
will be referred to henceforth using these names.
The WAITE model is a pioneering two-dimensional numerical dynamic
model that was developed at the Waite Agriculture Research Institute
(WAITE), Australia (Lungley, 1973). Simulation is based on individual
roots, and all parameters are kept constant for the whole simulation period.
Branching is restricted to the laterals of first and second order. Computations are performed in discrete time steps (1 day), and each root tip grows
individually for the entire duration of the simulation. It is interesting to note
that most of the recently published root simulation models largely follow
this approach.
ROOTMAP is a static morphogenetic three-dimensional model of the
growth and structure of fibrous root systems (Diggle, 1988a). Lengths and
locations of each root segment and location and age of each root tip and
each branch in the system are explicitly calculated. The growth of all
emerged root tips is simulated concurrently. In this model, soil conditions
and root growth, together with branching responses to those conditions,
remain unchanged during the course of a simulation. All growth rates and
development times are temperature dependent. Growing roots are tracked
by keeping a separate record for each root tip and each branch in
the root system. The model has recently been developed to allow simulated
root systems to respond to the supply of water and nutrients in the soil
environment (Dunbabin et al., 2002).
INRA Model
The INRA model is a three-dimensional architecture model of the maize
root system that was developed in France by INRA (Institut National de la
Recherche Agronomique) (Pagès et al., 1989). It simulates root architecture
in discrete time steps in terms of three basic processes: emergence of new root
axes from the shoot, extension, and branching. It takes into account the
kinetics of emergence of a specific primary root, with geometric representation of its location. The emergence and location of branches are estimated
only by spatial parameters. The model has been further extended to assess
the influence of assimilate availability on root growth and architecture
(Thaler and Pagès, 1998) and to simulate water uptake by root systems
(Doussan et al., 1998).
York Model
The York model is a topological three-dimensional root growth model
that was developed at the University of York (Fitter et al., 1991). It takes a
‘‘link’’ as the basic unit of root system classification, simulating the development of root systems using topology, branching angles, diameters, and link
lengths. Branching probabilities have been applied to determine branching
nodes. The model can estimate root system magnitude (the number of
external links), ‘‘altitude’’ (the number of links in the longest single path),
and the exploitation eYciency of each root system.
Davis Model
The Davis model is a three-dimensional simultaneous dynamic simulation
of root growth, soil water flow, solute transport, and uptake from the
University of California, Davis (Clausnitzer and Hopmans, 1994). The
model is linked to three-dimensional transient soil water flow and solute
transport submodels based on the finite element method. Root age has an
eVect on root water and solute uptake, and influences of nutrient deficiency
or excessive nutrient concentration on root growth are included. The model
also considers feedback functions for water uptake, as well as shoot production of assimilate required for root growth. When simulating root growth,
one of three levels of complexity concerning treatment of transpiration and
root water uptake must be selected. Each has diVerent requirements in terms
of both input parameters and specification of the form of interaction between
root and shoot growth and environmental factors. Somma et al. (1998)
expanded this model further by considering solute transport, nutrient uptake,
and the interaction between plant growth and nutrient concentration.
6. SimRoot
SimRoot, the simulation and visualization of root systems (Davis, 1993;
Lynch et al., 1997), focuses on the data structure of root segments produced
by a specified growth model. In addition to its function as a simulation
model, it can also be considered a platform for visualization of a root system
in three dimensions. The parameters for root growth are stored in a data
structure consisting of several components, while the output from the model
is stored in an ‘‘extensible tree’’ data structure. When operating in an
optional ‘‘solid rendering’’ visualization mode to display the root system,
the diameter of each root is determined on the basis of its position along the
root axis, order of the root, nutrient concentrations, etc. The incorporation
of kinematic functions in the model can explicitly treat spatial heterogeneity
of physiological processes in the root system. A simple carbon cost function
is also incorporated, based on measurements of respiration, C exudation,
and biomass deposition along root axes for Phaseolus vulgaris seedlings
under laboratory conditions (Nielsen et al., 1994). The model has since
been modified to include factors involved in competition among multiple
root systems (Rubio et al., 2001).
Frac-Root Model
The Frac-Root model is a static three-dimensional model of a root system
based on fractal theory, which is also from INRA, France (Ozier-Lafontaine
et al., 1999). The model was based on self-similarity and ‘‘pipe model’’
assumptions, together with observations of topology, branching rules, link
length, and diameter, as well as root orientation. The root length, diameter,
and angle between proximal roots are input parameters. Additionally, the
proportionality factor between total cross-sectional area of a root before and
after branching and an allocation parameter for partitioning biomass between the new segments after a branching need to be predefined. Because the
model fully considers the fractal character of the root system, the model is
time independent, and hence no root elongation rate is considered.
All the models describe the dynamics of a root system by considering root
architecture and developmental processes. Developmental schemes categorize roots into several types, and the terms used to describe roots in this
chapter are defined below. A root is called an axis when it has developed
from the seed (seminal axis) or the stem (nodal axis). Roots arising from the
axis are designated first-order laterals; those branching from first-order
laterals are designated second-order laterals, etc. Although the definition
of orders and the number of categories vary from model to model (Table I),
the topological structure is common to all models. Axes and the first two
orders are taken into account in most of the models reviewed here.
Table I
Root System Terminology Used in the Reviewed Modelsa
WAITE model
INRA model
York model
Seminal axis
and nodal axis
(or primary) axis
Davis model
Primary root
Proximal root
In order to facilitate discussion in the review, the varying terms used by diVerent model authors
have been replaced by a common set of terminology, as shown in the column headings.
Figure 1 General parameters used to describe root architecture. Each root is made up of an
axis and laterals (several orders). Apical non-branch distance, basal non-branch distance, and
inter-branch distance are used to specify emergence of higher order laterals.
Laterals emerge from each member of the next lower order in acropetal
sequence. The youngest lateral is generally separated from the apex by an apical
non-branch distance (called an external link in the York model); the oldest lateral
is normally located from the base point by a basal non-branch distance (Fig. 1).
The orientation of a newly emerged lateral is controlled by two variables: the
insertion angle and the radial angle. The insertion angle is the angle between the
mother root and the branch in the plane containing the two roots. The radial
angle is the angle between the branch direction and a specified reference direction
(analogous to north on a map), in the plane perpendicular to the mother root.
The branching process can be split into two parts for ease of simulation:
branching position on a root and branching orientation. The former focuses
mainly on the position of the new branch and the number of nodes that can
Table II
Parameters Used in the Models to Describe Branching Position and Branching Orientation
Model no.
Length of
apical nonbranching
Length of
the basal
number of
xylem poles
be created on an axis or a lateral root. The latter calculates the emergence
direction of a new branch. The parameters used in the models and their
values (if applicable) relating to branching position and orientation are
summarized in Table II.
Branching Position
All models treat the expansion of the branched zone in a strictly acropetal
way. The parameter inter-branching distance is unanimously considered
in all the models, either as an input parameter or as a parameter to be
estimated during simulation. In the Frac-Root model, link length is calculated from experimentally derived relationships between the distance, the
diameter, and the order of the segment. Four of the models take the length
of the apical non-branching zone as a parameter. Only the INRA model sets
the basal non-branch zone of a root as a parameter. The WAITE model,
ROOTMAP, and the Davis model simply treat it as an inter-branch distance (Tables II and III). If all three parameters related to the branch
position are included in a model, the number of branches on a root (Nb) is
expressed as
Ll < ðLanz þ Lbnz Þ
Nb ¼
Ll Lanz Lbnz
Ll ðLanz þ Lbnz Þ;
or alternatively, if only the distance of apical non-branch zone and interbranch distance are considered (so Lbnz ¼ Lib), then
Table III
Parameter Values in the Models for Branching Position and Branching Orientationa
Small grain cereal
Length of apical non-branching zone (cm)
First-order lateral
Second-order lateral
Third-order lateral
Length of the basal non-branched zone (cm)
First-order lateral
Second-order lateral
Interbranch distance (cm)
First-order lateral
Second-order lateral
Third-order lateral
100.0 (hr)
150.0 (hr)
Caropca bean
Leguminous tree
Gliricidia sepium
0.5 (0.6)
Model no.
Insertion angle (degree)
First-order lateral
Second-order lateral
Total number of xylem poles
First-order lateral
Second-order lateral
Values for the York model are not available.
Parameter values selected for dynamic assimilate allocation to shoot and root with transpiration rate dependent on current leaf area.
Based on Ge et al. (2000). Values not in parentheses in the axis row are for the taproot, and those in parentheses are for basal roots that arise from the
Values are estimated based on unimpeded growth rate at 158C and apical non-branching at 15 8C.
Values are calculated based on total number of xylem poles with Eq. (5).
Radial angle (degree)
First-order lateral
Second-order lateral
Nb ¼
Ll < Lanz
Ll Lanz ;
where (in both cases) Ll is the length of a lateral root, Lanz is the length of the
apical non-branching zones of a lateral root, Lbnz is the length of the basal
non-branching zones of a lateral root, and Lib is the inter-branch distance
along the lateral root. The symbol b c is a mathematical function representing the largest integer that is less than or equal to the quotient.
Potential nodes where branches might arise have been introduced into the
York model. This allows variations in inter-branch distance and the length
of the apical non-branch zone during simulation. The actual mean interbranch distances are always greater than the minimum value as set by the
user. In contrast to other models, inter-branch distance and apical nonbranch zones are outputs from the model rather than input parameters.
The probability of generating branches from a given potential node (derived
from Fitter, 1987) is calculated as
pðvÞ ¼ v max
bc v
bc ðiþ1Þ
where p(v) is the probability of branch generation, v is the root order of the
potential node (v ¼ 0 for the axis), and bc is a branching coeYcient. When bc
approaches zero, branching becomes equiprobable at all nodes.
The Frac-Root model sets the positions of new branches by the relationship between link length and root diameter. The mean inter-branch distance
for a given root order (Lib , cm) is calculated as
Lib ¼ 6:5136 ln ðd̄Þ þ 21:827;
where d̄ is the mean root diameter (cm) for a given root order.
Branching Orientation
The number of parameters required to determine branch orientation
varies in the models, depending on the number of dimensions considered
in a particular case. The WAITE model assumes that branch distribution
has radial symmetry in a vertical plane, a specified slope being assigned to
each segment produced in a discrete time step. For the other six models,
insertion angle (vertical angle or branch angle) has been used. The parameter
radial angle (relative to the horizontal, azimuth), is included in the INRA,
York, SimRoot, and Frac-Root models. In the Frac-Root model, the angles
are chosen randomly, depending on the diameter of the previous link, and a
negative radial angle is used to represent the angle for branch orientation
(Table III).
In some models, an additional parameter, the total number of xylem
poles, is used to derive the radial angle. The INRA, York, SimRoot, and
Frac-Root models share the same algorithm and use both the radial angle
and the total number of xylem poles to determine the branch orientation.
The radial angle g in the INRA, York, and SimRoot models is expressed as
where X is the total number of xylem poles and N is a randomly chosen
integer between 1 and X.
Use of a transformation matrix for the angles indicated the direction of
root branching in the SimRoot model.
The Davis model and ROOTMAP use only the insertion angle, modified
from its initial value by a random variation within a pre-defined range. Also
in ROOTMAP, the initial value of the insertion angle is always 908.
To describe the growth of a root system dynamically, the elongation rates
of various root types need to be known. The general quantitative expression
of growth rate, Va, has the following pattern:
Va ¼ Vp fðT; W ; S; UÞ;
where Vp is potential (maximum) growth rate for various root categories,
f(T, W, S, U ) is a response function for soil temperature (T ), soil moisture
(W), mechanical resistance (S), and nutrient status of the root (U ).
The function could be derived from the combination of individual eVects
in various ways. Among the models reviewed, only the Davis model
links elongation rates to a range of environmental factors, but ROOTMAP
includes responses to soil temperature. Parameters, together with
their values, used to describe elongation rates in the models are listed in
Table IV.
Elongation rates were set for diVerent root types as inputs in the models.
In the York model, the growth rate of the primary axis is specified as an
input (V0), together with fractions of that value for all higher order roots, in
order to be comparable with some published experimental results (May et al.,
1965; Schuurman and de Boer, 1970):
Table IV
Parameter Values for Root Elongation Rate in the Modelsa
Model no.
Primary growth (cm/d)
First-order lateral
Second-order lateral
Third-order lateral
Small grain
1.18 (2.04)
Maximum branch length (cm)
First-order lateral
Second-order lateral
Soil resistance parameter
Reference temperature (8C)
Temperature of zero growth (8C)
Values for the York model are not available.
Based on Ge et al. (2000). Value not in parentheses in the axis row is for the taproot, and that in
parentheses is for a basal root that arises from the taproot.
Parameter values selected for dynamic assimilate allocation to shoot and root with transpiration rate dependent on current leaf area.
< 2
Va ðvÞ ¼
> Va ð1Þ
ðv 1Þ2
ðv ¼ 1Þ
ðv 2Þ:
For the INRA model, the growth rate of primary roots is set as a constant
value of 2.0 cm d 1, and the root growth rates for the higher-order roots are
expressed as
Va ¼ Vp e
where t is the age of the root meristem in days and Vp is the potential
elongation rate, which was set at 6.4 and 1.5 for first-order and secondorder laterals, respectively. The constant k was set at 0.8 for both orders.
The ROOTMAP model assumes a temperature function previously used
by Porter et al. (1986), in terms of a time parameter similar to commonly
used soil degree-days (SDD). Elongation rates are adjusted by this time
parameter. The time parameter ( ft, d 1) is defined by the following linear
relationship, incorporating local actual soil temperature (T, 8C) when it falls
within the range between the temperature of zero growth (Tmin, 8C) and a
reference temperature (Tref, 8C):
fl ¼
T min
In the Davis model, the actual root elongation rate (Va, cm h 1) is
obtained by multiplying the unimpeded elongation rate (Vp, for a given
root age and root order, cm h 1) by three response functions: a soil strength
factor (imps), a soil temperature factor (impt), and a soil water solution
concentration factor (impc):
Va ¼ Vp ipms ipmt ipmc
ipms ¼
< 0
: 1:0
s smax
s < smax ;
where smax (MPa) is the soil strength at which growth ceases completely and
s is the current soil strength:
impt ¼
T Tmin >
Tmax Tmin
T Tmax >
Tmax Tmin
T > Tmax or T < Tmin
Topt ðTmin þ Tmax Þ
and Tmin T Tmax
Topt > ðTmin þ Tmax Þ
and Tmin T Tmax ;
< ln T opt TTmin
Topt ðTmin þ Tmax Þ
: ln max Topt
Tmax Tmin
Topt > ðTmin þ Tmax Þ
and Topt (8C) is the optimum temperature at which the temperature function
is unity, Tmin (8C) is the lower threshold temperature below which the
response is zero, and Tmax (8C) is the upper threshold temperature above
which the response is zero (Fig. 2A):
impc ¼
< c
c < cmin ; c > cmax
cmax c
c max coptu
cmin c < coptl
coptl c < coptu
coptu c cmax ;
where c (mmol) is the current soil water solution concentration, cmin and cmax
are minimum and maximum concentrations for plant growth, and coptl and
coptu are lower and upper limits of the optimum concentration range for
plant growth (Fig. 2B).
Figure 2 Response functions to temperature (A) and nutrient concentration (B) in
the Davis model. The temperature response curves are based on the following parameter values:
Tmin ¼ 0:08C; Tmax ¼ 40:08C; Topt ¼ 15:0 ½Topt < 1=2ðTmax þ Tmin ފ; 20:0 ½Topt ¼ 1=2 ðTmax þ
Tmin ފ; 28:08C ½Topt > 1=2ðTmax þ Tmin )]. The nutrient concentration curve is plotted with
cmin ¼ 0:005; cmax ¼ 2:0; coptl ¼ 0:1, and coptu ¼ 1:5 (mmol cm 3).
A kinematic approach was adopted from Silk et al. (1986) to calculate the
local relative rate of root volume change (dV) along a root axis:
dV ¼
@vz 2vz @r
r @z
where r is root radius, z is distance from the root tip, and vz is the longitudinal growth velocity. The first term on the right-hand side represents the local
relative elongation rate, and the second term represents the radial and
tangential components of the relative growth rate.
Elongation Direction
In the WAITE model, the root distribution assumes radial symmetry,
with roots constrained to grow in short, straight segments, each segment
having a specified slope. The slope of each root varies over the time course.
In all three-dimensional models, the elongation direction of a growing
root tip is based on at least two parameters: the previous elongation direction of the tip over the last time step, and an angle related to geotropism. The
INRA model uses an additional, third parameter to represent mechanical
constraints, either as a random value or as a user specified value (Table V).
In the ROOTMAP model, the growth direction of a root tip is determined
stochastically from a deflection angle (f, 8) from the previous direction, and
a deflection orientation (u, 8), defined as the angle between the elongation
direction and the vertical. These two angles are calculated on the basis of
user-defined values of the probability of occurrence (p) and two exponents
f ¼ 360 pI1
u ¼ 180 p1 Ig ;
where Id is the deflection index and Ig is the geopropism index, both
with a range of 0–1. Both these indices have to be specified for each branch
The York model allows the branch angle to decline progressively to a
user-defined final value to represent dynamic changes in growth direction.
Also, the value can be randomized by specifying a percentage range
around the existing value of branch angle by which it can vary, for any
growing tip.
In the INRA model, growth direction (D i ) is computed using three
directional components: the initial direction of the root at the previous
Table V
Parameters and Their Values to Determine the Direction of Root Growth in Some Modelsa
Model no.
Mechanical constraint in soil profile
25 cm depth
Below 25 cm
Soil strength gradient (cm/MPa)
First-order lateral
Second-order lateral
Deflection index(D)
First-order lateral
Second-order lateral
Geotropism index (G)
First-order lateral
Second-order lateral
max. 458
1.00 (0.80)
Values for the York model are not available. The elongation direction in the Frac-Root model
is chosen randomly from the known actual angle distribution that depends on the diameter of
the previous link.
Based on Ge et al. (2000). The value not in parentheses in the axis row is for the taproot and
that in parentheses is for a basal root that arises from the taproot.
Parameter values selected for dynamic assimilate allocation to shoot and root with transpiration rate dependent on current leaf area.
Values for seminal root and roots arising from it.
time step (D i 1 ), a vertical vector representing geotropism ( G ), and a vector
representing mechanical constraints ( S ). The length of the first vector is set
at 1, and the last two are expressed as the product of the elongation during
the current time step and weighting factors.
Di ¼ Di
! !
þS þG
The Davis and Frac-Root models have the same procedure as the INRA
model for calculating growth direction. However, in the Davis model, a
limited random deviation D i 1 is applied to give an approximate representation of the space-exploring nature of root system growth (Somma et al.,
1997). The Frac-Root model calculates the elongation direction of a new
growing tip by choosing an angle at random from their known actual
distribution, dependent on the diameter of the previous link, an insertion
angle, and a positive radial angle.
In the SimRoot model, the direction of a new segment is determined in
a separate submodel based on the direction of the previous segment. It
can be modified in order to take account of geotropism and a randomness
Dynamics of Root Diameter
In order to determine the influence of root architecture on root functions
and the volume of soil available for exploitation of resources, root diameter
(or radius) was estimated in some three-dimensional models.
In the York model, the radius of a growing tip is predefined by the user;
the radius (r, cm) of any segment other than a growing tip is calculated in
terms of the link magnitude, m (the number of growing tips derived from the
segment being considered) (Fitter, 1987).
r ¼ 0:2 þ 0:01m:
In the Frac-Root model, at a given node that is generating new links, the
diameter of a root of the same order (dl, mm) is calculated as
dl2 ¼
b 2
d ;
a bb
where b and a are allocation and proportionality factors and dbb (mm) is the
diameter of the previous segment (link).
The diameter of higher order roots is estimated as
dh2 ¼
1 b
d2 ;
ðn 1Þa bb
where n is the total number of new segments generated at a given node, and
the distribution of this number at any branching event is assumed to follow a
uniform distribution.
In the SimRoot model, simulation of root radius is built into the root
growth model. Its radius (r) is correlated to the length of the root (L) from
the root tip (Ge et al., 2000).
r ¼ a L 2;
where a is the root radius growth coeYcient.
Data structure is crucial when manipulating large databases in which
simulation results are stored to reproduce root architecture topologically
and visually. Most of the models in our selection use an ASCII format file to
store root system simulation output. Each record represents a root segment,
including segment location, segment length, connections, branch order, etc.
The SimRoot and ROOTMAP models use special data structures to save
simulated results.
SimRoot has an optimized data structure (extensible tree structure) to
store simulated results. For each node in the data structure, four pointers are
topologically defined to point to its ‘‘parent,’’ its ‘‘first-generation children,’’
and its ‘‘right and left siblings.’’ Meanwhile, information on the direction of
growth and the length and spatial origin of the segment grown during
consecutive time steps is kept for each node. This allows visualization of
root architecture and operation of the database to be carried out more
rapidly than with other approaches.
The ROOTMAP model uses a binary tree data structure to store information about root systems. In this structure, each root tip record and each
branch record are treated as nodes, and for each branch record there are
four pointers (Diggle, 1988b), as follows:
1. One pointing to the root tip of the same order as itself,
2. One pointing to the root tip that originated at the branch record, if it
3. One pointing to the next, younger branch record in the direction of the
root tip of the same order, if that branch record exists, and
4. One pointing to the next branch record in the direction of the root tip of
the succeeding order, if it exists.
The models reviewed here describe root architecture as it arises from root
growth processes such as extension and branching. Root diameter is also
described in some of the models. Although the WAITE model is twodimensional with constant parameter values, its diagrammatic description
of root architecture is adopted in almost all models developed subsequently.
Inter-branching distance and insertion angle are two essential parameters for
reproducing root architecture, whether a model is two- or three-dimensional.
In a three-dimensional model, an extra parameter, radial angle, must be
known in order to define the location of a branch relative to the root from
which it originates. Almost all the reviewed three-dimensional models
adopt the same algorithm as the INRA model to calculate the current
direction of a newly formed root segment, in relation to the previous elongation direction of the tips and an angle related to geotropism. The models
include some variation in treatment of the randomness of this direction.
Most reviewed models mimic the morphogenetic program to simulate the
topological characteristics of the root systems.
This review of the development of existing models shows the need for
extending the scope of root modelling to consider a number of additional
factors. This will facilitate the use of these models to describe real situations.
Root mortality processes lead to the disappearance of roots from a live
root system following an aging phase. The process of root mortality has a
very large and important influence on nutrient cycling, especially in perennial plants. It causes dynamic changes in both total root volume and root
system architecture. Simultaneously, dead root material is added into an
organic matter pool for subsequent decomposition. Unfortunately, the mortality process is ignored or simplified in current versions of nutrient cycling
models, perhaps because of the complexity of the senescence process. The
reviewed models were generally executed for only quite short periods compared to the life cycle of plants, which may have avoided the need to consider
senescence. In this chapter we are concerned with the factors influencing the
turnover of the fine root pool, rather than cases of catastrophic root death
induced by, for example, pathogenic fungi (Brown and Kulasiri, 1994) or
harvesting of cereal crops.
Individual fine roots tend to have relatively short lifespans and may
account for as much as 33% of global annual net primary productivity
(Jackson et al., 1997). Interactions between root mortality and endogenous
and exogenous environments are poorly understood. Pregitzer et al. (2000)
reviewed responses to temperature of fine roots of trees and concluded that
soil warming had the greatest eVect on root production and mortality. An
experiment reported by Majdi (2001) indicates that liquid fertilization in a
Norway spruce (Picea abies (L.) Karst.) stimulates the process of root
mortality significantly. In the Davis model, root maturity is controlled by
the parameter maximum root length for a given order. In each time step, it
scans the whole root system; if a branch’s length is equal or greater than
the correspondent maximum length, the branch will be removed from the
system. One model that does incorporate the senescence process is that of
Jourdan and Rey (1997), who used survival probability ( ps) to represent the
process based on field observations for the oil palm (Elaeis guineenis Jacq.)
root system. The survival probability ps(i) of an elementary length unit
(segment) in a given length category i, described by the percentage of the
maximum length of a given root type for each morphological root type that
is distinguished by development pattern and state of diVerentiation, is
expressed as
11=Ni Ni 1
ps ðiÞ ¼ 1
pm ðiÞ ¼ B1
where pm(i) is the probability of dying in length category i, fi is the proportion of dead roots in length category i in relation to the initial population,
and Ni is number of elementary length units from the base of the root in
length category i.
The main purpose of the models reviewed here was to reproduce root
architecture; few of them consider interaction between root growth, development, and the environment. This makes it diYcult to use them to accurately
predict root growth under field conditions. Even under controlled conditions,
their use is limited because of simplifying assumptions used in process descriptions and because of the interactions between these processes and the environment. When considering the eVects of environmental factors, the interaction
between the roots and the rest of the plant also needs to be considered, since
such factors influence both above- and below-ground plant components.
Although the Davis model linked elongation rates with environmental
factors (soil temperature, moisture, soil strength, and soil solute concentration), there are other factors that may dramatically aVect root growth and
development that have not been considered. For example, Aguirrezabal and
Tardieu (1996) pointed out that in field-grown sunflower, the elongation of a
root branch was related to photosynthetic photon flux density (PPFD) and
to the distance from the apex of the branch under study to the base of the
taproot. Aguirrezabal et al. (1993) concluded from reviewed papers that
carbon nutrition aVects not only the total root biomass and length, but
also the number of roots, the individual elongation rate of diVerent
branches, and the elongation rate of branches appearing on apical vs basal
parts of the taproot. Mycorrhizal colonization is another important
biological factor to be considered, as it can alter both root architecture
and root longevity (Atkinson et al., 2003; Durall et al., 1994; Espeleta
et al., 1999; Hooker et al., 1995). Continued development of root dynamics
models should thus include the eVect of mycorrhizal colonization on root
growth and development, although this may be challenging because of
limited understanding of the processes involved. In their toolkit to simulate
root system structure for ecosystem management applications, Brown and
Kulasiri (1994) included a representation of the spread of fungal populations
along the root system, with options of alternative functions to represent this
process to be selected by the user. However, only negative eVects of a fungal
population causing death of roots were considered (as discussed earlier),
rather than any beneficial eVects of mycorrhizal fungi.
While water and nutrient availability are major factors influencing the
growth of roots (as discussed in the following) and other plant components,
uptake (and in some cases also partitioning among plant components) is
generally modelled with the assumption of unlimited availability.
Root water uptake has been simulated by two approaches. The microscopic scale approach, first outlined by Gardner (1960), investigates water
movement toward an individual root and has been described in relation to
simulating root water uptake by Herkelrath et al. (1977) and Aura (1996). It
assumes that soil water flows radially through the soil to the plant root from
an imaginary thick-walled hollow cylinder of soil, with its outer radius
determined by the root density and its inner radius being the surface of the
root. In contrast, the macroscopic scale approach ignores the details of water
flow patterns toward individual roots. There are two main model groups for
a more detailed approach. In one group, root resistances and water potentials inside and at the root–soil water interface are used to calculate the water
uptake rate. In the other, plant transpiration is allocated to root uptake,
which is a function of root depth and water content.
Among the prototype versions of the reviewed models, only the Davis
model considers root water uptake and water loss to the atmosphere. Water
uptake rate (Wu) at a given time is estimated by the macroscopic scale
approach and is thus determined by an extraction function ( fe), a normalized potential root water uptake site distribution ( fnu), and a potential
transpiration rate (Tpot):
Wu ðx; y; z; tÞ ¼ fe ðx; y; z; tÞ fnu ðx; y; z; tÞ Tpot
fu ðx; y; z; tÞ
fnu ðx; y; z; tÞ ¼ P
fu ðx; y; z; tÞ
where fu is potential root uptake site distribution and fu is the integration of
the distribution over the complete soil domain.
The extraction function later introduced by Feddes et al. (1978) was used
to account for the local influence of soil water potential on the root water
uptake rate. In the further extended model (Somma et al., 1998), this
expression became one of two options. The second option (which became
the default function) used an expression from van Genuchten (1987) that
considered the combined eVects of matric and osmotic potential on the
uptake rate. The potential root uptake site distribution is constructed by
identifying the finite element that surrounds each growing apex, and subsequently setting each of eight nodes to the value of the inverse distance
between the apex and the respective node. This function is further modified
in the second option. The function value is equal to the inverse distance
between the center of the root segment and the respective corner and
is proportional to the segment length. To account for root age eVects,
the function is multiplied by a weighting factor ( fw), which is a piecewise
linear function of root segment age and branching order:
f u ðx; y; z; tÞ ¼
Di Ls
fw ;
where Di is the inverse distance,
indicates the summation of the inverse
distances of eight nodes, Ls is the root segment length, and Lt is total root
segment length over the complete soil domain.
An adaptation of the ROOTMAP model to include nutrient and water
transport and uptake was described by Dunbabin et al. (2002). In this
extended version of the model, root water uptake rate is also estimated
according to a macroscopic scale approach:
Wu ¼ Tpot fw RLD;
where RLD is root length density and fw is a weighting factor expressed as a
sigmoid curve with soil water potential ( ) ranging from the drainable upper
limit (field capacity) to the lower extraction limit (wilting point):
fw ¼
1 þ b exp ð ku Þ
where b and ku are parameters.
Modelling root nutrient uptake began with simulations of mass flow and
diVusion of nutrients to a uniform cylindrical root surface, as was the case
with water flow (Claassen and Barber, 1976; Nye and Marriott, 1969).
Mathematical simulation has been attempted according to various
approaches, and several reviews have been published (Gregory, 1996;
Mmolawa and Or, 2000; Rengel, 1993). Up to now, models have seldom
linked root architecture with nutrient uptake.
In the Davis model, a finite element grid simulated by summing nodal
sink values gives root solute uptake throughout the soil domain:
S 0 ¼ d Wu þ ð1
dÞ A;
where S0 is solute sink value for a given nodal, d is a partition coeYcient, Wu
is the water uptake rate for a given node, and A describes the rate of active
uptake for that node and is represented by the sum of Michaelis-Menten and
linear components:
J max
þ f Rd ;
Km þ c
where Jmax is maximum uptake rate, Km is a Michaelis-Menten constant, f is
a first-order rate coeYcient, Rd is the root area density, and c is the nodal
solute concentration.
The SimRoot model has a general formula to represent a local rate (Ur) for
nutrient uptake, CO2 respiration, exudation, and carbohydrate deposition,
based on Silk et al. (1986) and Sharp et al. (1990):
Ur ¼
dQ dðQ vz Þ dQ
þ vz dt
where Q is the local cumulative quantity of substance, z is the distance from
the root tip, and vz is the longitudinal growth velocity.
In the extended ROOTMAP model, there is an approximate representation
of root solute uptake by a randomly dispersed root system developing within a
finite volume of soil (Baldwin et al., 1973).
The fact that growth of a root system depends on carbohydrate supply
from above-ground plant organs has a profound eVect on root system
In the Davis model, the quantity of assimilate allocated to the root system
is limited by a piecewise linear function that determines the root/shoot allocation ratio of new assimilate. At a given growth time step, a tentative segment
length is calculated for each growing apex, and the potential requirement for
new root assimilates is estimated. If the quantity of assimilate allocated to the
root is smaller than the potential requirement, that allocated to all new
segments is scaled down by the same factor to make the actual assimilation
equal to that allocated to the root system. If the allocation exceeds the
requirement, the extra assimilate is assumed to be exuded by the root.
Thaler and Pagès (1998) used a source-sink relationship as an indicator to
represent the endogenous environment when they simulated rubber seedling
root growth and architecture using the INRA model as a framework. The
growth of each root was calculated as a function of its own growth potential
and of assimilate availability within the whole plant; the potential elongation
rate of a root was then estimated by its apical diameter as an indication of
the size of the meristem. The latter was evaluated by a monomolecular
function fitted to the upper limit of the observed apical diameter-elongation
rate scatter plot:
br ðdt d0 Þ
Vp ¼ Vmax 1 e Vmax ;
where Vp is potential elongation rate for a given root (cm d 1), Vmax is maximum
elongation rate for all roots (1.70 cm d 1), br is an initial slope of the curve
(40 d 1), d0 is the threshold diameter below which the root does not elongate
(0.025 cm), and dt is diameter of the root at time t (cm), measured at the distance
from the tip corresponding to the meristem level (0.3 cm). The dynamics of
root diameter was simulated by successive applications of the function:
dt ¼ cdt dt 1 ;
where dt is the diameter of a given root at time step t, dt 1 is that at the previous
time step, and cdi is the decrease/increase rate, which is controlled by the supply:demand ratio of carbon.
Field management factors, including nutrient application (fertilizer/manure), soil water management (irrigation/drainage), row spacing, cutting or
pruning of perennial plants, and grazing in grassland, all aVect plant assimilation and respiration and in turn influence photosynthate availability to
root growth and root architecture. Experiments on the relationship between
the fine root dynamics of sugar maple (Acer saccharum) and nitrogen availability suggest greater metabolic activity for roots in nitrogen-rich zones,
leading to greater carbohydrate allocation to these roots (Burton et al.,
2000). Arredondo and Johnson (1998) investigated the influence of cutting
on root architecture and morphology of three grass species and found that
root branching of a grazing-tolerant species decreased over time, while root
branching of a grazing-sensitive species increased over time. Up to now, such
management eVects have not been built into a root architecture model.
However, it will be important to incorporate these eVects into potential
applications for the optimization of resource management.
Interactions between roots and above-ground plant components also
need to be considered in relation to field management factors and cropping
system choices. Agroforestry systems are based on the spatial and temporal
complementarity between species that allows resource exploitation. This
applies both above- and below-ground, and there is wide acceptance of the
value not only of having spatially varied root activity between species, but
also of the potential nutrient cycling value of diVerent temporal dynamics.
These principles can also be applied to more traditional intercropped agricultural or horticultural situations. Hauggaard-Nielsen and Jensen (2001)
concluded that the competitiveness of the pea root system was an important
factor in the success of pea–barley intercrops. One application of the model
under development is in screening the most likely candidate species for field
trials of intercropping systems.
Similar principles apply to selecting the most appropriate varieties for the
agricultural systems of the future. Plant breeders constantly strive to produce
new crop varieties that will improve yield, yield stability, or specific attributes
such as disease resistance. New varieties are generally selected under high
nutrient conditions, and thus root topology of these varieties may therefore
not be optimal in lower nutrient conditions (Siddique et al., 1990). Future
farming scenarios across the world will include systems such as organic farming
and other lower input forms of agriculture in which nutrient availability may be
limited. Plant breeding has generally not focused on root characteristics; however, models may be a way of predicting how new varieties will perform under
diVerent farm management scenarios. Research to find markers linked to genes
controlling root architectural characteristics, e.g., deep rooting characteristics
from wild relatives of lettuce, is now underway (Johnson et al., 2000).
Future development of a model would allow the interaction of disease
and root system dynamics and architecture on nutrient flows to be investigated. Cook (2001) estimated that 25 to 50% of the root of wheat and barley
in the U.S. Pacific Northwest are aVected by disease. Nutrient placement,
especially of immobile nutrients such as P, is known to be critical for crop
nutrition in plants suVering from root disease (Cook et al., 2000).
The integration of dynamic models of above-ground growth, three-dimensional root system demography, and interactions between plant and environment into one single model is a major challenge because of the complexity of
the systems discussed in this chapter. Currently, each root elongation rate is
set as an input, taking no account of the source of carbohydrate and proteins.
In order to simulate root growth under field conditions, the fraction of
photosynthate allocated to the root system must also be known, which
requires development of an appropriate submodel. In order to understand
the interaction between a plant and the environment, it will be advantageous
to develop a model framework to integrate submodels that simulate various
plant and environmental components. For example, in the most recent
developments to the ROOTMAP model described by Dunbabin et al.
(2002), various interacting components of the rooting environment, including
nitrate and water flows, dynamic resource allocation, and root architectural
development, have been integrated within an object-oriented framework. At
the core of this framework is an ‘‘engine’’ that interacts with the relevant
components for information exchange. It is also straightforward to plug in
additional components or modules to the framework in order to investigate
potential mechanisms that control the response of a plant to its environment.
We have shown that a number of existing models adequately describe root
architecture arising from the processes of root extension and branching.
Procedures for computerized visualization of root architecture and growth
processes also exist. It would be appropriate to develop simulation tools to
avoid spending time repeatedly rewriting the same processes in each successive
model. The expressions describing branching orientation and position, for
example, are similar in all the reviewed models. It would be possible to write a
library for such processes so that model developers could simply treat these
processes as ‘‘plug-ins’’ for their own models. Future work should therefore
concentrate on integrating various other relevant processes into such a model,
with the ultimate objective of creating an interactively linked root growth and
soil nutrient cycling systems model. These additional processes include root
longevity and mortality, as well as the influence of environmental and management factors on root growth, extension, branching, and mortality. One
existing model has been developed within a framework suitable for adding in
further modules representing these additional processes and factors, so use
may be made of such a framework for the anticipated developments.
SAC receives financial support from The Scottish Executive Environment
and Rural AVairs Department.
Acock, B., and Pachepsky, Y. A. (1996). Convective-diVusive model of two-dimensional root
growth and proliferation. Plant Soil 180, 231–240.
Aguirrezabal, L. A. N., Pellerin, S., and Tardieu, F. (1993). Carbon nutrition, root branching
and elongation: Can the present state of knowledge allow a predictive approach at a wholeplant level? Environ. Exp. Bot. 33, 121–130.
Aguirrezabal, L. A. N., and Tardieu, F. (1996). An architectural analysis of the elongation of
field-grown sunflower root systems. Elements for modelling the eVects of temperature and
intercepted radiation. J. Exp. Bot. 47, 411–420.
Arredondo, J. T., and Johnson, D. A. (1998). Clipping eVects on root architecture and morphology of 3 range grasses. J. Range Manage 51, 207–213.
Asseng, S., Richter, C., and Wessolek, G. (1997). Modelling root growth of wheat as the linkage
between crop and soil. Plant Soil 190, 267–277.
Atkinson, D., Black, K. E., Forbes, P. J., Hooker, J. E., Baddeley, J. A., and Watson, C. A.
(2003). The influence of arbuscular mycorrhizal colonization and environment on root
development in soil. Eur. J. Soil Sci. 54, 751–757.
Aura, E. (1996). Modelling non-uniform soil water uptake by a single plant root. Plant Soil 186,
Baldwin, J. P., Nye, P. H., and Tinker, P. B. (1973). Uptake of solutes by multiple root systems
from soil. III—A model for calculating the solute uptake by a randomly dispersed root
system developing in a finite volume of soil. Plant Soil 38, 621–635.
Bengough, A. G., and Mullins, C. E. (1990). Mechanical impedance to root growth: A review of
experimental techniques and root growth responses. J. Soil Sci. 41, 341–358.
Berntson, G. M. (1994). Modelling root architecture: Are there tradeoVs between eYciency and
potential resource acquisition? New Phytol. 127, 483–493.
Berntson, G. M., and Woodward, F. I. (1992). The root system architecture and development of
Senecio vulgaris in elevated CO2 and drought. Funct. Ecol. 6, 324–333.
Bingham, I. J., and Bengough, A. G. (2003). Morphological plasticity of wheat and barley roots
in response to spatial variation in soil strength. Plant Soil 250, 273–282.
Bingham, I. J., Blackwood, J. M., and Stevenson, E. A. (1997). Site, scale and time-course for
adjustments in lateral root initiation in wheat following changes in C and N supply. Ann.
Bot. 80, 97–106.
Bland, W. L., and Dugas, W. A. (1988). Root length density from minirhizotron measurements.
Agron. J. 80, 271–275.
Bledsoe, C. S., Fahey, T. J., Day, F. P., and Ruess, R. W. (1999). Measurement of static root
parameters. Biomass, length, and distribution in the soil profile. In ‘‘Standard Soil Methods
for Long-Term Ecological Research’’ (G. P. Robertson, D. C. Coleman, C. S. Bledsoe, and
P. Sollins, Eds.), pp. 413–436. Oxford University Press, Oxford, UK.
Bobbink, R., Hornung, M., and Roelofs, J. G. M. (1998). The eVects of air-borne nitrogen
pollutants on species diversity in natural and semi-natural European vegetation. J. Ecol. 86,
Böhm, W. (1979). ‘‘Methods of Studying Root Systems’’. Springer-Verlag, Berlin.
Boukcim, H., Pagès, L., Plassard, C., and Mousain, D. (2001). Root system architecture and
receptivity to mycorrhizal infection in seedlings of Cedrus atlantica as aVected by nitrogen
source and concentration. Tree Physiol. 21, 109–115.
Brissette, J. C., and Chambers, J. L. (1992). Leaf water status and root system water flux of
shortleaf pine (Pinus echinata Mill.) seedlings in relation to new root growth after transplanting. Tree Physiol. 11, 289–303.
Brown, T. N., and Kulasiri, D. (1994). Simulation of Pinus radiata root system structure for
ecosystem management applications. Simulation 62, 286–294.
Bruinsma, J. (2003). ‘‘World Agriculture: Towards 2015/30, an FAO Perspective’’. FAO, Rome.
Burton, A. J., Pregitzer, K. S., and Hendrick, R. L. (2000). Relationships between fine root
dynamics and nitrogen availability in Michigan northern hardwood forests. Oecologia 125,
Carpenter, S. R., Caraco, N. F., Correll, D. L., Howarth, R. W., Sharpley, A. N., and Smith,
V. H. (1998). Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecol.
Appl. 8, 559–568.
Chiatante, D., Iorio, A. D., Maiuro, L., and Scippa, S. G. (1999). EVect of water stress on root
meristems in woody and herbaceous plants during the first stage of development. Plant Soil
217, 159–172.
Claassen, N., and Barber, S. A. (1976). Simulation model for nutrient uptake from soil by a
growing plant system. Agron. J. 68, 961–964.
Clausnitzer, V., and Hopmans, J. W. (1994). Simultaneous modeling of transient three-dimensional root growth and soil water flow. Plant Soil 164, 299–314.
Cook, R. J. (2001). Management of wheat and barley root diseases in modern farming systems.
Australas. Plant Path. 30, 119–126.
Cook, R. J., Ownley, B. H., Zhang, H., and Vakoch, D. (2000). Influence of paired-row spacing
and fertiliser placement on yield and root diseases of direct-seeded wheat. Crop Sci. 40,
Coutts, M. P. (1987). Developmental processes in tree root systems. Can. J. Forest Res. 17,
Davis, R. D. (1993). ‘‘Geometric Modeling and Computer Visualization of Botanical Root
Systems’’. Master Thesis, The Pennsylvania State University, University Park, PA.
Derner, J. D., Polley, H. W., Johnson, H. B., and Tischler, C. R. (2001). Root system response
of C4 grass seedlings to CO2 and soil water. Plant Soil 231, 97–104.
de Willigen, P., Heinen, M., Mollier, A., and van Noordwijk, M. (2002). Two-dimensional
growth of a root system modelled as a diVusion process. I. Analytical solutions. Plant Soil
240, 225–234.
Diggle, A. J. (1988a). ROOTMAP—A model in three dimensional coordinates of the growth
and structure of fibrous root systems. Plant Soil 105, 169–178.
Diggle, A. J. (1988b). In ‘‘ROOTMAP2.1—A Root Growth Simulation Program’’,
Miscellaneous Publication No. 6188. Western Australian Department of Agriculture,
Doussan, C., Pagès, L., and Vercambre, G. (1998). Modelling of the hydraulic architecture of
root systems: An integrated approach to water absorption-model description. Ann. Bot. 81,
Drennan, P. M., and Nobel, P. S. (1996). Temperature influences on root growth for Encelia
farinosa (Asteraceae), Pleuraphis rigida (Poaceae), and Agave deserti (Agavaceae) under
current and doubled CO2 concentrations. Am. J. Bot. 83, 133–139.
Dunbabin, V. M., Diggle, A. J., Rengel, Z., and van Hugten, R. (2002). Modelling the interactions between water and nutrient uptake and root growth. Plant Soil 239, 19–38.
Durall, D. M., Marshall, J. D., Jones, M. D., Crawford, R., and Trappe, J. M. (1994).
Morphological changes and photosynthate allocation in ageing Hebeloma crustuliniforme
(Bull.) Quel. and Laccaria bicolor (Maire) Orton mycorrhizas of Pinus ponderosa Dougl. ex.
Laws. New Phytol. 127, 719–724.
Espeleta, J. F., Eissenstat, D. M., and Graham, J. H. (1999). Citrus root responses to localized
drying soil: A new approach to studying mycorrhizal eVects on the roots of mature trees.
Plant Soil 206, 1–10.
Feddes, R. A., Kowalik, P. J., and Zaradny, H. (1978). ‘‘Simulation of Field Water Use and
Crop Yield’’. Simulation Monographs, Pudoc, Wageningen.
Fitter, A. H. (1982). Morphometric analysis of root systems: Application of the technique and
influence of soil fertility on root system development in two herbaceous species. Plant Cell
Environ. 5, 313–322.
Fitter, A. H. (1986). The topology and geometry of plant root systems: Influence of watering
rate on root system topology in Trifolium pratense. Ann. Bot. 57, 91–101.
Fitter, A. H. (1987). An architectural approach to the comparative ecology of plant root
systems. New Phytol. 106(Suppl.), 61–77.
Fitter, A. H., Graves, J. D., Wolfenden, J., Self, G. K., Brown, T. K., Dogie, D., and Mansfield,
T. A. (1997). Root production and turnover and carbon budgets of two contrasting grasslands under ambient and elevated atmospheric carbon dioxide concentrations. New Phytol.
137, 247–255.
Fitter, A. H., Stickland, T. R., Harvey, M. L., and Wilson, G. W. (1991). Architectural analysis
of plant root systems. 1. Architectural correlates of exploitation eYciency. New Phytol. 118,
Forbes, P. J., Black, K. E., and Hooker, J. E. (1997). Temperature-induced alteration to root
longevity in Lolium perenne. Plant Soil 190, 87–90.
Forbes, P. J., Ellison, C. H., and Hooker, J. E. (1996). The impact of arbuscular mycorrhizal
fungi and temperature on root system development. Agronomie 16, 617–620.
Foster, S. S. D., Bridge, L. R., Geake, A. K., Lawrence, A. R., and Parker, J. M. (1986). ‘‘The
Groundwater Nitrate Problem: A Summary of Research on the Impact of Agricultural
Land-Use Practices on Groundwater Quality between 1976 and 1985’’. Hydrogeological
Report No.86/2. British Geological Survey, Wallingford, UK.
Galloway, J. N. (1995). Acid deposition: Perspectives in time and space. Water Air Soil Poll. 85,
Gardner, W. R. (1960). Dynamic aspects of water availability to plants. Soil Sci. 89, 67–73.
Gavito, M. E., Curtis, P. S., Mikkelsen, T. E., and Jakobsen, I. (2001). Interactive eVects of soil
temperature, atmospheric carbon dioxide and soil N on root development, biomass and
nutrient uptake of winter wheat during vegetative growth. J. Exp. Bot. 52, 1913–1923.
Ge, Z., Rubio, G., and Lynch, J. P. (2000). The importance of root gravitropism for inter-root
competition and phosphorus acquisition eYciency: Results from a geometric simulation
model. Plant Soil 218, 159–171.
Gerwitz, A., and Page, E. R. (1974). An empirical mathematical model to describe plant root
systems. J. Appl. Ecol. 11, 773–781.
Goss, M. J., and Watson, C. A. (2003). The importance of root dynamics in cropping systems
research. J. Crop Prod. 8, 127–157.
Goulding, K. (2000). Nitrate leaching from arable and horticultural land. Soil Use Manage 16,
Grant, R. F. (1997). Changes in soil organic matter under diVerent tillage and rotation:
Mathematical modeling in ecosys. Soil Sci. Soc. Am. J. 61, 1159–1175.
Grant, R. F., and Robertson, J. A. (1997). Phosphorus uptake by root systems: Mathematical
modelling in ecosys. Plant Soil 188, 279–297.
Gregory, P. J. (1996). Approaches to modelling the uptake of water and nutrients in agroforestry systems. Agroforestry Syst. 31, 51–65.
Hackett, C. (1968). A study of the root system of barley. I. EVects of nutrition on two varieties.
New Phytol. 67, 287–299.
Hackett, C., and Rose, D. A. (1972). A model of the extension and branching of a seminal root
of barley, and its use in studying relations between root dimensions. I. The model. Aust.
J. Biol. Sci. 25, 669–679.
Hauggaard-Nielsen, H., and Jensen, E. S. (2001). Evaluating pea and barley cultivars for
complementarity in intercropping at diVerent levels of soil N availability. Field Crop Res.
72, 185–196.
Henderson, R., Ford, E. D., and Renshaw, E. (1983). Morphology of the structural root system
of Sitka spruce: 2. Computer simulation of rooting patterns. Forestry 56, 137–153.
Herkelrath, W. N., Miller, E. E., and Gardner, W. R. (1977). Water uptake by plant: II. The
root contact model. Soil Sci. Soc. Am. J. 41, 1039–1043.
Hooker, J. E., Black, K. E., Perry, R. L., and Atkinson, D. (1995). Arbuscular mycorrhizal
fungi induced alteration to root longevity of poplar. Plant Soil 172, 327–329.
Hooker, J. E., Munro, M., and Atkinson, D. (1992). Vesicular-arbuscular mycorrhizal fungi
induced alteration in poplar root system morphology. Plant Soil 145, 207–214.
Huang, B. R., Taylor, H. M., and McMichael, B. L. (1991). Growth and development of
seminal and crown roots of wheat seedlings as aVected by temperature. Environ. Exp.
Bot. 31, 471–477.
Huck, M. C., Ishihara, K., Peterson, C. M., and Ushijima, T. (1983). Soybean adaptation to
water stress at selected stages of growth. Plant Physiol. 73, 422–427.
Jackson, R. B., Mooney, H. A., and Schulze, E. D. (1997). A global budget for the fine
root biomass, surface area, and nutrient contents. Proc. Natl. Acad. Sci. USA 94,
Jenkinson, D. S. (2001). The impact of humans on the nitrogen cycle, with focus on temperate
arable agriculture. Plant Soil 228, 3–15.
Johnson, W. C., Jackson, L. E., Ochoa, O., van Wijk, R., Peleman, J., St Clair, D. A., and
Michelmore, R. W. (2000). Lettuce, a shallow-rooted crop, and Lactuta serriola, its wild
progenitor, diVer at QTL determining root architecture and deep soil water exploration.
Theor. Appl. Genet 101, 1066–1073.
Johnson, H., Bergström, L., Jansson, P. E., and Paustian, K. (1987). Simulated nitrogen
dynamics and losses in a layered agricultural soil. Agr. Ecosyst. Environ 18, 333–356.
Jones, C. A., Dyke, P. T., Williams, J. R., Kiniry, J. R., Benson, V. W., and Griggs, R. H.
(1991). EPIC: An operational model for evaluation of agricultural sustainability. Agr. Syst.
37, 341–350.
Jordahl, J. L., and Karlen, D. L. (1993). Comparison of alternative farming systems. III. Soil
aggregate stability. Am. J. Alternative Agr. 8, 27–33.
Joslin, J. D., Wolfe, M. H., and Hanson, P. J. (2000). EVects of altered water regimes on forest
root systems. New Phytol. 147, 117–129.
Jourdan, C., and Rey, H. (1997). Modelling and simulation of the architecture and development
of the oil-palm (Elaeis guineensis Jacq.). I. The model. Plant Soil 190, 217–233.
King, J., Gay, A., Sylvester-Bradley, R., Bingham, I., Foulkes, J., Gregory, P., and Robinson,
D. (2003). Modelling cereal root systems for water and nitrogen capture: Towards an
economic optimum. Ann. Bot. 91, 383–390.
King, J. S., Thomas, R. B., and Strain, B. R. (1997). Morphology and tissue quality of seedling
root systems of Pinus taeda and Pinus ponderosa as aVected by varying CO2, temperature,
and nitrogen. Plant Soil 195, 107–119.
Laboski, C. A. M., Dowdy, R. H., Allmaras, R. R., and Lamb, J. A. (1998). Soil strength and
water content influences on corn root distribution in a sandy soil. Plant Soil 203, 239–247.
Lungley, D. R. (1973). The growth of root systems—A numerical computer simulation model.
Plant Soil 38, 145–159.
Lynch, J. P., Nielsen, K. L., Davis, R. D., and Jablokow, A. G. (1997). SimRoot: Modelling and
visualization of root systems. Plant Soil 188, 139–151.
Majdi, H. (2001). Changes in fine root production and longevity in relation to water and
nutrient availability in a Norway spruce stand in northern Sweden. Tree Physiol. 21,
May, L. H., Chapman, F. H., and Aspinall, D. (1965). Quantitative studies of root development.
I. The influence of nutrient concentration. Aust. J. Biol. Sci. 18, 25–35.
Milchunas, D. G., Lee, C. A., Lauenroth, W. K., and CoYn, D. P. (1992). A comparison of C14,
Rb86, and total excavation for determination of root distributions of individual plants.
Plant Soil 144, 125–132.
Misra, R. K., and Gibbons, A. K. (1996). Growth and morphology of eucalypt seedling-roots,
in relation to soil strength arising from compaction. Plant Soil 182, 1–11.
Mitchell, C. C., and Entry, J. A. (1998). Soil C, N and crop yields in Alabama’s long-term ‘Old
rotation’ cotton experiment. Soil Till. Res. 47, 331–338.
Mmolawa, K., and Or, D. (2000). Root zone solute dynamics under drip irrigation: A review.
Plant Soil 222, 163–190.
Newton, C. D., Clark, H., Bell, C. C., Glasgow, E. M., and Campbell, B. D. (1994). EVects of
elevated CO2 and simulated seasonal changes in temperature on the species composition
and growth rates of pasture turves. Ann. Bot. 74, 53–59.
Nielsen, K. L., Lynch, J. P., Jablokow, A. G., and Curtis, P. S. (1994). Carbon cost of root
systems: An architectural approach. Plant Soil 165, 161–169.
Niemi, K., Häggman, H., and Sarjala, T. (2002). EVects of exogenous diamines on the interaction between ectomycorrhizal fungi and adventitious root formation in Scots pine in vitro.
Tree Physiol. 22, 373–381.
Norby, R. J., O’Neill, E. G., Hood, W. G., and Luxmoore, R. J. (1987). Carbon allocation, root
exudation and mycorrhizal colonization of Pinus echinata seedlings grown under CO2
enrichment. Tree Physiol. 3, 203–210.
Norse, D. (2003). Fertilizers and world food demand—Implications for environmental stresses.
IFA-FAO Agriculture Conference: Global Food Security and the Role of Sustainable
Fertilization, 26–28 March 2003 Rome, Italy. http://www.fertilizer.org/ifa/publicat/PDF/
Nye, P. H., and Marriott, F. H. C. (1969). A theoretical study of the distribution of substances around roots resulting from simultaneous diVusion and mass flow. Plant Soil 30,
Ozier-Lafontaine, H., Lecompte, F., and Sillon, J. F. (1999). Fractal analysis of the root
architecture of Gliricidia sepium for the spatial prediction of root branching, size and
mass: Model development and evaluation in agroforestry. Plant Soil 209, 167–180.
Pagès, L. (1999). Root system architecture: From its representation to the study of its elaboration. Agronomie 19, 295–304.
Pagès, L., Jordan, M. O., and Picard, D. (1989). A simulation model of the three-dimensional
architecture of the maize root system. Plant Soil 119, 147–154.
Parton, W. J., Schimel, D. S., Cole, C. V., and Ojima, D. S. (1987). Analysis of factors
controlling soil organic-matter levels in Great-Plains Grasslands. Soil Sci. Soc. Am. J. 51,
Persson, H. (1978). Root dynamics in a young Scots’s pine in Central Sweden. Oikos 30,
Peterson, B. J., Wolheim, W. M., Mulholland, P. J., Webster, J. R., Meyer, J. L., Tank, J. L.,
Marti, E., Bowden, W. B., Valett, H. M., Hershey, A. E., Mcdowell, W. H., Dodds, W. K.,
Hamilton, S. K., Gregory, S., and Morrall, D. D. (2001). Control of nitrogen export from
watersheds by headwater streams. Science 292, 86–90.
Pimentel, D., Harvey, C., Resosudarmo, P., Sinclair, K., Kurz, D., Mcnair, M., Crist, S.,
Shpritz, L., Fitton, L., SaVouri, R., and Blair, R. (1995). Environmental and economic
costs of soil erosion and conservation benefits. Science 267, 1117–1123.
Porter, J. R., Klepper, B., and Belford, R. K. (1986). A model (WHTROOT) which synchronizes root growth and development with shoot development for winter wheat. Plant Soil 92,
Pregitzer, K. S., King, J. S., Burton, A. J., and Brown, S. E. (2000). Responses of tree fine roots
to temperature. New Phytol. 147, 105–115.
Pulgarin, A., Navascués, J., Casero, P. J., and Lloret, P. G. (1988). Branching pattern in onion
adventitious roots. Am. J. Bot. 75, 425–432.
Rengel, Z. (1993). Mechanistic simulation models of nutrient uptake: A review. Plant Soil 152,
Rose, D. A. (1983). The description of the growth of root systems. Plant Soil 75, 405–415.
Rubio, G., Walk, T., Ge, Z. Y., Yan, X. L., Liao, H., and Lynch, J. P. (2001). Root gravitropism and below-ground competition among neighbouring plants: A modelling approach.
Ann. Bot. 88, 929–940.
Ruess, R. W., Cleve, K. V., Yarie, J., and Viereck, L. A. (1996). Contributions of fine root
production and turnover to the carbon and nitrogen cycling in taiga forests of the Alaskan
interior. Can. J. Forest Res. 26, 1326–1336.
Schaller, N. (1993). The concept of agricultural sustainability. Agr. Ecosyst. Environ 46, 89–97.
Schuurman, J. J., and de Boer, J. J. H. (1970). The developmental pattern of roots and shoots of
oats under favourable conditions. Neth. J. Agr. Sci. 18, 168–181.
Sharp, R. E., Hsiao, T. C., and Silk, W. K. (1990). Growth of the maize primary root at low
water potentials. II. Role of growth and deposition of hexose and potassium in osmotic
adjustment. Plant Physiol. 93, 1337–1346.
Shibusawa, S. (1994). Modeling the branching growth fractal pattern of the maize root-system.
Plant Soil 165, 339–347.
Siddique, K. H. M., Belford, R. K., and Tennant, D. (1990). Root:shoot ratios of old and
modern, tall and semi-dwarf wheats in a Mediterranean environment. Plant Soil 121,
Silk, W. K., Hsiao, T. C., Diedenhofen, U., and Matson, C. (1986). Spatial distributions of
potassium, solutes, and their deposition rates in the growth zone of the primary corn root.
Plant Physiol. 82, 853–858.
Smith, V. H., Tilman, G. D., and Nekola, J. C. (1999). Eutrophication: Impacts of excess
nutrient inputs on freshwater, marine, and terrestrial ecosystems. Environ. Pollut. 100,
Soil Association (2003). ‘‘Soil Association Organic Food and Farming Report 2003’’. Soil
Association, Bristol, UK.
Somma, F., Clausnitzer, V., and Hopmans, J. W. (1997). ‘‘An Algorithm for Three-Dimensional, Simultaneous Modeling of Root Growth, Transient Soil Water Flow, and Solute
Transport and Uptake, V. 2.1’’, Paper No. 100034. Department of Land Air, and Water
Resources, University of California, Davis, CA.
Somma, F., Hopmans, J. W., and Clausnitzer, V. (1998). Transient three-dimensional modeling
of soil water and solute transport with simultaneous root growth, root water and nutrient
uptake. Plant Soil 202, 281–293.
Steiner, R. A. (1995). Long-term experiments and their choice for the research study.
In ‘‘Agricultural Sustainability: Economic, Environmental and Statistical Considerations’’
(V. Barnett, R. Payne, and R. Steiner, Eds.), pp. 15–21. John Wiley & Sons Ltd., West
Sussex, UK.
Thaler, P., and Pagès, L. (1998). Modelling the influence of assimilate availability on root
growth and architecture. Plant Soil 201, 307–320.
Tierney, G. L., and Fahey, T. J. (2001). Evaluating minirhizotron estimates of fine root
longevity and production in the forest floor of a temperate broadleaf forest. Plant Soil
229, 167–176.
Tsutsumi, D., Kosugi, K., and Mizuyama, T. (2002). EVect of hydrotropism on root system
development in soybean (Glycine max): Growth experiments and a model simulation. J.
Plant Growth Regul. 21, 441–458.
United Nations (UN) (2001). ‘‘World Population Prospects, the 2000 Revision—Highlights’’,
Document No. ESA/P/WP. 165. The United Nations, New York, NY.
Upchurch, W. J., Kinder, R. J., Brown, J. R., and Wagner, G. H. (1985). ‘‘Sanborn Field—
Historical Perspective’’. Missouri Agricultural Experiment Station Research Bulletin
(1054). Missouri Agricultural Experimental Station, Columbia, MO.
van Bergen, P. F., Bull, I. D., Poulton, P. R., and Evershed, R. P. (1997). Organic geochemical
studies of soils from the Rothamsted classical experiments. 1. Total lipid extracts, solvent
insoluble residues and humic acids from Broadbalk wilderness. Org. Geochem. 26, 117–135.
van Genuchten, M. T. (1987). ‘‘A Numerical Model for Water and Solute Movement in and
Below the Root Zone’’, Research Report No. 121. U.S. Salinity Lab, ARS USDA,
Riverside, CA.
van Noordwijk, M., Dejager, A., and Floris, J. (1985). A new dimension to observations in
minirhizotrons—A stereoscopic view on root photographs. Plant Soil 86, 447–453.
Vincent, C. D., and Gregory, P. J. (1989). EVects of temperature on the development and
growth of winter wheat roots. II. Field studies of temperature, nitrogen and irradiance.
Plant Soil 119, 99–110.
Vitousek, P. M., Aber, J. D., Howarth, R. W., Likens, G. E., Matson, P. A., Schindler, D. W.,
and Schlesinger, W. H. (1997). Human alteration of the global nitrogen cycle: Sources and
consequences. Ecol. Appl. 7, 737–750.
Vogt, K. (1991). Carbon budgets of temperate forest ecosystems. Tree Physiol. 9, 69–86.
Waisel, Y., Eshel, A., and Kafkafi, U. (1996). ‘‘Plant Roots: The Hidden Half’’. Marcel Dekker,
Inc., New York, NY.
Watson, C. A., Ross, J. M., Bagnaresi, U., Minotta, G. F., RoY, F., Atkinson, D., Black, K. E.,
and Hooker, J. E. (2000). Environment-induced modifications to root longevity in Lolium
perenne and Trifolium repens. Ann. Bot. 85, 397–401.
Weaver, J. E. (1926). ‘‘Root Development of Field Crops’’. McGraw-Hill, New York, NY.
Williams, J. R., and Renard, K. G. (1985). Assessments of soil erosion and crop productivity
with process models (EPIC). In ‘‘Soil Erosion and Crop Productivity’’ (R. F. Follett and
B. A. Stewart, Eds.), pp. 67–103. American Society of Agronomy, Madison, WI.
This Page Intentionally Left Blank
R. J. Haynes
Discipline of Soil Science, School of Applied Environmental Sciences,
University of Natal, Pietermaritzburg,
Scottsville 3209, South Africa
I. Introduction
II. Total Soil Organic Matter
A. Attainment of Equilibrium
B. EVects of Agricultural Practice
III. Particulate Organic Matter
A. Method of Quantification
B. Nature of the Fraction
C. Amounts Present in Soils
D. Management-Induced Changes
E. Seasonal Fluctuations
F. Significance to Soil Quality
IV. Dissolved Organic Matter
A. Method of Extraction
B. Nature of the Fraction
C. Biodegradability of DOM
D. Adsorbed Organic Matter
E. Quantities of DOM
F. Management-Induced Changes
G. Seasonal Fluctuations
H. Significance to Soil Quality
V. Extractable Forms of Organic Matter
A. Hot Water-Extractable Organic Matter
B. Dilute Acid-Hydrolyzable C
C. Permanganate-Oxidizable C
VI. Potentially Mineralizable C and N
A. Method of Quantification
B. Nature of the Fraction
C. Relationship with Other Pools
D. Amounts Present in Soils
E. Management-Induced Changes
F. Seasonal Flunctuations
G. Significance to Soil Quality
Advances in Agronomy, Volume 85
Copyright 2005, Elsevier Inc. All rights reserved.
0065-2113/05 $35.00
VII. Synthesis and Conclusions
A. Significance of Labile Organic Matter Fractions
B. Practical Value of Labile Organic Matter Fractions
Total soil organic matter content is a key attribute of soil quality since it has
far-reaching eVects on soil physical, chemical, and biological properties. However, changes in contents of organic carbon (C) and total nitrogen (N) occur
only slowly and do not provide an adequate indication of important short-term
changes in soil organic matter quality that may be occurring. Labile organic
matter pools can be considered as fine indicators of soil quality that influence
soil function in specific ways and that are much more sensitive to changes in soil
management practice. Particulate organic matter consists of partially decomposed plant litter, and it acts as a substrate and center for soil microbial activity,
a short-term reservoir of nutrients, a food source for soil fauna and loci for
formation of water stable macroaggregates. Dissolved (soluble) organic matter
consists of organic compounds present in soil solution. This pool acts as a
substrate for microbial activity, a primary source of mineralizable N, sulfur (S),
and phosphorus (P), and its leaching greatly influences the nutrient and organic
matter content and pH of groundwater. Various extractable organic matter
fractions have also been suggested to be important, including hot water-extractable and dilute acid-extractable carbohydrates, which are involved in stabilization of soil aggregates, and permanganate-oxidizable C. Measurement of
potentially mineralizable C and N represents a bioassay of labile organic matter
using the indigenous microbial community to release labile organic fractions of
C and N. Mineralizable N is also an important indicator of the capacity of the
soil to supply N for crops. It is concluded that individual labile organic matter
fractions are sensitive to changes in soil management and have specific eVects on
soil function. Together they reflect the diverse but central eVects that organic
ß 2005 Elsevier Inc.
matter has on soil properties and processes.
Concerns regarding soil degradation and agricultural sustainability have
kindled interest in assessment of soil quality. Soil quality is simply defined as
the capacity of a soil to function, encompassing its living and dynamic
nature (Karlen et al., 1997). A more specific definition is the capacity of a
soil to function within ecosystem boundaries to sustain biological productivity, maintain environmental quality, and promote plant and animal health
(Carter et al., 1997; Doran and Parkin, 1994). An assessment usually
includes measurement of soil quality indicators that, in some way, influence
the function for which the assessment is being made. Such indicators can be
divided into chemical (e.g., pH, extractable nutrients, salinity), physical (e.g.,
aggregation, bulk density, hydraulic conductivity), and biological (e.g.,
microbial biomass C, basal respiration, earthworm numbers).
Soil organic matter is an extremely important attribute of quality since it
influences soil physical, chemical, and biological properties and processes. For
example, it is a source of energy and nutrients for soil biota, it is a plant nutrient
(N, S, and P) source via mineralization, and it aVects aggregate stability,
traYcability, water retention, and hydraulic properties. As a result, soil organic
matter content and quality are now regarded as key factors in the evaluation of
the sustainability of management practices (Gregorich et al., 1994, 1997a).
Changes in total soil organic matter content in response to alterations in
soil management practice are diYcult to detect because of the generally high
background levels and natural soil variability (Haynes and Beare, 1996). However, soil organic matter is a heterogeneous mixture of materials, ranging from
fresh plant and microbial residues to relatively inert humic compounds, with
turnover rates measured in millennia (Baldock and Nelson, 2000; Stevenson,
1994). Many attempts have been made to identify labile pools of organic matter
that are more sensitive to changes in management or environmental conditions
than total soil organic matter content. Examples include C and N held in the
microbial biomass and particulate organic matter and in water soluble, easily
extractable and potentially mineralizable fractions (Gregorich et al., 1997a;
Haynes and Beare, 1996; Janzen et al., 1997; Moore, 1997).
The level of our knowledge regarding the significance and applicability of
various labile fractions as indicators of soil quality diVers greatly. For
example, a number of workers have reviewed, in detail, the significance of
microbial biomass C and N levels (Carter et al., 1999; Dalal, 1998; Smith and
Paul, 1990; Sparling, 1997). By contrast, the nature and significance of the
non-living, labile organic matter pools are much less well understood. For
example, past research on soluble C and N has concentrated on forest soils,
and their significance to the quality of agricultural soils has only recently
been recognized. The significance of particulate organic matter has been
recognized for some time (Gregorich and Janzen, 1996), but that of the
mineralizable and extractable fractions is less well-known.
The objective of this chapter is to discuss the nature and significance and
interrelationship between these non-living labile organic matter fractions
and their value as indicators of the quality of agricultural soils.
Soil organic matter content is generally measured as organic C and/or
total N content. Although the organic fraction of soils typically accounts
for a small, but variable, proportion (typically 5–10%) of soil mass, it exerts
far-reaching eVects on soil properties. Indeed, soil organic matter has long
been suggested as the single most important indicator of soil productivity
(Allison, 1973; Campbell, 1978). This is because organic matter greatly
aVects chemical, physical, and biological properties and processes in soils.
Several workers have tabulated and discussed these eVects in detail (Baldock
and Nelson, 2000; Stevenson, 1994). The main chemical factors aVected are
charge characteristics, cation exchange capacity, buVering capacity, formation of soluble and insoluble complexes with metals, and interactions with
xenobiotics such as pesticides. Key physical properties that are influenced
include aggregate formation and stabilization, water retention, resistance
and resilience to compaction, and thermal properties. The most important
biological properties of organic matter are its role as a reservoir of metabolizable energy for soil microbial and faunal activity, its eVect in stabilizing
enzyme activity, and its value as a source of plant-available N, S, and P via
An equilibrium soil organic matter content is attained within a mature
natural ecosystem that is dependant upon the interaction of soil-forming
factors (i.e., climate, topography, parent material, and time) (Baldock and
Nelson, 2000; Haynes, 1986a). At this equilibrium level, the amount of
organic C accumulating in the soil is the same as the amount lost via
respiration as CO2.
In agricultural soils, changes in soil management practice aVect soil
organic matter content by (i) altering the annual input of organic matter
from above- and below-ground plant litter and (ii) altering the rate at which
the decomposer community degrades organic matter and releases organic
C to the atmosphere as CO2. Under any particular long-term soil management practice, soil organic matter content reaches a new steady-state level
where organic matter accumulation is balanced by losses as CO2. This
balance of soil C is shown schematically in Fig. 1.
The input of C to the soil occurs mainly as above-ground plant litter,
turnover of root material, and exudation of carbonaceous material from
roots (Cadisch and Giller, 1997; Paustian et al., 1997). This C originates
from atmospheric CO2 that has been photosynthetically fixed and
incorporated into organic compounds in plants. Once the organic residues
are added to the soil, they are decomposed by the combined actions of soil
fauna and microorganisms. During this process, the bulk of the residue
C (about 70%) is returned to the atmosphere as CO2 through faunal and
microbial respiration (Jenkinson et al., 1991). The remainder of added C,
including that incorporated into the microbial biomass, undergoes further
Harvested C
Plant C
Labile C
Stabilized C
Figure 1 A schematic diagram of the C cycle in agricultural soils. Reprinted from ‘‘Soil
Quality for Crop Production and Ecosystem Health’’, 1997, Gregorich et al. (Eds.), pp. 277–291,
Janzen et al.: Soil organic matter dynamics and their relationship to soil quality, with permission
from Elsevier.
transformations with the eventual formation of relatively recalcitrant
humic substances. These materials may be further stabilized by formation
of complexes with soil mineral surfaces (Sollins et al., 1996).
Soil organic C is shown in Fig. 1 as being composed of two major pools: a
labile and a stabilized fraction. This is a convenient division, although, in
fact, soil organic matter includes a continuum of materials ranging from
highly decomposable to very recalcitrant. The labile fraction consists of
material in transition between fresh plant residues and stabilized organic
matter. Much of it is plant and microbial tissue in various states of decomposition. It generally is considered to have a short turnover time (less than 10
years) (Janzen et al., 1997). Pools of organic matter that have been identified
as part of the labile fraction include particulate organic matter, microbial
biomass C, soluble C, potentially mineralizable C, and that extractable with
various reagents. Each of these pools defines an aspect of the labile fraction
and their significance is discussed in detail in the following sections of this
Stabilized organic C is composed of organic materials that are highly
resistant to microbial decomposition because of their chemical structure
and/or their association with soil minerals. It consists mainly of humic
substances, which are complex systems of high-molecular-weight organic
molecules made up of phenolic polymers produced from the products of
biological degradation of plant and animal residues and the synthetic
activity of microorganisms (Baldock and Nelson, 2000; Stevenson, 1994).
Humic substances make up 70–80% of the soil organic matter content of
most mineral soils. The complex structure of humic substances is largely
responsible for their stability, although other factors such as the formation
of biologically stable clay–organic matter complexes and physical inaccessibility of organic matter within soil aggregates are also important (Sollins
et al., 1996; Stevenson, 1994).
The most dramatic eVect of agricultural practice occurs when soil under
native vegetation is converted to arable agriculture. Typically, organic matter levels decline rapidly in the first 10–20 years and then stabilize at a
new equilibrium level after 30–100 years (Fenton et al., 1999; Haynes and
Beare, 1996; Paustian et al., 1997). A number of factors contribute to the
losses of organic matter, including (i) a much lower allocation of carbonaceous residues to the soil (due to the relatively wide spacing of crop plants,
removal of harvested products, and burning or removal of crop residues); (ii)
tillage-induced aggregate disruption and exposure of physically protected
organic material to microbial action, thus hastening decomposition rates;
(iii) more favorable conditions for decomposition (e.g., tillage-induced
aeration, irrigation, fertilizer and lime additions); and (iv) greater losses of
surface soil by wind and water. It is important to recognize that long-term
arable agriculture characteristically results in an increase in bulk density
compared with that under native vegetation or pasture (Dominy and
Haynes, 2002). It is therefore important to compare organic matter measurements on a volume basis (i.e., kg ha 1 to a stipulated depth) as well as on
a mass basis (Gregorich et al., 1994). This also applies to the measurements
of labile organic matter fractions discussed below. In some cases, trends in
organic matter content with land use calculated on a volume basis can diVer
significantly from those presented on a mass basis (Dominy and Haynes,
Factors that increase organic matter inputs, and thus that tend to increase
soil organic matter content, under arable agriculture include (i) a decreasing
proportion of fallow in rotation, (ii) an increase in the proportion of cereal
compared to root crops, (iii) an increasing proportion of perennial crops
(forage legumes and grasses) in rotation, (iv) the return of crop residues to
the soil rather than burning or removal, (v) fertilizer and irrigation additions
that promote increased yields and thus greater organic matter returns, and
(vi) additions of organic manures or other organic wastes (Fenton et al.,
1999; Janzen et al., 1997, 1998a,b; Johnston, 1986; Paustian et al., 1997). The
most common method of attempting to reduce the rate of organic matter
decomposition is to create less disturbance to the soil by conversion from
conventional to minimum or zero tillage. Although this characteristically
results in accumulation of organic matter in the surface 5 cm, the total
organic matter in the soil profile often remains unchanged (Haynes and
Beare, 1996; Janzen et al., 1998a).
Large perturbations to the soil system, such as conversion of native vegetation to arable agriculture, cause large changes in organic C or total N content.
These are reflected in sizeable decreases in the size of both the labile
and stabilized organic matter fractions, although the decrease is more pronounced, and occurs first, in the labile fractions. Changes in soil management
within agricultural systems usually cause more subtle changes in the balance
between inputs and losses of soil organic matter and thus in total soil organic matter content. Because of the relatively large quantity of background
organic matter already present, changes are diYcult to detect and are usually
demonstrated in long-term (e.g., >25 years) experiments (Campbell et al., 1997;
Christensen and Johnston, 1997; Janzen, 1995; Johnston, 1986). That is, as
already noted, the stabilized fraction makes up the bulk of the soil organic
matter, and it has turnover times measured in thousands of years. As a result, its
content is largely unaVected by management practices imposed on the soil.
By contrast, the labile fraction, with its much greater turnover time, is
aVected much more rapidly by management-induced changes in organic
matter inputs or losses. Janzen et al. (1998a), for example, measured a
progressive decline in soil organic C content between 1910 and 1953 that
was more pronounced under wheat fallow than continuous wheat (Fig. 2).
Radiocarbon dating of soil organic C showed that the mean residence time
increased with time; this was particularly marked in the wheat fallow system
(Fig. 2). Thus, the loss of soil organic C occurred largely by depletion of the
young, labile fractions so the mean residence time of the remaining soil
organic C increased. In contrast, at another site, applications of fertilizer
N were shown to induce an increase in organic matter content (due to
increased yields and greater organic matter returns), and the mean residence
time was decreased (Janzen et al., 1998a). This was due to a disproportionate
accumulation of C in young, labile organic C fractions.
Similarly, many field experiments have shown that management-induced
changes in soil organic matter status occur much more rapidly in the labile
pools (e.g., microbial biomass, particulate organic matter, soluble organic
matter) than in organic C or total N (Campbell et al., 1999a,b; Graham et al.,
2002). Thus, labile pools can be used as early indicators of changes in total
organic matter that will become more obvious in the longer term (Gregorich
et al., 1994, 1997a). In addition, the labile fraction has a disproportionately
large eVect on nutrient-supplying capacity and structural stability of soils
(Haynes and Beare, 1996; Janzen et al., 1997). These properties of the labile
Figure 2 Changes in the concentration and mean residence time of organic C in the surface
15 cm of soil at a long-term site in Alberta. Rotation sequence: W, continuous wheat; FWW,
fallow–wheat–wheat. ‘‘Soil Processes and the Carbon Cycle’’ by Lal et al. Copyright 1998 by
CRC Press LLC. Reproduced with permission of CRC Press LLC in the format Other Book via
Copyright Clearance Center.
fraction are the major reason why it has been the subject of much research in
recent years.
Particulate organic matter (POM) is a transitory pool of organic matter
between fresh plant residues and humified organic matter (Gregorich and
Janzen, 1996). It is typically enriched in C and nutrients, and although it
represents only a small portion of the soil mass, it is an important attribute
of soil quality since its short turnover time makes it an important source of
C and nutrients.
Particulate organic matter can be separated from soils by two distinct
methods resulting in two diVerent terms: light fraction (LF) organic matter
and sand-sized fraction (SSF) organic matter. Light fraction is isolated by
collection of dispersed soil materials that float on heavy liquids of densities
typically between 1.5 and 2.0 g cm 3. Commonly, soils are dispersed in NaI
solution having a specific gravity of about 1.7 g cm 3 (Gregorich and Ellert,
1993). The fractionation is based on the fact that the density of soil minerals
is typically >2.0 g cm 3, so free organic matter floats on these dense liquids.
Sand-sized fraction is defined as organic matter associated with sand-sized
organic matter (>20 mm diameter for European and >53 mm diameter for
American particle size classification systems). It is isolated by sieving a
dispersed soil.
Studies regarding aggregate stability have suggested that stable
macroaggregates tend to have cores of POM and that aggregates form
around particles of decaying plant residues (Golchin et al., 1994, 1998).
Thus, two forms of POM may exist in soils: (i) POM that is free and without
any significant association with mineral particles and (ii) occluded POM that
is buried within soil aggregates and/or strongly associated with mineral
particles. In recent times, some workers have diVerentiated these two
forms. The extent to which soil aggregates are disrupted and dispersed will
determine the relative amounts of free and occluded material that are
released. Thus, the free POM is usually extracted by flotation following
shaking of soil samples with a heavy liquid (density 1.6–1.8 g cm 3) for 5
or 10 minutes (Besnard et al., 1996; Gregorich et al., 1997b). The occluded
fraction is then released either by shaking for 16–18 hours or through
sonification, and is then isolated as LF by density fractionation or as SSF
by sieving.
Particulate organic matter is composed primarily of plant debris with a
recognizable cellular structure, but microscopic examination has revealed
that it also contains fungal hyphae, spores, seeds, faunal skeletons, and
charcoal (Skjemstad et al., 1990; Spycher et al., 1983). It contains a portion
of the soil microbial biomass (involved in decomposing the plant residue) as
well as humified material (produced during the decomposition of the plant
residue) (Baldock et al., 1992; Ladd et al., 1977).
Although both LF and SSF organic matter contain mostly plant residues,
their chemical and biological properties are not identical. For example,
Gregorich et al. (1995) measured the decrease in soil organic matter when
forest soils were converted to continuous maize for 25 years. By measurement of total organic C and natural 13C abundance, they showed that
mineralization of LFC was faster than that of SSFC, with the result
that after 25 years, 70% of the C in the LF was derived from maize,
compared to only 45% for the SSF. Similarly, Carter et al. (1998) investigated loss of soil organic matter when forest soils were converted to arable
agriculture. Whereas there was a 72% decrease in LFC, SSFC was not
greatly influenced by arable cultivation.
The SSFC generally represents a much higher proportion of soil C than
LFC, particularly in agricultural soils (Carter et al., 1998). In a survey of 20
forest and cropped soils, it was found that the SSF contained more organic
C and had a lower C:N ratio than LF, suggesting that the former was more
decomposed (Gregorich and Janzen, 1996). Using 13C nuclear magnetic
resonance (13C NMR) and pyrolysis-field ionization mass spectrometry
(Py-FIMS), Gregorich et al. (1996) confirmed that SSF is more decomposed
than LF. The mass spectra showed fewer lignin monomers, and dimers,
lipids, and alkyl-aromatic compounds were present in the SSF compared
with the LF. 13C NMR data indicated that the SSF contained relatively
lesser amounts of carbohydrates and aliphatic compounds and had a higher
degree of aromaticity than the LF.
These diVerences can be attributed to diVerences in the methodology of isolating the two forms of particulate organic matter. In fractions
collected on the basis of particle size, humified organic materials bound
strongly to large inorganic particles, and organic debris coated with mineral
particles will be retained on sieves and included in the SSF (Baldock
and Nelson, 2000). Much of the organic matter present as coatings on
sand grains may be more decomposed and humified than that which is
floated oV the sand fraction as LF organic matter. In addition, LFC is
separated from silt- and clay-sized particles as well as that from sand-sized
Because of these diVerences, several workers have concluded that density
fractionation is more eVective than particle size fractionation in separating
labile and non-labile organic matter fractions (Dalal and Mayer, 1987;
Gregorich and Janzen, 1996). Certainly, the less aromatic nature and more
rapid turnover time of LFC compared to SSFC confirms that the LF is a
more labile pool.
In agricultural soils, the LF typically contains 20–30% C and 5–20% N and
makes up 2–18% of total C and 1–16% of total N contained in the whole soil
(Gregorich and Janzen, 1996). The C:N ratio of the LF is normally intermediate
between that of whole soil and plant tissue (Gregorich et al., 1997a).
As noted previously, SSF organic matter typically accounts for a proportion of total soil organic matter content considerably larger than the LF. In
general, SSF makes up 20–45% of total organic C and 13–40% of total
N found in the whole soil (Bowman et al., 1999; Cambardella and Elliott,
1992; Carter et al., 1994, 1998; Doran et al., 1998; Franzluebbers and
Arshad, 1997; Hussain et al., 1999).
The LF typically accounts for a considerably higher percentage of total
organic C in undisturbed soils under native vegetation than in cultivated
agricultural ones (Carter et al., 1998; Skjemstad et al., 1986). Conversion of
undisturbed sites to arable agriculture typically results in a disproportionate
decrease in LF organic matter (Dalal and Chan, 2001; Janzen et al., 1998a).
Carter et al. (1998), for instance, showed that in comparison with forested
sites, arable cultivation caused a mean decrease in organic C content of 24%
but a decrease of 72% for LFC. The explanation for this is that upon
conversion to arable agriculture, litter inputs are greatly decreased, and
their rate of decomposition is increased by factors such as tillage, irrigation,
and fertilizer inputs.
Similarly, agricultural practices that aVect the amount of residue input and/or the rate of residue decomposition have a much greater and
earlier eVect on LF than on whole-soil organic matter content (Biederbeck
et al., 1994; Bremer et al., 1994; Janzen et al., 1992). The greater responsiveness of LF to changes in management compared to total soil organic matter
content has included increases due to continuous cropping compared to
frequent summer fallow (Janzen et al., 1992), cropping with grasses, legumes,
and continuous pastures rather than arable row crops (Angers et al., 1999;
Bremer et al., 1994; Carter et al., 1998), conversion from conventional to
zero tillage (Alvarez et al., 1998; Bolinder et al., 1999), and addition of
fertilizers, thus increasing crop growth and residue inputs in both arable
(Gregorich et al., 1997b) and grassland (Nyborg et al., 1999) systems.
As illustrated in Fig. 3, Gregorich et al. (1996) showed that the increase
in LFC in response to long-term fertilization of maize was much more
Figure 3 Quantities of total organic C and C4 (maize)-derived C in the whole soil and light
fraction in the surface 10 cm of fertilized and unfertilized soil following 32 years of maize
cropping. Data from Gregorich et al. (1996). Reprinted from ‘‘Soil and Tillage Research’’,
Vol. 47, 1998, pp. 181–195, Janzen et al.: Management effects on soil C storage on the Canadian
prairies, with permission from Elsevier.
pronounced than that for organic C. Using 13C techniques to discriminate
between native soil C and maize-derived C, they also showed that the gain in
LFC was predominantly derived from C4 (maize) residues.
Similarly, on a long-term sugarcane trash management experiment
(Fig. 4), the increase in organic C between unfertilized pre-harvest burnt
(BR) and trashed (T) treatments was 1.2-fold while that for LFC was 2.7fold. The lower values for LFC in the fertilized treatment compared to the
unfertilized trashed treatment (Fig. 4) illustrate another important point:
that the LF is a transient pool, and its size and composition will fluctuate
depending on the time of crop residue inputs and their rate of decomposition. Fertilizer applications promoted more rapid decomposition of crop
residues (trash) and LF so that at the time of sampling (8 months after trash
deposition), LFC was lower in fertilized plots.
Although SSFC accounts for a considerably greater proportion of organic C than LFC, there is still a disproportionate loss of SSFC, compared with
total organic C content of the whole soil, when undisturbed vegetation is
converted to arable agriculture (Cambardella and Elliott, 1992). Similarly,
there is a greater increase in SSFC than in organic C when conventionally
tilled soils are converted to zero tillage (Franzluebbers and Arshad, 1997;
Hussain et al., 1999; Needelman et al., 1999), when continuous cereal
cropping replaces the use of a summer fallow (Bowman et al., 1999), and
when forage grasses rather than cereals or row crops are grown (Doran et al.,
1998; Franzluebbers et al., 2000).
Figure 4 EVects of long-term trash management and fertilizer applications under sugarcane
on soil organic C, light fraction C, K2SO4-extractable C, and potentially mineralizable C. Grass,
undisturbed grassed area; BR, burnt with harvest residues removed; B, burnt with harvest
residues left on plots; T, green cane harvested with retention of a trash blanket; FO, unfertilized;
and F, fertilized annually with N, P, and K. Means associated with the same letter are not
significantly diVerent (P 0.05). Reprinted from ‘‘Soil Biology and Biochemistry’’, Vol. 34,
2002, pp. 93–102, Graham et al.: Soil organic matter content and quality: Effects of fertilizer
applications, burning and trash retention on a long-term sugarcane experiment in South Africa,
with permission from Elsevier.
Although some workers have noted only minor variations (Bremer et al.,
1994), substantial seasonal fluctuations in LF material have been observed
by a number of workers (Boone, 1994; Campbell et al., 1999a,b; Conti et al.,
1992; Spycher et al., 1983). Campbell et al. (1999a) noted that temporal
variability in LFC and LFN was associated with changes in soil moisture,
temperature, rainfall and, in some cases, rhizodeposition of root material
during anthesis of wheat. Whenever conditions favored rapid decomposition
in situ (e.g., high moisture, temperature, and/or precipitation), Campbell
et al. (1999a,b) obtained low values for LF material in subsequent laboratory measurements. Boone (1994) concluded that the seasonality of organic
matter inputs to the soil was the main factor aVecting the amounts of LF
extracted from soils under maize.
Particulate organic matter contributes to soil function in a number of
ways. First, it is the pathway through which C is returned to the soil from
plant litter (either as above-ground residues or from root turnover). It is,
therefore, the precursor for formation of other forms of organic matter (e.g.,
microbial biomass, soluble, nonhumic, and humic) and thus is a key attribute of soil quality. The POM is the major source of cellular C and energy
for the heterotrophic microbial biomass. As a result, microflora are concentrated on and around the POM rather than being distributed homogeneously throughout the soil volume (Gregorich and Janzen, 1996). Kanazawa and
Filip (1986), for example, reported that 34–42% of bacteria and 33% of fungi
in soil were associated with organic particles (<1 g cm 3), as was a large
portion of soil enzyme activity. Thus, overall soil microbial activity is greatly
aVected by POM additions.
Not only is POM deposition the major pathway by which C enters the
soil, but it is also the major pathway by which nutrients are recycled from
crop residues back to the soil. Some of the nutrients are rapidly leached
from the POM (e.g., K), while others, present mainly in organic form (e.g.,
N and S), are released by mineralization during decomposition of POM.
Entry of POM into the soil is extremely important in relation to maintenance of soil structure and, particularly, macroaggregation. Electron
microscopy has demonstrated that decomposing POM becomes encrusted
with clay and silt particles via mucilagenous material (Golchin et al., 1994;
Waters and Oades, 1991), and other studies have shown that labelled plant
materials added to soils are initially incorporated into macroaggregates
(Puget et al., 1995). The large, active microbial community associated
with the decomposing POM produces binding agents such as exocellular
mucilagenous polysaccharides. The decomposing POM acts as a center
for intense microbial activity, and aggregations of mineral particles are
stabilized (Golchin et al., 1994, 1998). The size of the macroaggregates
formed is dependent on the size and geometry of the deposited POM.
However, as decomposition of POM continues, the organic cores holding
the macroaggregates together are broken down into smaller pieces and
macroaggregate stability declines. Thus, macroaggregate stabilization by
POM is a transient, dynamic process, and maintenance of a high level of
macroaggregation is dependant on the continued addition of POM to the
Particulate organic matter is a major food and energy source for endogeic
(soil-dwelling) soil fauna, including many earthworms and termites (Lavelle
et al., 1998). Endogeic earthworms, for example, ingest soil but have a
selective preference for materials with a high organic matter content. Particulate organic matter within the soil matrix originating from crop residues is a
major source of energy and nutrients to such soil fauna (Curry, 1998; Lavelle
et al., 1998). Due to this selective feeding, their casts have much higher
contents of organic matter and nutrients and higher microbial activity than
the surrounding soil (Edwards and Bohlen, 1996; Lee, 1985). Endogeic
earthworms seldom come to the soil surface but continually burrow through
the topsoil and backfill their burrows and other soil voids with their casts
(Lavelle, 1988). They are extremely important in relation to both maintenance of soil porosity and nutrient cycling by increasing nutrient availability
(particularly that of N) (Blair et al., 1995b).
Thus, POM is associated with a multitude of soil processes and functions
and is, therefore, a key attribute of soil quality.
Dissolved organic matter (DOM) is the organic material present in dissolved form in soil solution. It originates as leachates from plant litter,
exudates from soil microflora, and roots and hydrolysis of insoluble soil
organic matter. It consists of a wide range of organic compounds. The vast
bulk of research into the nature and significance of this fraction has been
concentrated in forest soils (Herbert and Bertsch, 1995). In very recent times
it has been used as a labile organic matter fraction and soil quality indicator
in agricultural soils.
While DOM in soil leachates, extracted soil solutions, and saturation
paste extracts have been studied, for simplicity, many workers have extracted
it from field-moist soils with distilled/deionized water (1:2 w/v extractant
ratio for 1 hour) (Burford and Bremner, 1975). Following centrifugation,
the extract is often passed through a membrane filter (e.g., 0.45 mm)
to remove dispersed soil material. This method was originally developed
to extract a pool of C readily available to the heterotrophic biomass that
could be used to help predict the potential for denitrification.
A number of salt-extractable organic matter fractions have also been used
as measures of dissolved organic C (DOC) and N (DON). Salt extracts are
used mainly for ease of extraction since they cause flocculation of soil
colloids. Many workers have used the amount of C extracted from fieldmoist soils with 0.5 M K2SO4 as a measure of ‘‘labile’’ or ‘‘soluble C’’ (Chan
and Heenan, 1999; Graham et al., 2002; Haynes, 1999; Jensen et al., 1997).
The amount of organic N in the extract has also been used as a measure of
soluble organic N (McNeill et al., 1998). This is an attractive option because
microbial biomass C and N are often routinely measured by the direct
extraction method (Vance et al., 1987). Such an analysis is based on the
diVerence between C and N extracted with 0.5 M K2SO4 from chloroformfumigated and unfumigated soils. Thus, where microbial biomass C and
N are measured so too is K2SO4-extractable C and/or N.
Exchangeable mineral NðNHþ
N) is routinely extracted from
4 and NO3
soils with 2 M KCl (Keeney and Nelson, 1982). A number of workers
investigating the N economy of agricultural soils have analyzed the organic
N content of these KCl extracts as a measure of soluble organic N (Bhogal
et al., 2000; Murphy et al., 2000).
Care should be exercised when using salt solutions such as 0.5 M K2SO4
or 2 M KCl as extractants for DOC and DON since the concentration of
DOM measured may well diVer significantly from that measured by water
extraction, which, in turn, could diVer from that present in displaced soil
solutions. This is because DOM is in equilibrium with that adsorbed to clay
colloids. The pH of extraction, ionic strength, and dominant species of
anions and cations present will all aVect adsorption/desorption of organic
matter and therefore the concentration of DOM present in solution. These
aspects are discussed in more detail in a following section.
Several workers have noted that, in most soils, water extracts contain
more DOC than 0.5 M K2SO4 (Bolan et al., 1996; Haney et al., 1999). This
has been attributed to a number of factors. First, due to a salt eVect, the pH
of K2SO4 extracts can be 1 to 2 units lower than that of water extracts (Bolan
et al., 1996). Second, Haney et al. (1999) suggested that flocculation of
soil colloids in K2SO4 extracts may favor readsorption of solubilized
C onto the colloids. The relationship between water-extractable DON and
2 M KCl extractable DON appears, as yet, unknown.
Normally, DOM is extracted from field-moist soils. Air drying soils prior
to extraction of dissolved C results in a substantial increase in its concentration (Bolan et al., 1996; Davidson et al. 1987; Haynes, 2000). Haynes (2000)
reported 1.6- to 3.8-fold increases in agricultural soils, while Davidson et al.
(1987) noted 2- to 10-fold increases for forest soils from North Carolina.
Such increases are partially derived from lysis of the microbial biomass that
is killed by desiccation (West et al., 1992). Solubilized C can also be derived
from soil humic material since drying causes the macromolecules of
soil organic matter to change into a highly condensed state. Such shrinkage
of organic matter results in disruption of organo-mineral associations, with
the subsequent release of some low-molecular-weight humic components
(Haynes and Swift, 1991).
The organic N solubilized by drying a soil is thought by some to be an
important pool of labile, readily mineralizable organic N. For example,
organic N extracted from air-dried soils with 0.01 M CaCl2 has been closely
correlated with plant N uptake in pot (Appel and Mengel, 1990, 1993;
Nunan et al., 2001) and field experiments (Appel and Mengel, 1992; Groot
and Houba, 1995). The organic N solubilized by drying, and subsequently
extractable with CaCl2, is mainly from the non-microbial biomass, soil
organic N pool (Appel, 1998; Nunan et al., 2001).
Dissolved organic matter may originate from leaching from above- or
below-ground plant litter and/or the synthetic activity of soil microflora
involved in decomposition of the litter and/or native soil organic matter.
In addition, the organic matter in soil solution is in equilibrium with three
main pools of solid-phase organic matter: (i) that adsorbed to the surface of
Al and Fe hydrous oxides, (ii) that precipitated due to coagulation in the
presence of divalent and trivalent cations, and (iii) the insoluble organic
matter present in the topsoil (Reemtsma et al., 1999). It consists of a wide
range of organic compounds, including simple aliphatic organic acids, phenols, phenolic acids, free amino acids, sugar acids, carbohydrates, and
complex humic molecules of various molecular weights (Stevenson, 1994).
The chemical nature of DOM in forest soils, leachates, and ground and
surface waters has conventionally been characterized based on its adsorption
to nonionic and ion-exchange resins that separate acids, bases, and neutral
substances with hydrophobic and hydrophilic groups (Aiken and Leenheer,
1993; Herbert and Bertsch, 1995; Leenheer, 1981; Qualls and Haines, 1991).
In general, the major part of DOC (approximately 40–60%) is present as
hydrophobic acids (Jandl and Sollins, 1997; Qualls and Haines, 1991; Qualls
et al., 1991; Smolander et al., 2001; Vance and David, 1991). This fraction is
thought to contain mainly humic substances. Weak hydrophobic acids are
mainly phenols and account for 10–20% of DOC, while hydrophilic acids
account for 20–30% and consist of humic and nonhumic substances with low
molecular weight and with a high content of carboxylic acids (Cook and
Allan, 1992a; Smolander et al., 2001). Hydrophilic neutrals consist of free
carbohydrates and alcohols and make up 3–20% of DOC (Qualls et al., 1991;
Smolander et al., 2001). While the hydrophobic acid fractions contain the
highest proportion of DOC, the dissolved organic N, and more particularly
S and P, are concentrated predominantly in the hydrophilic acid fractions
(Kaiser, 2001; Qualls and Haines, 1991).
In agreement with the finding that much of the DOM is present as soluble
humic material, both Guggenberger and Zech (1994) and Huang et al. (1998)
showed that DOM is a highly oxidized form of organic matter. Using
pyrolysis mass spectrometry and pyrolysis gas chromatography mass spectrometry, these workers showed that in comparison with bulk soil organic
matter, DOM shows an accumulation of more oxidized lignin and aromatic
structures, especially those containing carboxylic and dicarboxylic acids,
which are typical of those found in humic and fulvic acids. The polysaccharide-type compounds in the DOM were found to be more modified (greater
abundance of furan structures) and had significantly lower molecular
weights and more diverse structures than those in soils, suggesting that
they were of highly modified plant and/or microbial origin.
The importance of soil humic material in supplying the DOC fraction
was demonstrated by Gregorich et al. (2000), who used the 13C technique
to examine the origin of DOC and microbial biomass C in maize-cropped
soils. They showed that after 30 years of maize cropping, 10–23% of
humus C, 30–50% of WSC, and 82–92% of microbial biomass C was C4
(maize)-derived. Thus, native, humus-derived C still dominated the isotopic
composition of the DOC pool.
Dissolved organic matter is generally considered a labile substrate for soil
microbial activity. Laboratory incubation studies have shown a strong
positive correlation between CO2 respired and the amount of water-extractable C in a wide range of soils (Burford and Bremmer, 1975; Cook and
Allan, 1992a,b; Davidson et al., 1987; Seto and Yanagiya, 1983; Zak et al.,
1990), and Brooks et al. (1999) found a close link between heterotrophic soil
microbial activity and water-soluble C at a catchment scale.
Despite this, generally less than 50% (typically only 10–40%) of DOC has
been observed to be readily degradable as estimated by solution incubations
(Boissier and Fontvieille, 1993; Boyer and GroVman, 1996; Jandl and
Sollins, 1997; Qualls and Haines, 1992; Wagai and Sollins, 2002; Yano
et al., 2000). Boyer and GroVman (1996), for example, found that roughly
22% (deciduous forest) and 29% (maize field) of water-extractable C from
topsoils were degraded within 2 weeks. Wagai and Sollins (2002) found
that the percentage of biodegradable DOC in forest soils was 13–16% in
leachates and 18–27% in water extracts. The biodegradability of DON is
similar to that of DOC (Qualls and Haines, 1992). The poor degradability
is thought to be due to the presence of substantial quantities of soluble
humic substances, which are relatively recalcitrant. Using the classical fractionation procedure for DOM, both Jandl and Sollins (1997) and Qualls and
Haines (1992) demonstrated that biodegradable DOC is present mainly in
the carbohydrate-rich hydrophilic neutral fraction and that the biodegradability of DOC is strongly related to the initial content of this fraction. Thus,
the readily metabolizable fraction of DOC will be extremely important in
relation to soil microbial activity. This is likely to be mainly carbohydrate
and of both plant and microbial origin. Since this fraction is highly labile, it
is the flux of readily metabolizable C passing through it that is a major
determinant of soil microbial activity.
Adsorption/desorption of organic matter by reactive soil surfaces has
substantial eVects on concentrations of DOM present in soil solution. The
capacity of soils to adsorb organic matter is largely determined by their Al and
Fe oxide content (Jardine et al., 1989; Kaiser et al., 1996; Moore et al., 1992).
Negatively charged, high-molecular-weight, humic polymers can form
strong bonds with metal hydrous oxide surfaces through both electrostatic
bonding (anion exchange) and specific adsorption, i.e., ligand exchange
(Stevenson, 1994). Anion exchange is possible because positive sites exist on
the amphoteric Fe and Al oxide surfaces. Ligand exchange occurs through
displacement of OH20.5þ and OH0:5 groups at the oxide surface by OH and
COOH groups on the humic molecules and results in strong binding of the
humic material to hydrous oxides. Physical (Van der Waals) forces are also
involved in the interaction of humic molecules with oxide surfaces (Stevenson, 1994). Low-molecular-weight organic acids such as citrate, malate, and
oxalate are also known to be specifically adsorbed to Fe and Al hydrous
oxide surfaces by ligand exchange (Jones and Brassington, 1998).
Factors that decrease adsorption and favor desorption of organic matter
also favor increased concentrations of dissolved organic matter in solution.
These factors include (i) increasing pH, (ii) decreasing ionic strength
in electrolyte solution, (iii) decreasing valency of cations present in solution
(Al3þ < Ca2þ < Naþ), and (iv) decreasing strength with which anions in
solution can be adsorbed by metal oxide surfaces (H2PO4 > SO24 >
NO3 ¼ Cl ) (Kaiser and Zech, 1997; Reemtsma et al., 1999; Styllberg
and Magnusson, 1995; Vance and David, 1991). For the above reasons,
the use of various salt solutions to extract DOM is likely to influence
the amounts present, particularly in soils with an appreciable content of
Al and Fe oxides.
Dissolved organic C in field-moist soils typically accounts for
only 0.05–0.40% SOC in agricultural soils (Campbell et al., 1999a,b;
Chantigny et al., 1999; Haynes and Williams, 1999; Lundquist et al., 1999;
Sarathchandra et al., 1988). In forest soils, it is often in the range of
0.25–2.0% but can be considerably higher (Boyer and GroVman, 1996;
Cook and Allen, 1992a; Smolander et al., 2001). Less is known regarding
the proportion of total N present as DON. However, Smolander et al. (2001)
found that in Norway spruce stands DON represented 0.15% of total N;
under clear cutting, it represented 0.34%. In agricultural soils, Haynes (2000)
found it accounted for 0.15–0.19% of total N, while in undisturbed pasture
soils it represented 0.15–0.61%, and following cultivation of pastures,
0.22–2.8% (Bhogal et al., 2000).
The responsiveness of DOM to changes in agricultural practice is
not well documented at present. Nevertheless, it has been shown to be
increased more markedly than organic C or total N by addition of crop
residues (Graham et al., 2002; Jensen et al., 1997), replacement of wheatfallow systems by continuous wheat (Campbell et al., 1999a,b), conversion
of conventional management to an organic system (Lundquist et al., 1999),
conversion of an arable system to pasture (Haynes 1999, 2000), ploughingin a pasture (Murphy et al., 2000), and stock camping by grazing animals (Haynes and Williams, 1999), and Chantigny et al. (1999) found it
was decreased by increasing fertilizer N rates. As shown in Fig. 4, Graham
et al. (2002) found that the increase in 0.5 M K2SO4-extractable C induced
by conversion from preharvest burning (BR) of sugarcane to green cane
harvesting with trash retention (T) was greater (1.7-fold) than that for
organic C (1.2-fold).
Due to the labile nature of DOC, seasonal fluctuations in its concentration are commonly encountered (McGill et al., 1986; Rolston and Liss,
1989). Under arable systems in Canada, DOC was reported to increase
from spring to summer and decrease from summer to autumn (Campbell
et al., 1999a,b). Similarly, in an arable system in Denmark, Jensen et al.
(1997) found that 0.5 M K2SO4-extractable C and N were higher in spring
and summer than in autumn and winter. In arable soils in Britain, Murphy
Figure 5 Seasonal changes in water-extractable C in the surface 7.5 cm of soil under a
permanent pasture over a 2-year period. Reprinted from ‘‘Biology and Fertility of Soils’’,
Vol. 6, 1988, pp. 328–335, Sarathchandra et al.: Seasonal changes and the effects of fertilizer
on some chemical, biochemical and microbiological characteristics of high-producing pastoral
soil, with permission from Springer-Verlag.
et al. (2000) observed an increase in 2 M KCl-extractable organic N in spring
and then a decrease during summer. Because of the shortage of available C in
arable soils, DOC probably increases in spring due to deposition of root C by
the growing crop. This soluble C is then metabolized by the microbial
biomass in late summer, when soil temperature and moisture conditions
favor high microbial activity.
In C-rich pastoral systems, DOC has been observed to be low in
summer and high during late winter (Dormaar et al., 1984; Sarathchandra
et al., 1988). The peaks in DOC under a permanent pasture over the
late winter–spring period (August–September) are clearly evident in Fig. 5.
This increase has been attributed to greatly decreased soil microbial
activity and/or death and lysis of microbial cells during the cold conditions.
Soluble organic matter decreases during spring because conditions are
more favorable for microbial activity; this results in use of DOC by the
microbial biomass.
Dissolved organic matter is considered the most dynamic C fraction in
soils, and a portion of it is a readily available substrate for microbial activity
(McGill et al., 1986). It therefore represents a mobile source of energy in
soils. It is also a primary source of mineralizable N, S, and P (Haynes, 2000)
and so can make an important contribution to nutrient availability and
cycling. It behaves as a reactive component of soil solution and can form
soluble complexes with multivalent cations (Stevenson, 1994), thus influencing
their bioavailability and/or movement within the soil profile. For example,
complexation of monovalent Al in soil solution by soluble organic matter
renders it essentially nonphytotoxic (Haynes and Mokolobate, 2001). In addition, DOM can contribute to soil acidity through the presence of low-molecular-weight organic acids (Dijkstra et al., 2001), and its leaching contributes to
the nutrient and organic matter status and pH of ground and surface waters
(McDowell and Likens, 1988; Moore, 1997; Qualls and Haines, 1991).
In many respects, DOM has been found to be more important in its role
in the N than the C cycle (Qualls et al., 1991). Indeed, in general, DON (as
well as S and P) is more mobile in soils than SOC since it is preferentially
concentrated in the lower molecular weight, more mobile humic fractions
(Kaiser, 2001; Kaiser and Zech, 1997). Half, or more, of the soluble N in soil
solution is in organic form in most forested ecosystems (Bergmann et al.,
1999; Casals et al., 1995; Cortina and Romana, 1992; Seely et al., 1998;
Smolander et al., 1995). Furthermore, DON is the major form of N exported
from most forested watersheds (Qualls et al., 1991; Smolander et al., 2001;
Sollins and McCorison, 1981). Smolander et al. (2001) studied concentrations of DON under Norway spruce stands over a 5-year period. As shown
in Fig. 6, under the forested plot, organic N in soil solution averaged less
than 2 mg L 1 but amounted to 77% of total N in solution. Clear-cutting
increased the amount of both mineral and organic N in solution, but the
percentage of total N present in organic form still averaged 65%.
In the past 5 years, attention has turned to the role of DON in agricultural
soils. Bhogal et al. (2000) found that DON accounted for 20–90% of 2 M
KCl-extractable N in pastoral soils, with exceptionally high values observed
in recently cultivated pastures. Murphy et al. (1999) found that it accounted
for 33–60% of total soluble N in arable and grass ley soils, while in 12 arable
soils it was observed to make up 40–50% of soluble N (Murphy et al., 2000).
In a grassland soil in northern Ireland, DON accounted for up to 55 and
20% of annual N losses via drainage from plots receiving 100 and 500 kg
N ha 1 year 1, respectively (Watson et al., 2000), and concentrations of
SON in drainage water exceeded the European Community maximum
admissible concentration for drinking water (1.0 mg N L 1). Thus, DON
seems to be an important pool of N in agricultural soils.
The very small size of the DOM pool and its highly labile nature has led
some to question the validity of its use as a soil quality indicator (Baldock
and Nelson, 2000). That is, it is flux of readily available substrates through
the DOM that is important in relation to the size and activity of the
microbial biomass and nutrient availability. Because the concentration
of readily metabolizable organic compounds in solution is kept low by
microbial assimilation and/or mineralization, the size of the pool does not
Figure 6 Concentrations of dissolved organic N and percentage of total dissolved N present
as organic N in soil solution collected at a depth of 10 cm over a 5-year period after clear-cutting
a Norway spruce stand or leaving it under forest. Columns show annual means (SD).
Reprinted from ‘‘Biology and Fertility of Soils’’, Vol. 33, 2001, pp. 190–196. Smolander et al.:
Dissolved soil organic nitrogen and carbon in a Norway spruce stand and in an adjacent
clear-cut, with permission from Springer-Verlag.
necessarily reflect the flux through it. Nonetheless, as discussed above, it has
been used successfully as an indicator of changes in soil management, and its
role in leaching of N is becoming increasingly recognized.
Many diVerent chemical extractants have been used in attempts to extract
a labile portion of organic matter from soils. For example, many chemical
indices of potentially mineralizable soil N have been proposed (Goh
and Haynes, 1986; Keeney, 1982). These can be divided into three broad
groups: (i) weak (hot water, hot 0.01 M CaCl2, hot 1 M or 2 M KCl, 0.01 M
NaHCO3), (ii) intermediate (alkaline permanganate, Na2CrO4 plus H3PO4,
1 M NaOH), or (iii) strong (6 N H2SO4, K2Cr2O7-H2SO4) extractants (Goh
and Haynes, 1986). Numerous other reagents have been used to extract a
labile fraction of organic matter, including NaOH, Na2CO3, Na2P2O7,
acetylacetone, acetylaldehyde, acetone (Stevenson, 1994), and chelating
resins (Dormaar, 1972).
A detailed discussion of the use of various extractants is beyond the scope
of this chapter. However, three diVerent extractants that have been used
recently to evaluate labile organic matter for soil quality evaluation are
discussed in the following sections.
Two diVerent approaches have been employed to extract the hot waterextractable fraction. The first is to extract with boiling water for about
1 hour (Keeney and Bremner, 1966; Leinweber et al., 1995; Redl et al.,
1990), and the second is to extract at 80 8C for about 16 hours (Chan
and Heenan, 1999; Ghani et al., 2000; Sparling et al., 1998). Hot waterextractable C extracted by either method accounts for about 1–5% of soil
organic C (Chan and Heenan, 1999; Leinweber et al., 1995; Sparling et al.,
The determination of an easily mineralizable N by extraction with hot
water was pioneered by Keeney and Bremmer (1966). Later, Körschens et al.
(1990) suggested that the fertility status of soils could be characterized by
determinations of hot water-extractable C and N. Leinweber et al. (1995)
used solid state 13C-NMR and pyrolysis-field ionization mass spectrometry
to show that hot water-extracted organic matter was largely composed of
carbohydrates and N-containing compounds, amino-N species and amides
in particular. Ghani et al. (2000) observed that 45–60% of C extractable
with hot water was carbohydrates. Several workers have suggested that hot
water extracts contain organic substances that are mainly of microbial origin
(Redl et al., 1990), and the monosaccharide content of the carbohydrate
component confirms this (see below). Hot water-extractable C is usually
extracted from air-dried soils, so much of the microbial biomass has
been desiccated and the cells lysed. Thus, a substantial portion of the hot
water-extractable C and N may originate directly from the microbial biomass (Sparling et al., 1998). It may also originate from root exudates and
lysates, organic matter weakly adsorbed to soil minerals, that bound to, or
trapped, in humic molecules and that involved in bonding soil aggregates
The carbohydrate content of hot water extracts has been used as an index of
soil quality, particularly in relation to soil aggregation (Haynes and Beare,
1996). Carbohydrates are very important bonding agents for soil aggregates
(Degens, 1997; Haynes and Beare, 1996), but measurement of total acid hydrolyzable carbohydrates does not diVerentiate between total carbohydrates and
the more specific pool that is involved in aggregation. A promising alternative is
to extract an active fraction that is involved in binding aggregates. The hot
water-extractable carbohydrate fraction has been suggested as such a pool
(Haynes and Swift, 1990). Indeed, a number of workers have observed that
the hot water-extractable carbohydrate fraction is more closely related to
aggregate stability than total carbohydrates or total organic C of soils (Angers
et al., 1993a; Ball et al., 1996; Haynes and Beare, 1997; Haynes and Francis,
1993; Haynes and Swift, 1990; Haynes et al., 1991). This fraction accounts for
about 6–13% of the total carbohydrate content of soils (Haynes and Francis,
1993; Haynes et al., 1991; Puget et al., 1999). The monosaccharide content of
hot water extracts has been analyzed in order to examine its origin. The
[galactose (G) þ mannose (M)]:[arabinose (A) þ xylose (X)] ratio is typically
low (<0.5) for plant polysaccharides and high (>2.0) for microbial polysaccharides (Oades, 1984). In bulk soils, the (G þ M):(A þ X) ratio in hot water
extracts generally ranges from 1.3 to 1.7 (Ball et al., 1996; Debrosz et al., 2002;
Puget et al., 1999), and in rhizosphere soil, Haynes and Francis (1993) and
Haynes and Beare (1997) found a ratio of 1.9–2.3. This suggests that the extracts
are dominated by mucigel of microbial origin but that plant polysaccharides
are also present.
Haynes et al. (1991) found that hot water-extractable carbohydrate
changed much more rapidly in response to short-term pasture than organic
C. As shown in Table I, 4 years of pasture in an arable pasture rotation
Table I
EVect of Previous Cropping History on Aggregate Stability, Organic C, Hot Water-Extractable
Carbohydrate and Biomass C Content of a Soil from the South Island of New Zealand
18 years pasture
4 years pasturea
1 year pasture
1 year arable
4 years arable
10 years arable
(MWD, mm)
C (%)
Hot water extractable
(mg C g 1)
C (mg C g 1)
The 1-year and 4-years pasture and 1-year and 4-years arable soils come from a cropping
rotation of 4 years arable followed by 4 years pasture.
(Data from Haynes et al., 1991.)
resulted in a substantial increase in hot water-extractable carbohydrate,
microbial biomass C, and aggregate stability, while organic C content did
not change significantly. Other workers have also recorded short-term
increases in hot water-extractable carbohydrate and microbial biomass
C in response to rhizodeposition of organic matter where no changes in
organic C were detectable (Haynes and Beare, 1997; Haynes and Francis,
A labile fraction of soil C, more specifically, carbohydrate C, has been
extracted by dilute acid hydrolysis (0.5 M–2.5 M H2SO4) (Angers and
Mehuys, 1989; Angers et al., 1993b; Carter et al., 1994; Chan and Heenan,
1999; Shepherd et al., 2001). Chan and Heenan (1999) found that hydrolysis
with 1.5 M H2SO4 released 32–37% of total organic C, while Rovira and
Vallejo (2002) found hydrolysis with 5 N H2SO4 extracted between 22 and
45% of total organic C. Dilute acid hydrolysis commonly extracts 5–16 times
as much carbohydrate as hot water (Angers et al., 1993a; Puget et al., 1999;
Shepherd et al., 2001); it extracts about 65–85% of the total carbohydrate
content of soils (Puget et al., 1999).
Acid hydrolyzable carbohydrates have been found to change more rapidly in response to changes in management than organic C content. Angers
et al. (1993b) found that the ratio of acid hydrolyzable carbohydrate C to
total organic C was greater under zero than conventional tillage, while
Angers and Mehuys (1989) found that the ratio increased in the order of
bare soil < maize < barley < lucerne. Nevertheless, Angers et al.
(1993a) observed that aggregate stability was more closely correlated with
hot water-extractable than dilute acid hydrolyzable carbohydrates.
Blair et al. (1995a) suggested that a fraction of organic C oxidizable
with 333 mM KMnO4 for 1 hour was a useful index of labile soil
C. This fraction encompasses all those organic components that can be
readily oxidized by KMnO4, including labile humic material and polysaccharides (Conteh et al., 1999). Blair (2000) showed that neither CaCO3
nor charcoal contributes significantly to labile C measured by oxidation.
The KMnO4-oxidizable organic C fraction accounts for 5–30% (often
after 15–20%) of organic C (Blair, 2000; Blair et al., 1995a, 1998;
Conteh et al., 1999; Graham et al., 2002; Whitbread et al., 1998).
This oxidizable fraction is usually more sensitive to soil management than
total organic C content. It has been shown to be sensitive to conversion from
grassland to arable agriculture (Blair et al., 1995a; Lefroy et al., 1993) or
conversion from burning to crop residue retention (Blair, 2000; Blair et al.,
1998; Conteh et al., 1999). Blair et al. (1998), for example, reported no
significant change in organic C in the top 7.5 cm of soil following conversion
of sugarcane from burning to trash retention, but there was a significant
increase in oxidizable C.
Mineralizable C is usually measured by incubating a sample of field-moist
soil in a sealed chamber containing an alkali trap. The CO2-C accumulated
in the trap is measured by acid titration (Öhlinger, 1996a; Zimbilske, 1994).
The incubation period commonly ranges from 10 to 30 days, and the
chamber is opened and the trap periodically replaced to allow gas exchange
and thus maintenance of aerobic conditions. The CO2 accumulated in the
headspace can also be measured using a CO2 analyzer (a gas chromatograph
or infrared gas analyzer), and various continuous flow automated methods
have been developed to allow simultaneous aeration and periodic gas sampling (Alef, 1995a; Öhlinger, 1996b; Zimbilske, 1994). Mineralizable C (and
N) is usually calculated in mg kg 1 soil. Since the CO2 evolved is produced
by microbial respiration, it can also be presented as basal respiration
rate (mg CO2-C g 1 day 1). Mineralizable N can be measured in
closed or open incubation systems. A closed incubation is the same as
that described above, and mineralizable C and N can be measured simultaneously. The quantity of exchangeable (2 M KCl-extractable) mineral N
4 plus NO3 N) in the soil is measured before and after incubation, and
mineralizable N is calculated by the diVerence (Alef, 1995b; Drinkwater
et al., 1996).
In an open incubation, soil is typically mixed with sand (to maintain
porosity and hydraulic conductivity) and incubated in a leaching column
for 8–30 weeks (Bundy and Meisinger, 1994; Stanford, 1982). The soil
is leached with 0.01 M CaCl2 periodically, and NHþ
4 and NO3 N in leachates
is measured. A nutrient solution (minus N) is applied after each leaching,
and the soil is then drained to a known tension and reincubated. In recent
times, open incubation systems have been used in preference to closed
systems since they are thought to simulate the eVect of continual removal
of mineralized NHþ
4 and NO3 N by plant uptake. Their disadvantage
is that they are time consuming in comparison with short-term closed
Short-term anaerobic (waterlogged) laboratory incubation systems have
also been used to measure mineralizable N (Keeney, 1982), while C and
N mineralization can both be measured in field incubations (Anderson,
1982; Raison et al., 1987).
It is important to note here that potentially mineralizable C and N are not
analagous measurements. The CO2 evolved during incubation indicates
the total metabolic activity of the heterotrophic microorganisms in the soil
that are decomposing organic matter, using substrate C as an energy source,
and respiring CO2. Potentially mineralizable N is a measure of the net flux of
mineral N released from the mineralizable organic fraction in the soil.
However, mineralization and immobilization of N occur simultaneously.
That is, a portion of the mineral N released during gross N mineralization
is assimilated by the soil microflora, and the excess not required by them
accumulates in the soil as NHþ
and NO3 N. As a result, the magnitude and
patterns of potentially mineralizable C and N do not necessarily correspond.
For example, if the soil contains an available substrate with a wide C:N
ratio, then during its decomposition, and release of CO2, mineral N will be
assimilated from the surrounding soil (immobilized) by the decomposer
microflora. As a result, there may be no immediate release of mineral
N into the soil.
Potentially mineralizable C and N have been observed to be positively
correlated with microbial biomass C and N (Angers et al., 1993b; Campbell
et al., 1991, 1997; Franzluebbers et al., 1994; Hassink, 1995; Janzen et al.,
1992), light fraction C and N (Barrios et al., 1996; Campbell et al., 1999a,b;
Hassink, 1995; Janzen et al., 1992; Wander and Bidard, 2000), and soluble
C and N (Campbell et al., 1999a,b). The strong linear relationship between
LFC and soil respiration rate (i.e., potentially mineralizable C) for soils from
a long-term soil management trial in Saskatchewan is shown in Fig. 7. Such
results are not surprising because the light fraction, microbial biomass, and
soluble organic matter can all be substrates for mineralization of C and N. In
addition, the microbial biomass is the agent for mineralization. In this
regard, it is interesting to note that Campbell et al. (1999b) observed a
negative relationship between seasonal fluctuations in microbial biomass
Figure 7 Relationship between basal soil respiration and light fraction C on a long-term
crop rotation experiment in Saskatchewan. Reprinted from ‘‘Soil Science Society of America
Journal’’, Vol. 56, 1992, pp. 1799–1806. Janzen et al.: Light-fraction organic matter in soils from
long-term crop rotations, with permission from the Soil Science Society of America.
C and potentially mineralizable C. They suggested that conditions in the
field that were optimal for development of a large microbial population also
favored greater in situ mineralization. As a result, there is less substrate left
in the soil to be mineralized in a subsequent laboratory incubation.
Often, light fraction C is more closely correlated with potentially mineralizable C than light fraction N is with potentially mineralizable N (Campbell et al.,
1997, 1999b; Janzen et al., 1992). This is thought to be the case because the wide
C:N ratio of the light fraction can induce temporary N immobilization (Janzen
et al., 1992; Whalen et al., 2000). On a long-term crop rotation experiment,
Biederbeck et al. (1994) found that a multiple regression including microbial
biomass C and light fraction C accounted for 98% of the variability in potentially mineralizable C while at another similar site these two parameters
accounted for 82% of variability (Campbell et al., 1997).
The positive correlations of (i) light fraction organic matter, (ii) microbial
biomass, and (iii) soluble organic matter with potentially mineralizable C and
N in soils do not necessarily mean that these three fractions contribute
most of the mineralizable C and N in soils. For example, in most studies,
light fraction N has been shown to contribute considerably less to net
mineralization than the remaining humified heavy fraction (Boone, 1994;
Whalen et al., 2000; Yakovchenko et al., 1998). This is because the light
fraction comprises a relatively small proportion (e.g., 1–16%) of total soil
nitrogen, and only a small proportion of that (1–5%) is readily mineralizable
(Barrios et al., 1996; Imhof et al., 1996; Yakovchenko et al., 1998). Boone
(1994) calculated that the light fraction contributed between 2 and 13% of
net soil N mineralization.
The salient points here are that these labile pools are dynamic and that
it is the flux of C and nutrients flowing through them that is important rather
than the amount held in them at any one time. The size of these pools may,
however, be indicative of the flux through them. Nitrogen present in
the humic component in easily degradable forms may well have recently
passed through the labile light fraction, soluble, and microbial biomass
pools. Recently immobilized N is normally more readily mineralized
than the bulk of native soil organic N, and immobilized N becomes increasingly recalcitrant with time as it becomes more strongly incorporated into
complex humic molecules (Haynes, 1986b).
The quantities of mineralizable C and N measured are dependent on
many factors, including temperature, moisture content, aeration, sample
pretreatment (particularly air-drying), and the duration and measurement
interval of incubation (Goh and Haynes, 1986; Keeney, 1982). Due to the
above reasons, and the lack of accepted standard conditions and duration of
incubation, it is often not possible to compare absolute quantities of potentially mineralizable C and N reported between studies. However, changes in
mineralizable organic matter due to alterations in land management are
generally evident regardless of the magnitude of absolute values. As a
broad generalization, potentially mineralizable C and N account for between
0.8 and 12% (often 1.5–5.0%) of total organic C and N (Franzluebbers et al.,
1996; Gregorich et al., 1994; Hassink, 1994; Haynes, 1999; Sollins et al.,
1984; Whalen et al., 2000).
In general, mineralizable C and N show a greater responsiveness
to changes in soil management than do organic C or total N (Campbell
et al., 1997; Gregorich et al., 1997a). Disproportionately greater increases in
mineralizable than in total organic matter have been observed in response to
decreases in the amount of fallow in cereal rotations (Biederbeck et al., 1994;
Bremer et al., 1994; Campbell et al., 1999a,b), cropping with grasses
rather than cereals (Biederbeck et al., 1994; Campbell et al., 1997), conversion from conventional to zero tillage (Carter and Rennie, 1982;
Needelman et al., 1999), and long-term fertilization (Campbell et al.,
1997). A disproportionate increase in potentially mineralizable C (1.9-fold)
compared with organic C (1.2-fold) in response to trash retention rather
than preharvest burning of sugarcane is shown in Fig. 4. In general,
mineralizable N concentrations are less responsive to soil management
than those for C (Biederbeck et al., 1994; Campbell et al., 1997). This is
because temporary immobilization of N can occur concomitantly with
C mineralization and CO2 evolution.
It has been suggested that the amount of C rendered mineralizable
following air-drying is a good indicator of labile organic C (Franzluebbers
et al., 1996, 2000). Franzluebbers et al. (2000) found that this fraction
was positively correlated with soil microbial biomass C, DOC, and POC
and was a more sensitive measure than total organic C to the eVects of
conversion from conventional to zero tillage, use of forage crops in rotation,
and long-term fertilizer applications.
Seasonal fluctuations in potentially mineralizable C and N occur in
field soils (Bonde and Rosswall, 1987; Boone, 1994; Campbell et al.,
1999a,b; Franzluebbers et al., 1995) and can usually be related to rhizodeposition root material during crop growth and/or inputs of litter
and crop residues. Fluctuations generally appear less pronounced for
mineralizable N than for C (Campbell et al., 1999a,b). In wheat fields
in Swift Current, Saskatchewan, Campbell et al. (1999a) found that potentially mineralizable C increased over the growing season (due to inputs of
root material) and then declined in autumn as C mineralization in situ
increased due to favorable moisture conditions. By contrast, early in the
growing season, potentially mineralizable N either remained constant or
tended to decrease. This was interpreted as deposition of root material
with a wide C:N ratio, causing release of CO2 but concomitant temporary
N immobilization.
Under field conditions, rates of C and N mineralization are often limited
by moisture and temperature restraints. Thus, mineralization of C and
N measured in a laboratory incubation under optimum temperature and
moisture conditions represents a maximum potential rate. It gives no indication of what proportion of that mineralizable pool will be mineralized
during the growing season.
It is also important to recognize that sample pretreatment prior to incubation favors mineralization. That is, the soil is normally sieved (<2 mm) and
sometimes air-dried and then subsequently rewetted. Sieving breaks up soil
aggregates and exposes organic matter that was previously inaccessible to
microbial attack (Haynes and Beare, 1996). Drying and rewetting further
increases the size of the potentially mineralizable pool (Haynes, 1986b).
The amount of mineralizable C and N released during a laboratory incubation may well be the result of previous accumulation of organic matter and
net N immobilization under field conditions. For example, when an arable
field is converted to pasture, soil organic matter content characteristically
increases, and as a result there is net immobilization of N into the organic
fraction. Because there is more labile organic C and N in this pasture soil,
when it is removed from the field, sieved, and incubated, more C and
N are mineralized. Thus, potentially mineralizable C and N levels measured
in laboratory incubations are more likely an indication of the size of the labile
pool of C and N in the soil rather than a reflection of the potential for net
mineralization to occur under undisturbed field conditions. Despite this,
under disturbed soil conditions (i.e., arable crop production), laboratory
indices of potentially mineralizable N are routinely used in some localities
to aid in making fertilizer N recommendations (Goh and Haynes, 1986;
Keeney, 1982). Fertilizer recommendations are normally adjusted downward
depending on the size of the mineralizable pool.
The amount of potentially mineralizable N in soils can also be of environmental significance. For example, leaching losses of NO3 over winter
from arable systems often originate predominantly from mineralization of
soil organic N over the later summer–autumn period (MacDonald et al.,
1989), and denitrification losses of N2 and N2O over the same period often
originate from the same source (Addiscott and Powlson, 1992).
Measurement of potentially mineralizable C and N can be considered as a
bioassay for labile organic matter since it uses the indigenous microbial
biomass to release C and N from soil organic matter. The chemical and, to
a lesser extent, physical conditions during incubation are specific to the soil
being used. Thus, potentially mineralizable C and N can provide an integration of chemical, physical, and microbial aspects of the soil (Sparling, 1997)
and therefore can be considered a good soil quality indicator.
Total soil organic matter content can be considered as a course indicator
of soil quality. However, as discussed throughout this chapter, changes in
the content of organic C and total N occur only slowly and do not provide
an adequate indication of important changes in soil organic matter quality
that may be occurring. In order to evaluate such changes, measurement of
labile organic matter pools (that make up a relatively small proportion of
total organic matter) is required. These pools are fine indicators of soil
quality that influence soil function in specific ways. They are typically
much more sensitive to changes in soil management practice than total soil
organic matter content.
The relationships between the various pools of organic matter are shown
in Fig. 8, and the typical quantities present and their significance are summarized in Table II. Although the microbial biomass was considered only
superficially in this chapter, it is included here for completeness. In agricultural soils, the input of organic matter comes in the form of crop residues
that are returned to the soil surface, mostly during harvest, from root
turnover during crop growth and from root material left in the soil after
harvest. As this plant material decomposes, POM is formed. This partially
decomposed material can be extracted as LF by density fractionation or as
the SSF (>53 mm fraction fraction) by dispersion and sieving. The chemical
and biological natures of these two fractions diVer in that the SSF makes
up a greater proportion of soil organic C and total N than does the LF
(Table II). The LF also tends to be less aromatic, and it has a more rapid
turnover time than the SSF.
Particulate organic matter is a readily decomposable substrate for
microorganisms and a short-term resevoir of nutrients. A large part of the
Figure 8 Schematic diagram showing the relationship between various organic matter
Organic fraction
Typical quantities
Organic C ¼ 7–60 g C kg
Particulate organic
LF ¼ 2–18% of organic C,
1–16% of total N SSF ¼ 20–45%
of organic C and 13–40% of total N
Microbial biomass
1–5% of organic C and 1–6% of total N
Soluble organic
About 0.05–0.40% organic C and total N
Extractable organic
C and N
Variable amount of organic C (1–40%)
depending on the extractant
C and N
About 1–5% of organic C and total N
Nature and significance
Sum of organic material (both living and dead) present in soil excluding
living plant material. Single most important factor involved in soil
productivity. Has massive eVects on chemical, physical, and biological
properties and processes in soils.
Partially decomposed plant litter isolated by density fractionation (LF) or
sieving (SSF). Substrate and center for soil microbial activity, short-term
resevoir of nutrients, food source for earthworms, and other soil fauna
and focci for formation of water stable aggregates.
Organic material associated with cells of living soil microorganisms. Agent for
transformation and cycling of organic matter and nutrients, formation and
decay of humic material, dynamic source and sink of plant nutrients, and
an agent involved in formation and stabilization of aggregates.
Water soluble organic compounds present in soil solution, including simple
compounds of plant and microbial origin as well as humic material.
Available substrate for microbial activity, primary source of mineralizable
N, S, and P, its leaching greatly influences nutrient and organic matter status
and pH of groundwater.
Organic C and N solubilized/hydrolyzed/oxidized by various chemical reagents.
The hot water-extractable fraction is dominated by microbial carbohydrates
and is believed to be involved in aggregate stabilization. Acid-hydrolyzable
carbohydrates are also thought to be involved in aggregation.
Permanganate-oxidizable C is a non-specific labile fraction.
Quantities of organic C and N released by indigenous soil microflora during a
laboratory incubation. Values are the result of an integration of physical,
chemical, and microbiological properties of the soil. Indicator of the N fertility
of soils and their ability to supply N to crops.
Total organic
C and N
Table II
Nature, Significance, and Typical Quantities of Selected Organic Matter Fractions Present in Soils
microbial community and thus respiratory and enzyme activity in soils is
associated with the LF, and it also acts as the center for the formation of
water-stable aggregates. While fungal hyphae permeate the decomposing LF
organic matter, bacteria live in water films over its surface. The LF is also
important as a food source for earthworms and other soil fauna. Increases in
POM usually reflect greater organic matter inputs in the form of aboveand/or below-ground plant litter. As such, they can be expected to be
translated into a higher soil organic matter content in the longer term.
However, if high POM levels are the result of factors that are temporarily
reducing decomposition rate, then increases in organic matter content are
unlikely. The small, transitory nature of POM means that changes in
C supply and/or rate of decomposition induced by changes in soil management practice are generally reflected by earlier, more pronounced changes in
particulate than in total soil organic matter content.
The microbial biomass mainly consists of bacteria and fungi and makes
up about 1–5 and 2–6% of organic C and total N, respectively, in soils. The
diverse soil microbial community acts as an agent for the transformation
and cycling of organic matter and nutrients and also as a sink (during
immobilization) and source (following microbial death) of nutrients. It is
important in soil aggregation through the binding and gluing actions of
exocellular polysaccharides and enmeshing eVects of fungal hyphae. Because
of its high turnover rate, relative to total soil organic matter, the microbial
biomass can change rapidly in response to changes in soil chemical and
physical properties induced by changes in soil management. It is recognized
as a useful, sensitive early indicator of changes in organic matter status
induced by changes in soil management.
Dissolved organic matter consists of a wide range of organic components
including simple organic acids, phenols, carbohydrates, amino sugars, and
complex humic molecules. Although it represents a key labile substrate for
microbial activity, only about 10–40% of it is readily degradable. This
fraction is rich in carbohydrates, while the recalcitrant part consists mainly
of relatively resistant soluble humic substances. Dissolved organic matter
originates from leaching from plant litter, the products of decomposition
of the litter, the synthetic activity of decomposer microflora, hydrolysis of
insoluble organic polymers, and desorption of organic matter adsorbed to
soil colloids. As well as being an important microbial substrate, it is important in the terrestrial C and N cycles. Its leaching contributes to the nutrient
and organic matter status and pH of ground and surface waters. It is the
major form of N leached from forests, and it may also account for 50% or
more of N leached from agricultural soils. The fact that DOM is extracted
with water and sometimes dilute salt extractants makes comparisons of
diVerent studies diYcult since the relationship between these various forms
of DOM is, at present, unclear.
Several extractable organic matter fractions have also been proposed as
important indicators of soil quality, including those extractable with hot
water, dilute acid, and permanganate. The hot water-extractable C fraction
accounts for 1–5% of soil organic C, and about 50% of this is thought to be
present as carbohydrate C. Because it is extracted from air-dried soils, much
of it originates from desiccated microbial cells, but it also includes exocellular polysaccharides, root exudates, lysates, and humic material. The hot
water-extractable carbohydrate fraction has been used as a sensitive indicator of changes in organic matter status induced by changes in soil management and as a C fraction closely involved in aggregation and aggregate
stability. Dilute, acid-hydrolyzable C, or carbohydrate C, has also been
used as an indicator of changes in organic matter status and aggregate
stability induced by changes in soil management. Generally, it is less well
correlated with aggregate stability than the hot water-extractable fraction.
Similarly, permanganate-oxidizable C accounts for about 15–20% of
organic C and has been used as a relatively sensitive indicator of changes
in organic matter status induced by changes in soil management.
The measurement of potentially mineralizable C and N is a bioassay of
labile organic matter using the indigenous microbial community to release
labile fractions of C and N from soil organic matter. It is diYcult to compare
mineralizable C and N between studies because of diVerences in moisture
content, temperature, and length of incubation period. Mineralizable
C indicates the total metabolic activity of the heterotrophic microflora in
releasing labile organic C as CO2-C. However, mineralizable N is a measure
of the net flux of mineral N released from soil organic N. Nitrogen mineralization and immobilization occur simultaneously, and the quality of organic
residues in soils (particularly their C:N ratio) can greatly influence the
magnitude of the net release of mineral N.
Despite these complications, measurements of potentially mineralizable
C and N use the indigenous microflora to release organic C and N under
laboratory conditions, where chemical and physical conditions are largely
determined by inherent soil properties. Thus, these measurements represent
an integration of physical, chemical, and microbiological properties of
the soil. Potentially mineralizable N has been used for many years as an
indicator of the N fertility of soils and their ability to supply N for crop
growth. In addition, it is an indicator of the potential supply of soil nitrate
that can be lost to the atmosphere via denitrification or to groundwater via
nitrate leaching.
Quantities of microbial biomass, light fraction, and water-soluble C and
N in soils are commonly positively correlated with levels of mineralizable
C and N. This does not, however, necessarily mean that these pools contribute most of the C and N to the potentially mineralizable fraction. Much
of this probably originates from easily degradable humified and partially
humified organic material. It is important to recognize that pools such as
soluble, light fraction, and microbial biomass are dynamic and that it is the
rate of transfer of C and N through them that is of equal or greater
importance than their size at any one time. Nonetheless, the size of these
pools may well be indicative of the flux through them, and as a result, they
are valuable indicators of soil quality.
Taken together, labile organic matter fractions reflect the diverse, but
central, roles that organic matter have regarding soil properties and processes and thus the ability of the soil to function. This multifunctional role of soil
organic matter means that a suite of labile fractions is typically required to
provide an overview of major soil functions, including soil structural integrity, nutrient availability and turnover, and soil biological activity. The fact
that these labile fractions are transient and highly sensitive means that values
are subject to substantial seasonal variability; thus, sampling needs to be
carried out at the same time each year. Otherwise, temporal variability may
obscure important diVerences. When changes in soil management have
resulted in significant alterations in bulk density, comparisons of values for
total and labile organic matter on a volume basis are important.
In undisturbed natural ecosystems, the quantities of total and labile
organic matter present vary greatly from system to system and are a function
of soil-forming factors such as climate, topography, parent material, and
time. Similarly, under a given agricultural land management practice, the
organic matter status attained can diVer appreciably depending upon factors
such as climate, clay content, and mineralogy. As a result, threshold levels of
total and labile organic matter are not available and, indeed, probably diVer
substantially for diVerent soils and their use under diVerent management
purposes. Therefore, it is important that a comparison be made to a reference or baseline soil that has remained unaVected by agricultural management (i.e., a sample taken from under undisturbed native vegetation). In
addition, investigation of the rate of change in values for labile organic
matter fractions over time is generally more meaningful than considering
absolute values.
Labile organic matter fractions can be and are being used to monitor
changes in soil quality. However, it is important that a mechanistic view
of agricultural soil management is maintained. Because diVerent fractions
reflect diVerent key functions of organic matter, their use is extremely valuable in investigating how various agricultural management strategies influence the biological, chemical, and physical properties of soils and ultimately
the sustainability of such strategies. With a mechanistic understanding of
how soil management practices aVect soil properties and processes, new
and innovative management strategies can be devised that will improve
agricultural sustainability.
Addiscott, T. M., and Powlson, D. S. (1992). Partitioning losses of nitrogen fertilizer between
leaching and denitrification. J. Agric. Sci. Camb. 118, 101–107.
Aiken, G., and Leenheer, J. (1993). Isolation and chemical characterization of dissolved and
colloidal organic matter. Chem. Ecol. 8, 135–151.
Alef, K. (1995a). Soil respiration. In ‘‘Methods in Applied Soil Microbiology and Biochemistry’’
(K. Alef and P. Nannipieri, Eds.), pp. 214–219. Academic Press, London.
Alef, K. (1995b). Nitrogen mineralization in Soils. In ‘‘Methods of Soil Microbiology and
Biochemistry’’ (K. Alef and P. Nannipieri, Eds.), pp. 234–238. Academic Press, London.
Allison, F. E. (1973). ‘‘Soil Organic Matter and its Role in Crop Production’’. Elsevier,
Alvarez, C. R., Alvarez, R., Grigera, M. S., and Lavado, R. S. (1998). Associations between
organic matter fractions and the active soil microbial biomass. Soil Biol. Biochem. 30,
Anderson, J. P. E. (1982). Soil respiration. In ‘‘Methods of Soil Analysis, Part 2. Chemical and
Microbiological Properties’’ (A. L. Page, R. H. Miller, and D. R. Keeney, Eds.),
pp. 831–871. American Society of Agronomy, Madison, WI.
Angers, D. A., Bissonnette, N., Légère, A., and Samson, N. (1993b). Microbial and biochemical
changes induced by rotation and tillage in a soil under barley production. Can. J. Soil Sci.
73, 39–50.
Angers, D. A., Edwards, L. M., Sanderson, J. B., and Bissonnette, N. (1999). Soil organic
matter quality and aggregate stability under eight potato cropping sequences in a fine sandy
loam of Prince Edward Island. Can. J. Soil Sci. 79, 411–417.
Angers, D. A., and Mehuys, G. R. (1989). EVects of cropping on carbohydrate content and
water-stable aggregation of a clay soil. Can. J. Soil Sci. 69, 373–380.
Angers, D. A., Samson, N., and Légère, A. (1993a). Early changes in water-stable aggregation induced by rotation and tillage in a soil under barley production. Can. J. Soil Sci. 73,
Appel, T. (1998). Non-biomass soil organic N—The substrate for N mineralization flushes
following soil drying-rewetting and for organic N rendered CaCl2-extractable upon drying.
Soil Biol. Biochem. 30, 1445–1456.
Appel, T., and Mengel, K. (1990). Importance of organic nitrogen fractions in sandy soils,
obtained by electro-ultrafiltration of CaCl2 extraction, for nitrogen mineralization and
nitrogen uptake of rape. Biol. Fertil. Soils 10, 97–101.
Appel, T., and Mengel, K. (1992). Nitrogen uptake of cereals grown on sandy soils as related to
nitrogen fertilizer application and soil nitrogen fractions obtained by electroultrafiltration
(EUF) and CaCl2 extraction. Eur. J. Agron 1, 1–9.
Appel, T., and Mengel, K. (1993). Nitrogen fractions in sandy soils in relation to plant nitrogen
uptake and organic matter incorporation. Soil Biol. Biochem. 25, 685–691.
Baldock, J. A., and Nelson, P. N. (2000). Soil organic matter. In ‘‘Handbook of Soil Science’’
(M. E. Sumner, Ed.), pp. B25–B84. CRC Press, Boca Raton, FL.
Baldock, J. A., Oades, J. M., Waters, A. G., Peng, X., Vassallo, A. M., and Wilson, M. A.
(1992). Aspects of the chemical structure of soil organic materials as revealed by solid-state
C NMR spectroscopy. Biogeochemistry 16, 1–42.
Ball, B. C., Cheshire, M. V., Robertson, E. A. G., and Hunter, E. A. (1996). Carbohydrate
composition in relation to structural stability, compactibility and plasticity of two soils in a
long-term experiment. Soil Tillage Res. 39, 143–160.
Barrios, E., Buresh, R. J., and Sprent, J. I. (1996). Nitrogen mineralization in density fractions
of soil organic matter from maize and legume cropping systems. Soil Biol. Biochem. 28,
Bergmann, C., Fischer, T., and Hüttl, R. F. (1999). Significance of litter- and humus-layer
quality for rates and forms of N cycling in moder- to rawhumus—moder profiles under
Scots pine (Pinus sylvestris L.). Plant Soil 213, 11–21.
Besnard, E., Chenu, C., Balesdent, J., Puget, P., and Arrouays, D. (1996). Fate of particulate
organic matter in soil aggregates during cultivation. Eur. J. Soil Sci. 47, 495–503.
Bhogal, A., Murphy, D. V., Fortune, S., Shepherd, M. A., Hatch, D. J., Jarvis, S. C., Gaunt,
J. L., and Goulding, K. W. T. (2000). Distribution of nitrogen pools in the soil profile of
undisturbed and reseeded grasslands. Biol. Fertil. Soil 30, 356–362.
Biederbeck, V. O., Janzen, H. H., Campbell, C. A., and Zentner, R. P. (1994). Labile soil
organic matter as influenced by cropping practices in an arid environment. Soil Biol.
Biochem. 26, 1647–1656.
Blair, G. T., Chapman, L., Whitbread, A. M., Ball-Coelho, B., Larsen, P., and Tiessen, H.
(1998). Soil carbon changes resulting from sugarcane trash management at two
locations in Queensland, Australia, and in North-East Brazil. Aust. J. Soil Res. 36, 873–881.
Blair, G. J., Lefroy, R. D. B., and Lisle, L. (1995a). Soil carbon fractions based on their degree
of oxidation, and the development of a carbon management index for agricultural systems.
Aust. J. Soil Res. 46, 1459–1466.
Blair, J. M., Parmelee, R. W., and Lavelle, P. (1995b). Influences of earthworms on biogeochemistry. In ‘‘Earthworm Ecology and Biogeography in North America’’ (P. F. Hendrix,
Ed.), pp. 127–158. Lewis, Boca Raton, FL.
Blair, N. (2000). Impact of cultivation and sugar-cane green trash management on carbon
fractions and aggregate stability for a Chromic Luvisol in Queensland, Australia. Soil
Till. Res. 55, 183–191.
Boissier, J. M., and Fontvieille, D. (1993). Biodegradable dissolved organic carbon in seepage
waters from two forest soils. Soil Biol. Biochem. 25, 1257–1261.
Bolan, N. S., Baskaran, S., and Thiagarajan, S. (1996). An evaluation of the methods of
measurement of dissolved organic carbon in soils, manures, sludges, and stream water.
Commun. Soil. Sci. Plant Anal. 27, 2723–2737.
Bolinder, M. A., Angers, D. A., Gregorich, E. G., and Carter, M. R. (1999). The response of soil
quality indicators to conservation management. Can. J. Soil Sci. 79, 37–45.
Bonde, T. A., and Rosswall, T. (1987). Seasonal variation of potentially mineralizable nitrogen
in four cropping systems. Soil Sci. Soc. Am. J. 51, 1508–1514.
Boone, R. D. (1994). Light-fraction soil organic matter: Origin and contribution to net nitrogen
mineralization. Soil Biol. Biochem. 26, 1459–1468.
Bowman, R. A., Vigil, M. F., Nielsen, D. C., and Anderson, R. L. (1999). Soil organic matter
changes in intensively cropped dryland systems. Soil Sci. Soc. Am. J. 63, 186–191.
Boyer, J. N., and GroVman, P. M. (1996). Bioavailability of water extractable organic carbon
fractions in forest and agricultural soil profiles. Soil Biol. Biochem. 28, 783–790.
Bremer, E., Janzen, H. H., and Johnston, A. M. (1994). Sensitivity of total light fraction and
mineralizable organic matter to management practices in a Lethbridge soil. Can. J. Soil Sci.
74, 131–138.
Brooks, P. D., McKnight, D. M., and Bencala, K. E. (1999). The relationship between soil
heterotrophic activity, soil dissolved organic carbon (DOC), leachate, and catchment-scale
DOC export in headwater catchments. Water Resour. Res. 35, 1895–1902.
Bundy, L. G., and Meisinger, J. J. (1994). Nitrogen availability indices. In ‘‘Methods of Soil
Analysis Part 2’’ (R. W. Weaver, S. Angle, P. Bottemly, D. Bezdicek, S. Smith, A.
Tabatabai, and A. Wollum, Eds.), pp. 951–984. Soil Science Society of America, Madison,
Burford, J. R., and Bremner, J. M. (1975). Relationships between denitrification capacities of
soils and total water-soluble and readily decomposable soil organic matter. Soil Biol.
Biochem. 7, 389–394.
Cadisch, G., and Giller, K. E. (1997). ‘‘Driven by Nature: Plant Litter Quality and Decomposition’’. CAB, Wallingford, UK.
Cambardella, C. A., and Elliott, E. T. (1992). Particulate soil organic matter changes across a
grassland cultivation sequence. Soil Sci. Soc. Am. J. 56, 777–783.
Campbell, C. A. (1978). Soil organic carbon nitrogen and fertility. In ‘‘Soil Organic Matter’’
(M. Schnitzer and S. U. Khan, Eds.), pp. 173–271. Elsevier, New York.
Campbell, C. A., Biederbeck, V. O., Zentner, R. P., and Lafond, G. P. (1991). EVects of crop
rotations and cultural practices on soil organic matter, microbial biomass and respiration in
a thin Black Chernozem. Can. J. Soil Sci. 71, 363–376.
Campbell, C. A., Janzen, H. H., and Juma, N. G. (1997). Case studies of soil quality in the
Canadian prairies: Long-term field experiments. In ‘‘Soil Quality for Crop Production and
Ecosystem Health’’ (E. G. Gregorich and M. R. Cater, Eds.), pp. 351–397. Elsevier,
Campbell, C. A., Biederbeck, V. O., Wen, G., Zentner, R. P., Schoenau, J., and Hahn, D.
(1999a). Seasonal trends in selected soil biochemical attributes: EVects of crop rotation in
the semiarid prairie. Can. J. Soil Sci. 79, 73–84.
Campbell, C. A., Lafond, G. P., Biederbeck, O., Wen, G., Schoenau, J., and Hahn, D. (1999b).
Seasonal trends in soil biochemical attributes: EVects of crop management on a Black
Chernozem. Can. J. Soil Sci. 79, 85–97.
Carter, M. R., and Rennie, D. A. (1982). Changes in soil quality under zero tillage farming
systems: Distribution of microbial biomass and mineralizable C and N potentials. Can.
J. Soil Sci. 62, 587–597.
Carter, M. R., Angers, D. A., and Kunelius, H. T. (1994). Soil structural form and stability,
and organic matter under cool-season perennial grasses. Soil Sci. Soc. Am. J. 58,
Carter, M. R., Gregorich, E. G., Anderson, D. W., Doran, J. W., Janzen, H. H., and Pierce, F. J.
(1997). Concepts of soil quality and their significance. In ‘‘Soil Quality for Crop Production
and Ecosystem Health’’ (E. G. Gregorich and M. R. Carter, Eds.), pp. 1–19. Elsevier,
Carter, M. R., Gregorich, E. G., Angers, D. A., Beare, M. H., Sparling, G. P., Wardle, D. A.,
and Voroney, R. P. (1999). Interpretation of microbial biomass measurements for soil
quality assessment in humid temperature regions. Can. J. Soil Sci. 79, 507–520.
Carter, M. R., Gregorich, E. G., Angers, D. A., Donald, R. G., and Bolinder, M. A. (1998).
Organic C and N storage, and organic C fractions, in adjacent cultivated and forested soils
of eastern Canada. Soil Till. Res. 47, 253–261.
Casals, P., Romanya, J., Cortina, J., Fons, J., Bode, M., and Vallejo, V. R. (1995). Nitrogen
supply rate in Scots pine (Pinus sylvestris L.) forests of contrasting slope aspect. Plant Soil
169, 67–73.
Chan, K. Y., and Heenan, D. P. (1999). Microbial-induced soil aggregate stability under
diVerent crop rotations. Biol. Fertil. Soils 30, 29–32.
Chantigny, M. H., Angers, D. A., Prévost, D., Simard, R. R., and Chalifour, F. P. (1999).
Dynamics of soluble organic C and C mineralization in cultivated soils with varying
N fertilization. Soil Biol. Biochem. 31, 543–550.
Christensen, B. T., and Johnston, A. E. (1997). Soil organic matter and soil quality—Lessons
learned from long-term experiments at Askov and Rothamsted. In ‘‘Soil Quality for Crop
Production and Ecosystem Health’’ (E. G. Gregorich and M. R. Carter, Eds.), pp. 399–430.
Elsevier, Amsterdam.
Conteh, A., Blair, G. T., Lefroy, R. D. B., and Whitbread, A. M. (1999). Labile organic carbon
determined by permanganate oxidation and its relationships to other measurements of soil
organic carbon. Humic Subst. Environ. J. 1, 3–15.
Conti, M. E., Palma, R. M., Arrigo, N., and Giardino, E. (1992). Seasonal variations of the light
organic fractions in soils under diVerent agricultural management systems. Commun. Soil.
Sci. Plant Anal. 23, 1693–1704.
Cook, B. C., and Allan, D. L. (1992a). Dissolved organic carbon in old field soils: Compositional
changes during the biodegradation of soil organic matter. Soil Biol. Biochem. 24, 595–600.
Cook, B. D., and Allan, D. L. (1992b). Dissolved organic carbon in old field soils: Total
amounts as a measure of available resources for soil mineralization. Soil Biol. Biochem.
24, 585–594.
Cortina, J., and Romana, J. (1992). Nutrients transfer from forest floor to mineral soil in a
Pinus radiata D.Don stand. In ‘‘Responses of Forest Ecosystems to Environmental
Changes’’ (A. Teller, P. Mathy, and J. N. R. JeVers, Eds.), pp. 767–768. Elsevier, London.
Curry, J. P. (1998). Factors aVecting earthworm abundance in Soils. In ‘‘Earthworm Ecology’’
(C. A. Edwards, Ed.), pp. 37–64. St. Lucie Press, Boca Raton, FL.
Dalal, R. C. (1998). Soil microbial biomass—What do the numbers really mean? Aust. J. Exp.
Agric. 38, 649–665.
Dalal, R. C., and Chan, K. Y. (2001). Soil organic matter in rainfed cropping systems of the
Australian cereal belt. Aust. J. Soil. Res. 39, 435–464.
Dalal, R. C., and Mayer, R. J. (1987). Long-term infertility of soils under continuous cultivation
and cereal cropping in Southern Queensland. VI. Loss of total nitrogen from diVerent
particle-size and density fractions. Aust. J. Soil Res. 25, 83–93.
Davidson, E. A., Galloway, L. F., and Strand, M. K. (1987). Assessing available carbon:
Comparison of techniques across selected forest soils. Commun. Soil Sci. Plant Anal. 18,
Debrosz, K., Vognsen, L., and Labouriau, R. (2002). Carbohydrates in hot water extracts of soil
aggregates as influenced by long-term management. Commun. Soil Sci. Plant Anal. 33,
Degens, B. P. (1997). Macro-aggregation of soils by biological bonding and binding mechanisms
and the factors aVecting these: A review. Aust. J. Soil Res. 35, 431–460.
Dijkstra, F. A., Geibe, C., Holmstrom, S., Lundström, U. S., and van Breemen, N. (2001). The
eVect of organic acids on base cation leaching from the forest floor under six North
American tree species. Eur. J. Soil Sci. 52, 205–214.
Dominy, C. S., and Haynes, R. J. (2002). Influence of agricultural land management on organic
matter content, microbial activity and aggregate stability in the profiles of two Oxisols.
Biol. Fertil. Soils 36, 298–305.
Doran, J. W., and Parkin, T. B. (1994). Defining and assessing soil quality. In ‘‘Defining Soil
Quality for a Sustainable Environment’’ (J. W. Doran, D. C. Coleman, D. F. Bezedick, and
B. A. Stewart, Eds.), pp. 3–21. Soil Science Society of America, Madison, WI.
Doran, J. W., Elliott, E. T., and Paustian, K. (1998). Soil microbial activity, nitrogen cycling,
and long-term changes in organic carbon pools as related to fallow tillage management.
Soil Till Res. 49, 3–18.
Dormaar, J. F. (1972). Chemical properties of organic matter extracted from a number of Ah
horizons by a number of methods. J. Soil. Sci. 52, 67–77.
Dormaar, J. F., Johnston, A., and Smoliak, S. (1984). Seasonal changes in carbon content, and
dehydrogenase, phosphatase, and urease activities in mixed prairie and fescue grassland Ah
horizons. J. Range Manage 37, 31–35.
Drinkwater, L. E., Cambardella, C. A., Reeder, J. D., and Rice, C. W. (1996). Potentially
mineralizable nitrogen as an indicator of biologically active soil nitrogen. In ‘‘Methods of
Assessing Soil Quality’’ (J. W. Doran and A. J. Jones, Eds.), pp. 217–229. Soil Science
Society of America, Madison, WI.
Edwards, C. A., and Bohlen, P. J. (1996). ‘‘Biology and Ecology of Earthworms’’. Chapman &
Hall, London.
Fenton, T. E., Brown, J. R., and Mausbach, M. J. (1999). EVects of long-term cropping on
organic matter content of soils: Implications for soil quality. In ‘‘Soil Quality and Soil
Erosion’’ (R. Lal, Ed.), pp. 95–124. CRC Press, Boca Raton, FL.
Franzluebbers, A. J., and Arshad, M. A. (1997). Particulate organic carbon content and
potential mineralization as aVected by tillage and texture. Soil Sci. Soc. Am. J. 61,
Franzluebbers, A. J., Haney, R. L., Hons, F. M., and Zuberer, D. A. (1996). Determination of
microbial biomass and nitrogen mineralization following rewetting of dried soil. Soil Sci.
Soc. Am. J. 60, 1133–1139.
Franzluebbers, A. J., Hons, F. M., and Zuberer, D. A. (1994). Seasonal changes in soil microbial
biomass and mineralizable C and N in wheat management systems. Soil Biol. Biochem. 26,
Franzluebbers, A. J., Hons, F. M., and Zuberer, D. A. (1995). Tillage and crop eVects on
seasonal soil carbon and nitrogen dynamics. Soil. Sci. Soc. Am. J. 59, 1618–1624.
Franzluebbers, A. J., Stuedemann, J. A., Schomberg, H. H., and Wilkinson, S. R. (2000). Soil
organic C and N pools under long-term pasture management in the Southern Piedmont
USA. Soil Biol. Biochem. 32, 469–478.
Ghani, A., Dexter, M., Sarathchandra, U., Perrott, K. W., and Singleton, P. (2000). Assessment
of extractable hot-water carbon as an indicator of soil quality on soils under long-term
pastoral, cropping, market gardening and native vegetation. Proc. Aust. NZ. Joint Soils
Conf. 2000 2, 119–120.
Goh, K. M., and Haynes, R. J. (1986). Nitrogen and agronomic practice. In ‘‘Mineral
Nitrogen in the Plant-Soil System’’ (R. J. Haynes, Ed.), pp. 379–468. Academic Press,
Orlando, FL.
Golchin, A., Baldock, J. A., and Oades, J. M. (1998). A model linking organic matter decomposition, chemistry, and aggregate dynamics. In ‘‘Soil Processes and the Carbon
Cycle’’ (R. Lal, J. M. Kimble, R. F. Follett, and B. A. Stewart, Eds.), pp. 245–266. CRC
Press, Boca Raton, FL.
Golchin, A., Oades, J. M., Skjemstad, J. O., and Clark, P. (1994). Soil structure and carbon
cycling. Aust. J. Soil. Res. 32, 1043–1068.
Graham, M. H., Haynes, R. J., and Meyer, J. H. (2002). Soil organic matter content and quality:
EVects of fertilizer applications, burning and trash retention on a long-term sugarcane
experiment in South Africa. Soil Biol. Biochem. 34, 93–102.
Gregorich, E. G., and Ellert, B. H. (1993). Light fraction and macro-organic matter in mineral
Soils. In ‘‘Soil Sampling and Methods of Analysis’’ (M. R. Carter, Ed.), pp. 397–407.
Canadian Society of Soil Science, Lewis, Boca Raton, FL.
Gregorich, E. G., and Janzen, H. H. (1996). Storage of soil carbon in the light fraction and
macro-organic matter. In ‘‘Structure and Soil Organic Matter Storage in Agricultural Soils’’
(M. R. Carter and B. A. Steward, Eds.), pp. 167–190. CRC Press, Boca Raton, FL.
Gregorich, E. G., Carter, M. R., Angers, D. A., Monreal, C. M., and Ellert, B. H. (1994).
Towards a minimum data set to assess soil organic matter quality in agricultural soils. Can.
J. Soil Sci. 74, 367–385.
Gregorich, E. G., Carter, M. R., Doran, J. W., Pankhurst, C. E., and Dwyer, L. M. (1997a).
Biological attributes of soil quality. In ‘‘Soil Quality for Crop Production and
Ecosystem Health’’ (E. G. Gregorich and M. R. Carter, Eds.), pp. 81–113. Elsevier,
Gregorich, E. G., Drury, C. F., Ellert, B. H., and Liang, B. C. (1997b). Fertilization eVects on
physically protected light fraction organic matter. Soil Sci. Soc. Am. J. 61, 482–484.
Gregorich, E. G., Ellert, B. H., and Monreal, C. M. (1995). Turnover of soil organic matter and
storage of corn residue carbon estimated from natural 13C abundance. Can. J. Soil Sci. 75,
Gregorich, E. G., Liang, B. C., Drury, C. F., Mackenzie, A. F., and McGill, W. B. (2000).
Elucidation of the source and turnover of water soluble and microbial biomass carbon in
agricultural soils. Soil Biol. Biochem. 32, 581–587.
Gregorich, E. G., Monreal, C. M., Schnitzer, M., and Schulten, H. R. (1996). Transformation of
plant residues into soil organic matter: Chemical characterization of plant tissue, isolated
soil fractions, and whole soils. Soil Sci. 161, 680–693.
Groot, J. J. R., and Houba, V. J. G. (1995). A comparison of diVerent indices for nitrogen
mineralization. Biol. Fertil. Soils 19, 1–9.
Guggenberger, G., and Zech, W. (1994). Dissolved organic carbon in forest floor leachates:
Simple degradation products or humic substances? Sci. Total Env. 152, 37–47.
Haney, R. L., Franzluebbers, A. J., Hons, F. M., and Zuberer, D. A. (1999). Soil C extracted
with water or K2SO4: pH eVect on determination of microbial biomass. Can. J. Soil Sci. 79,
Hassink, J. (1994). EVects of soil texture and grassland management on soil organic C and N and
rates of C and N mineralization. Soil Biol. Biochem. 26, 1221–1231.
Hassink, J. (1995). Density fractions of soil macro-organic matter and microbial biomass as
predictors of C and N mineralization. Soil Biol. Biochem. 27, 1099–1108.
Haynes, R. J. (1986a). Origin, distribution, and cycling of nitrogen in terrestrial ecosystems.
In ‘‘Mineral Nitrogen in the Plant-Soil System’’ (R. J. Haynes, Ed.), pp. 1–51. Academic
Press, Orlando, FL.
Haynes, R. J. (1986b). The decomposition process: mineralization, immobilization, humus
formation, and degradation. In ‘‘Mineral Nitrogen in the Plant-Soil System’’ (R. J. Haynes,
Ed.), pp. 52–126. Academic Press, Orlando, FL.
Haynes, R. J. (1999). Labile organic matter fractions and aggregate stability under short-term,
grass-based leys. Soil Biol. Biochem. 31, 1821–1830.
Haynes, R. J. (2000). Labile organic matter as an indicator of organic matter quality in arable
and pastoral soils in New Zealand. Soil Biol. Biochem. 32, 211–219.
Haynes, R. J., and Beare, M. H. (1996). Aggregation and organic matter storage in mesothermal, humid Soils. In ‘‘Structure and Organic Matter Storage in Agricultural Soils’’
(M. R. Carter and B. A. Stewart, Eds.), pp. 213–262. CRC Press, Boca Raton, FL.
Haynes, R. J., and Beare, M. H. (1997). Influence of six crop species on aggregate stability and
some labile organic matter fractions. Soil Biol. Biochem. 29, 1647–1653.
Haynes, R. J., and Francis, G. S. (1993). Changes in microbial biomass C, soil carbohydrate
composition and aggregate stability induced by growth of selected crop and forage species
under field conditions. J. Soil Sci. 44, 665–675.
Haynes, R. J., and Mokolobate, M. S. (2001). Amelioration of AI toxicity and P deficiency in
acid soils by additions of organic residues: A critical review of the phenomenon and the
mechanisms involved. Nutr. Cycling Agroecosys. 59, 47–63.
Haynes, R. J., and Swift, R. S. (1990). Stability of soil aggregates in relation to organic
constituents and soil water content. J. Soil Sci. 41, 73–83.
Haynes, R. J., and Swift, R. S. (1991). Concentrations of extractable Cu, Zn, Fe and Mn in a
group of soils as influenced by air- and oven-drying and rewetting. Geoderma 49, 319–333.
Haynes, R. J., and Williams, P. H. (1999). Influence of stock camping behaviour on the soil
microbiological and biochemical properties of grazed pastoral soils. Biol. Fertil. Soils 28,
Haynes, R. J., Swift, R. S., and Stephen, R. C. (1991). Influence of mixed cropping rotations
(pasture-arable) on organic matter content, water stable aggregation and clod porosity in a
group of soils. Soil Tillage Res. 19, 77–87.
Herbert, B. E., and Bertsch, P. M. (1995). Characterization of dissolved and colloidal organic
matter in soil solutions: A review. In ‘‘Carbon Forms and Functions in Forest Soils’’ (J. M.
Kelley and W. W. McFee, Eds.), pp. 63–88. Soil Science Society of America, Madison, WI.
Huang, Y., Eglington, G., van der Hage, E. R. E., Boon, J. J., Bol, R., and Ineson, P. (1998).
Dissolved organic matter and its parent organic matter in grass upland soil horizons
studied by analytical pyrolysis techniques. Eur. J. Soil Sci. 49, 1–15.
Hussain, I., Olson, K. R., and Ebelhar, S. A. (1999). Long-term tillage eVects on soil chemical
properties and organic matter fractions. Soil Sci. Soc. Am. J. 63, 1335–1341.
Imhof, H., Ehaliotis, C., Giller, K., Miranda, C., Pereira, J. M., Urquiaga, S., Boddey, B., and
Cadisch, G. (1996). Light fraction mineralization potentials of humid tropical Soils.
In ‘‘Progress in Nitrogen Cycling Studies’’ (O. Van Cleemput, G. Hofman, and A.
Vermoesen, Eds.), pp. 63–67. Kluwer, Dordrecht.
Jandl, R., and Sollins, P. (1997). Water extractable carbon in relation to belowground carbon
cycle. Biol. Fertil. Soils 25, 196–201.
Janzen, H. H. (1995). The role of long-term sites in agroecological research: A case study. Can.
J. Soil Sci. 75, 123–133.
Janzen, H. H., Campbell, C. A., Brandt, S. A., Lafond, G. P., and Townley-Smith, L. (1992).
Light-fraction organic matter in soils from long-term crop rotations. Soil Sci. Soc. Am. J.
56, 1799–1806.
Janzen, H. H., Campbell, C. A., Ellert, B. H., and Bremer, E. (1997). Soil organic matter dynamics
and their relationship to soil quality. In ‘‘Soil Quality for Crop Production and Ecosystem
Health’’ (E. G. Gregorich and M. R. Carter, Eds.), pp. 277–291. Elsevier, Amsterdam.
Janzen, H. H., Campbell, C. A., Gregorich, E. G., and Ellert, B. H. (1998b). Soil
carbon dynamics in Canadian agroecosystems. In ‘‘Soil Processes and the Carbon
Cycle’’ (R. Lal, J. M. Kimble, R. F. Follett, and B. A. Stewart, Eds.), pp. 57–80. CRC
Press, Boca Raton, FL.
Janzen, H. H., Campbell, C. A., Izaurralde, R. C., Ellert, B. H., Juma, N., McGill, W. B., and
Zentner, R. P. (1998a). Management eVects on soil C storage on the Canadian prairies. Soil
Till. Res. 47, 181–195.
Jardine, P. M., Weber, N. L., and McCarthy, J. F. (1989). Mechanisms of dissolved organic
carbon adsorption on soil. Soil Sci. Soc. Am. J. 53, 1378–1385.
Jenkinson, D. S., Adams, D. E., and Wild, A. (1991). Model estimates of CO2 emissions from
soil in response to global warming. Nature 351, 304–306.
Jensen, L. S., Mueller, T., Magid, J., and Nielsen, N. E. (1997). Temporal variation of C and
N mineralization, microbial biomass and extractable organic pools in soil after oilseed rape
straw incorporation in the field. Soil Biol. Biochem. 29, 1043–1055.
Johnston, A. E. (1986). Soil organic matter eVects on soils and crops. Soil Use Manage. 2,
Jones, D. L., and Brassington, D. S. (1998). Sorption of organic acids in acid soils and its
implications in the rhizosphere. Eur. J. Soil Sci. 49, 447–455.
Kaiser, K. (2001). Dissolved organic phosphorus and sulphur as influenced by sorptive interactions with mineral subsoil horizons. Eur. J. Soil Sci. 52, 489–493.
Kaiser, K., and Zech, W. (1997). Competitive sorption of dissolved organic matter fractions to
soils and related mineral phases. Soil Sci. Soc. Am. J. 61, 64–69.
Kaiser, K., Guggenberger, G., and Zech, W. (1996). Sorption of DOM and DOM fractions to
forest soils. Geoderma 74, 281–303.
Kanazawa, S., and Filip, Z. (1986). Distribution of microorganisms, total biomass, and enzyme
activities in diVerent particles of brown soil. Microb. Ecol. 12, 205–215.
Karlen, D. L., Mausbach, M. J., Doran, J. W., Cline, R. G., Harris, R. F., and Schuman, G. E.
(1997). Soil quality: A concept, definition, and framework for evaluation. Soil Sci. Soc. Am.
J. 61, 4–10.
Keeney, D. R. (1982). Nitrogen-availability indices. In ‘‘Methods of Soil Analysis, Part 2.
Chemical and Microbiological Properties’’ (A. L. Page, R. H. Miller, and D. R. Keeney,
Eds.), pp. 711–733. American Society of Agronomy, Madison, WI.
Keeney, D. R., and Bremner, J. M. (1966). Comparison and evaluation of laboratory methods
of obtaining an index of soil nitrogen availability. Agron. J. 58, 498–503.
Keeney, D. R., and Nelson, D. W. (1982). Nitrogen-inorganic forms. In ‘‘Methods of Soil
Analysis, Part 2. Chemical and Microbiological Properties’’ (A. L. Page, R. H. Miller, and
D. R. Keeney, Eds.), pp. 643–698. American Society of Agronomy, Madison, WI.
Körschens, M., Schulz, E., and Behm, R. (1990). Der Statische Versuch Lauchstädt.
In ‘‘Dauerfeldversuche’’ (M. Korschens, Ed.), pp. 7–23. Akademie der Landwirtschaftswissen-Schaften, Berlin.
Ladd, J. N., Amato, M., and Parsons, J. W. (1977). Studies of nitrogen immobilization and
mineralization in calcareous soils. 111. Concentration and distribution of nitrogen derived
from soil biomass. In ‘‘Soil Organic Matter Studies’’, Vol. 1, pp. 301–311. International
Atomic Energy Agency, Vienna.
Lavelle, P. (1988). Earthworm activities and the soil system. Biol. Fertil. Soils 6, 237–251.
Lavelle, P., Pashanasi, B., Charpentier, F., Gilot, C., Rossi, J.-P., Derouard, L., Andre, J.,
Ponge, J.-F., and Bernier, N. (1998). Large-scale eVects of earthworms on soil organic
matter and nutrient dynamics. In ‘‘Earthworm Ecology’’ (C. A. Edwards, Ed.), pp. 103–121.
St. Lucie Press, Boca Raton, FL.
Lee, K. E. (1985). ‘‘Earthworms: Their Ecology and Relationships with Soils and Land Use’’.
Academic Press, New York, NY.
Leenheer, J. A. (1981). Comprehensive approach to preparative isolation and fractionation of
dissolved organic carbon from natural waters and waste waters. Environ. Sci. Technol. 15,
Lefroy, R. D., Blair, G. T., and Strong, W. M. (1993). Changes in soil organic matter with
cropping as measured by organic carbon fractions and 13C natural isotope abundance.
Plant Soil 155, 399–402.
Leinweber, P., Schulten, H. R., and Körschens, M. (1995). Hot water extracted organic matter:
Chemical composition and temporal variations in a long-term field experiment. Biol. Fertil.
Soils 20, 17–23.
Lundquist, E. J., Jackson, L. E., Scow, K. M., and Hsu, C. (1999). Changes in microbial
biomass and community composition, and soil carbon and nitrogen pools after incorporation of rye into three California agriculture soils. Soil Biol. Biochem. 31, 221–236.
MacDonald, A. J., Powlson, D. S., Poulton, P. R., and Jenkinson, D. S. (1989). Unused
fertilizer nitrogen in arable soils—Its contribution to nitrate leaching. J. Sci. Food. Agric.
46, 407–419.
McDowell, W. H., and Likens, G. E. (1988). Origin, composition, and flux of dissolved organic
carbon in the Hubbard Brook Valley. Ecol. Mono. 58, 177–195.
McGill, W. B., Cannon, K. R., Robertson, J. A., and Cook, F. D. (1986). Dynamics of soil
microbial biomass and water-soluble organic C in Breton L after 50 years of cropping to
rotations. Can. J. Soil Sci. 66, 1–19.
McNeill, A. M., Sparling, G. P., Murphy, D. V., Braunberger, P., and Fillery, I. R. P. (1998).
Changes in extractable and microbial C, N, and P in a western Australian wheatbelt soil
following simulated summer rainfall. Aust. J. Soil Res. 36, 841–854.
Moore, T. R. (1997). Dissolved organic carbon: sources, sinks, and fluxes and role in the soil
carbon cycle. In ‘‘Soil Processes and the Carbon Cycle’’ (R. Lal, J. M. Kimble, R. F.
Follett, and B. A. Stewart, Eds.), pp. 281–292. CRC Press, Boca Raton, FL.
Moore, T. R., de Souza, W., and Koprivnjak, J. F. (1992). Controls on the sorption of dissolved
organic carbon by soils. Soil Sci. 154, 120–129.
Murphy, D. V., Fortune, S., Wakefield, J. A., Stockdale, E. A., Poulton, P. R., Webster, C. P.,
Wilmer, W. S., Goulding, K. W. T., and Gaunt, J. L. (1999). Assessing the importance
of soluble organic nitrogen in agricultural Soils. In ‘‘Managing Risks of Nitrates to
Humans and the Environment’’ (J. R. Wilson, Ed.), pp. 65–86. Royal Society of Chemistry,
Cambridge, UK.
Murphy, D. V., Macdonald, A. J., Stockdale, E. A., Goulding, K. W. T., Fortune, S., Gaunt,
J. L., Poulton, P. R., Wakefield, J. A., Webster, C. P., and Wilmer, W. S. (2000). Soluble
organic nitrogen in agricultural soils. Biol. Fertil. Soils 30, 374–387.
Needelman, B. A., Wander, M. M., Bollero, G. A., Boast, C. W., Sims, G. K., and Bullock,
D. G. (1999). Interaction of tillage and soil texture: Biologically active soil organic matter in
Illinois. Soil Sci. Soc. Am. J. 63, 1326–1334.
Nunan, N., Morgan, M. A., Brennan, D., and Herlihy, M. (2001). Organic matter extracted
with 0.01 M CaCl2 or with 0.01 M NaHCO3 as indices of N mineralization and microbial
biomass. Biol. Fertil. Soils 34, 433–440.
Nyborg, M., Malhi, S. S., Solberg, E. D., and Izaurralde, R. C. (1999). Carbon storage and light
fraction C in a grassland dark grey Chernozem soil as influenced by N and S fertilization.
Can. J. Soil Sci. 79, 317–320.
Oades, J. M. (1984). Soil organic matter and structural stability mechanisms and implications
for management. Plant Soil 76, 319–337.
Öhlinger, R. (1996a). Soil respiration by titration. In ‘‘Methods in Soil Biology’’ (F. Schinner, R.
Öhlinger, E. Kandeler, and R. Margesin, Eds.), pp. 95–98. Springer-Verlag, Berlin.
Öhlinger, R. (1996b). Soil respiration by infrared gas analysis. In ‘‘Methods in Soil Biology’’
(F. Schinner, R. Öhlinger, E. Kandeler, and R. Margesin, Eds.), pp. 98–102. SpringerVerlag, Berlin.
Paustian, K., Collins, H. P., and Paul, E. A. (1997). Management controls on soil carbon.
In ‘‘Soil Organic Matter in Temperate Agroecosystems’’ (E. A. Paul, K. Paustian, E. T.
Elliot, and C. V. Cole, Eds.), pp. 15–49. CRC Press, Boca Raton, FL.
Puget, P., Angers, D. A., and Chenu, C. (1999). Nature of carbohydrates associated with waterstable aggregates of two cultivated soils. Soil Biol. Biochem. 31, 55–63.
Puget, P., Chenu, C., and Balesdent, J. (1995). Total and young organic matter distribution in
aggregates of silty cultivated soils. Eur. J. Soil. Sci. 46, 449–459.
Qualls, R. G., and Haines, B. L. (1991). Geochemistry of dissolved organic nutrients in water
percolating through a forest ecosystem. Soil Sci. Soc. Am. J. 55, 1112–1123.
Qualls, R. G., and Haines, B. L. (1992). Biodegradability of dissolved organic matter in forest
throughfall, soil solution, and stream water. Soil Sci. Soc. Am. J. 56, 578–586.
Qualls, R. G., Haines, B. L., and Swank, W. T. (1991). Fluxes of dissolved organic nutrients and
humic substances in a deciduous forest. Ecology 72, 254–266.
Raison, R. J., Connell, M. J., and Khanna, P. K. (1987). Methodology for studying fluxes of soil
mineral-N in situ. Soil Biol. Biochem. 19, 521–530.
Redl, G., Hübner, C., and Wurst, F. (1990). Changes in hot water soil extracts brought about by
nitrogen immobilization and mineralization processes during incubation of amended soils.
Biol. Fertil. Soils 10, 45–49.
Reemtsma, T., Bredow, A., and Gehring, M. (1999). The nature and kinetics of organic matter
release from soil by salt solutions. Eur. J. Soil. Sci. 50, 53–64.
Rolston, D. E., and Liss, H. J. (1989). Spatial and temporal variability of water-soluble organic
carbon in a cropped field. Hilgardia 57, 1–19.
Rovira, P., and Vallejo, V. R. (2002). Labile and recalcitrant pools of carbon and nitrogen in
organic matter decomposing at diVerent depths in soil: An acid hydrolysis approach.
Geoderma 107, 109–141.
Sarathchandra, S. U., Perrott, K. W., Boase, M. R., and Waller, J. E. (1988). Seasonal changes
and the eVects of fertilizer on some chemical, biochemical and microbiological characteristics of high-producing pastoral soil. Biol. Fertil. Soils 6, 328–335.
Seely, B., Lajtha, K., and Salvucci, G. D. (1998). Transformation and retention of nitrogen in a
coastal forest ecosystem. Biogeochemistry 42, 325–343.
Seto, M., and Yanagiya, K. (1983). Rate of CO2 evolution from soil in relation to temperature
and amount of dissolved organic carbon. Jpn. J. Ecol. 33, 199–205.
Shepherd, T. G., Saggar, S., Newman, R. H., and Dando, J. L. (2001). Tillage-induced changes
to soil structure and organic carbon fractions in New Zealand soils. Aust. J. Soil Res. 39,
Skjemstad, J. O., Dalal, R. C., and Barron, P. F. (1986). Spectroscopic investigations of
cultivation eVects on organic matter of Vertisols. Soil. Sci. Soc. Am. J. 50, 354–359.
Skjemstad, J. O., Le Feuvr, R. P., and Prebble, R. E. (1990). Turnover of soil organic matter
under pasture as determined by 13C natural abundance. Aust. J. Soil Res. 28, 267–276.
Smith, J. L., and Paul, E. A. (1990). The significance of soil microbial biomass estimations.
In ‘‘Soil Biochemistry’’ (J. M. Bollag and G. Stotzky, Eds.), Vol. 6, pp. 357–396. Marcel
Dekker, New York, NY.
Smolander, A., Kitunen, V., and Mälkönen, E. (2001). Dissolved soil organic nitrogen and carbon
in a Norway spruce stand and in an adjacent clear-cut. Biol. Fertil. Soils 33, 190–196.
Smolander, A., Kitunen, V., Priha, O., and Mälkönen, E. (1995). Nitrogen transformations in
limed nitrogen fertilized soil in Norway spruce stands. Plant Soil 172, 107–115.
Sollins, P., and McCorison, F. M. (1981). Nitrogen and carbon solution chemistry of an old growth
coniferous forest watershed before and after cutting. Water Resour. Res. 17, 1409–1418.
Sollins, P., Homann, P., and Caldwell, B. A. (1996). Stabilization and destabilization of soil
organic matter: Mechanisms and controls. Geoderma 74, 65–105.
Sollins, P., Spycher, G., and Glassman, C. A. (1984). Net nitrogen mineralization from lightand heavy-fraction forest soil organic matter. Soil Biol. Biochem. 16, 31–37.
Sparling, G. P. (1997). Soil microbial biomass, activity and nutrient cycling as indicators of soil
health. In ‘‘Biological Indicators of Soil Health’’ (C. E. Pankhurst, B. M. Doube, and
V. V. S. R. Gupta, Eds.), pp. 97–119. CAB, Wallingford, UK.
Sparling, G., Vojvodic-Vukovic, M., and Schipper, L. A. (1998). Hot-water-soluble C as a
simple measure of labile soil organic matter: The relationship with microbial biomass C.
Soil Biol. Biochem. 30, 1469–1472.
Spycher, G., Sollins, P., and Rose, S. (1983). Carbon and nitrogen in the light fraction of a forest
soil: Vertical distribution and seasonal fluctuations. Soil Sci. 135, 79–87.
Stanford, G. (1982). Assessment of soil nitrogen availability. In ‘‘Nitrogen in Agricultural Soils’’
(F. J. Stevenson, J. M. Bremner, R. D. Hauck, and D. R. Keeney, Eds.), pp. 651–688.
American Society of Agronomy, Madison, WI.
Stevenson, F. J. (1994). ‘‘Humus Chemistry: Genesis, Composition, Reactions’’. John Wiley,
New York, NY.
Styllberg, U. S., and Magnusson, T. (1995). Cations adsorbed to soil organic matter—A
regulatory factor for the release of organic carbon and hydrogen ions from soils to waters.
Water Air Soil Poll. 85, 1095–1100.
Vance, E. D., Brookes, P. C., and Jenkinson, D. S. (1987). An extraction method for measuring
microbial biomass C. Soil Biol. Biochem. 19, 703–707.
Vance, G. F., and David, M. B. (1991). Chemical characteristics and acidity of soluble organic
substances from a northern hardwood forest floor, central Maine, USA. Geochim. Cosmochim. Acta 55, 3611–3625.
Wagai, R., and Sollins, P. (2002). Biodegradation and regeneration of water-soluble carbon in a
forest soil: Leaching column study. Biol. Fertil. Soils 35, 18–26.
Wander, M. M., and Bidard, M. G. (2000). Tillage practice influences on the physical protection, bioavailability and composition of particulate organic matter. Biol. Fertil. Soils 32,
Waters, A. G., and Oades, J. M. (1991). Organic matter in water stable aggregates. In ‘‘Advances
in Soil Organic Matter Research: The Impact on Agriculture and Environment’’ (W. S.
Wilson, Ed.), pp. 163–175. Royal Society of Chemistry, Cambridge, UK.
Watson, C. J., Jordan, C., Lennox, S. D., Smith, R. V., and Steen, R. W. J. (2000). Organic
nitrogen in drainage water from grassland in Northern Ireland. J. Environ. Qual. 29,
West, A. W., Sparling, G. P., Feltham, C. W., and Reynolds, J. (1992). Microbial activity and
survival in soils dried at diVerent rates. Aust. J. Soil Res. 30, 209–222.
Whalen, J. K., Bottomley, P. J., and Myrold, D. D. (2000). Carbon and nitrogen from light- and
heavy-fraction additions to soil. Soil Biol. Biochem. 32, 1345–1352.
Whitbread, A. M., Lefroy, R. D. B., and Blair, G. J. (1998). A survey of the impact of cropping
on soil physical and chemical properties in north-western New South Wales. Aust. J. Soil
Res. 36, 669–681.
Yakovchenko, V. P., Sikora, L. J., and Millner, P. D. (1998). Carbon and nitrogen mineralization of added particulate and macroorganic matter. Soil Biol. Biochem. 30, 2139–2146.
Yano, Y., McDowell, W. H., and Aber, J. D. (2000). Biodegradable dissolved organic carbon in
forest soil solution and eVects of chronic nitrogen deposition. Soil Biol. Biochem. 32,
Zak, D. R., Grigal, D. F., Gleesan, S., and Tilman, D. (1990). Carbon and nitrogen cycling
during old-field succession: Constraints on plant and microbial biomass. Biogeochemistry
11, 111–129.
Zimbilske, L. M. (1994). Carbon mineralization. In ‘‘Methods of Soil Analysis, Part 2.
Microbiological and Biochemical Properties’’ (R. W. Weaver, S. Angle, P. Bottomly, D.
Bezdicek, S. Smith, A. Tabatabai, and A. Wollum, Eds.), pp. 835–863. Soil Science Society of
America, Madison, WI.
Yadvinder-Singh,1 Bijay-Singh1 and J. Timsina2
Department of Soils, Punjab Agricultural University, Ludhiana 141 004, India
CSIRO Land and Water, Griffith NSW 2680, Australia
Availability of Crop Residues in Rice-Based Cropping Systems
Management Options for Crop Residues
Crop Residue Decomposition
A. Kinetics of Crop Residue Decomposition
B. Factors AVecting Residue Decomposition
C. Fallow Period and Crop Residue Management
Crop Residue Management EVects on Nutrient Availability in Soils
A. Nitrogen
B. Phosphorus
C. Potassium
D. Sulfur
E. Micronutrients
EVect of Crop Residues on Soil Properties
A. Soil Fertility
B. Chemical Properties
C. Physical Properties
D. Biological Properties
E. Crop Residues for Reclamation of Salt-AVected Soils
Biological Nitrogen Fixation
Phytotoxicity Associated with Crop Residue Incorporation into
the Soil
Weed Control and Herbicide EYciency
Emission of Greenhouse Gases
A. Methane
B. Nitrous Oxide
C. Mitigation Strategies
Agronomic Responses to Crop Residue Management
A. Rice–Wheat Cropping System
B. Rice–Rice Cropping System
C. Rice–Legume Cropping System
D. Other Rice-Based Cropping Systems
Advances in Agronomy, Volume 85
Copyright 2005, Elsevier Inc. All rights reserved.
0065-2113/05 $35.00
XII. Summary and Conclusions
XIII. Research Needs
Rice is the most important crop in Asia, where more than 90% of all rice is
grown and consumed (Blake, 1992). About half of the total area planted to
rice is irrigated, and it accounts for the three-fourths of global production
(Huke and Huke, 1997). In tropical Asia, rice–rice constitutes an important
annual crop rotation. In the subtropical Asia, rice and wheat are grown in
rotation in more than 13 million hectares in the Indo-Gangetic plains of
South Asia (India, Bangladesh, Nepal, Pakistan) and on similar hectarages
in the basin of the Yangtze river in China (Timsina and Connor, 2001). In
addition to wheat, other crops grown in rotation with rice are barley, oats,
maize, sorghum, legumes (mung bean, peanuts, soybean, lentil, chickpea),
oilseeds (mustard, rapeseed), potato, sugarcane, and cotton.
Nutrient cycling in the soil–plant ecosystem is an essential component of
sustainable productive agricultural enterprise. Although during the last three
decades fertilization practices have played a dominant role in the rice-based
cropping systems, crop residues—the harvest remnants of the previous
crop—still play an essential role in the cycling of nutrients. Incorporation
of crop residues alters the soil environment, which in turn influences the
microbial population and activity in the soil and subsequent nutrient transformations. It is through this chain of events that management of crop
residues regulates the eYciency with which fertilizer, water, and other
reserves are used in a cropping system. Another feature of rice-based cropping system in the tropics is the inherent conflict between maximizing shortterm production at minimum cost versus providing sustainable health and
long-term productivity of the soil. One reason for this conflict is the general
below-average economic condition of the farmers practicing rice-based cropping systems. In the tropics, crop residues have, in fact, played a pivotal role
in the maintenance of soil resources at acceptable levels because these are the
major sources of C inputs.
Tropical agricultural ecosystems are distinct from temperate ones in
terms of biological degradation of soils, which results in reduction in organic
matter content due to decline in the amount of C inputs from biomass
(Stewart and Robinson, 1997). Tropical soils vary widely in their properties
and are generally poor in native soil fertility and productivity. The removal
of crop residues leads to low soil fertility and thereby decreased crop
production. Organic materials such as crop residues oVer sustainable and
ecologically sound alternatives for meeting the nutrient requirements of
crops. In addition to their role as the primary source of C inputs, crop
residues, and the way they are managed, have a significant impact on soil
physical properties (Boyle et al., 1989).
Future increase in food production in the tropics will only be possible
through improvement in soil productivity. Increased concern for the environment and increased emphasis on sustaining soil productivity has resulted
in major interest in the maintenance and improvement in soil organic matter
in recent years. Proper management and utilization of crop residues and
other agricultural wastes will constitute an important factor in achieving this
objective. With widespread use of combine harvesters, crop residues (mainly
rice and wheat) largely remain in the field and must be managed to provide
the greatest advantages.
The increasing constraints of labor and time under intensive agriculture
have led to the adoption of mechanized farming in rice-based cropping
systems. For example, under highly intensive rice–wheat cropping system in
northwestern India, combine harvesting of rice and wheat fields, which leaves
large amounts of crop residues in the fields, is now a common practice. As
crop residues interfere with tillage and seeding operation for the next crop,
farmers often prefer to burn these residues. In addition to causing environmental pollution, burning results in large losses of organic carbon and plant
nutrients. In recent years, the concept of soil quality has been suggested as a
tool for assessing the long-term sustainability of agricultural practices
at local regional, national, and international levels. Crop residue management is known to aVect either directly or indirectly most of the soil quality
indicators—chemical, physical, or biological. It is perceived that soil
quality is improved by the adoption of sound crop residue management
practices. For example, Karlen et al. (1994) evaluated several soil quality
indicators and developed a soil quality index, which had values of 0.45, 0.68,
and 0.86 for removal, normal, and double residue treatments, respectively.
In comparison with green manures and legume residues, cereal straws are
relatively poor with respect to N and P content. Thus, crops sown immediately after the incorporation of residues of cereal crops suVer due to deficiency of plant-available N. Addition of fertilizer N to the decomposing
residues only partially oVsets the immobilization process. Therefore, a major
problem encountered in the profitable utilization of cereal crop residues is
the occurrence of microbial immobilization of soil and fertilizer N (Mary
et al., 1996). Suitable manipulations of processes such as nutrient immobilization are an important component of an eYcient crop residue management
program. For example, allowing adequate time for decomposition of crop
residues before planting the next crop can be beneficial in alleviating adverse
eVects due to N immobilization and phytotoxicity.
According to one estimate, more than 1000 million tonnes of cereal
residues are being produced annually in the developing world (FAO,
1999). If crop residues could be better managed, this would directly improve
crop yields by increasing soil nutrient availability, decreasing erosion, improving soil structure, and increasing soil water holding capacity. Crop
residues can also lead to negative eVects on crop production in the short
term because of N immobilization and possible release of phytotoxic compounds. Considerable research has been conducted in the last few decades
relating residue management to soil chemical, physical, and biological properties and consequent fertilizer management practices needed for successful
crop production. In this chapter, we have tried to use this knowledge to
make recommendations and conclusions for crop residue management in
rice-based cropping system. We have not attempted to review all available
literature; only pertinent data sets have been used to substantiate diVerent
conclusions that emerge from the literature. There may be two diVerent
systems of crop residue recycling: (1) when residues are applied directly to
the soil and (2) when residues are first allowed to decompose and are used as
compost. We have focused our attention mainly on the in situ incorporation
of crop residues left naturally in the field under rice-based cropping systems.
A special situation is created by the planting of green manure crops in a crop
rotation and increasing the amount of crop residues at planting time of the
next crop. The use of crop residues as green manures has already been
reviewed extensively (Buresh and De Datta, 1991; Yadvinder-Singh et al.,
1991) and is not included in this chapter. Similarly, depending on the scale
considered, although manures, diVerent organic by-products, animal and
human wastes, and food processing wastes originate mainly from harvested
plants, these are not categorized as crop residues in this chapter. The
challenge is to (1) scientifically understand the short- and long-term eVects
of diVerent crop residue management on the cycling of C, N, and other
nutrients, and (2) develop technologies for crop residue management that
are agronomically beneficial, environmentally friendly, and do not add extra
Rice, wheat, corn, soybean, barley, rapeseed, and potato are the major
residue-producing crops. Asia is the major producer of crop residues—
52.6% of the world residues production occurs in Asia. Rice, wheat, and
corn are the major crops, contributing about 84% of the total production of
crop residue in Asia. On a global basis, these seven crops produced 2956
Table I
Residue Production (103 t) by Rice and DiVerent Crops Grown in Rotation with Rice in the
Tropics in 1998a
South America
Data pertaining to residue production was computed by multiplying grain yield data reported
by FAO (1999) with straw:grain ratios reported by Larson et al. (1978) for South America and
by Bhardwaj (1995) and Beri and Sidhu (1996) for Asia and Africa.
million tons of residues in 1998; rice residues were around 1000 million
tons (FAO, 1999). Production of residues by diVerent crops that can be
grown in rotation with rice in diVerent countries is shown in Table I. Rice
contributes about 34.3% of the total residue production, which is 1.2 times
more than wheat. Reliable quantitative estimates of crop residues in tropical
countries are, however, lacking. Below-ground residue production has often
been ignored due to the diYculty in measuring it. In India, an attempt has
been made to arrive at a figure based on estimates of crop yields and
knowledge of the harvest index of diVerent crops. For example, an estimate
for India was made by Bhardwaj and Gaur (1985) by assuming that all
residues generated were left in the field and that nutrient availability from
this component followed mineralization of 50% per cropping cycle. Average
yields of irrigated rice will have to increase from 4.9 t ha 1 in 1991 to about
8 t ha 1 in 2025 (Cassman and Pingali, 1995). If rice cropping is intensified at
this scale, grain yield and total biomass production will increase by about
60% during the next 30 years. Our rough estimates indicate that the expected
increase in biomass production will potentially increase the amount of
C remaining in straw and roots by about 90 million t year 1 and that of
N by about 1.8 million t year 1 (Doberman and Witt, 2000). This represents
an enlarged sink for CO2 but also a greater potential source for CH4
Globally, about 31, 26, and 154% of N, P, and K, respectively, of the
fertilizer consumption in 1998 were found in crop residues (FAO, 1999).
Residues of seven leading crops in all the continents contained about
18.8 million tons of N, 2.9 million tons of P, and 24.0 million tons of
K. An estimate of the quantity of N, P, and K contained annually in
Table II
Estimates of N, P, and K ( 103 t) in Residue Produced by DiVerent Crops Grown in Rotation with
Rice in the Tropics in 1998a
Africa and
South America
Estimates of N, P, and K in crop residues were computed by multiplying residue yield data
given in Table I with N, P, and K contents in straw reported by Larson et al. (1978) for South
America and by Bhardwaj (1995) and Beri and Sidhu (1996) for Asia and Africa.
the residues of major crops grown in rice-based rotations in diVerent continents is presented in Table II. These estimates are based on average
nutrient concentrations in crop residues as reported by Larson et al.
(1978) for Europe, South and Central America, and Oceania and by Beri
and Sidhu (1996) and Bhardwaj (1995) for Asia and Africa. The values
reported in Table II do not include nutrients contained in roots. The crop
residue N is available to the extent of 41% in Asia followed by 28% in Northcentral America, 15% in Europe, 11% in South America, 4% in Africa, and
only 1% in Oceania. In addition to N, P, and K, crop residues also contain
substantial amounts of secondary and micronutrients. The fertilizer equivalent value of field residues for nine Indian crops worked out to be 760,000
tons, a sizeable and significant figure (Bhardwaj and Gaur, 1985). In China,
straw yield of cereals has been calculated as 621.6 million tons per year, and
20–30% is commonly returned to fields following harvest (Compilatory
Committee, 1990).
There exist several options for managing crop residues. These include
being removed from the field, left on the soil surface, incorporated into the
soil, burned in situ, composted, or used as mulch for succeeding crops.
Throughout the tropics there is little recycling of crop residues in the
field—these are either harvested for fuel, animal feed, or bedding or are
burned in the field. Crop residues removed from the field can also be used
as bedding for animals, a substrate for composting, biogas generation or
mushroom culture, or as a raw material for industry. Local conditions
determine the disposal method. Currently, in China, North Vietnam,
India, Bangladesh, and Nepal, complete removal of straw from the field is
widespread in areas with hand harvest and great demand for straw as fodder,
as fuel, or for industrial purposes, causing large nutrient export from rice
fields. Open-field burning of rice straw is predominant in areas with combine
harvesting (northern India, Thailand, parts of China) or where manual
thrashing is done in the field (Indonesia, Malaysia, Myanmar, Philippines,
southern Vietnam). In many parts of the tropics, crop residues are burned in
the field due to the ignorance of farmers about their value and lack of proper
technology for in situ incorporation of residues (Samra et al., 2003). For
example, in the intensive rice–wheat cropping system in the Indo-Gangetic
plains of South Asia, crop residues, particularly rice straw, are not used as
animal feed and are disposed of by burning. This is a cost-eVective method
of straw disposal and helps to reduce pest and disease populations resident in
the straw biomass, but it also causes pollution by releasing CO2, N2O, NH3,
and particulate, leading to global warming and health concerns (Kirkby,
1999). It also reduces the number and activity of soil microbes. The magnitude of C and nutrient loss during burning is influenced by the quantity of
residue burned and the intensity of the fire. Complete burning of rice straw
at 470 8C in muZe furnace resulted in 100, 20, 20, and 80% losses of N, P, K,
and S, respectively (Sharma and Mishra, 2001). The corresponding losses
due to burning of wheat straw were 100, 22, 22, and 75%. The losses of
nutrients were less due to incomplete burning of the crop residues in open air
under field conditions: 88.6% N, 1.8% P, 17.5% K, and 25.3% S for wheat
straw and 89.2% N, 5.5% P, 19.9% K, and 20.5% S for rice straw, as
compared to complete burning. No loss of micronutrients was noted
during incomplete burning of straw. The temperature of heating was more
important than the duration of heating.
In the Philippines, Indonesia, and parts of China, heaping of rice straw in
the field at threshing sites is common. Heaping the straw in successive
quadrants of a field each season is recommended to even out nutrient
distribution. The straw decomposes slowly, largely aerobically, and can be
easily spread and incorporated into the soil at the beginning of the next
season. Not much is known about the rate at which straw in heaps decomposes or about the loss of N via denitrification or loss of N and K through
Because of air pollution concerns and nutrient losses, the burning of
residues is now being reconsidered in many regions of the world (Ocio
et al., 1991; Miura and Kanno, 1997). However, in double- or triple-cropped
rice-based systems with sustained flooding, incorporating straw may reduce
yields (Cassman et al., 1995). The crop residues can impede seedbed preparation and contribute to disease and weed problems. There are currently few
options for rice straw because of its poor quality for forage, bioconversion,
and engineering applications (Jenkins et al., 1997). Rice growers are therefore seeking alternative disposal options, such as incorporation of the straw
into the soil. The incorporation of rice residues and continuous flooding has
become common in tropical soils through intensification of rice cropping
practices (Cassman and Pingali, 1995).
In addition to introducing an extra cost, rice straw incorporation
in association with flooding likely impacts soil fertility through nutrient
and pest interactions (Cassman et al., 1995, 1997; Olk et al., 1996) and
environmental quality through greenhouse gas emissions (Bossio et al.,
1999; Delwiche and Cicerone, 1993). In rice–wheat cropping systems, too,
management of rice straw, rather than wheat straw, is a serious problem,
because there is very little turn-around time between rice harvest and wheat
Incorporating the crop residues into the soil and allowing them to decompose returns to the soil almost all the nutrients in the straw. The common
practice of burning the residues can have a net short-term beneficial eVect on
the N supply to subsequent crops but a deleterious eVect on overall N supply
and soil C. In North America and Europe, incorporation of cereal straw is
being considered as an alternative to burning because of concern over the
adverse environmental impacts of burning (Prasad and Power, 1991).
With the advent of direct drilling, there is now much interest in the
possibility of direct drilling of wheat into rice stubble—either the full stubble, or after removal or burning of the header tailings. Current research in
Punjab, India shows that sowing into stubble using no-till seed drill is
impaired by blockages with the loose straw and inadequate closure of the
seed slots. Bed planting provides new opportunities and challenges for
stubble management in rice–wheat systems, which need to be addressed.
Conservation tillage and mulch farming techniques have proven useful
in the highly erodible soils of the Loess Plateau of China (Zhiqiang et al.,
1999). Keeping in mind both socioeconomic and biophysical factors, there is
a need to develop conservation tillage systems for a wide range of ricebased cropping systems, soils, and agroecological environments. Use of
crop residues as mulch is important to the development of soil-specific
conservation tillage systems in the upland soils of the tropics.
DiVerent residue management technologies or strategies need to be developed at a regional level to fit diVerent rice-based cropping systems and to
accommodate the management diversity required within a single farming
enterprise. Estimates of relative costs of diVerent options must be developed,
as the most attractive choices might have significant impacts on environmental quality through their eVects on microbial processes that determine
the magnitude of C storage in the soil, methane emission into the atmosphere, and long-term soil fertility. Incorporation of straw in the soil is the
management option dealt in detail in this chapter. After the crop is harvested, the straw is spread on the land and incorporated into the soil by
disking or plowing.
Decaying of crop residues starts as soon as the residues come into contact
with the soil. The process of decomposition is controlled by the interaction
of three components: the soil organisms or biological processes, the quality
of crop residues, and the physical and chemical environment. The combination of these components determines not only the rate of decomposition of
crop residues but also the end product of the decomposition process. The
amount of plant materials decomposed in the soils is determined by the loss
of dry weight of these plant materials buried in the soil or by the evolution
of CO2 from plant materials, either unlabeled or 14C or 13C. Burying of
rice straw in soil has been reported to accelerate the decomposition in
comparison with placing the straw on the soil surface (Kumar and Goh,
2000). Residues are managed diVerently; e.g., residues can be placed on the
surface, mixed into the soil, or confined in mesh bags within the soil. Surface
placement or heterogeneous distribution reduces the residue–soil contact as
compared with a homogenous distribution. This may aVect the decomposition dynamics. Knowledge of such eVects is important when results from
diVerent studies are being compared and is essential when developing
and calibrating decomposition models. It is also important when assessing
the eVects of tillage practices resulting in diVerent degrees of residue–soil
contact, e.g., no-till ploughing and rotovating. The degree of contact between crop residues and the soil matrix, as determined by the method of
residue incorporation, aVects decomposition dynamics under both natural
and experimental conditions. A dearth of information exists regarding straw
decomposition under upland conditions and its eVect on long-term
N availability in temperate regions.
Crop residues left in the field after harvest are the raw materials for
humus formation and may represent a significant supply of nutrients to
subsequent crops. Knowledge about residue decomposition is, therefore,
essential for management of agroecosystems. Most of the work on decomposition of crop residues has been carried out in temperate soils (Kumar and
Goh, 2000). Soil C content depends on the amount of C that leaves the soil
through decomposition, erosion, or leaching. Under normal circumstances,
most of the C is lost from the system through decomposition. Kinetic models
of decomposition have commonly used some form of the first-order equation
(Molina et al., 1983). Although a single rate constant has been used to
describe decomposition over the long term in the field or laboratory (Havis
and Alberts, 1993; Schomberg et al., 1994b; Kuo et al., 1997), most shortterm laboratory studies have shown that crop residues contain two or more
decomposition fractions (Gilmour et al., 1985; Ajwa and Tabatabai, 1994).
In cases in which two or more fractions exist, decomposition can be described by a sequence of first-order equations that allow all fractions to
decompose at the same time (simultaneous model).
The change of C in the soil can be expressed mathematically in one kinetic
rate constant of decomposition:
Ct ¼ C0 e
k1 t
þ Ca e
k2 t
where Ct is the amount of soil C at time t, C0 is the amount of soil C at time
0, k1 is the decomposition rate constant (day 1) of the total soil C pool
before amendment of C added, Ca is the amount of C (crop residue) added,
and k2 is the added C. The decomposition process is often viewed as a series
of first-order reactions for various C fractions, each with its own size and
decomposition rate decomposition rate constant (day 1) (Jenkinson and
Rayner, 1977; Parton et al., 1988). The rapid and slow fractions with a
characteristic slope and intercept can be mathematically represented as two
simultaneous first-order reactions:
% decomposed ¼ % rapidð1
exp½ k1 tŠÞ þ
% rapidÞð1
exp½ k2 tŠÞ;
where % rapid is the amount of crop residue organic C in the rapid fraction,
(100 % rapid) is the amount of crop residue organic C in the slow fraction, k1
is the rapid-fraction first-order rate constant, k2 is the slow-fraction
first-order rate constant, and t is the elapsed time. The percentage of the
crop residue C remaining is 100 minus decomposed in Eq. (2).
The rate constant increased with temperature and was significantly lower
under flooded conditions (Devevre and Howarth, 2000). The rate of decomposition of rice straw at 25 8C under flooded conditions was as low as the
rate of decomposition of rice straw at 5 8C under nonflooded conditions.
Under nonflooded conditions at 15 and 25 8C, the model described two pools
of decomposable C (C1/k1 and C2/k2). The first pool (C1) in the nonflooded
treatment at 25 8C represented 30% of the straw C with a turnover time
of 12.5 days; the second pool (C2) represented 36% of the straw C with
a turnover time of 100 days. The corresponding values at 15 8C were
24% with a turnover time of 17 days and 53% with a turnover time of 333
Under flooded conditions, only one pool of C could be described using
the equation. At 25 8C, C mineralized represented 52% of the straw C with a
turnover time of 50 days against 46% of the straw C at 15 8C with a turnover
time of 100 days. The most apparent reason for only one pool of decomposable C in flooded treatment was that recycling of waste products from
fermentative metabolism extended the availability of labile sources of C.
From the amount of C mineralized in the flooded treatments, it is evident
that the C2 pool (recalcitrant C compounds: cellulose, lignin, and microbial
melanins) was partially degraded and contributed to the total C mineralized.
The study indicated that the conversion of C2 straw components to C1
components under fermentive conditions most likely increased the C utilization eYciency of the more recalcitrant C2 pool under flooded conditions.
The larger biomass, the simultaneously lower total amount of C mineralized,
and the higher eYciency of substrate C conversion to microbial biomass
(yield factor) in the flooded soil supported these observations (Devevre and
Howarth, 2000). These researchers reported values of C1 and C2 of 1086 and
1324 mg C g 1 and k1 and k2 values of 0.08 and 0.01 day 1 under nonflooded
conditions, respectively. The C1 and k1 values for flooded soil were 1916 mg
C g 1 and 0.02 day 1 incubated at 25 8C.
If management changes are desired to achieve an increase in total soil C or
soil organic matter (SOM), Eq. (2) allows for two major options: increase
C input or reduce decomposition rates. Virtually no studies have explored a
reduction in decomposition rates in intensive lowland rice systems, and
hence no literature review can be presented.
Table III shows that rates of crop residue decomposition depend on
residue type, length of decomposition period, and climatic conditions.
Cheng and Wen (1998) studied the decomposition of rice straw over a
10-year period in two soils with diVerent mineralogical characteristics
in fields under upland and submerged conditions in China. Using the
first-order equation for residue decomposition, they calculated annual mineralization rates (k) of 0.127 under upland and 0.106 under submerged
conditions in yellow-brown soil (pH 7.7). The corresponding rates in red
soil (pH 4.6) were 0.159 and 0.0948. The half-lives of residual C in the two
soils were 4.4–5.5 years under upland and 6.5–7.3 years under submerged
conditions. The percentage of organic C retained in two soils under upland
and submerged conditions was 29.3–31.3% after year 1 and 7.92–11.6% in
year 10 in the yellow-brown soil and 33.5–35.2% in year 1 and 7.48–13.0%
in year 10 in the red soil. Mineralization of residual organic N followed
the same pattern as residual C. More N from plant material was retained in
Table III
Decomposition of DiVerent Crop Residues in DiVerent Laboratory and Field Studies
Cookson et al.
Saini (1989)
Reddy et al.
Ghidey et al.
Type of
Method used
and experimental
Litter bag
Litter bag
Litter bag
Saviozzi et al.
Kaboneka et al.
Martin et al.
Litter bag
Yadvinder-Singh Field
et al. (2004b)
Litter bag
Wheat straw
Barley straw
Rice straw
Rice straw
90 days
90 days
197 days
330 days
Wheat straw
Wheat straw
Rice straw
10 months
10 months
10 months
1 year
1 year
1 year
30 days
30 days
14 days
90 days
1 year
10 days
20 days
40 days
77 days
Rice straw
the yellow-brown soil than in the red soil. Rice straw mineralized more
slowly under submerged conditions than under upland conditions.
Buyanovsky and Wagner (1997) described the C decomposition from
wheat straw using nonlinear regression to fit a two-component exponential
model. This relationship could be used to calculate the percentage of the
residue C remaining in the soil at any specified time. The first component
represents the rate of mineralization of the readily decomposable fraction.
This includes the simple sugars, soluble proteins, hemicellulose, and cellulose. The second component represents the mineralization of the resistant
products of microorganisms. The half-life of the first component was 18 days
and that of second component was 433 days. The humification coeYcient for
wheat straw was 0.24.
Mishra et al. (2001a) studied the C and N mineralization from wheat
straw using the nylon mesh bag technique in a silty loam paddy soil. Wheat
straw decomposed at a faster rate initially for 2 weeks, and the rate of
decomposition (measured as loss in weight of straw) slowed down thereafter,
probably when soluble carbohydrates and proteins were exhausted. Within
2 weeks, 30.7 and 25.3% of wheat straw in the bags had decomposed in the
wheat straw plus green manure and wheat straw alone plots, respectively.
The green manuring accelerated the decomposition of the wheat straw by
lowering the C:N ratio of the decomposing material and by stimulating the
microbial population to carry out the decomposition. By the end of 22 weeks,
82–86% of the wheat straw was decomposed. The practice of wheat straw
incorporation in conjunction with green manure holds promise for improving the soil productivity in rice–wheat cropping systems. The decomposition
of wheat straw followed the first-order kinetic model. The decay rate constant (k) for the wheat straw was 0.013 day 1 and the half-life was
60 days. The C loss could account for 48–49% of the weight loss during
22 weeks. The other constituent of weight loss from wheat straw could
be soluble components such as K, Cl, and organic substances released as
intermediate products during decomposition. Up to 90% of K can be lost
from the crop residues within a few weeks after incorporation into the soil.
Zhu et al. (1988) reported that about 75% of wheat straw was decomposed in
about 150 days under field conditions in China.
Using the nylon mesh bag technique, Mishra et al. (2001b) noted three
phases in the decomposition of rice straw in a silty loam soil during wheat
growing season. The first phase lasted for 5 weeks, during which the rate of
decomposition was relatively faster (38% of the rice residue decomposed),
followed by the second phase of slow rate of decomposition from the 6th to
the 15th weeks (22% of the rice residue decomposed), which may partly be
attributed to the prevailing low air temperatures, followed by the third phase
of fast rate of decomposition up to 23 weeks (19% of the rice residue
decomposed) due to the rising air temperatures. By the end of 23 weeks,
79% of the rice straw was decomposed. The decomposition of rice straw
was satisfactorily described by the Douglas and Rickman (1992) model
(R2 ¼ 0:97). The computed values of fN and k were 1.356 and 0.00045
CDD 1, respectively, where CDD is cumulative degree days. Witt et al.
(1998), using the litter bag technique, reported a 56% decrease in rice crop
residues within 56 days after incorporation at IRRI, Philipines.
Kanazawa and Yoneyama (1980) observed that the decomposition of rice
straw occurred in two phases in a clayey soil under flooded and upland
conditions in a laboratory at 30 8C. During the first 2 to 4 months of
incubation, the dry weight decreased by half. This was followed by a long
period of very gradual weight decrease. After 12 and 24 months, about 70
and 75% of the initial weight of rice straw, respectively, was lost under both
flooded and upland conditions. The above decay pattern has two phases:
rapid C loss in the first few months and then slow loss in the subsequent long
period. The period of rapid decay seems to be almost coincident with the time
of high population of bacteria and fungi. This study showed that the fungi
are the main agents in the decomposition of organic materials under upland
conditions. In contrast, anaerobes are the main agents in the course of
residue decomposition under flooded conditions, and the breakdown of
cellulose and lignin may be slow under such decomposing systems. Under
field conditions, soil moisture conditions may change from time to time, and
this may cause the shifting participation of diVerent microorganisms during
plant residue decomposition. The rate of decomposition was slightly higher
under flooded than under upland conditions, but the pattern of residue
decomposition was similar under the two moisture regimes. The residue
decomposition trend was closely related to the changes in the microbial
population (bacteria, fungi, and actinomycetes) under flooded conditions.
On the other hand, under upland conditions, significant correlations were
observed between residue weight loss and bacteria or fungi, but not between
residue weight and actinomycetes. This suggested that the fresh crop residues
added to soil are a good substrate for microbial activities. Though the
numbers of microbes were small under flooded compared to upland conditions, aerobic microbes may play some role in the decomposition process of
rice residues in the early incubation periods when the O2 concentration in the
soil is high. In the early incubation period, addition of rice straw caused an
increase in the microbial numbers under both flooded and upland conditions. The decomposing ability of the anaerobes under flooded conditions
may be high enough to decompose the plant materials to a similar extent as
that of the aerobes under upland conditions. The numbers of microorganisms in upland soil were larger than under flooded conditions, and the
addition of rice residues brought about a vigorous increase in the numbers,
especially of actinomycetes and fungi.
In tropical systems, mineralization rates are potentially higher because of
high soil temperatures during cropping season, particularly at the time of
incorporation of residues. Several reviews have summarized the factors
aVecting crop residue decomposition, particularly in temperate climates
(Kumar and Goh, 2000; Parr and Papendick, 1978; Prasad and Power,
1991; Smith et al., 1992). We include here only a brief commentary on the
pertinent factors of crop residue decomposition that have relevance to ricebased cropping systems in the tropics. Carbon and nitrogen cycling are
mainly caused by changes in the frequency, amount, type, and mode of
recycling of crop residues; the frequency, length, and intensity of wetting
and drying cycles, that is, disturbances that cause severe shifts in microbial
activities and also aVect soil physical properties; and the O2 supply to the soil
during rice growth, that is, the amount of irrigation water percolating
through the soil and the intensity of soil reduction processes.
There are three main factors that aVect crop residue decomposition in the
soil: (1) crop residue factors, (2) edaphic factors, and (3) management
factors. The development of an eVective crop residue management program
depends on a thorough understanding of the ways in which these factors
influence the decomposition process. It has been recognized that organic
residue decomposition and hence soil organic matter dynamics are a
direct result of the physiocochemical environments, e.g., aeration (aerobic/
anaerobic, soil structure) and the quality of the resource acting through their
regulation of the decomposer community.
Crop residue chemical composition plays an important role in determining decomposition rates. Thus, in order to predict the decomposition and
nutrient mineralization patterns of plant residues, it is essential to understand their constitution in terms of soluble and resistant fractions. Early in
the decomposition process, rapid loss of simple sugars and amino acids may
occur within a few hours to a few days, while polysaccharides, proteins, and
lipids decompose at much slower rates. Lignin makes up 5 to 30% of crop
residue material and is more resistant to decomposition than other plant
constituents. Lignin is an important substrate for soil humus formation due
to its resistance to decomposition.
Janzen and Kucey (1988) found that diVerences in decomposition rates of
crop residues were positively correlated with crop N content. There was no
significant relationship reported between decomposition rates and C:N ratio,
water-soluble C, lignin, hemicellulose, and cellulose content of crop residues.
Using a perfusion system, Villegas-Pangga et al. (2000) observed that the
CO2 release rates in 30 rice varieties varied; the percentage of C released
from straw ranged from 15.4 to 38.4% in 42 days. There was an inverse
relationship (R2 ¼ 0.6) between cumulative C release and C:N ratio and a
direct relationship between digestible organic matter (DOM) and cumulative C release. A straw quality index (SQI) was developed to describe the
decomposition rate of the rice straw as follows:
56:85 þ ð11:68 % NÞ þ ð1:25 % DOMÞ þ ð2:59 % ligninÞ;
R ¼ 0:81:
These findings suggested that SQI is a practical tool for assaying the quality
of the straw materials to predict their usefulness in crop residue management
systems. Despite a twofold diVerence between varieties in the amount of
C evolved over 20 days, the proportion of nutrient release did not diVer
significantly between them.
The availability to microbes of C and N contained in crop residues along
with lignin content greatly influence decomposition rates and N availability
to plants (Vigil and Kissel, 1991). It is generally accepted that residues with
low N content or a high C:N ratio decompose more slowly than those with
a low C:N ratio or high N content (Magid et al., 1997; Parr and Papendick,
1978). Christensen (1986) found that 44% of wheat straw containing 0.92%
N decomposed during the first month but only 7% of the straw containing
0.4% N decomposed during the same period of incubation. Luo and Cheng
(1991) found that the number of days required for 50% mass loss of crop
residues was significantly correlated with the N content of the residue.
Decomposition rates are normally greater for legume residues (low C:N
ratio) than those for cereal residues (high C:N ratio) (Ladd and Foster,
Although N content and C:N ratio are useful in predicting residue decomposition rates, they should be used with some caution. Reinertsen et al.
(1984) and Stott and Martin (1989) indicated that the C:N ratio of straw was
not a good decomposition index. De Haan (1977) found no relationship
between percentage of N in added plant residue and the rate of decomposition. Gilmour et al. (1998) observed that initial (0–2 weeks) decomposition
was related to crop residues N and C:N ratio, while subsequent decomposition was not related to these factors. Since C:N ratio does not indicate
the availability of the C and N to the microorganisms, crop residue decomposition based on available C and N seems to relate more closely to
field observations than decomposition based on total C and N contents
(Mtambanengwe and Kirchmann, 1995).
The concentration of polyphenol is generally greater in mature residues than in green leaves (Fox et al., 1990; Palm and Sanchez, 1991).
The rate of plant residue breakdown depends on the relative proportion
of these fractions. Hagin and Amberger (1974) estimated the half-life of
sugars, hemicellulose, cellulose, and lignin as 0.6, 6.7, 14.0, and 364.5 days,
respectively. Other factors such as lignin, hemicellulose, and polyphenol
content should also be considered for predicting decomposition of crop residues. Lignin is known to be a recalcitrant fraction and is highly resistant to
microbial decomposition (Mellilo et al., 1982). Many workers have found
that increasing lignin concentration reduces the decomposition rate and
nutrient release from plant residues (Fox et al., 1990; Tian et al., 1992).
Saini et al. (1984) reported that the rates of decomposition of stubbles of
rice, wheat, and rape were lower than those of their straws due to high
ash and lignin contents. Polyphenol concentration in plant tissue also
reduces its rate of decomposition by binding to protein and forming complexes resistant to decomposition (Vallis and Jones, 1973). Since polyphenols
have diVerent properties with respect to binding N-containing compounds
depending upon their molecular weight (Scalbert, 1991), these govern decomposition and N release in some studies but not in others (Vanlauwe et al.,
The decomposition rate of plant residues cannot be predicted from a
single property of the organic material. When considered simultaneously,
these properties can predict the decomposition rate from a wide range of
plant residues. In some studies, polyphenol:N and (lignin þ polyphenol):
N ratios have been correlated with residue decomposition and nutrient
release (Constantinides and Fownes, 1994; Fox et al., 1990; Palm and
Sanchez, 1991; Tian et al., 1995). It has been suggested that the polyphenol:N ratio may serve as a short-term index for green manures, while the
(lignin þ polyphenol):N ratio could be used for more mature or woody plant
materials (Palm, 1995).
Residue Particle Size
The accessibility of plant residues to soil microbes is of primary importance in their rate of decomposition. The particle size of the residue can
provide diVerent degrees of accessibility, which in turn aVect residue decomposition rates as well as the mineralization-immobilization process. Generally, small particles decompose faster than large particles because the
increased surface area and better distribution in soil will increase the susceptibility to microbial attack (Jensen, 1994). Angers and Recous (1997) studied
the eVect of particle size (0.03 to 10 cm) of wheat straw (C:N ¼ 270) on the
decomposition in a silt loam soil incubated at 15 8C. Early decomposition
(3–17 days) was faster for the small-sized particles (0.06–0.1 cm), followed by
the large-sized particle (5 and 10 cm). After 102 days, the very fine particles
(<0.1 cm) showed the greatest and the intermediate-sized classes (0.5–1.0 cm)
the lowest amount of C mineralization. On the other hand, finely ground
particles of ryegrass (C:N ¼ 9) decomposed at a lower rate than intermediate-sized classes. It was hypothesised that greater accessibility and availability of N were responsible for the higher rate of decomposition observed for
finely ground wheat straw, while a physical protection of finally ground
residues was probably involved in the observed reverse eVect of ryegrass
with a low C:N ratio. Puig-Gimenez and Chase (1984), however, observed
no significant eVect of length of straw on its decomposition. The eVect of
plant residue particle size on C and N mineralization may thus be an
interaction between clay content, secondary metabolic products, plant residue chemical composition, period of decomposition, and fanual activity
(Kumar and Goh, 2000).
Environmental Factors
Environmental conditions can aVect residue decomposition rates (Parr
and Papendick, 1978). Generally, decomposition rates are faster in tropical
areas and decrease as water availability and temperature decrease. Maximum activity of decomposers is near 30 to 35 8C and thus supports maximum
residue decomposition in this range (Roper, 1985; Stott and Martin, 1989).
Singh et al. (1995) found that 24.8–29.0% and 39.5–43.4% of the applied
C through rice and wheat residues was decomposed in 60 days at 258 and
40 8C, respectively.
Soil water content can dramatically influence crop residue decomposition
and nutrient cycling (Doel et al., 1990). The optimum water potential for
residue decomposition lies in soil water potentials between 30 and 100
kPa. Soil dried to a water potential of 10 MPa evolved CO2 at about one-half
the rate of soils incubated at optimum water content ( 20 to 50 KPa)
(Sommers et al., 1981). Pal and Broadbent (1975) showed that the maximum
rate of decomposition of plant residues occurred at 60% water holding
capacity (WHC) and the rates decreased at either 30 or 150% WHC. The
lower rates of straw decomposition under lowland conditions are probably
due to limited aeration for microbial activity. However, Villegas-Pangga et al.
(2000) reported that there was 27–45% reductions in C evolution in rice straw
under anaerobic conditions compared to aerobic systems. These results suggested that under flooded conditions, depletion of O2 decreases the decomposition rate of straw but the initial rate of nutrient release is unaVected. This
uncoupling of C and nutrient release appears to be related to the more labile
components of the nutrients present in the plants and their physiological role.
Devevre and Horwath (2000) reported that flooding had a tendency to
reduce C mineralization, and the study showed that anaerobes recycled
fermentation waste products during the long-term incubation, resulting in
a lower net residue C mineralization in flooded systems compared to nonflooded conditions. As a result, similar microbial production was observed
under flooded and nonflooded conditions even though anaerobes decomposed less straw C than aerobes. These results indicate that a significant
amount of decomposition occurred under flooded conditions, but because
substrate use eYciency was higher, less straw C was mineralized than under
aerobic conditions. Kinetic analyses of C mineralization curves confirmed
that C mineralized in the flooded treatment was mainly from labile pools,
with significant amounts coming from more recalcitrant pools, such as
cellulose and lignin depending on temperature.
Li and Lin (1993) reported that in the Wuxi province of China, decomposition rates of rice straw were similar under upland and submerged conditions. Drying and re-wetting conditions encountered under field situations
may also influence the decomposition of plant residues. Thus, Gestel et al.
(1993) found that soil drying and wetting increased the turnover of
C-labeled plant material. Nyhan (1976) summarized published data,
which showed that Q10 (the change in the rate of reaction for each 10 8C
change in temperature) for plant material decomposition averaged about 2.6
for temperatures from 12.5 to 40 8C. Later on, Howard and Howard (1993)
reported Q10 values from about 2.0 to 2.8 for soil respiration within the
10 to 20 8C temperature range. They also noted that the eVect was similar
across various soil moisture regimes.
Management Factors
The eVect of management is predominant in controlling the amount and
kind of plant residues returned to the soil and in determining the degree of
soil disturbance through tillage. Residues are managed diVerently; e.g.,
residues may be placed on the surface, mixed into the soil, or confined in
mesh bags within the soil. Surface placement or heterogeneous distribution
reduces the residue–soil contact as compared with a homogenous distribution. This may aVect the decomposition dynamics. Knowledge of such
eVects is important when results from diVerent studies are being compared
and is essential when developing and calibrating decomposition models. It
is also important when assessing eVects of tillage practices resulting in
diVerent degrees of residue–soil contact, e.g., no-till ploughing and rotovating. The degree of contact between crop residues and the soil matrix, as
determined by the method of residue incorporation, aVects decomposition
dynamics under both natural and experimental conditions.
Evidence from laboratory and field studies has suggested that the rates of
the decomposition of plant materials added to soil are proportional to the
amounts initially added (Larson et al., 1972). Generally, small amounts of
crop residues will decompose more rapidly than large amounts (Novak,
1974). However, Ladd et al. (1983) found the amount of 14C-labeled plant
material (Medicago littorallis) added was greater; the proportion of residual
organic C and N in soil was smaller.
Residue placement may be paramount in controlling the rate of crop
residue decomposition and nutrient cycling. Crop residue placement influences soil temperature, and water regimes indirectly influence microbial
activity and residue decomposition. Residue management practices that
involve intensive tillage and incorporation of residues increase residue decomposition rates and loss of SOM. Residue incorporation eVects are diYcult to separate from tillage eVects because incorporation is accomplished
through some type of tillage operation. Decomposition rates of incorporated
residues (rice and wheat straw) faster than those of surface residues resulted
from greater soil–residue contact, a more favorable and stable microenvironment, particularly soil moisture regime, and increased availability
of exogenous N for decomposition by microorganisms (Cogle et al., 1987;
Schomberg et al., 1994a). Important secondary eVects of tillage on the soil
microclimate include the influence of surface residue coverage on soil temperature, water interception and infiltration, and the eVects of tillageinduced changes in porosity and the eVects of soil structure on soil aeration
and water relations. Observed increases in organic matter with reduced
tillage systems are largely attributed to reduced decomposition rates of
surface residues compared to the rapid decomposition of incorporated residues (Schomberg et al., 1994b). However, the rate and degree of organic
matter accumulation associated with surface residues vary widely due to
diVerences in climate, soil type, and residue quality. Under conditions of
warm temperatures and increased water availability, SOM accumulation
from surface residues is reduced.
The depth of residue incorporation has also been shown to aVect
the decomposition of residues. Increasing the depth of residue incorporation from 50 to 200 mm resulted in a decrease in breakdown rate due to
less biological activity (Kanal, 1995). Likewise, the decomposition rate of
rice straw was reduced by 13% by increasing the depth of incorporation
of residues from 0–10 cm to 20–30 cm (V. Beri, Department of Soils, PAU,
Ludhiana, personal communication). In contrast, Breland (1994) found that
increasing the depth of incorporation up to 300 mm increased the decomposition rate of residues, due to more favorable moisture regimes in lower
layers. Puig-Gimenez and Chase (1984), however, reported that straw
decomposition was not aVected by the depth of placement.
Crop residues managed previously can also significantly aVect the decomposition of freshly applied crop residues. For example, Cookson et al. (1998)
observed that on a soil where wheat, barley, or lupin residues were previously managed for 3 years, wheat straw from the incorporated treatment had a
decomposition rate 50% greater than that from the burned or removed
treatment during 90 days.
Availability of Nutrients
Incorporation of crop residues increases the populations of all types of
macro- and microorganisms, and the new cells require all the essential
nutrients for their growth and activity. Residues of cereal crops such as
wheat, rice, barley, and maize have larger C:N ratios (low N contents) that
may require the addition of exogenous N in order for decomposition to
proceed. Mary et al. (1996) concluded that mineral N availability in soil is an
important factor controlling decomposition under field conditions. According to Jenkinson (1981), it is unusual for nutrients other than N to limit the
decomposition of plant materials in normal soils. Enhanced decomposition
rates of crop residues due to the application of N have been reported by
several workers (Bangar and Patil, 1980; Debnath and Sinha, 1993). In
contrast, no or even negative eVects of N added on decomposition and
microbial activity have also been reported (Cheshire and Chapman, 1996;
Clay et al., 1990; Hassink, 1994). The reduction in decomposition in the
latter studies was probably due to the adverse eVects of NH3 toxicity on
Soil Properties
Soil texture may aVect the decomposition of plant materials. Decomposition of crop residues was more rapid in soils with less clay content because
clay protected the organic matter from decomposition (Jenkinson, 1977;
Merckx et al., 1985; Ladd et al., 1996). As clay content increases, soil surface
area also increases which results in increased organic C stabilization potential. The role of clay in stabilizing organic matter appears to be more
important in warm soils, where decomposition occurs at higher rates.
Skene et al. (1997) showed that for high-quality substrates, physical protection by inorganic materials is a major limiting factor to decomposition,
whereas for low-quality substrates, chemical protection is the major limiting
factor. Hassink et al. (1993) reported that the texture influences the soil
physical environment, which further aVects microbial activity. Li and Lin
(1993) also stated that soil properties controlled the decomposition rates of
crop residues during the first 2 years at two diVerent locations in China. On
the other hand, Amato et al. (1987) could not observe a significant eVect of
soil properties on decomposition of wheat residues.
Soil pH aVects both the nature and the size of the microbial population,
both of which ultimately aVect the residue decomposition (Paul and Clark,
1989). In general, decomposition of crop residues proceeds more rapidly in
neutral than in acid soils. Consequently, the treatment of acid soils with lime
enhances the decomposition of plant residues (Condron et al., 1993). Likewise, soil salinity also aVects the residue decomposition through its direct
influence of osmotic potential on microbial activity or through alterations of
pH, soil structure, aeration, and other factors (Nelson et al., 1996). In the
absence of pH and aeration eVects, sodicity increased and salinity decreased
the decomposition of plant residues, but with no significant interaction
(Nelson et al., 1996). Jenkinson (1971) reported that the decomposition
pattern is not substantially aVected by soil properties or crop residue type
included in the study, and the patterns obtained in diVerent climates can
essentially be superimposed by using an appropriate rate constant factor.
Using the buried mesh bag technique, Henriksen and Breland (2002)
found that soil type had no eVect on decomposition of the easily degradable
clover residues, but cumulative C mineralization of barley straw after 52
days was less in the subsoil than in the topsoil by 12% of initial C. This study
showed the necessity of ensuring realistic conditions for the decomposer
microflora when studying eVects of substrate quality on decomposition
and when extrapolating the results for hemicellulose rich residues to field
conditions. The diVerences in decomposition as aVected by residue–soil
contact in laboratory experiments could be due to soil treatment before
incubation, for example, sieving, which seems to damage fungi in particular.
Thus, the diVerences may not be distinct under natural field conditions with
an undisturbed microflora.
In a traditional single-crop rice system, fallow periods between crops are
long and the soil is allowed to dry, resulting in re-oxidation of reduced
substances, more complete decomposition of added organic matter, and
large shifts in microbial communities. Similar changes occur in rice followed
by wheat or other upland crop systems. In contrast, in a double- or triplecrop rice system, fallow periods are short, the soil is not allowed to dry and
re-oxidize completely, and large amounts of crop residues are returned to the
field more than only once per year. Fallow periods may range from very dry,
to dry-wet, to completely wet conditions and diVer in their length, but only
under drier conditions did a shallow early tillage accelerate soil N mineralization during early growth stages of the rice crop grown thereafter
(Dobermann and Witt, 2000). It is therefore likely that years of intensive
rice–rice cropping lead to a decline in the steady-state soil redox potential
along with a gradual accumulation of reduced substances (Fe and organic
compounds), a change in the qualitative composition of SOM toward more
phenolic compounds, and possibly even an increase acidification of the
rice rhizosphere resulting from greater root-induced Fe2þ oxidation. Unfortunately, we lack long-term experiments with rice-based systems of diVerent
cropping intensity in which such subtle changes with time have been
measured over periods more than 10 years.
In systems in which declining SOM content is a concern (rice-upland
crop), straw incorporation may be the only choice for maintaining or increasing SOM content. In rice–rice (-rice) systems with short aerobic periods, organic matter management must focus on managing the quality of
soil organic matter, that is, avoiding the accumulation of highly complex
organic matter with slow N mineralization rates. In this case, the time of
incorporation of organic materials is more important than the amount.
Compared with the traditional method of wet incorporation shortly
before planting of the next rice crop, the potential benefits of shallow
incorporation of the rice straw shortly after the harvest of rice include
(1) accelerated aerobic decomposition of crop residues (about 50% of the
carbon within 30–40 days), leading to increased N availability (Witt et al.,
1998) and reduced CH4 emission; (2) re-oxidation of ferrous iron and other
Figure 1 EVect of decomposition period on the mass remaining of litter bag rice residue.
reduced substances, leading to increased P availability; (3) reduced weed
growth; and (4) savings in irrigation water during land soaking for rice by
reducing cracking and bypass flow water losses in heavy clay soils. Early
incorporation of residues allows additional time for phenol degradation to
occur under aerobic conditions, thus possibly altering the rates of soil
organic matter formation and subsequent decomposition.
Using the litter bag technique, Yadvinder-Singh et al. (2004b) observed
that time of incorporation had a large eVect on the decomposition of rice
residue during the fallow phase (October–November) after rice harvest
(Fig. 1) in rice–wheat rotation in Punjab, India. At wheat seeding, the
mass loss of rice residue was 51% for a 40-day decomposition period,
compared with 35% for a 20-day decomposition treatment and 25% for a
10-day decomposition treatment. The amount of mass loss remained significantly higher for the 40-day decomposition period compared to the 10-day
or 20-day period up to 72 days after seeding of wheat. At the end of the
study, no significant diVerence, however, was noted among the three
The data for all three decomposition periods could be best described
using a single logarithmic equation:
Y ¼ 14:9 lnðX Þ
R2 ¼ 0:951 ðn ¼ 15Þ;
where Y is the total decomposition period (days) and X is the percent mass
loss of rice residue. Equation (4) may prove useful in predicting residue
decomposition under soil and environmental conditions similar to those in
this study. Rice straw decaying for 0–10 days lost 2.45% of its initial mass
each day. In comparison, the values were 1.0% day 1 for 10–20 days and
0.8% day 1 for 20–40 days during fallow.
The N concentration of the rice residue increased continuously during
the 190-day decomposition period, indicating loss of C as CO2 and/or N
immobilization in the residue by microorganisms, which build up new microbial protein from plant and soil N. The N content of rice straw at the time of
incorporation was 5.6 g kg 1, which increased to 14.8 g kg 1 at the end of the
study period. At the time of wheat sowing, residue N was significantly
lower (7.0 g kg 1) in the 10-day decomposition treatment than in the 20-day
(7.9 g kg 1) or 40-day (8.1 g kg 1) treatment. At the end of the study (150 days
after wheat sowing), the N concentration in rice residue was observed to be
similar (14.8 g kg 1) for 10-, 20-, or 40-day decomposition treatments.
Yadvinder-Singh et al. (2004b) observed substantial immobilization of
fertilizer N with straw incorporation at 10 days after fertilizer application in
treatments in which rice straw was incorporated at 0 and 10 days before
application of fertilizer compared to the no-straw treatment (Fig. 2). The
magnitude of immobilized N was influenced by the decomposition period of
rice straw prior to fertilizer application. Interestingly, immobilization of N in
the treatment where fertilizer N was applied concurrently with straw incorporation (0 day) always remained lower than the treatment without straw (Fig. 2).
Mineral N in the soil was significantly higher under the 20- and 30-day
pre-decomposition periods than under the no-straw treatment at all sampling
times. These data clearly demonstrated that incorporation of rice straw at
20 days or more before wheat sowing will minimize any adverse eVects on
crop growth due to N immobilization after straw incorporation (Fig. 2). This
Figure 2 EVect of pre-decomposition period of rice straw on mineral N (NH4 þ NO3)
dynamics in soil amended with 100 mg N kg 1 and incubated at 75% field capacity moisture
regime at 30 8C.
study suggested a lower amount of N immobilization by rice straw containing
6.7 g N kg 1 when allowed to decompose for 20 or 30 days before fertilizer
N application compared with that reported for wheat and barley straw by Mary
et al. (1996). The results from the decomposition and N mineralization studies
suggested that rice residue is likely to have little adverse eVects on N availability
in the soil when it is allowed to decompose under aerobic conditions for at least
10 days before sowing of the next upland crop.
Rice straw is characterized by a high C:N ratio and abundant K, Si, and
C (Ponnamperuma, 1984). Wheat straw has comparable properties except for
low Si and low K concentration. The successful utilization of crop residues as
a nutrient source relies on manipulating the biological processes in the soil so
as to optimize nutrient availability with respect to plant demand.
A simplified model of the regulation of nutrient flux in the agoecosystem is
presented in Fig. 3. This conceptual model depicts the flow of carbon and
nutrients among organic residues, organic and inorganic pools in soil, and the
plant. Pathways of loss are also included. Decomposition and mineralization
of plant residue are mediated by both soil faunal and microbial populations.
Some of the carbon and associated nutrients are mineralized immediately
(pathway 1a) or are immobilized in the soil microbial pool (pathway 2a), later
to be transformed into other soil organic pools via microbial by-products
(3a). Recalcitrant plant material also may enter the soil organic pools directly
(3b). The carbon and nutrients held in the various soil organic matter pools
are subsequently decomposed and assimilated by soil biomass, resulting in
Figure 3 Conceptual model of nutrient pathways in crop residue amended soils (Myers
et al., 1994).
additional mineralization (1b). The inorganic nutrients released by mineralization may be assimilated by soil biota via immobilization (2). Immobilization
occurs simultaneously with mineralization, and the rate at which nutrients are
available for plant uptake depends on the net balance between mineralization
(1a plus 1b) and immobilization (2). The inorganic nutrients may also be taken
up by plants (pathway 3), lost by leaching or volatilization (pathway 4), or
remain in the soil (Myers et al., 1994). The size of the inorganic pool depends
on the balance of the various processes that add to the pool (mineralization)
and those that subtract (immobilization, plant uptake, and losses).
The proportion of N transferred from the residue to the plant and the rate
at which it occurs are determined by the balance between the rates of the
various processes represented by these flux pathways. This balance is regulated by a hierarchy of factors. Environment, which includes climate and
soil, is an overriding control and determines the rate of the transfer between
pools. The rates also vary depending on the quality of the decomposing
substrate. By manipulating the quality of crop residues, it should be possible
to manage nutrient release to coincide with the time course of the nutrient
requirements of the crop (Swift, 1987). When low-quality crop residues (low
N and P, high lignin or polyphenol contents) are incorporated into the moist
soil, nutrients become available to the plants. With high-quality residues,
nutrients are initially released rapidly in excess of plant demand with a risk
of nutrients such as N being lost via leaching or denitrification or a nutrient
such as P becoming chemically unavailable (Anderson and Swift, 1983).
1. Kinetics of Nitrogen Mineralization–Immobilization
Mineralization and immobilization of N occur simultaneously in the soil.
Net rates of mineralization and immobilization are an integration of a
number of soil N processes and a number of factors, which act on the
interacting processes as well as directly on mineralization–immobilization.
Residues from rice, wheat, barley, maize, and other small grains with large
C:N ratios are noted for their initial N immobilization, which can negatively
aVect crop yields (Cassman et al., 1997; Christensen, 1986). Microbes
using the crop residues as an energy source compete with crop plants for
available N.
The addition of cereal residues often results in a net N immobilization
phase followed by a net re-mineralization phase (Müller et al., 1988; Nieder
and Richter, 1986; Robin, 1994; Yadvinder-Singh et al., 1988). Generally,
mineralization of N from low-N residues occurs only after 50–60% of
residues is decomposed or after the C:N ratio is below 30 (Christensen,
Table IV
Maximum Immobilization of Soil Mineral N Measured During Decomposition of Wheat/Rice
Straw under Laboratory Conditions
Simon (1960)
Guiraud (1984)
Reinertsen et al. (1984)
Bakken (1986)
Nieder and Richter (1986)
Robin (1994)
Beri et al. (1995)
Immobilized N (mg N g
Without N.
With mineral N.
Modified from Mary et al. (1996).
1986). A large number of papers have focused on N immobilization immediately after straw incorporation under laboratory conditions and indicate a
narrow range of N immobilization values (Table IV). Under field conditions,
Mary et al. (1996) derived N immobilization values of 13.0, 24.3, and 32.0 mg
N g 1 added carbon under N0, N180, and N330 treatments, respectively. The
latter value was equal to the immobilization potential found with straw in
incubation studies. Mishra et al. (2001a) reported that the N content of
wheat straw increased from the initial level of 0.50% to 0.73% after 10 weeks
and to 0.88% after 22 weeks. The C:N ratio of wheat straw decreased from
90 to as low as 32 by the end of 22 weeks of decomposition period. During
first 2 weeks, 24% of the total N in wheat straw was released into the soil.
There was net immobilization between 3 and 10 weeks. After 10 weeks, net
N mineralization was noted in the study. By 22 weeks, apparent
N mineralization from wheat straw was 71%. The rate of N mineralization
was lower than that of C mineralization from wheat straw. It suggested that
incorporation of wheat straw at about 8 to 10 weeks before rice transplanting can help to alleviate the adverse eVect of wheat straw on rice growth due
to N immobilization. In another study, Mishra et al. (2001b) noted that
during the decomposition of rice straw, N content increased from 0.60 to
1.21% and the C:N ratio decreased from 70 to 21.3 with time. About 4.5% of
the total N present in rice straw was released within 5 weeks of its incorporation into the soil, and apparent N mineralization after 23 weeks was 57.5%
of the total N in the rice straw, of which more than 50% was mineralized
during the third phase.
Similarly, Yadvinder-Singh et al. (2004b) reported that the N content of
rice straw at the time of incorporation was 5.6 g kg 1, which increased to
14.8 g kg 1 at the end of the 190-day study period. At the time of wheat
sowing, residue N concentration was significantly lower (7.0 g kg 1) in the
10-day decomposition treatment than in the 20-day (7.9 g kg 1) or 40-day
(8.1 g kg 1) decomposition treatment. Rice residue appeared to have a brief
initial period of N release after incorporation. This study showed that for
every 10% increase in mass loss, there was about 2.75% (1.07 kg N ha 1)
release from the applied residue N. In this study, despite a substantial mass
loss of 69% residue, the amount of N release was small (6–9 kg N ha 1).
Residues go through several phases in their decomposition, with N dynamics
related to the stage or extent of mass loss. Using 15N-labeled rice residue,
Yoneyama and Yoshida (1977a) reported that 8% of the N in the rice leaf
sheath (8 g N kg 1) was mineralized in 30 days at 30 8C under upland
conditions in the laboratory.
In the tropics, the turnover of organic matter in flooded soils can be as
fast as that in aerobic soils. This causes net immobilization of N in flooded
rice soils after rice or wheat straw incorporation. A review of work on
interaction between decomposition of plant residues and N cycling in soils
showed that the amount of N immobilization can be large, and the intensity
and kinetics of N immobilization and subsequent mineralization depend on
the nature of plant residues and the type of associated decomposers (Mary
et al., 1996).
In a laboratory incubation study, rapid production of NHþ
4 –N occurred
during the first week of incubation of untreated flooded soils, while in a soil
treated with rice straw only traces of NHþ
4 –N were formed (Phongpan,
1987). However, during the subsequent 6 weeks, the accumulation of
4 –N was considerably higher in soil treated with rice straw than in
untreated soil. This suggests that NHþ
4 –N was immobilized in the soil
amended with rice straw during the early stages followed by remineralization
in later stages. Similar observations have been made by Nishio et al. (1993),
who estimated a remineralization constant of 0.023 day 1 when 0.25 mM of
ðNH4 Þ2 SO4 was applied along with rice straw. Azmal et al. (1997) observed
that in rice straw amended soil (200 mg C per 100 g soil applied every 6
weeks) mineral N was immobilized immediately after each application of
rice straw due to its high C:N ratio. The immobilization increased until the
third application. This behavior was ascribed to the gradual accumulation in
soil of a portion of undecomposed organic N. Nieder and Richter (1989)
observed immobilization of about 30–40 kg N ha 1 in the soil after straw
Toor and Beri (1991) observed almost compelete immobilization of native
as well as applied N (120 mg kg 1) by the seventh day in a soil amended with
rice straw. At the end of 60 days of incubation, however, about 40 mg
mineral N kg 1 soil was remineralized. Patel and Sarkar (1993) observed
initial rapid rate and magnitude of immobilization of 15N-labeled urea in
three soils amended with wheat straw, and remineralization of immobilized
N occurred by 60 days. Singh et al. (1992) also found that about half of
N immobilized by rice straw was remineralized in 90 days after straw
incorporation. Yadvinder-Singh et al. (1988) observed that remineralization
of immobilized N in a soil amended with rice and wheat straw occurred
4 weeks after incorporation of residues at 35 8C under laboratory conditions.
Using 15N-labeled rice straw, Zhu et al. (1988) reported that about 1% of the
rice straw N was mineralized in 112 days.
Kawaguchi et al. (1986) observed that eventual mineralization of
rice straw N during 28 weeks of incubation period ranged from 17 to 24%.
Addition of rice straw also increased the mineralization of soil N due to a
priming eVect, which was estimated to be 30–100 mg N kg 1 soil. The degree
of the priming eVect correlated with the amount of mineralized rice straw N.
Incorporation of crop residue with low cellulose and high water-soluble
N contents results in more N being mineralized than had been added in
the residues, demonstrating this priming eVect (Bending et al., 1998).
Using 15N-labeled rice and maize straw, Wang et al. (2001) reported
that after a 112-day incubation period under submerged conditions, net
recovery of mineralized N from soil only accounted for 1.0–1.3% of the
straw N added to diVerent soils. About 2.0–4.3% of straw N was immobilized by microbial biomass, and 0.2–14.2% was fixed by clay minerals.
Vertisol fixed a markedly higher amount of straw N than Ultisol. The loss
of straw 15N ranged from 29.7 to 46.3% during the incubation period. The
total amount of straw N mineralized during the 122-day incubation ranged
from 47.7 to 51.7%.
Kushwaha et al. (2000) reported that in residue removal treatment,
N mineralization rates were maximal during the seedling stage of crops
and then decreased through the crop’s maturity. In residue-retained treatments, however, N mineralization rates were lower than in the residueremoved treatments at seedling stage of both crops. At grain-forming
stage, the N mineralization rates in residue-retained treatments considerably
exceeded the rates in corresponding residue-removed treatments. Microbial
immobilization of available N during the early phase of crops and its pulsed
release later during the period of greater N demand of crops enhanced the
degree of synchronization between crop demand and N supply.
Using the first-order kinetics relationship, Mary et al. (1996) calculated
the maximum net immobilization at 51 kg N ha 1 and a rate constant of
0.031 day 1 in soil amended with wheat straw under field conditions. Islam
et al. (1998) estimated N mineralization potential (N0) under flooded conditions at 35 8C for 12 weeks. The values of No for rice straw were 7 to 15 times
that of the control soil. The first-order rate constants for unamended soils
ranged from 0.35 to 0.52 mg N kg 1 week 1, and for rice straw-amended
soil, from 0.75 to 1.22 mg N kg 1 week 1.
Dynamic computer models can be used to distinguish between immobilization and mineralization of N and between availability and loss through
leaching and denitrification in soils treated with crop residues under field
Factors Affecting N Mineralization
The residue quality and availability of soil N are important determinants
of N mineralization–immobilization occurring during residue decomposition. Mineralization of organic N depends on the N requirements of the soil
microbial population, the biochemical composition of the decomposing crop
residue, and several soil and environmental factors. Crop residue management can aVect N immobilization and stabilization processes important to
eYcient utilization of N from fertilizers, crop residues, and soil organic
matter. The availability of nutrients from crop residues depends to a great
extent on mineralization of nutrients from the crop residues in relation to
crop demand. Nitrogen mineralization is a crucial process of nutrient dynamics in the soil–plant system. It has been reported that the larger and
sustained microbial biomass found under flooded compared to aerobic
conditions may act to immobilize more N and make it less available for
plant uptake. Together with N losses, such as denitrification and leaching,
sustained flooding of soil may reduce the size of available N pools, as seen in
some areas of the tropics (Cassman et al., 1995).
a. Crop Residue Quality. It is well established that the chemical
composition of crop residue influences its rate of decomposition and
mineralization–immobilization in the soil and subsequent N uptake by the
crops (Jensen, 1997; Magid et al., 1997; Mary et al., 1996). The quality of
organic residues has been assessed by measuring diVerent biochemical properties that have been shown to delay or enhance the decomposition and
N mineralization processes. Incorporation of residues with low N contents,
such as rice, wheat, and barley, may result in microbial immobilization of
soil and fertilizer N, eVectively reducing N availability to plants.
Using 15N-labeled crop residues, Norman et al. (1990) found that N
mineralized from the time of residue incorporation until the rice harvest
from rice, wheat, and soybean straw was 9, 38, and 52%, respectively. The
respective amounts of residue N recovered by rice was 3, 37, and 62%
from those residues. Several workers have reported higher rates of N mineralization from crop residues with a lower C:N ratio and higher total
N content (Bending et al., 1998; Vigil and Kissel, 1991). Regression analysis
of the data showed that 72–75% of the variability in the measurement of
N mineralization from crop residues can be explained by using either the
C:N ratio or the square root transformation of the N concentration of
residues (Vigil and Kissel, 1991). The break point between net N mineralization and net N immobilization was calculated to occur at a C:N ratio of 40,
which corresponds to 10 g N kg 1 residue (assuming residue carbon as 400 g
kg 1). The prediction of N mineralization was improved when regression
analysis included total N and the lignin:N ratio as independent variables.
The fitted equations provided an estimate of the maximum amount of N that
will potentially mineralize in a season from the incorporation of crop residues of diVerent N contents. Generally, mineralization of N from low-N
residues occurs only after 50 to 60% is decomposed or after the C:N ratio is
below 30.
But others have found quite diVerent results (Haynes, 1986). It is to be
expected that the significance of diVerent chemical parameters will depend
strongly on the condition (temperature and soil moisture content) under
which the experiments were conducted, the range of decomposability and
N concentration in the plant material, the N availability in the soil, and
whether the shorter or longer results are considered. When plant materials
with very diVerent decomposabilities are compared, lignin content is likely
to be an important parameter, as reported by Neely et al. (1991), whereas
when comparing young plant materials from young green manures it is
unlikely to show much relation to N release.
Polyphenol compounds have been shown to react with residue N and
render it unavailable for plant uptake (Palm and Sanchez, 1991), while lignin
is a not easily decomposable cell wall polymer. Azmal et al. (1996) observed
a negative relationship between N mineralized from crop residues and
the amount of cellulose and hemi-cellulose of added organic material. So
far, particle size has not been considered a quality parameter, but needs
attention, as a smaller particle size leads to higher substrate-decomposer
community contact, resulting in high decomposition and N mineralization.
Wide variation in the proportion of various constituents in crop residues
and their diVerential behavior in the soil after incorporation makes the
description of N release solely based on the C:N ratio of the residue too
simplistic (Jarvis et al., 1996). In a flooded soil, net N mineralization from
crop residues (legumes as well as cereals) was not correlated with N and the
C:N ratio, but it was correlated with the lignin:N ratio of the residue under
greenhouse conditions (Becker et al., 1994b). Similarly, initial soil NHþ
4 –N
accumulation rates under field conditions were higher from residues
with a relatively low lignin:N ratio, suggesting that the lignin:N ratio of
the applied residue may be a suitable index for predicting N mineralization
rates in flooded soils. Quality components controlling N mineralization from
crop residues also change during decomposition (Bending et al., 1998).
While water-soluble phenolic content significantly correlated with net
N mineralization at early stages, C:N ratios and total N content were
correlated with net N mineralization toward the end of incubation
(6 months) only. In a field experiment, Clement et al. (1995) observed that
immediately after incorporation into wetland rice soil, N mineralization was
positively correlated with crop residue N content. However, at tillering, the
tannin:N ratio was best correlated with the rate of N release. Grain yield was
best predicted by the (lignin þ polyphenol):N ratio. This study emphasized
the importance of the interaction among chemical constituents of crop
residues in the dynamics of N release and uptake by rice.
b. Environmental Factors. Nitrogen mineralization is profoundly influenced by temperature changes that are normally encountered under field
conditions. The majority of soil microorganisms are mesophyllic and prefer
moderate temperatures with optimum activity between 25 and 35 8C. Thus,
turnover of nutrients in plant residues would generally be accelerated in
tropical soils. Pal et al. (1975) suggested that the initial Q10 (2–3 days) for
N mineralization of crop residues recently incorporated into soil is > 2. Vigil
and Kissel (1995) reported that measured Q10 for N mineralization depended
on the C:N ratio of the residue and incubation time, indicating that for
predictive purposes a single Q10 value is inadequate for describing the eVect
of temperature on crop residue N mineralization. Honeycutt and Potaro
(1990) field tested the application of thermal units for predicting N mineralization from crop residues. It was found that thermal units are valid for
predicting commencement of net mineralization of N from crop residues,
despite the harsh environmental conditions and wide temperature variations
to which these residues and soils were subjected.
Nitrogen transformations in flooded soils under rice are markedly diVerent from those taking place in upland soils. The diVerence in the behavior of
N in upland and submerged soils is due to the diVerence in activity of
microorganisms functioning under aerobic and anaerobic conditions. In an
incubation experiment, Yoneyama and Yoshida (1977b) found that net
mineralization of soil N was depressed by the addition of rice straw, except
that the addition of leaf blade under lowland conditions gave more mineral
N at later stage than the unamended control (Table V). Under lowland
conditions, the amount of N immobilized in soil amended with rice straw
was small during the first week but increased substantially after 2 to 3 weeks.
Under upland conditions, the immobilized N reached its maximum during
the first week, but the amount of N immobilized was smaller than that under
lowland conditions. At 30 days of incubation, 26, 20, and 17% of total
N under lowland conditions and 14, 7, and 8% under upland conditions
were mineralized in leaf blades, stems, and leaf sheaths, respectively. This
suggests that mineralization of rice residue N takes place throughout the
decomposition of residue even if the net mineralization of N was not observed by the incorporation of residue low in N content. The amount of
absorbed soil N in rice residue (influx) and the remaining original rice
Table V
Mineralization of Rice Residue in Soil under Upland and Lowland Conditions
Rice residue
Incubation time (days)
Total N
(mg kg 1)
Lowland conditions
Leaf sheath
Leaf blade
Upland conditions
Leaf sheath
Leaf blade
From Yoneyama and Yoshida (1977b).
residue N (outflux) is more vigorous under lowland than under upland
conditions. Therefore, rice yields and N uptake will be greater under lowland
than under upland conditions. Kanazawa and Yoneyama (1980) observed
that mineral N in soil amended with 15N-labeled rice straw under upland
conditions remained at low levels compared with untreated control throughout the 24 months of incubation. In flooded soil, the mineral N was lower in
straw-treated soil during the first 4 months, and the diVerences were small
between unamended and amended soil thereafter.
It is generally found that N mineralization is higher under anaerobic
conditions than under aerobic conditions (Ono, 1989). According to Liu
et al. (1996), higher mineral N levels in rice straw amended soil under
anaerobic compared to aerobic conditions possibly occurred because the
minimum need of microorganisms for release of ammonium N from crop
residues in flooded soil is about 0.5% compared with 1.7% in aerobic
systems. Thus, inorganic N is released in larger quantities in anaerobic
than in aerated soils, although the release rate may be slower. Mineralization
rate of N in crop residues is reduced at low soil water contents.
The eVect of soil temperature and water content on N mineralization can
be calculated by using a relationship derived by Andren and Paustin (1987).
A normalized time (equivalent to Qsum) was calculated as
TðnormalizedÞ ¼ tðrealÞ fðTÞ gð*Þ;
where f(T) is a correction factor due to soil temperature and g(*) is a
reduction factor due to soil water potential. The factor f(T) is a multiexponential function of temperature. It is set at 1 at a temperature of
25 8C. The eVect of soil moisture on N mineralization was described as
an exponential function of soil water potential. The combined eVect was
calculated as the product of the two terms, assuming that there were no
interactions between temperature and moisture. This approach enables the
comparison of field experiments diVering in climatic conditions to laboratory experiments conducted under constant temperature and moisture. Mary
et al. (1996) found that when the normalized days were substituted to
real days, the diVerence in the kinetics of net N immobilization and
C decomposition in the soils where wheat straw was incorporated under
field conditions were not significant between two years (Fig. 4).
c. Placement of Crop Residues. Using 15N-labeled crop residues, Smith
and Sharpley (1990) found that surface placement of residues reduced
N availability as compared to soil incorporation, but the diVerences
were only equivalent to 1 to 7 kg N ha 1. Residue placement influences
N mineralization through an eVect on the microclimate of the residue.
Slower decomposition rates of surface residues may result in greater potential for immobilizing N for longer periods than for incorporated residues.
Schomberg et al. (1994b) reported that N immobilization period was longer
than 1 year for surface applied wheat and sorghum residues and about
4 months for buried residues. The maximum value for N immobilization
was 50% lower for buried residues. Although greater N immobilization may
occur with surface residues, subsequent N mineralization can occur within a
period that is optimum for crop utilization.
Residue incorporation with conventional tillage agroecosystems can
be characterized as bacterial-based food webs with fast rates of litter
Figure 4 (A) EVect of wheat straw on net immobilization of soil mineral N versus ‘‘normalized’’ time. (B) Wheat straw C (fraction > 1 mm) remaining in soil in the field experiments versus
‘‘normalized’’ time (Mary et al., 1996).
decomposition and nutrient mineralization, while surface residues under no
tillage systems support fungal-based food webs that result in slower decomposition and greater nutrient retention (Beare et al., 1996). Placement of
residues may play an important role in determining availability of soil N to
subsequent crops during the N immobilization-mineralization process.
d. Soil Type. Soil texture controls mineralization by (1) influencing
aeration/moisture status, (2) aVecting the physical distribution of organic
materials and hence potential for degradation, and (3) conferring some
degree of ‘‘protection’’ through an association of organic materials with
clay particles (Hassink et al., 1993). Becker et al. (1994b) observed that
residue N release in clayey soil was approximately twice that of sandy soil.
DiVerences in mineralization rates between soils would have an impact on
the fertilizer N requirement of the subsequent crop and the potential for
N loss due to leaching or denitrification. Using 15N-labeled wheat straw and
legume residues, Amato et al. (1987), however, observed that eVect of soil
properties and climate on the residual organic 15N was small. Decomposition and mineralization of crop residues, however, are inhibited under
strongly acidic conditions. For example, Fu et al. (1987) indicated that
N mineralization increased as soil pH increased from 5 to 7.
e. Soil and Fertilizer Nitrogen. Cereal residues generally possess low
N content and may require addition of exogenous N for decomposition to
proceed. From a series of experiments, Yoshida et al. (1973) inferred that
N mineralization in soil amended with rice straw increased with increasing
soil, but N mineralization
4 –N concentration up to 300 mg N kg
decreased when rates greater than 300 mg N kg 1 were applied. Mary et al.
(1996) concluded that immobilization intensity of crop residues expressed
per unit of mineralized carbon is reduced and N remineralization is delayed
in soils with low mineral N concentrations. Nitrogen availability in soil can
therefore strongly modify the mineralization–immobilization kinetics by a
feedback eVect. On bare plots, immobilization of mineral N by wheat straw
incorporation increased markedly by the addition of mineral N throughout
the decomposition. A better prediction of the evolution of mineral N in
soil may, therefore, require description and modeling of the respective
localization of both organic matter and mineral N in soil aggregates.
Effect of Crop Residues on Utilization of N by Crops
Availability of N from crop residues to subsequent crops is highly dependent on decomposition rate, residue quality, and environmental conditions
(Fox et al., 1990). Application of crop residues has been shown to depress
the NHþ
4 –N concentration in soil and flood water due to N immobilization
and consequential lower N uptake by rice compared with control (Huang
and Broadbent, 1988; Nagarajah et al., 1989). Rice is known to take up more
organic N than any other crop because (1) it takes up NHþ
4 , amino acids, or
relatively high molecules of organic N, preferentially; (2) it has stronger
activity in competing with soil microorganisms than the other crops; (3)
it secretes organic substrates that support multiplication of microfauna,
resulting in rapid decomposition of organic residues; and (4) it has superior
Km (Michaelis constant), Vmax (maximum uptake velocity), and Cmin
(minimum concentration of a nutrient) for N uptake (Yamagata et al.,
1996). Thus, rice is expected to respond to crop residue N better than
other crops. Yoneyama and Yoshida (1977a) found that N uptake by rice
from residue-amended soil was at its peak during the intermediate stages of
growth, and N uptake from the fertilizer was rapid during early growth.
They recorded 25% N recovery from straw N by rice plants in 130 days.
Although contribution of 5 t straw ha 1 to the current N needs of rice is
relatively small, the long-term eVects may be substantial. For example,
Tanaka (1974), Chatterjee et al. (1979), and Kosuge and Zulkarnani (1981)
reported that continuous application of straw builds up soil organic matter
and ensures high N content and uptake and partial substitution of straw
N for fertilizer N.
Jiang et al. (1998) observed that N utilization by wheat in the presence of
wheat straw (4.5 t ha 1) was highest when N was applied in three equal splits
at sowing, tillering, and stem elongation. Guirad and Berlier (1971) reported
that the reduction in the N uptake from Ca(NO3)2 in the wheat strawamended plots was due to higher losses of NO3–N by denitrification, and
from (NH4)2SO4 it was caused by immobilization of N in the soil.
Malik et al. (1998) found that incorporation of wheat straw along
with green manure enhanced nutrient availability; and synchrony between
N release and plant uptake was best achieved in soil receiving straw
along with green manure. A temporary lag in N immobilization and
mineralization provided a N-conserving mechanism for the system. Broadbent and Nakashima (1965) followed mineralization and plant uptake of
N immobilized by application of straw. When N was added with the straw,
there were indications that remineralization of immobilized N was faster
than mineralization of N in the unamended soil. However, when no N was
applied with the straw, the results did not support the synchrony concept.
Support for the synchrony concept is found in the results of a field experiment with flooded rice (Amarasiri and Wickramsinghe, 1988) in which rice
receiving a 60 kg N ha 1 fertilizer along with straw yielded about the same as
that receiving 90 kg N ha 1 as fertilizer alone. This role of straw may be
interpreted as one of N recycling in a system where losses from the mineral
N pool are potentially large and as such is a type of synchrony.
Tanaka and Nishida (1996) observed that wheat straw decreased N uptake
by rice and increased the amount of 15N remaining in the soil at 17 days after
transplanting. At the booting stage, 6 days before heading, N uptake was
higher and the 15N remaining in the soil was lower in the treatments in which
wheat straw was applied than in unamended control treatment. It was
concluded that decrease in N uptake by wheat straw was caused by
N uptake inhibition and not by N deficiency in the early stages of rice
In a greenhouse study using 15N-labeled fertilizer, Masayna et al. (1985)
found that rice plants recovered 50–69% of applied fertilizer in the unamended soil and 45–53% in the rice straw-incorporated soil. In the second and
third crops of rice, recovery of residual N was slightly higher from rice strawamended soil than from unamended soil. Islam et al. (1998) found that large
amounts of mineral N pools were lost during the incubation that could not
be accounted for by microbial immobilization under field conditions. To the
contrary, Xu (1984) reported higher fertilizer utilization eYciency in rice
straw-amended soil (75.5 and 82.6%) than that in unamended light clay and
sandy loam soils (51.8 and 47.7%). This could be due to the increased
N immobilization and decreased losses of N via denitrification in residueamended soil (Craswell, 1978). Available data suggest that 10 to 20% of
N freshly supplied through cereal residues with a high C:N ratio (rice and
wheat straws) is assimilated by the rice crop, 10 to 20% is lost through
various pathways, and 60 to 80% is immobilized or stored in the soil under
field conditions (Koyama, 1981).
Losses of N
The presence of crop residues with high C:N ratios may also lead to
transformation of fertilizer or soil N into slowly available forms, which
may act as slow-release fertilizer and thereby improve N use eYciency.
Bird et al. (2001) reported that the total loss of N fertilizer, based on the
N isotope balance, was approximately 50% and was largely independent of
straw management practice. An increase in total soil microbial biomass in
combination with a large amount of added straw could have led to a
temporary strong sink for N fertilizer. The ensuing immobilization process
could lead to lower N fertilizer losses. Eagle et al. (2001) reported a decrease
in fertilizer N use eYciency with a concomitant increase in the plant available soil N following change in straw management from burning to incorporation. Only 1.8 kg ha 1 (3.5%) straw N was directly available to the
crop in the year following incorporation, and total N uptake increased by
23 kg N ha 1 5 years after straw incorporation. Huang and Lu (1996)
reported that heavy application of rice straw in combination with
N fertilizer at a C:N ratio greater than 40 would have a determental eVect on
the rice growth. Using 15N–labeled fertilizer, it was observed that total
recovery of N was reduced from 40.8% in no straw to 6.1% in straw
treatment with a C:N ratio of 40, but the total N loss was decreased from
13.7 to 5.5%. It was concluded from this study that for eYcient management
of rice straw and N fertilizer in flooded rice cultivation, it is advisable to
incorporate rice straw with a C:N ratio adjusted to <25.
a. Urea Hydrolysis and Ammonia Volatilization. Urea, the major fertilizer N source, can be significantly less eYcient when used under soil and
climatic conditions conducive to NH3 volatilization. The rate of urea hydrolysis, a process mediated through soil enzyme urease, has a direct bearing on
losses of N via NH3 volatilization. The incorporation of crop residues causes
a significant increase in the soil urease activity (Gill et al., 1998b; Phongpan,
1987). Khind and Bajwa (1993) also observed that urea hydrolysis proceeded
more rapidly in the crop residue-amended soil than in the control soil. The
rate of urea hydrolysis increased with the increasing rate of crop residue and
with the length of decomposition period. The first order rate constants for
urea hydrolysis ranged from 0.021 to 0.024 h 1 after the application of
200 mg N kg 1 in the unamended soil and from 0.071 to 0.250 h 1 in the
rice straw-amended soil, depending upon the length of decomposition period. Gill et al. (1998a) reported that in wheat straw-amended soil, hydrolysis
of urea was completed in 6 days, compared to 12 days in the unamended soil
under flooded conditions.
The increased rate of urea hydrolysis on residue-amended soil may lead to
increased loss of N via NH3 volatilization. In a silt loam soil, application of
soybean straw increased the urea hydrolysis from 42.8% in control to 90.0%
2 days after incubation. The ammonia loss in the first 4 days was 12.3 and
28.2% of the applied N in the control and residue-amended Guthrie soil,
respectively (Caramona et al., 1990). The cumulative N losses were, however, similar (45 and 44%) under the two treatments. McInnes et al. (1986)
found that cumulative loss of NHþ
4 –N was 7.6 and 16.6% of the N applied
from unamended and wheat-straw amended soil, respectively. Wheat straw
was found to have a urease activity of about 1830 mg urea kg 1h 1, a pH of
near 8, and a H ion buVering capacity of 53 mmol kg 1 (pH unit) 1. These
factors undoubtedly contributed to NH3 loss from upland soils. Singh and
Singh (1991), however, observed lower cumulative NH3 volatilization loss
from a saturated calcareous soil amended with rice straw than that from
unamended soil due to the lowering of soil pH by rice straw.
Gill et al. (1998a) measured more losses of N via NH3 volatilization from
flooded soils amended with wheat straw (11.6%) than from unamended soils
(7.1%) in 16 days. Application of nBTPT, a urease inhibitor, to the soils
amended with wheat straw reduced losses of NH3, although much less than
in the soil not amended with organic materials. These studies showed that
soils amended with crop residues would require two to four times more
nBTPT than the soils not amended with crop residues for eVectively reducing NH3 volatilization from flooded soils. Caramona et al. (1990) has also
drawn similar conclusions.
Residue left on the surface under no-till agriculture may also aVect NH3
volatilization. McGarity and Hoult (1971) found that plant material on the
soil surface influenced NH3 volatilization and ascribed this to urease associated with the ureolytic phylloplane and litter microorganisms. Bacon et al.
(1986) reported that stubble management involving cultivation lost 7–8 kg
N ha 1, while zero cultivation treatments lost an average of 15 kg N ha 1.
They further reported that partial burial of urea reduced ammonia volatilization from 36 kg under broadcast onto the surface of stubble retention
plots to 7 kg N ha 1. Retaining the stubble resulted in higher soil water
content increasing urea hydrolysis, leading to greater change in pH and
ammonium concentration. Additionally, some urea prills are retained within
the stubble above the soil surface. Under favorable conditions, urease activity associated with residue surfaces would enable urea hydrolysis in prills
held above the soil. Ash of crop residues at the soil surface also increased
ammonia volatilization due to its high pH (Bacon et al., 1986).
b. Leaching and Denitrification. Nitrogen immobilization by high C:N
ratio residues represents a potential temporary sink to reduce N loss from
leaching (Savant and De Datta, 1982). Immobilization of soil N within
surface residues may have a positive influence on subsequent crop growth
in that N remains near the root zone. However, leaching and denitrification
losses of N with in the soil profile may increase where surface-placed residues
result in increased water infiltration and reduced evaporation rates.
When residues are incorporated, depending on the placement and type, or
soil texture and water content, the potential for denitrification can be
increased dramatically (Aulakh et al., 1992). Generally, in lowland soils,
the rate of NO3 formation rather than available C limits denitrification.
Bacon et al. (1989) observed that incorporation of rice stubble in wheat
increased apparent denitrification of fertilizer N from an average of 34 to
53 kg N ha 1. The N loss occurred over several months, suggesting that
denitrification was maintained by continuous release of metabolizable carbohydrates from the decomposing rice stubble. Nugroho and Kuwatsuke
(1992b) found that under upland conditions (60% WHC), denitrification
hardly occurred when the level of NO3–N in soil was <5.5 mg g 1 soil.
The rate of denitrification substantially increased with the increase in the
level of NO3-N in the rice straw-amended soil, while in the soil without rice
straw, it only slightly increased with the increase in the level of NO3–N. The
maximum rate of denitrification was as high as 6.5 mg N2O–N g 1 day 1 in
the rice straw-amended soil. In a laboratory incubation study, Patrick and
Gotoh (1974) recorded reductions in fertilizer N losses by 30–40 mg kg 1
from the soil amended with rice straw. In pot and microplot studies, Xu
(1987) observed that application of rice straw with inorganic N enhanced
immobilization and reduced N losses via denitrification. Denitrification
losses from soil can be influenced by residue N content. Aulakh et al.
(1991) observed that denitrification losses from the soil at 90% water-filled
pore space were 87 to 127% of initial soil NO3 and increased further with
increasing residue N content.
In a 2-year field study in subtropical Australia, Cogle et al. (1987) observed that after 15 months, only 44% of applied 15N urea was recovered
from wheat residue incorporation treatment, compared to 55% from surface
retained treatment. The greater losses in incorporated straw treatment were
possibly due to greater availability of carbon to the denitrifying population
compared with treatment where wheat straw was retained on the soil
Phosphorus mineralization from crop residues is determined by the rate
of residue decomposition and microbial immobilization (Stevenson, 1986).
The activity of enzymes (phosphatases) that mineralize P is influenced by the
same factors that aVect microbial activity. Phosphorus content of the added
residue is perhaps the most important factor in regulating the mineralization
of P in crop residues. In general, net immobilization of P occurs following
addition of crop residues with less than 0.2 to 0.3% P, while net mineralization occurs with higher P contents (Stevenson, 1986). During early stages of
residue decomposition, net immobilization of P can conserve a substantial
amount of P in slowly available organic forms. Cycling of P in soil is not
easily measured, since mineralized P may be removed from the soil solution
via adsorption to colloidal surfaces or precipitation as Ca, Fe, or Al phosphates or immobilized into organic P (Stevenson, 1986). Availability of P to
plants in crop residue-amended soils is, therefore, a function of organic
matter turnover, concentration of inorganic P in soil solution, and P
requirements of microorganisms.
Black and Reitz (1972) and Qiu and Ding (1986) reported that application
of wheat straw decreased the NaHCO3-extractable P in all the soils used in
the study. In other studies, application of rice and wheat straw (with C:P
ratio >300) caused immobilization of P during the first 15 days and then
progressively increased the available P content in soil from the 30th day
onward (Mukherjee et al., 1995; Yadvinder-Singh et al., 1988). Addition
of crop residues to waterlogged soil, however, significantly increased the
Olsen-P content (Yadvinder-Singh et al., 1988). Mishra et al. (2001b)
reported that during the decomposition of rice straw, P content
increased from 0.10 to 0.195% with time. About 22.5 and 59.4% of the
total P present in rice straw was released within 5 and 23 weeks, respectively,
after its incorporation into the soil. McLaughlin et al. (1988) indicated that
crop residue P may not significantly contribute to the nutrition of the
subsequent crop but becomes incorporated into organic P forms. In that
study, only 5.4% of the legume residue (Medicago truncatula L.) P was
recovered by wheat plants, while 22 to 28% was recovered in microbial
biomass. The microbial biomass, therefore, controlled the rate of organic
P accumulation.
In a rice–wheat cropping system, Hundal and Thind (1993) reported that
incorporation of wheat straw (6 t ha 1) depressed labile P and dissolved
P but enhanced organic P content during the initial stages of plant growth.
In a rice-potato-groundnut rotation, application of crop residues increased
the available P content and reduced the depletion of P reserve of soil
(Chatterjee and Mondal, 1996).
In a typic xerofluvant soil, application of barley straw and other organic
materials decreased the P sorption of the soil (Berton and Pratt, 1997). Ohno
and Erich (1997) reported that management systems that return crop residues back into the soil may increase the availability of P by decreasing the
adsorption of P on the soil surface in an acid soil. The release of aluminum
into the soil increased linearly with increasing rates of crop residues. On the
other hand, P adsorption increased with addition of rice straw in several soils
during the first week of incubation under flooded conditions (Phongpan,
1989). Willet and Higgins (1978) also observed increase in P sorptivity of two
rice soils amended with rice straw under flooded conditions. The increase in
P sorptivity was due to a rise in the levels of oxalate Fe under waterlogged
conditions and was dependent on the free iron oxides content of the soils. In
a long-term study (1979–1991), soils amended with crop residues exhibited
more P adsorption and a greater P adsorption maxima than crop residues
that were removed or burned (Table VI). However, the value of the aYnity
coeYcient, i.e., the bonding strength with which P is adsorbed onto the soil
surface, was the least for the residue incorporation and largest for the residue
removal. Similarly, the Frendulich adsorption isotherm showed the largest
amount adsorbed P but the slowest rate of P adsorption when crop residues
were incorporated.
Organic matter in the surface of no-tillage soils has been shown to
influence P distribution and the availability of P in cropped soils. Higher
P availability in the upper layers in notillage soil was attributed to the
absence of mixing of added fertilizer P, increased quantities of organic P,
Table VI
Langmuir and Freundilich P Adsorption Factors for DiVerent Crop Residue
Management Practices (Average of 6.0 and 8.0 t ha 1 of Wheat and Rice Straw Applied
for 11 Years, Respectively)
Langmur adsorption
Freundlich adsorption
Crop residue
(mg g 1)
(mg mL 1)
Extent of
(mg g 1)
Rate of
(mg mL 1)
From Beri et al. (1995).
and possibly shielding P adsorption sites on soil colloids (Schomberg et al.,
1994a). Additionally, surface application and reduced mixing of P fertilizer
in no-tillage systems may reduce P fixation, thus allowing the accumulation
of unreacted phosphate under those conditions.
Residue management plays an important role in determining distribution
and availability of P in cropped soils. Microbial activity, climate factors, and
soil chemical status all influence the cycling of P in soils. Future research on
residue influences on P availability to plants should consider changes in
organic P, microbial P, and inorganic P transformations in soil. Continued
research on interactions between these pools should help improve the
eYcient utilization of P in cropping systems.
Phosphorus is a costly plant nutrient, since both rock phosphate and
S sources are scarce in developing countries such as India. Because of the
low grade of rock phosphate deposits, about 260 million t in India, it is not
recommended for soils with pH > 7.0. Narayanasamy and Biswas (1998)
reported that application of organic matter along with rock phosphate
increased the P eYciency. The suggested reasons were (1) formation of
plant-assimilable phosphorus-humic compounds, (2) anion replacement of
P ion by humate ion, and (3) coating of sesquoxide particles by humus,
which reduces P fixation. Sharma et al. (2001) reported that Mussoorie rock
phosphate (MRP) and diammonium phosphate (DAP) proved equally
eYcient in increasing grain yield and P uptake of rice when residues of the
preceding wheat were incorporated before rice transplanting or rice residue
was incorporated before sowing of the preceding wheat (Table VII). Without
residue incorporation, MRP (8.1% total P, 12% as citrate soluble) had no
significant eVect on grain yield and P uptake of rice. Similarly, MRP and
DAP proved equally eYcient in increasing wheat yield on residue-amended
plots. Available P in soil did not diVer under no P control and MRP when
Table VII
EVect of Crop Residue Management and Phosphorus Source on Crop Yields in Rice–Wheat
Rotation in India
Residue management
LSD ( p ¼ 0.05)
Rice yield
Wheat yield
P source
LSD ( p ¼ 0.05)
DAP, diammonium phosphate; MRP, mussoorie rock phosphate.
From Sharma et al. (2001).
Table VIII
EVect of Crop Residue Management and Phosphorus Source on Olsen-P in Rice–Wheat
Rotation in India
P source (kg ha 1)
Straw treatment
Both straws removed
Wheat straw incorporated
Rice straw incorporated
Both straws incorporated
LSD. ( p ¼ 0.05)
Residue management P source ¼ 5.95
DAP, diammonium phosphate; MRP, mussoorie rock phosphate.
From Sharma et al. (2001).
residues were removed, but the incorporation of crop residues resulted in
similar levels of available P in soil under DAP and MRP. The eVect was
more pronounced when both the residues were incorporated as compared to
incorporation of rice or wheat straw alone (Table VIII).
Biswas and Narayanasamy (2002) evaluated composts prepared by mixing rice straw with diVerent sources of rock phosphates collected from within
India. Cow dung slurry and Trichoderma viridii, a cellulytic fungus, were
inoculated to hasten the composting process. A phosphorus-solubilizing
microorganism (Aspergillus awamori) was also introduced 1 month after
the start of composting. The study showed that composting enhanced the
mobilization of P from rock phosphate as evidenced through increases in
water-soluble, citrate-soluble, and organic P fractions. Verma and Mathur
(1990) found that incorporation of rice straw along with cellulytic microorganisms and rock phosphate at 15 days before wheat sowing resulted in a
significant increase in wheat yield over recommended fertilizer management
practices. Tian and Kolawole (1998) reported that application of diVerent
crop residues increased the P uptake by Crotolaria ochsolenca from rock
phosphate. For eYcient use of P from rock phosphate in the low-fertility
soils, it is suggested to apply plant residues with high polyphenol and low
lignin contents.
Crop residues contain large quantities of potassium, and their recycling
can markedly increase K availability in soils (Chatterjee and Mondal, 1996;
Ning and Hu, 1990; Patil et al., 1993; Sarkar et al., 1989). Recycling of crop
residues can improve crop yields at low rates of K application and decrease
the crop response to the K applications. The role of crop residue recycling in
K balance in the rice–wheat cropping system has been dealt with in detail by
Bijay-Singh et al. (2003).
Yadvinder-Singh et al. (2004b) reported that release of K from rice straw
occurred at a fast rate, and within 10 days after incorporation, available soil
K contents increased from 50 mg K kg 1 in the untreated control to 66 mg
K kg 1 in straw-amended treatments. Tian et al. (1992) reported that most
of K in the rice residue was released in less than 41 days. The amount of
K released from organic materials in the first month was highly correlated
with the water-soluble K (Patil et al., 1993; Sarkar et al., 1989). Potassium is
not bound in any organic compound in the plant material, and thus its
release does not involve microorganisms.
Mishra et al. (2001b) reported that during the decomposition of rice straw,
K contents decreased from 1.30 to 0.28%. About 79% of the total K present
in rice straw was released within 5 weeks after its incorporation into the soil,
and 95.3% of K from straw was mineralized by the end of 23 weeks.
Sulfur is a critical nutrient for crop growth, and its deficiency is accentuated in soils of the tropics by intensive agricultural practices, less use of
organic manures, removal of crop residues, and leaching of SO4 by heavy
rains. It is generally accepted that plants assimilate S almost entirely in the
form of SO4, which is produced by the mineralization of organic S. Unlike
phosphates, sulphates are easily leached. Incorporating crop residues into
the soil is one way of reducing S losses by leaching.
Mineralization of S in soils is mediated by biological activity. Very limited
information is available on S mineralization rates and potentials of wetland
soils amended with crop residues. In an incubation study, application of
wheat and barley straw to two soils (pH > 7.0) increased the S concentration
in equilibrium solution, suggesting that the addition of crop residues to soil
would increase available S (Choi and Rossi, 1978). Organic materials with
high C:S ratios such as wheat straw and rice husk caused considerable
immobilization of S, particularly during the early stages of decomposition
(Somani and Saxena, 1975). Addition of inorganic S fertilizers may,
therefore, be necessary. Islam and Dick (1998b) observed that addition of
wheat straw with a low C:S ratio (100:1) had a significantly higher accumulation of SO4–S than the control or the higher C:S ratio (400:1) wheat
straw treatment. The cumulative amount of C mineralized was linearly
related to S mineralization. Islam and Dick (1998a) reported that the
S mineralization from crop residues followed first-order kinetics and that
the amount of SO4 in flooded soils amended with crop residues would
depend on the soil type, the nature of the crop residues, and the time of
decomposition. Crop residue management is a major determinant of longterm S fertilizer requirements. Singh and Sharma (2000) observed a significant increase in the availability of S in soil with the incorporation of crop
residues. The burning of straw or straw removal from rice paddies increases
the demand of the cropping system and will lead to increases in
S requirements in long term. Whitbread et al. (1999) reported an improvement in the S balance with the incorporation of rice straw over removal.
Long-term studies are needed to enable measurement of the eVects of
recycling crop residues and the impact of environmental inputs on
S dynamics in the soil–plant system.
A ton each of rice and wheat removes 96, 777, 745, 42, 55, and 4 g ha 1 of
Zn, Fe, Mn, Cu, B, and Mo, respectively. The total crop residue production
in India stands at 105 million tons, and based on micronutrient contents of
the residues, the micronutrient potential associated with crop residues would
be about 35.4 thousand tons (Prasad, 1999). About 50 to 80% of Zn, Cu,
and Mn taken up by rice and wheat crops can be recycled through residue
incorporation (Prasad and Sinha, 1995b). Therefore, recycling of crop
residues can help improve the availability of micronutrients in soil.
Iron and Manganese
The application of crop residues to flooded soils leads to a reduced redox
potential (Eh) and, as a consequence, increases the Fe and Mn concentrations in the soil solution. Katyal (1977) observed that not only did the
maximum concentrations of Fe and Mn occur earlier but also their concentrations were significantly higher in flooded soils amended with rice straw
compared to control. Yodkeaw and De Datta (1989) also noted that application of rice straw increased Fe2þ and Mn2þ concentrations in soil solution,
resulting in increased uptake of Fe and Mn by rice crop. Under controlled
Eh and pH conditions, Atta et al. (1996) observed that at an Eh value of
330 mV, soil suspension contained approximately double the amount of
water-soluble plus exchangeable Fe as compared with at Eh values of 150
to þ300 mV. Addition of wheat straw to soil suspension decreased the
exchangeable Fe fraction at pH 8.0, while it increased the same fraction at
both pH 6.0 and 7.0. Exchangeable and water-soluble Mn fractions were
reduced due to application of wheat straw at pH 8.0, while the easily
decomposable fraction increased at pH 7.0 and 8.0 and decreased at pH
6.0. In a greenhouse experiment, Sharma et al. (1989) measured significantly
higher leaching losses of Mn2þ and Fe2þ with increasing rates of rice straw
and percolation rate. As much as 111 kg Mn ha 1 and 110 kg Fe ha 1 were
lost through leaching in one cropping season.
Kang (1988) observed that the availability of Zn in diVerent pools (watersoluble, exchangeable, weakly and tightly complexed to organic matter) was
reduced by straw application at soil pH of 8.0. Several other workers (Yoon
et al., 1975; Dikshit et al., 1976; Raj and Gupta, 1986; Nagarajah et al.,
1989) have also reported that application of rice or wheat straw decreased
the Zn concentration in both flooded and upland soils. As a consequence, Zn
uptake and dry matter production were reduced compared to that in untreated control. Saviozzi et al. (1997), however, observed no significant eVect
of wheat straw (applied at 2% by weight) on the content and distribution of
Zn and Cu in soils. Even so, rice straw application has been found to
increase the Zn content of rice plants, possibly through its amelioritic
eVect on soil pH and ESP.
In calcareous soils, application of crop residues decreased the capacity
factor due to organic acids converting solid-phase labile Zn to soluble Zn
complexes (Prasad and Sinha, 1995a). The diVusion coeYcient of Zn was
increased with the addition of crop residues due to the presence of chelating agents released during their decomposition and thereby increasing the
concentration of total diVusible Zn. The diVusion coeYcient and Zn uptake
by rice in calcareous soils are related linearly.
Crop residues are an important constituent in nutrient cycling. The straw
of most cereal crops contains about 35, 10, and 80% of the total N, P, and
K taken up by the crop (Barnard and Kristoferson, 1985). Apart from the
straw is plant root material, which in most crops adds a substantial amount
of C to the soils. Long-term straw incorporation improves the fertility and
productivity of soils (Ponnamperuma, 1984). Soil organic matter has been
identified by many workers as a key factor in maintaining soil fertility and
crop production. Its maintenance is an essential requirement for increasing
and maintaining productivity. In most of Asia, rice straw incorporated into
the soil is the main source of SOM in the rice-based cropping systems. Since
the maintenance of soil nutrient status is an important aspect of sustainability, the management of crop residues and fertilizer to maintain soil
fertility is necessary.
Soil Organic Matter
In tropical soils, SOM plays a major role in soil productivity because
it represents the dominant reservoir and source of plant nutrients. It
also influences pH, cation exchange capacity, anion exchange capacity,
and soil structure. Its level in soil was used as a general indicator of soil
productivity. A major factor contributing to the level of SOM is annual
input of plant residues. Residue managment impacts on SOM and long-term
fertility are becoming more relevant in the context of soil quality in tropical
The prominent means of maintaining SOM in irrigated rice-based cropping systems in tropical countries have historically been the incorporation of
green manures, animal waste, or crop residues. In recent years, though, the
significance of green manures and animal wastes has been dramatically
altered by the increased use of mineral N fertilizers and other economic
considerations. More recently, crop residues including roots have become a
more common source of organic material added to the soil in many countries
in the tropics, where the use of combine harvesters is increasing (Flinn and
Marciano, 1984). For a given climatic region and soil type, the rate of
addition of carbon inputs is an important factor determining the amount of
organic matter that can be maintained in the soil. The soils tend to reach
equilibrium provided farming techniques and crop residue management
practices stay the same over a long enough period. Under conditions of
warm temperatures and increased water availability, organic matter
accumulation from residues is reduced.
Soil organisms use residues as a source of energy and nutrients, thereby
releasing CO2, inorganic compounds, and recalcitrant molecules, which
contribute to the formation of soil humus. Decomposition of crop residues
releases about 55–70% of the C to the atmosphere as CO2, 5–15% is
incorporated into microbial biomass, and the remaining C (15–40%) is
partially stabilized in soil as new humus (Stott and Martin, 1989). Because
the amount of carbon in soils is large and changes rather slowly, the
implications of a particular management system on the soil carbon may be
apparent only after several years to decades. Numerous calculations have
been made of the amount of residues needed to maintain organic matter at a
particular level (Paustian et al., 1997).
There exist only limited studies on the long-term eVect of crop residue
management on organic matter and N content of soils under rice-based
cropping systems in tropical and sub-tropical countries (Tables IX and X).
In these studies, increases in organic matter content due to crop residue
recycling are relatively small compared to those reported from temperate
regions (Prasad and Power, 1991). Incorporation of both residues increased
organic C and total N compared to removal or burning of straw (Dhiman
et al., 2000). When only rice or wheat straw was incorporated, organic C
content did not diVer significantly from removal or burning of straw. Rice
straw was more eVective in increasing total N content of soil than wheat
straw. Raju and Reddy (2000) reported that in rice–rice rotation, incorporation of rice straw to supply 25% of the recommended N fertilizer dose for
rainy season crop for 6 years significantly increased organic C content from
Table IX
EVect of Straw Management on the Nutrient Status of Mahaas Clay and Grain Yield Averaged for
Five Cultivars after the 16th Cropa
Organic C
Total N
Olsen P
(mg kg 1)
K (mg kg 1)
Grain yield
(t ha 1)
In a column, figures followed by a common letter are not significantly diVerent.
From Ponnamperuma (1984).
0.98% in straw removal treatment to 1.29%. Sharma (2001) reported that
organic C content increased from 0.56% in straw removal to 0.66% when
both the residues were incorporated for 2 years in rice–wheat rotation.
Burning and removal of crop residues were at par for their eVect on organic
C content.
Yadvinder-Singh et al. (2004b) reported that rice residue incorporation
increased organic carbon content of the sandy loam soil more significantly
than straw burning or removal after 7 years (Table X). Carbon sequestration
derived from changes in soil C content in the soil from rice residue applied at
7.1 t ha 1 annually for 7 years averaged 14.6%. In another long-term study,
Yadvinder-Singh et al. (2004a) reported that wheat straw incorporation in
rice increased organic C content from 0.40% in straw removal treatment to
0.53% in straw incorporation treatment after 12 years of experimentation on
a loamy sand soil. The values after 6 years were 0.38 and 0.49%, respectively,
suggesting smaller increases in organic C between 6 and 12 years than during
0–6 years. Carbon sequestration derived from changes in soil C content in
the soil from wheat straw incorporation for 12 years represented 10% of
the added carbon. The rate of increase in organic C with straw incorporation
is generally smaller in coarse-textured soils than in fine-textured soils.
For example, Verma and Bhagat (1992) and Dhiman et al. (2000) observed marked increases in organic C in sandy clay loam soils with residue
incorporation after 4–5 years.
Naklang et al. (1999) observed no significant eVect of rice straw incorporation for 3 years on total and labile C content of a sandy soil. In a
rice-barley rotation under dryland conditions in northern India, Kushwaha
et al. (2000) observed a significant increase (28%) in soil organic carbon and
33% increase in total N with the incorporation of crop residues compared to
their removal after one annual cycle. It was suggested that for soil fertility
enhancement in dryland agroecosystems, postharvest retention of crop residues (20–40% aboveground biomass) of previous crop and its incorporation
in soil through minimum tillage in the succeeding crop should be followed.
Application of rice straw at 10 t ha 1 to an upland sandy soil caused a net
increase in soil C by 0.31 t ha 1 over no rice straw treatment (Ono, 1989).
The increase in C represented 8% of the C applied in rice straw. At higher
rates of straw addition, the net increase in soil C was increased but the
percent C increase did not change significantly. The soil C buildup in the soil
was significantly positively correlated with %N and negatively correlated
with C:N ratio.
Using data from a 24-year long-term experiment at IRRI, Los Banos,
Alberto et al. (1996) showed that straw incorporation improved organic
C, total N, available P, and exchangeable K above that of the burned
straw and no straw treatments. There was an average increase of 0.4 t ha 1
in rice yield with straw incorporation, while burning the straw resulted in
Table X
EVect of Crop Residue Management on Organic Carbon and Total N Content of Soil
Reference and country
Type of crop residue and soil
Beri et al. (1995), India
Rice straw in wheat and wheat
straw in rice; sandy loam
Sharma et al. (1987), India
Rice straw in wheat and wheat
straw in rice, silty clay loam
Rice straw in rice–rice rotation;
IRRI (1986), Philippines
Liu and Shen (1992), China
Milk vetch green manure or
milk vetch þ rice straw in
rice–rice rotation
Zia et al. (1992), Pakistan
Rice straw in rice in rice–wheat
rotation; loam
Wheat straw, green manure and
wheat straw þ green manure
in rice in rice–wheat rotation;
loamy sand
Yadvinder-Singh et al.
(2004a), India
C (%)
Green manure
Green manure
þ rice straw
Straw removed
Straw incorporated
Wheat straw þ
green manure
of study
Dhiman et al. (2000), India
Ponnamperuma (1984),
Rice straw in rice in rice–rice
rotation; clayey
Kumar et al. (2000), India
Prasad et al. (1999), India
Verma and Bhagat (1992),
19 crops
Yadvinder-Singh et al.
(2004a), India
Rice straw in wheat and wheat
straw in rice in rice–wheat
rotation; clay loam
Mustard straw in rice in rice–
mustard rotation, acidic
sandy clay loam
Wheat straw in rice in rice–
wheat rotation; sandy clay
Rice straw in wheat in rice–
wheat rotation; sandy clay
Rice straw in wheat in rice–
wheat rotation; silty clay
Rice straw in wheat in rice–
wheat rotation; sandy loam
negligible improvements. This reported increase occurred over a 14-year
period and highlights the time frame over which SOM increases occur.
A small increase in SOM associated with improved residue management
demonstrates how diYcult it is to improve SOM levels and, consequently,
nutrient levels in coarse-textured soils of the tropics. Incorporation of rice
straw at 5 t ha 1 year 1 for 12 years showed only a small increase in organic
C and total N content of soil with 2% initial organic C level (IRRI, 1986).
Straw removal or burning caused a decline in organic matter content during
the first 3 years of the study, while the straw incorporation maintained the
original level.
Field experiments on a rice–wheat cropping system in India showed that
incorporation of crop residues as compared to burning or removal increased
organic carbon and total N contents (Table X). Adiningsih (1984) reported
that incorporation of rice straw into the soil for 4 years increased the soil
organic matter content from 2.4 to 3.9% and total N content from 0.25 to
0.33% over straw removal in Indonesia. In China, Liu and Weng (1991)
found that returning rice straw to rice fields for 2 years usually increased soil
organic matter content by 0.03 to 0.05%. From a long-term field experiment
in Japan, Gotoh et al. (1984) estimated that 13 to 25% of the organic matter
returned to soil through rice straw was incorporated into the soil organic
matter in a slowly permeable grey lowland soil. In a 3-year study on a barleyearly rice-late rice cropping sequence in China, He and Liu (1992) reported
that addition of organic materials (green manure, crop residues, and FYM)
resulted in a mean increase (average of six experiments) of 0.053% organic
C compared to loss of 0.04% under inorganic fertilizer treatment. They
calculated that supply of 3.2 to 4.6 t ha 1 (mean of 3.8 t ha 1) of crop
residues ha 1 year 1 would be needed to maintain the soil health and to
improve productivity.
In a long-term study on a rice–wheat cropping system in northwestern
India, the incorporation of crop residues along with green manure in rice
increased soil organic carbon and total N contents as compared to straw
removal, but the increase was almost similar to that when crop residues were
applied alone. These data suggested little eVect of green manure on soil
organic matter content in semi-arid climates, particularly in coarse-textured
soils (Table IX). In a long-term study (1981–1990) in China, Liu and
Shen (1992) studied the eVect of milk vetch green manure in early rice
and milk vetch plus rice straw in late rice in a rice–rice cropping system.
The increases in organic matter and total N concentrations in soil were in
the decreasing order: green manure plus rice straw, green manure, and
inorganic fertilizers. Further, mixed application of green manure and crop
residues improved the quality of soil organic matter (Table IX). Vityakon
et al. (2000) reported that application of rice straw at 10 t ha 1 increased
organic C content of upland soil by 0.31 t ha 1 year 1 over no straw in loam
soil in Thailand.
Naklang et al. (1999) used the two indices to calculate a carbon management index (CMI). They measured two fractions of organic carbon in soil.
The more labile fraction (CL) was measured by oxidation with 333 mM
KMnO4, and the nonlabile C (CNL) plus the C not oxidized by 333 mM
KMnO4, (i.e., CT-CL). The total C (CT) was measured by combustion. On
the basis of changes in CT between a reference site and the cropped site, a
carbon pool index (CPI) was calculated:
CPI ¼ CTcropped =CTreference
On the basis of changes in the proportion of CL in the soil (labiality ¼ L ¼
CL/CNL), a labile index (4) was determined.
CMI ¼ CPI LI 100
Incorporation of leaf litters increased the CMI from 9 in 1992 (initial) to
about 20 after 3 years in 1996 and CMI in no-litter treatment increased to 13.
Straw incorporation did not significantly aVect the CT (4.44 versus 4.11 mg
g 1) and CL (0.78 versus 0.79) compared to straw removal treatments. The
measurement of CL is a more sensitive indicator of SOM dynamics. Total
C measurement is still required to estimate bulk soil C change; however, CL
more accurately and quickly detects the impact of management on soil
C. Calculation of the CMI takes into account the change in CT pool size
and its lability and gives a more definitive picture of soil C dynamics than
when only a single parameter is used.
The studies on soil organic matter dynamics suggest that soil texture,
C inputs, and climatic conditions are the primary factors controlling stabilization of soil C. Simulation models allow us to account for such interacting
factors and thus can be profitably used to understand the dynamics of soil
organic matter in crop residue-amended soils on a long-term basis. Most of
these models predictions have not been tested using observed data, and there
is a need to revalidate these models for rice-based cropping systems. There is
no single fixed quantity of SOM that can be considered as optimal for all
soils. All other factors held constant, an increase of 1% in SOM content
will have greater eVects for a sandy soil than for a clay-loam soil on the
overall productivity level. Benefits of increased SOM will also depend on
land use. For example, improved physical properties in clay soils might be
more useful for upland crops than lowland rice, as the common practice of
puddling rice soils is intended to destroy soil structure.
Soil organic matter levels tend to be stable or increase under irrigated
rice double cropping (Cheng, 1984; Nambiar, 1994; Witt et al., 2000).
Organic matter content is generally lower in rice-upland crop rotations
such as rice–wheat or rice-maize (Cheng, 1984; Nambiar, 1994; Witt et al.,
2000). The reduced soil C sequestration in the rice-upland rotation resulted primarily from an increased amount of microbially mediated
C mineralization compared to the C mineralization rate in the rice–rice
system (Witt et al., 2000). Carbon sequestration with continuous rice cropping would also be favored by the accumulation of phenolic end products
that appears to occur when crop residues decompose under anoxic conditions in lowland rice systems (Olk et al., 1996). When crop residues are not
regularly incorporated in the lowland-upland crop rotations, the amounts of
labile SOM can decrease to the point of reducing the continuous supply of
available N through mineralization–immobilization turnover (Stevenson and
Kelley, 1985), which could lead to lower grain yield. In the light textured soils,
nutrients and soluble C compounds may move down the profile, thus resulting
in very slow, or no, long-term increase in soil fertility when residues are added
(Naklang et al., 1999).
Management of crop residues might also carry longer term impacts on the
chemical nature of SOM. The eVects of crop management on SOM quantity
in lowland rice soils have received more attention than have their eVects on
SOM quality. Few or no studies have examined the eVects of agronomic
practices on the quantity or quality of SOM and nutrient supply in intensive
continuous rice or rice–wheat rotations.
The quantity of SOM is not the sole factor that should be considered
when devising management practices to optimize the agronomic benefits of
SOM. A higher quantity of SOM does not automatically lead to a higher
quality of SOM. If most SOM-bound nutrients are in SOM fractions that
have low turnover rates, that is, high residence times, their roles in nutrient
supply will remain marginal. If the soil in question is a sandy soil, for
example, and if the crop obtains the bulk of its nutrients through decomposition of the various SOM pools rather than through the exchange of
nutrients present on CEC complexes, the nutrient supply power of the soil
will remain low. Ultimately, it remains the quality rather than the quantity
of SOM that will lead to improved soil quality, and hence a more sustainable
cropping system, in particular for those agrosystems that are prone to land
It remains a diYcult task to identify and quantify the intrinsic quality of
an SOM pool in terms of nutrient supply power, microbial activity, or
physical or chemical indices. Labile SOM pools are key suppliers of nutrients to the crop, whereas other SOM pools are more recalcitrant in nature
and will provide fewer nutrients, but their chemical and physical properties
provide stability to the soil. The relative sizes of the labile versus more
recalcitrant pools that make up total SOM might have pronounced eVects
on the indigenous nutrient supply and perhaps even yield (Biederbeck et al.,
1984; McGill et al., 1986), illustrating the complexities of managing SOM
quality. The chemical nature of humic acid fractions changed with an
increasing number of annual irrigated rice crops (Olk et al., 1996). The
increasingly phenolic nature of the humic acid was speculated to be a
contributing factor to an apparent decline in soil N supply and grain yield
in continuously cropped lowland rice soils, as phenols are known to stabilize
nitrogenous compounds under controlled conditions (Haider et al., 1965).
The eVect of rotating upland crops with rice on SOM quality indicated that
the phelonic nature of labile SOM extracted from rice–wheat soils is more
similar to that of labile SOM from lowland rice–rice soils than that from
upland rice soils (Olk et al., 2000). Again, the agronomic significance of this
finding is not clear.
Bird et al. (2002) examined the five soil organic matter fractions from soil
samples obtained after 4 to 6 years of rice residue management treatments
using 15N-labeled urea. After 4 years of straw management treatments, soil
incorporation of straw increased mobile humic acid (MHA) and light
fraction (LF) carbon and N compared with burned straw. Immobilization
of fertilizer N peaked in all soil organic matter fractions after one growing
season (120 days) and was greater in the MHA over the 2-year study.
Nitrogen fertilizer sequestration was in MHA and LF and was greatest
with straw incorporation compared with straw burned. Turnover of immobilized 15N fertilizer was fastest in the labile MHA and MFA (mobile fulvic
acid) fractions (7–9 years half-life) compared with a half-life of the moderately resistant MAHA (metal-associated humic acid) fraction (53 years) and
most stable humic (HUM) fraction (153 years). The MHA and LF fractions
represented the primary active sink and source of sequestered N, aVecting
both short- and long-term soil fertility. A study by Devevre and Howarth
(2000) suggested that it is not primarily the accumulation of degradation
byproducts that may sequester N in SOM, as suggested by Olk et al. (1996).
The larger and sustained microbial biomass found under flooded compared
to aerobic conditions may act to immobilize more N and make it less
available for plant uptake.
The composition and dynamics of SOM are generally the same in temperate and tropical soils, except that turnover rates in tropical soils usually
are higher than in colder climates. Therefore, many results from temperate
soils can be used to explain SOM dynamics and control in tropical soils. The
main transformations occurring during residue decomposition and humification are the loss of polysaccharides and phenolic moieties, modification of
lignin structures, and enrichment in recalcitrant, non-lignin aromatic structures (Zech et al., 1997). The rates of these transformations are controlled
primarily by climatic factors and only to a lesser extent by chemical factors
such as pH, C:N ratio, or litter quality. Soil organic matter stabilization by
interaction with minerals probably is more important in tropical than
temperate soils because of the more favorable climatic conditions for decomposition of organic matter. The protective eVect of minerals is most
pronounced for labile constituents such as polysaccharides or proteins.
Studies on the contribution of labile fraction to SOM dynamics in tropical
ecosystems are very scarce.
Crop residue management can aVect N immobilization and stabilization
processes important to eYcient utilization of N from fertilizers, crop residues, and soil organic matter. Bird et al. (2002) reported that a consistently
larger soil microbial biomass N and C pool was observed when straw was
incorporated than when it was burned. Because soil microbial biomass is a
prime source of available N for the crop, the incorporation of straw led to an
increase in the crop-available soil N. Although total soil N content had not
changed after 5 years of straw incorporation or burning, a significant increase had taken place in the more labile soil N pools (humic substances).
The more labile soil N pools remain key sources of readily available N for
crop utilization.
2. Total N
About 70% of the rice lands in south and south-east Asia contain
<0.2% N and are considered N deficient (Ponnamperuma, 1984). Incorporation of crop residues enhances the N content of several wetland rice
fields (Tables IX and X). Within 3 years of incorporating the rice straw at
6–7 t ha 1, total N content in soil increased by 0.021% over the straw
removal treatment. At IRRI, in situ incorporating the straw twice a year
caused an increase of 48 kg N ha 1 per season, averaged for two experiments
lasting 7 years (Ponnamperuma, 1984). In another study with three soils over
a 5-year period, the increase due to straw incorporation was computed to be
40 kg N ha 1 per season, about 10 kg N ha 1 per season more than the straw
N content. The extra N probably came from N fixation stimulated by straw
acting as an energy source for heterotrophs and as a CO2 supplement to
surface phototrophs (Ponnamperuma, 1984).
Available P and K
A number of studies (Alberto et al., 1996; Gangaiah et al., 1999;
Ponnamperuma, 1984; Prasad et al., 1999; Sharma, 2001; Singh and Sharma,
2000) (Tables IX and XI) showed a slight or no increase in the available P in
the soil and P uptake of rice and wheat in soils amended with rice or wheat
straw. In an 11-year field experiment on a loamy sand in Punjab (India),
incorporation of residues of both crops in rice–wheat rotation increased the
Table XI
EVect of Crop Residue Management on the Available P, K, and S Contents in Soil
and country
Beri et al.
Sharma et al.
Verma and
Prasad et al.
YadvinderSingh et al.
YadvinderSingh et al.
Type of crop
residue and soil
of study
Available nutrients
(mg kg 1)
Rice straw in
wheat and
wheat straw on
rice, sandy loam
Wheat straw in
rice, sandy clay
Rice straw in
wheat, sandy
clay loam
Rice straw in
wheat, sandy
clay loam
Wheat straw in
rice, sandy clay
Wheat straw in
rice, loamy
Wheat straw
þ GM
Rice straw in
wheat, sandy
total and available P and K contents in soil over removal of residues
(Table XII). Similarly, in a long-term study in Bihar (India), Misra et al.
(1996) observed increases in availability of P and K in soil with the incorporation of crop residues in rice–wheat rotation. Application of wheat straw
continuously for 12 years in a loamy sand soil caused only a small increase in
available K while taking into consideration the total amount of total
K recycled through the straw (Yadvinder-Singh et al., 2003a). In another
study on a sandy loam soil, incorporation of rice residue caused a smaller
but more significant increase in available K content in the soil than did the
residue removal treatments (Table XI). On average, rice residue added about
175 kg K ha 1 annually in treatments. Despite such large additions, the
increase in K availability in the soil was small. One possible reason for the
small increases observed in residue-amended plots may be the loss through
Table XII
EVect of Crop Residue Management on Soil Fertility of a Loamy Sand Soil over 11 Years of
Rice–Wheat Cropping System
Crop residue management
Soil property
Total P (mg kg 1)
Total K (%)
Olsen P (mg kg 1)
Available K (mg kg 1)
Available S (mg kg 1)
From Beri et al. (1995).
leaching of a significant proportion of residue K during rice cultivation on
highly permeable coarse-textured soils. Beaton et al. (1992) reported no
significant eVect of crop residues on the available K concentration in soil.
The K added through crop residues was possibly leached from the soil,
converted to unavailable forms, or taken up by the subsequent crops. Prasad
et al. (1999) reported an increase of 7 mg K kg 1 soil with straw incorporation compared to straw removal after 2 years of study in a sandy clay loam
soil. In a 6-year study by Raju and Reddy (2000) in rice–rice rotation,
application of rice straw to supply 25% of the recommended N fertilizer
(60 kg N ha 1) in rainy season rice caused a small increase in Olsen-P
compared to straw removal, but available K increased from 113 mg K kg 1
in straw removal to 143 mg kg 1 in straw-incorporated treatment.
The benefits of straw incorporation are reflected not only in the increase
in soil K but also in plant uptake (Beye, 1977; Ponnamperuma, 1984). In a
field experiment on rice-potato-mungbean rotation, Chatterjee and Mondal
(1996) observed that crop residues applied at a rate to supply 25% of the
recommended N requirement (75% supplied through fertilizer) increased
the available K in soil and lowered the depletion of non-exchangeable
K compared to the inorganic fertilizer alone treatment. Gill and Adiningsih
(1986) observed a marked response to K application in two crops of rice and
one soybean crop when crop residues were removed at harvest on Orthoxic
Troppudult soil high in Al and low in K. Recycling of crop residues dramatically improved the yields at low rate of K application and decreased crop
response to applied K. Magbanua et al. (1988) reported that there was a
strong K depletion when residues of previous crops were removed. Rice
straw incorporation improved nutrient (P, K, S) balance over straw removal.
Whitbread et al. (1999) also reported that removal of rice stubble resulted in
negative K and S balances in rice.
Available Micronutrients and Silicon
Incorporation of residues decreased the availability of Zn but had no
significant eVect on Cu and Mn availability in soil (Bijay-Singh et al., 1992).
In another study over a 5-year period on a silt loam soil in Himachal Pradesh
(India), incorporation of rice straw in wheat, however, caused a slight
increase in the availability of P, Mn, and Zn and a marked increase in the
availability of K (Verma and Bhagat, 1992). Likewise, incorporation of
crop residues on a long-term basis has been found to increase the DTPA
extractable Zn, Cu, Fe, and Mn contents in soil (Meelu et al., 1994).
Rice straw contains large amounts of Si that on incorporation can increase its availability in soils. According to Sumida and Ohyama (1991),
application of rice straw increased the Si content of rice plants, which helped
increase the lodging resistance in rice.
Redox Potential and Electrical Conductivity
Incorporation of crop residues is unlikely to have a significant eVect on
the changes in Eh of aerobic soils due to fast O2 diVusion into soil, except in
large aggregates. Fermentation of crop residues in waterlogged soils results
in low Eh (Beye et al., 1978) and high pCO2, which leads to initial increase in
the concentrations of Fe2þ and Mn2þ in soil solution followed by precipitation of carbonates. In rice straw-amended flooded soils, not only did the
peaks of CO2 occur earlier, but the concentration also increased as compared to control (Katyal, 1977). Murty and Singh (1976) observed an
increase of 0.04 atm of pCO2 due to the addition of wheat straw to flooded
rice. Yoo et al. (1990) observed that surface application of rice straw suppressed the formation of an oxidized layer and increased reducing conditions. Rice straw also reduced O2 concentration followed by an increase in
the accumulation of CO2.
Straw incorporation hastened and intensified soil reduction and also
increased pH and electrical conductivity (EC) of acid soils (Beye et al.,
1978) and decreased pH and EC of alkaline soils (Yodkeaw and De Datta,
1989). Katyal (1977) noted the acceleration and intensification of Eh and pH
changes and the achievement of peak concentrations of water soluble Fe,
Mn, and CO2 in three soils amended with crop residues. Several researchers
(Murty and Singh, 1976; Wu, 1996) observed increases in EC of soil with the
application of crop residues under anaerobic conditions.
2. Soil pH
Crop residues can influence soil pH through accumulation of CO2 and
organic acids during their decomposition in the soils. The reduction in Eh of
flooded soils after the incorporation of crop residues may increase soil pH
due to consumption of protons during the reduction of Fe and Mn oxides.
A sharp decrease in soil pH of flooded soils due to application of rice or
wheat straw has been recorded by Wu (1996) and Murty and Singh (1976).
Under controlled Eh levels, Atta et al. (1996) observed a slight decrease in
pH of soil suspension during 5 days of incubation. A slight decrease in
soil pH with the application of wheat or rice straw was also reported by
Saviozzi et al. (1997), Gangaiah et al. (1999), and Kushwaha et al. (2000).
However, Verma and Bhagat (1992) noted no significant eVect of rice straw
management on pH of an acidic sandy clay loam soils after 5 years of
Seki et al. (1989) observed that successive applications of rice straw and
wheat straw decreased the pH of the surface layer in coarse-textured grey
lowland and fine-textured yellow soils with a sharp decline in the first year.
In a silty clay loam soil (pH 7.3) amended with rice straw, Sharma et al.
(1989) recorded lower pH values of soil solution between 2 and 6 weeks after
transplanting compared to those from untreated soil. Retention of crop
residues on the soil surface can result in a decline in soil pH to levels that
may adversely aVect crop production (Schomberg et al., 1994a). In longterm field experiments (over 10 years), Beri et al. (1995) and Bellakki et al.
(1998), however, observed no significant eVect of incorporation of crop
residue on soil pH. In contrast, Yoo et al. (1990) recorded a significant
increase in the pH of floodwater with the application of rice straw.
The eVect of application of plant materials on soil pH depends on the
composition of plant material used (Yan et al., 1996). The mineralization of
N-rich compounds followed by nitrification produces protons, resulting in
the acidification of the soil. The influence of residues on soil pH appears to be
rather small, but the method of residue management may greatly influence
the soil reaction. The potential for pH changes with surface-managed residues
is greatest in response to application of fertilizers or high-N residues to the
surface and the absence of mixing soil amendments through the soil profile.
Crop residues play an important role in improving soil physical characteristics, but the degree of improvement depends on particle size distribution. Sandy soils with low SOM contents lack substantial structure and are
prone to severe erosion. Adding crop residues or manure will increase
microbial activity, which in some studies has led to the buildup of SOM and
formation of macro- and micro aggregates (Sparling et al., 1992; Angers
et al., 1993). DiVerences in aggregate stability also depend on the sources of
the organic materials, such as fungal hyphae versus microbial polysaccharides (Tisdall, 1991). On the other end of the particle spectrum, heavy clay
soils are often characterized by poor structure and aeration, but they can be
improved through the addition of organic amendments. Therefore, the
positive eVect of SOM on soil structure will be more pronounced for a
clay soil than for a silty soil.
In most climates, removal or burning of crop residues leads to deterioration of soil physical properties (Kladivko, 1994; Prasad and Power, 1991). In
rice, puddling of soil by cultivation in standing water could adversely aVect
soil structure through destruction of aggregates and peds (Sharma and De
Datta, 1985) and leads to formation of a pan of low permeability immediately
below the cultivated layer, particularly on fine-textured soils. This hard pan
could be detrimental for the productivity of the upland crop (say wheat) after
rice (Moorman and Van Breeman, 1978; Sur et al., 1981). A recent review has,
however, shown that puddling may or may not be detrimental to the succeeding non-rice crops and soil (Connor et al., 2003). Recycling of crop
residues influences soil structure, crusting, bulk density, moisture retention,
and water infiltration rate and may help reduce adverse eVects of hard pan
formation in rice-based cropping systems, which may play an important role
in the upland crop (such as wheat or maize) after rice than the rice crop.
The role of soil organic matter in aggregate stability is summarized in Fig. 5.
Straw incorporation helps the formation and stability of aggregates through
increase in microbial cells, and excrets microbial products and decomposition
products released during the death of the microorganisms (Lynch and Elliott,
1983). The soil organic matter in turn is protected within aggregates for
decomposition (Dalal and Bridge, 1996). The amount and chemical composition of organic residues, temperature, and moisture conditions are the major
factors determining aggregation in soil (Prasad and Power, 1991). Thus, easily
decomposable plant residues such as green manure and grain legume residues
provide transient and temporary aggregate stabilizing agents, while cereal crop
residues provide persistent aggregate stabilizing agents (Elliott and Lynch,
1984). Chaudhary and Ghildhyal (1969) obtained a close relation (r ¼ 0.76)
between organic C increased by organic materials addition and aggregate
stability of soil under wetland rice. Likewise, Elliott and Lynch (1984) found
that the eVect of straw on aggregation in a silt loam soil decreased with
increasing straw N content in the range of 0.25 to 1.09%.
Figure 5 A generalized summary of soil aggregates stabilization by various sources of
organic matter (Dalal and Bridge, 1996).
Several researchers have reported an improvement in soil aggregation
after incorporation of crop residues into the soil under rice-based cropping
systems (Bhagat et al., 2003; Liu and Shen, 1992; Liu et al., 1990; Meelu
et al., 1994; Oh, 1984). In a 10-year study on a rice–rice cropping system on a
vertisol, application of rice straw incorporated to meet either 25 or 50% of
recommended fertilizer N requirement increased the water stable aggregates
(Table XIII). In a rice–wheat cropping system on a loamy sand soil, incorporation of wheat straw over a 5-year period in rice promoted formation of
soil aggregates, particularly 1–2 mm size, and mean weight diameter (Table
XIV). A mixed application of green manure and crop residues was more
eVective compared to their separate applications. Similarly, in a long-term
experiment (1981–1990) on rice–rice rotation in China, Liu and Shen (1992)
noted that application of crop residues promoted aggregation. The contents
of micro-aggregates (0.25–1.0 mm) were increased from 10.9% in inorganic
fertilizer treatment to 12.1% in milk vetch green manure and to 13.6% in
green manure plus rice straw treatment.
In a 4-year barley-early rice-late rice crop rotation in China, Rixon
et al. (1991) found that addition of 3 t ha 1 of crop residues in late rice did
not significantly aVect distribution and stability of aggregates and moisture
retention characteristics of a gleyed paddy soil. However, after 5 years of the
above study, a continuous improvement in soil structure, volume weight,
porosity, aggregation, and plasticity was observed (Zhu and Yao, 1996).
The eVect of crop residues on aggregation also depends on the aggregation potential of the soil. Datta et al. (1989) have shown that when clay
Table XIII
EVect of Rice Straw Application on Soil Physical Properties in Rice–Rice Cropping System over a
10-Year Period on a Clayey Soil
Treatment to
summer rice
Rice straw to
meet 50% N
Rice straw to
meet 25% N
Green leaf
manure to
meet 50% N
(kg kg 1)
33 kPa
1.5 MPa
(kg kg 1)
(Mg m 3)
(cm h 1)a
HC, hydraulic conductivity.
From Bellakki et al. (1998).
Table XIV
EVect of Green Manure and Crop Residues on Soil Aggregation and Bulk Density in a Rice–Wheat
Cropping System on a Loamy Sand Soil after 5 Years
Water stable
aggregates (%)
Residue removed
Residue incorporated
Green manure (GM)
Crop residue þ GM
Bulk density
(Mg m 3)
0–10 cm
10–20 cm
From Meelu et al. (1994).
content in soil was low, burying of straw had a more favorable eVect on the
stability of aggregates, especially of crumbs 3–5 mm in diameter, than in soil
with 27% clay content. Likewise, Verma and Singh (1974) observed that
wheat straw caused a marked influence on soil aggregation in four diVerent
soils varying in texture. Maximum aggregation occurred in the sandy loam,
with minimum aggregation in alkali soil. Application of rice straw to alkali
clayey soil significantly increased water stable aggregates >0.25 mm. Total
organic C also increased, which resulted in a marked increase of macropores
as well as the aggregate size in the 2.0–0.84 mm size fractions (El Samanoudy
et al., 1993). In a friable self-mulching clay of the vertisol group, 34 years of
either stubble burning or incorporation had, however, little eVect on soil
structure (Dexter et al., 1982). The nature of plant material also plays an
important role in the development of soil structure. For example, Dhoot
et al. (1974) recorded the highest percentage of water-stable aggregates in
pearl millet-amended soil followed by rice straw or wheat straw and sesbania
green manure.
In a long-term field study in China, rice straw incorporation increased the
porosity and formation of large micro-aggregates and decreased the bulk
density of paddy soils (Li et al., 1986; Xu and Yao, 1988). Rice straw and
rape straw were more eVective in increasing porosity of soils than sesbania
green manure or pig manure (Li et al., 1986). Bellakki et al. (1998) and
Bhagat et al. (2003) noted a significant increase in the porosity of finetextured soils after the application of rice straw and lantana residues. He
and Liu (1992) observed that in rice straw-amended soil, porosity (>200 mm)
increased quickly after drying, which is favorable for land preparation and
sowing of upland crop in time after rice harvest. Beaton et al. (1992)
reported that addition of rice straw (6 t ha 1) over a 68-year period compared to inorganic fertilizers reduced the volume weight and increased the
porosity of paddy soils in Japan.
Hydraulic Conductivity and Infiltration Rate
Crop residues aVect hydraulic conductivity and infiltration by modifying
soil structure, proportion of macropores, and aggregate stability. Marked
increases in hydraulic conductivity and infiltration have been reported in
treatments where crop residues were retained on the surface or incorporated
by conventional tillage over the treatments where residues were either
burned or removed (Murphy et al., 1993; Valzano et al., 1997). In a 6-year
rice–wheat cropping system on a clay loam soil in India, Sharma et al. (1987)
noted increased cumulative infiltration of 7.39 cm h 1 under residue incorporation over 5.70 cm h 1 under residue removal. Similarly, in long-term
experiments on rice–wheat cropping system, incorporation of both rice and
wheat straw, as compared to their burning or removal, increased both
Figure 6 EVect of crop residue and green manure application on infiltration characteristics
of a loamy sand soil (Meelu et al., 1994).
infiltration rate and cumulative infiltration in sandy loam soils (Singh et al.,
1996; Walia et al., 1995). In another 5-year study on a rice–wheat cropping
system on a loamy sand soil, Meelu et al. (1994) observed increased rates of
infiltration on soil amended with green manure and crop residues (Fig. 6).
A mixed application of green manure and crop residues was more eVective in
increasing infiltration compared to their separate applications. On an alkali
clayey soil application, rice straw significantly increased hydraulic conductivity and total and quick drainage pores (El Samanoudy et al., 1993). In a
long-term rice–rice cropping system on a vertisol, Bellakki et al. (1998)
also noticed a significant increase in hydraulic conductivity of soil from
incorporation of rice straw (Table XIII).
Bulk Density, Compaction, and Penetration Resistance
In general, incorporation of crop residues into the paddy soils reduced
bulk density, penetration resistance, and compaction of soils under both
rice–rice and rice–wheat cropping systems (Bellakki et al., 1998; Meelu et al.,
1994; Singh et al., 1996; Walia et al., 1995). Xie et al. (1987) also reported
that continuous return of rice straw to a paddy field for 7 years resulted in a
soil bulk density decrease of 0.17 Mg m 3. In another long-term field
experiment over 25 years, incorporation of crop residues improved the
porosity and decreased penetration resistance of a gleyed soil (Roppongi
et al., 1993). Likewise, combined application of cereal crop residues and
green manure has proved to be more eYcient in reducing bulk density,
penetration resistance, and crusting of surface soil layers over their separate
applications (Liu and Shen, 1992; Meelu et al., 1994; Verma and Singh,
Bhushan and Sharma (2002) reported that with the application of lantana
residues to a silty loam soil continuously for 10 years in rice–wheat rotation, clods <2 cm in diameter increased while those 2–4 cm and 4–6 cm in
diameter decreased with straw additions. The mean weight diameter of clods
varied between 2.15 and 2.34 cm in lantana-treated soil versus 2.83 cm in the
control. The bulk density and breaking strength of soil clods were lower in
lantana-treated soil by 4–9% and 29–42% than in control, respectively.
About 23–47% less energy was required to prepare seed bed in lantanatreated soil than in control soil. The long-term addition of residues caused a
fundamental change in soil structural processes. A significant change in soil
consistency and the related physical properties such as surface cracking and
clod formation occurred after the addition of residues continuously for 10
years. Lantana-treated soil would become friable relatively soon, thereby
decreasing the turnaround time after rice harvest.
Soil Moisture Characteristics
In a long-term (1920–1988) study in Japan, Beaton et al. (1992) found
that addition of rice straw increased the water retention in paddy soil at 0 to
3.2 pF moisture tension, while no eVect was observed at pF 4.2. Rice straw,
thus, improved the supply of readily available water. In a 10-year field study
on a vertisol, the application of rice straw significantly improved the water
retention characteristics of paddy soil (Bellakki et al., 1998) (Table XIII).
Likewise, Lanjewar et al. (1992), Kushwaha et al. (2000), and Bhagat et al.
(2003) reported significant increase in the water holding capacity of soil after
straw incorporation compared with removal or burning. Pandey et al. (1985)
also observed that incorporation of rice and wheat straw for 5 years
increased soil water retention at 33 k Pa over straw removal.
Application of lantana residue for 10 years on a silty clay soil in a rice–
wheat cropping system in northwest India significantly increased the liquid
limit, plastic limit, shrinkage limit, and plasticity index (Bhushan and
Sharma, 2002). The friability limit of residue-treated soil decreased from
8.9 to 7.8–8.2% gravimetric moisture content of soil, but soil became friable
at a relatively higher moisture content. Soil cracking changed from wide,
deep cracks in a hexagonal pattern to a close-spaced network of cracks. The
cracks of sizes <5 mm increased and those of 10–20 mm and wider decreased
with residue additions.
The role of soil organisms as primary agents of decomposition, energy
flow, and nutrient cycling has become the subject of increased interest. Crop
residues provide energy for growth and activities of microbes and substrate
for microbial biomass, and provide conditions for a source-sink of nutrients.
Crop residue management alters soil environment, thereby influencing
microbiological populations and activity in soil and subsequent nutrient
Microbial Biomass
Microbial biomass, a small (1–5% by weight) but active fraction of soil
organic matter, is of particular concern in soil fertility considerations because it is more susceptible to management practices than the bulk organic
matter (Janzen, 1987). Soil microbial biomass (SMB) acts as a reservoir of
plant nutrients and is a major determinant for governing the nutrient (like N,
P, and S) availability in soils. Although SMB values are only a small portion
of total C and N in soils, this living portion of soil contains a substantial
amount of nutrients needed for crop growth. The amount of microbial
biomass and microbial activity depends on the supply of organic substrates
in soil. Therefore, regular addition of a suYcient amount of organic materials such as crop residue is important in the maintenance of microbial
biomass and improvement of soil fertility.
Several researchers (Azmal et al., 1996; Sridevi et al., 2003) have reported
a marked increase in microbial biomass following incorporation of crop
residues. After straw incorporation, microbial biomass-C (MBC) increased
by two- to fivefold in 10 days and reached the highest value by 30 days. For
example, Ocio and Brooks (1990) observed that straw addition, compared to
control, increased the microbial biomass by 87.5% in a sandy loam soil and
by about 50% in a clay soil. Malik et al. (1998) found that application of
wheat straw and green manure in a rice–wheat cropping system caused a
large increase in microbial biomass during the initial phases of rice crop. An
increase in microbial biomass was sustained throughout the growing season
of rice and resulted in synchronization between N release and N uptake.
Patra et al. (1992) found more biomass C in wheat straw than in cowpea
residue-amended soil, but the amount of microbial biomass N (MBN) was
significantly higher in the latter. Azmal et al. (1997) reported that the
amount of microbial biomass C and N increased immediately after rice
straw incorporation into a clay loam soil incubated under aerobic conditions, reached maximum values after 1 week of each application (2 g C as
rice straw kg 1 soil after every 6 weeks), and decreased thereafter. The level
of maximum biomass formation reached a ceiling after the second application, suggesting that soil has a certain capacity to hold biomass.
Singh (1991) reported that microbial C was maximum in the wheat straw
(10 t ha 1) plus fertilizer treatment (408–420 mg g 1) followed by straw (360–
392 mg g 1) and fertilizer treatments (238–246 mg g 1) in rice-lentil (Lens
esculenta) crop rotation under dryland conditions. With time, straw plus
fertilizer treatment accumulated 77% more microbial biomass C over control. The initial flush of microbial activity probably results from rapid
catabolism of simple soluble C compounds initially present in crop residues.
Biomass N is in a constant state of turnover and represents a significant
proportion of the total N, which is relatively constant throughout the year.
Application of labeled wheat straw to a clay loam soil increased the biomass
N from 46 mg N in control to 80 mg N g 1 soil by day 5 and remained at this
level by day 20 (Ocio et al., 1991). Bremer and Van Kessel (1992) studied the
dynamics of microbial C and N following the addition of 14C- and 15Nlabeled lentil and wheat straw in a sandy loam soil under field conditions.
Microbial 15N accounted for 65 to 81% of the added residue 15N. The results
suggested that microbial biomass may reduce losses of N and other nutrients
during the periods of low crop demand and may act as a source of nutrients
during the active crop growth.
Kushwaha et al. (2000) reported that straw incorporation increased the
SMB carbon during crop growth from 214–264 mg g 1 in straw removal to
368–503 mg g 1 in straw incorporation treatment after two annual ricebarley crop cycles. The MBC was increased by 48% and N by 60% in residue
retention over residue removal plots. Along with residue retention, tillage
reduction from conventional to zero increased the levels of MBC and MBN
over control. Addition of plant residues with a high C:N ratio may facilitate
transformation of fertilizer or soil N into a slowly available N and thus may
improve N use eYciency. Microbial biomass may act as slow-release fertilizer. It has been reported that the larger and sustained microbial biomass
found under flooded compared to aerobic conditions may act to immobilize
more N and make it less available for plant uptake, as seen in the some areas
of the tropics (Cassman et al., 1995).
Bird et al. (2001) observed that soil microbial biomass was always significantly greater when straw was incorporated than when it was burned.
Because soil microbial biomass is a prime source of available N for the
crop, the incorporation of straw led to an increase in the crop-available
soil N. Although the total N content did not change after 5 years of straw
incorporation or burning, a significant increase was noted in the more labile
soil N pools (humic substances).
Witt et al. (1998), however, observed no significant eVect of residue
incorporation on microbial biomass C and N, suggesting that microbial
biomasses are not sensitive indicators of processes governing net
N mineralization. Instead, incorporation of crop residues may have led to
enhanced microbial activity rather than microbial growth.
Microbial Population and Activity
Residue incorporation into the soil leads to increased bacterial and
fungal activities (Beare et al., 1996; Doran, 1980). For example, proteindecomposing microorganisms increased during the early stages of incubation of rice straw under waterlogged conditions (Fujii et al., 1972), which
was followed by an increase in the population of cellulose-decomposing
microorganisms. Sulphate-reducing microorganisms then increased after a
lag phase. Nugroho and Kawatskka (1992a) observed that application of
rice straw (C:N ¼ 52:1) increased all the microbial populations. In that
study, simultaneous application of rice straw and NHþ
4 –N to soil under
upland conditions increased the number of denitrifiers but significantly
depressed the N2 fixation activity. Beri et al. (1992) also observed that soil
treated with crop residues inhabited 5–10 times more aerobic bacteria and
1.5–11 times more fungi than the soil for which residues were either burned
or removed. Fujii et al. (1970), in contrast, found that with a short-term
incubation period (10 or 20 days) in an aerobic soil, the population of
nitrifying bacteria was higher in the absence of rice straw, but the reverse
was true with longer incubation periods (60 or 90 days). Ladatko and
Emtsev (1984) observed marked increase in the growth of Clostridium spp.
in a soil amended with rice straw. The increase in the growth of anaerobic
microorganisms was due to the formation of artificial anaerobic microsites
around the straw particles. Last but not least, residue quality may aVect the
microbial population, as smaller bacteria and fungal populations are greater
on cereal residues compared to those on legumes. As compared to bacteria,
fungi are more influenced by residue quality (Wardle, 1995).
Earthworms and micro-arthropods play a dominant role in organic matter decomposition and nutrient cycling associated with diVerent crop residue
management systems (Prasad and Power, 1991; Tian et al., 1993). Although
enough information is not available from rice-based cropping systems,
residue quality greatly influences macroorganism populations in the soil.
For example, the earthworm population was negatively correlated with the
C:N ratio, lignin:N ratio, and polyphenol concentration of the plant material (Tian et al., 1992), and the population of ants was significantly correlated
to N concentration of plant residues (Tian et al., 1993).
4. Enzyme Activities
Barreto and Westerman (1989) and Gill et al. (1998b) observed a significant increase in urease activity in surface soils after incorporation of wheat
straw. Likewise, Guan (1989) reported that application of wheat straw
increased the invertase activity of soil by 40–90 times compared to the
control treatment in both laboratory and field experiments and that the
activities of urease and alkaline phosphatase were also increased by wheat
straw additions. Gialhe et al. (1976) observed that dehydrogenase activity
increased with rice and wheat straw incorporation and was further increased
by N application. Goyal and Chander (1998) also reported an increase in the
microbial biomass and dehydrogenase and alkaline phosphates activities
with the addition of wheat straw to a sandy loam soil.
Organic materials have been used as amendments for reclaiming saline
and sodic soils. Incorporation of crop residues can bring about favorable
changes in the physico-chemical properties of such soils. Puttaswamygowda
and Pratt (1973) reported that addition of straw to sodic soil prior to
submergence for 130 days substantially lowered the pH and exchangeable
sodium percentage and increased the Na, Ca, Mg, and Fe2þ ions and
electrical conductivity. Xie et al. (1987) observed that incorporation of
wheat straw with and without sesbania green manure (GM) to a silt loam
saline-alkali soil decreased salt concentration of the top soil in the first 3
years. Similarly, Swarup (1992) found that incorporation of wheat straw and
rice husk alone or in combination with GM into a sodic soil significantly
reduced the pH and exchangeable sodium percentage, and increased PCO2
and exchangeable Ca þ Mg and extractable Fe, Zn, Mn, and P. The degree
of improvement was in the order GM þ rice husk > GM þ wheat straw >
GM > rice husk > wheat straw > control. The increased availability of
nutrients resulted in improved yields and nutrient uptake by rice. Marked
decreases in pH, exchangeable sodium percentage, and electrical conductivity of salt-aVected soils amended with crop residues have also been reported
by many other workers (Abdul-Wahid et al., 1998; Hussain et al., 1996;
Illayas et al., 1997; More, 1994).
In a lysimeter study using calcareous sandy loam soil under a rice–
wheat–maize fodder system, Sekhon and Bajwa (1993) reported that
irrigation with sodic water caused precipitation of Ca and increased the
accumulation of Na in the soil and adversely aVected the crop yields.
Incorporation of rice straw decreased the precipitation of Ca and carbonates
increased the removal of Na in drainage water, decreased pH and electrical
conductivity of the soil, and improved crop yields. The release of organic
acids during decomposition of residues possibly mobilized the soil Ca. The
quantity of gypsum required for controlling the harmful eVect of sodic
irrigation water on soil properties can be considerably reduced in the
presence of crop residues.
Incorporation of wheat straw into a saline soil at 7 t ha 1 for 3 years
improved soil physical properties such as bulk density, pore volume, and soil
water retention and improved soil productivity (Wang et al., 1988). Improvement in soil physical properties (bulk density, porosity, and hydraulic
conductivity) due to addition of crop residues was also reported by Hussain
et al. (1996). Thus, recycling of crop residues on salt-aVected soils is likely to
have greater benefits than on normal soils (Swarup, 1992; Abdul-Wahid
et al., 1998).
Naturally occurring heterotrophic and phototrophic bacteria use the
straw either directly by the use of hemicellulose and simple carbohydrates
or indirectly following the decomposition of cellulose by decomposer microorganisms. Asymbiotic N2-fixing bacteria can use crop residues for energy
through the use of some hemicellulose components (Halsall et al., 1985) or
products of straw decomposition (Roper and Halsall, 1986). The heterotrophic diazotrophs depend on carbon for energy. Since most N2-fixing bacteria
are unable to use cellulose directly as a substrate for N2 fixation, cellulose
must be degraded to simpler intermediates before being used by diazotrophs.
Adachi et al. (1989) showed the existence of linkage between anaerobic
cellulytic bacteria and anaerobic N2-fixing bacteria during the decomposition of straw. The role of crop residues in biological N2 fixation by heterotrophic and phototrophic bacteria has been reviewed in detail by Roper and
Ladha (1995).
Anaerobic conditions and a decrease in inorganic N content of soil
following incorporation of straw favor N2 fixation by heterotrophic and
phototrophic bacteria in waterlogged soils (Yoneyama et al., 1977). Under
laboratory conditions, a wide range of values of N2 fixation (0.8 to 7.07 mg
N fixed per g of straw in 14 to 56 days) have been obtained due to diVerences
in the form and amount of straw, time of incubation, and methods used for
quantification (Roper and Watanabe, 1986). Only a few quantitative data
on the amount of N2 fixed or N gained following straw application in
greenhouse or field conditions are available.
Enhanced N2 fixation in flooded soils amended with straw has been
reported by Rice and Paul (1972) and Charyulu and Rao (1981). Rao
(1980) estimated that N2 fixation in 30 days in flooded soil amended with
chopped straw at 5 or 10 t ha 1 and planted to rice was two to four times
that of the unamended control. Based on a per hectare furrow slice of 0.7 106 kg dry soil ha 1, extrapolation of the values of 15N incorporation in
straw-amended soil in a 30-day experiment indicates N2 fixation of about
7 kg ha 1 in the unamended soil and 25 kg N ha 1 in straw-amended soil.
Santiago-Ventura et al. (1986) measured twice the N gain following straw
incorporation equivalent to 10 t ha 1 after the three consecutive rice crops
compared with control pots; N gain ranged from 2 to 4 mg N fixed g 1 straw
added. Nugroho and Kwatsuka (1992b) found maximum rates of N2 fixation as stimulated by rice straw amendment to be as high as 220 mg g 1
day 1 when the level of NHþ
4 –N in the soil was below 7.8 mg N kg
High levels of NH4–N (98–298 mg kg soil) inhibited the initial N2 fixation
activity. When denitrification occurred at high rates, N2 fixation was suppressed and vice versa. Yoo et al. (1990) reported that surface application of
rice straw increased the pH of the floodwater to an optimum level for the
growth of N2-fixing microorganisms, and thereby increased the N2 fixation
by phototrophic bacteria and blue-green algae.
In aerobic soils, intese microbial activity during the decomposition of
crop residues results in the development of anaerobic and microaerobic
microsites in soils, including surface soils (Hill et al., 1990). These sites can
support N2 fixation by a wide range of free-living, diazotrophic bacteria,
including anaerobic bacteria. In situ measurements of N2 fixation associated
with wheat straw indicated amounts fixed (based on the acetylene reduction
technique) ranging from 1 kg N ha 1 in 31 days to 12.3 kg N ha 1 in 22 days
(Roper, 1983). The amount of wheat straw added to soil ranged from 4.3 to
7.2 t ha 1 under conditions where moisture was not limiting (i.e., field
capacity). In a laboratory incubation study, Saha et al. (1995) observed
that berseem (Trifolium alexandrinum) and rice straw significantly increased
aerobic nonsymbiotic N2-fixing bacteria, phosphate-solubilizing bacteria,
and S-oxidizing microorganisms, resulting in greater availability of N, P,
and S in the soil.
Crop residue-associated N2 fixation is modified by mineral N, temperature, moisture, oxygen concentration, soil characteristics, and straw management techniques (Roper and Ladha, 1995). In fact, straw decomposition
is also directly aVected by these factors. In a field experiment, Roper (1983)
observed a positive correlation (r ¼ þ0.98) between nitrogenase activity and
wheat straw decomposition. As already discussed, the N2 fixation rates in
straw-amended soils are higher under waterlogged conditions than under
upland conditions (Rao, 1976). Roper et al. (1994) found that nitrogenase
activity under field conditions was the highest with straw incorporation and
the activity decreased in the order straw incorporation > straw mulched >
no tillage. The depth of straw incorporation into soil also aVected the
nitrogenase activity. Straw mixed lightly with the soil near the surface
produced significantly higher nitrogenase activity than soil in which straw
was incorporated throughout the plough layer (Roper et al., 1989). Kanungo
et al. (1997) recorded higher nitrogenase activity in the top 1–2 cm soil layer
after the placement of organic residues, while residue placement in 2–6 cm
layers significantly reduced nitrogenase activity, irrespective of soil type. The
high nitrogenase activity in the topsoil was associated with larger populations of Azospirillum, Azotobacter, and anaerobic N2 fixers and favorable
redox potential supporting growth of N2 fixers.
The adverse eVects of substances originating from decomposing crop
residues have long been considered as a cause of poor growth and yield of
many crops (Patrick et al., 1963). Since breakdown of cellulose occurs
readily, many of the adverse eVects of residues occur within a relatively
short time after the incorporation of residues and the sowing of the following crop. Warmer climates further accelerate the breakdown of crop residues. Thus, incorporation of crop residues can have adverse eVects on
subsequent crops other than rice if anaerobic conditions develop (Cannell
and Lynch, 1984). However, anaerobic decomposition of crop residues with
no-tillage may have adverse eVects on seedling establishment of rice. Lynch
(1977) reported that under certain conditions, substances toxic to cereal
seedlings are produced by cereal residues that decay near the seedlings.
These findings assume greater importance when crops are grown immediately
after cereals and with minimal cultivation. When seed drills operate in soils
where crop residues are placed on the soil surface or are only shallowly
incorporated, seed and residue can be placed in close contact, particularly in
fine-textured soils. Wet conditions that lead to anaerobic decomposition of the
residues can adversely aVect seedling growth (Elliott et al., 1978; Kimbler,
1973). Kimbler (1973) reported that the degree of inhibition of growth of
wheat by wheat straw depended on the length of decomposition period and
was greatest when the period was only 2–6 days. Surface retention leads to
slow decomposition, and incorporation is recommended as soon as possible.
Phytotoxic substances (e.g., phenolic acid and acetic acid) are produced
from degrading crop residues preferentially under anaerobic soil conditions
(at least in localized zones) and seldom accumulate in aerobic soil because of
rapid metabolization by microorganisms. Gaur and Pareek (1974), however,
detected a larger number of phenolic and aliphatic acids under aerobic
than under anaerobic conditions. In a laboratory incubation study, the
addition of rice or wheat straw produced large amounts of acetic acid
under anaerobic conditions 4 to 8 days after the incorporation of straw
(Bhat, 1991). Tanaka et al. (1990) reported that straw incorporation resulted
in accumulation of reducing substances and various aliphatic aromatic
acids in soil, which can inhibit rice root growth. Low temperature and
acidity further favor the production and persistence of fatty acids (Cho
and Ponnamperuma, 1971). At temperatures over 30 8C, these acids disappear within 2–3 weeks of straw incorporation. The organic acids are phytotoxic in the millimolar concentration range and can cause significant crop
losses, which can be between 13 and 29% in heavy clay soils when seed is
direct drilled in the presence of wheat straw in winter (Graham et al., 1986).
Studies on homogenous slurries of a soil in a chemostat showed that the
formation of organic acids from plant residues is primarily linked to Eh; the
critical Eh being about zero (Lynch and Gunn, 1978).
Goodlass and Smith (1978) observed that evolution of C2H4 from soils
under anaerobic conditions was stimulated by amending soils with barley or
wheat straw. Temporary anaerobic conditions resulted in large increases in
C3 and C4 hydrocarbons. The association between degradation products and
C2H4 suggests that both may be implicated when root growth is adversely
aVected by the anaerobic decomposition of plant residues. Wu et al. (1997)
observed that application of rice straw increased the level of reducing
substances in soil at 20 days after application and reduced rice plant weight
at 30 and 70 days after planting. In a greenhouse study, Sharma et al. (1989)
found that total water soluble organic acids extracted from the root zone of
rice plants (100 mm soil depth) increased with increasing amounts of rice
straw (Fig. 7), but the acid production decreased with increasing rate of
percolation. Highest acid concentration (364 m mol L 1) was obtained with
the addition of 20 t rice straw ha 1 and a percolation rate of 15 mm day 1.
The organic acids formed at 2 weeks after transplanting did not persist in
soil solution; rather, they disappeared rapidly and the rice yields were same
under all the treatments.
The toxic eVects of aliphatic acids on rice growth have been widely
studied. Most investigations have been of short term and on young
plants. Nevertheless, in several instances, quite low concentrations of acetic
acid, propionic acid, and butyric acid have killed rice seedlings (Rao and
Mikkelsen, 1977). The injury caused by monobasic aliphatic acids depends
on the type of acid present and its concentration. The inhibitory eVect on
rice seedlings generally increases with increasing molecular weight, increasing with order formic, acetic, propionic, and butyric acid (Chandrasekaran
and Yoshida, 1973). Tanaka et al. (1990) observed that rice root elongation
was markedly inhibited by the solution extracted from flooded soil
with incorporated wheat straw; the extract contained aliphatic and phenolic acids under acidic conditions. Huang and Lu (1996) reported that
Figure 7 Temporal changes in volatile organic acid concentrations in the soil solution
collected at 100 mm soil depth as aVected by added rice straw (mean of three percolation rates)
(Sharma et al., 1989).
pre-flooding after rice straw incorporation for 2 weeks is suYcient for
oVsetting any adverse eVect due to phytotoxicity and N deficiency in rice.
Wallace and Whitehead (1980) have reported that volatile fatty acids are
more toxic than nonvolatile aliphatic acids between 0.5 and 1.0 mM concentrations and that the organic acids produced at one site do not diVuse
very far onto the soil. Therefore, the establishing crop roots must not come
into close contact with decomposing residues.
Adverse eVects of decomposing residues on crops under aerobic conditions have been widely reported (Bhowmik and Doll, 1982). Phenolic acids
such as ferulic, p-coumaric, and p-hydroxybenzaldehyde released from living
or dead tissues of variety of plant species caused adverse eVects on the
growth of crops (Nelson, 1996). Elliott et al. (1981) could not demonstrate
the phytotoxicity to winter wheat on plots when wheat straw was mixed into
the soil. N immobilization during straw decomposition rather than phytotoxicity appeared to be the primary factor adversely aVecting yield because
yield decline was largely overcome by high rate of N application. Chung
(2001) identified p-hydroxy benzoic acid (6.34–6.87 mg kg 1), p-coumaric
acid (0.34 mg g 1) and ferulic acid (0.05 mg g 1) during the decomposition
of rice straw. The nature and composition of allelopathic compounds
depended on the type of crop residue or variety. P-hydroxy benzoic acid
(10 3 M) showed the greatest inhibitory eVect on barnyard grass seed
germination, seedling growth, and dry weight.
In the rice field, the concentration of phytotoxins varies during the growth
period of the crop and is probably greater in the early stages of flooding. It
may also vary spatially within the rhizosphere. The rice crop can also show
considerable compensatory growth from adverse eVects on early growth after
rice straw has been ploughed into the soil (Gotoh and Onikura, 1971).
Organic acids accumulated around straw only in the early stages of decomposition, and hence if straw decomposition could be accelerated by any means,
the danger period for seedling could be reduced. The concentration of organic
acids in flooded soils in the tropics receiving 5–10 t ha 1 of straw is not toxic
to rice (Ponnamperuma, 1984). Witt et al. (2000) noted no evidence that
late residue incorporation caused phytotoxic eVects as a result of reduced
organic compounds or toxins produced during residue decomposition.
It is recommended to plough crop residues shortly after harvest is completed, because the decomposition of the straw occurs early after incorporation, the phytotoxicity occurring in the initial period of growth of the rice
plant can be alleviated, and stable yields can be obtained.
Weeds are a major problem in the productivity of rice-based cropping
systems. Depending upon their type and intensity, 20 to 50% or even greater
losses in grain yields of rice and wheat are common due to competition from
the weeds (Walia and Brar, 2003). Most studies in weed control in rice and
other crops have been confined to evaluating the eVects of herbicide, tillage,
water, and their interactions (Bhagat et al., 1999; Gajri et al., 1999). Few
studies have been conducted on the dynamics of weed population and
herbicide eYciency under residue management in rice-based cropping systems. Such information is needed in weed control strategies for rice-based
cropping systems to improve their productivity. Kumar and Goh (2000)
reported that crop residues can suppress weeds in many ways, for example,
(1) through their physical presence on the soil surface as mulch and by
restricting solar radiation reaching below the mulch layer, (2) by direct
suppression caused by allelopathy, and (3) by controlling N availability.
Burning of residues can help in eVective removal of weed seeds and weeds.
The major disadvantage of incorporation of rice straw compared to burning
is the increase in weed and possible pest pressure. Roeder et al. (1998)
reported that compared with farmers’ traditional burning of crop and
weed residues, mulching reduced rice yield by 43% in one out of four
comparisons and increased weed biomass by 19–100%.
In addition to influencing the weed growth and population, crop residue
management and tillage practices also influence the eYciency of soil-applied
pre-emergence herbicides (Kumar and Goh, 2000). Because pre-emergence
herbicides are applied to the soil, the amount and quality of residues and ash
content left behind after residue burning might aVect their activity. Continuous burning or incorporation of residues of both crops over years results in
buildup of ash or organic matter in the soil. The eYciency of soil-applied
herbicides may decline because of increased absorption capacity of soil. Brar
et al. (1998), however, observed that there was no significant eVect of
burning or incorporation of crop residues on the eYcacy of butachlor
applied to rice and isoproturon applied to wheat in a rice–wheat cropping
system. Mt. Pleasant et al. (1992) observed that mulching residues had little
eVect on weed control and crop yields were always higher when residues
were incorporated in a rice-based cropping system. The reports on the eVect
of crop residue management practices on weed growth and herbicide
eYciency are not conclusive and need further investigation to improve the
productivity of rice-based cropping systems.
The eVects of crop residue management on the pests and diseases in
rice-based cropping systems in the tropics have not received much attention.
Methane (CH4) and nitrous oxide (N2O) are important greenhouse gases,
N2O being about 300 and CH4 being 15 times more radiatively active than
CO2 (mass basis, considering residence time in the atmosphere) (Rodhe,
1990). Flooded rice soils are a major source of atmospheric CH4, contributing about 10% of the total global emissions of CH4 (Mitra et al., 1999; Neue
and Sass, 1996; Rennenberg et al., 1992; Sass et al., 1990; Wassmann et al.,
1998). Global methane emission from flooded rice fields has been estimated
at 20–100 Tg year 1 (Neue, 1993). In comparison, the total agricultural
sources of N2O are quite small, ranging from 0.03 to 3.0 Tg N year 1
(IPCC, 1996).
Incorporation of organic materials (crop residues, green manures, compost) to regenerate depleted soil resources and promote sustainable food
productions in the tropics should significantly increase CH4 emissions. Thus,
residue management strategies may create conflicts between the goals of
sustainable agriculture and mitigation of greenhouse gases when used in
flooded rice-based systems. Soil properties, water management, organic
amendment, and temperature have been reported as the major factors
controlling the amount of CH4 emitted from rice fields (Sass et al., 1991;
Schütz et al., 1989). It has been estimated that CH4 emissions from rice
cultivation in India (45 million ha) should not exceed 2.5 t year 1. The main
reason for low CH4 emissions from rice fields in India is that the soils have
very low organic C or receive very little organic amendments (Jain et al.,
2000). The burning of crop residues also contributes to the global CH4
budget. For each ton of crop residue burned, 2.3 kg CH4 is emitted (Grace
et al., 2003). In rice–wheat cropping system, 0.14 t year 1 will be emitted, if
one-half of the 12 million ha under rice–wheat cropping system is burned.
Organic C from added crop residues, organic manures, soil organic
matter, or rice plant roots is the major driving force for CH4 production in
rice-based agriculture systems (Wang et al., 1992; Yagi and Minami, 1990).
Numerous studies from all over the world have demonstrated that added
crop residues, composts, and green manures enhance CH4 fluxes relative to
unamended controls (Bossio et al., 1999; Chen et al., 1993; Chidthaisong
et al., 1996; Glissmann and Conard, 1999; Neue et al., 1994; Rath et al.,
1999; Wassmann et al., 1993). The seasonal emissions from paddy rice with
organic additions ranged from 1.1 to 148 g CH4 m 2 and increased methane
emissions 1.2- to 32-fold over unamended control soils. Crop residues serve
as a substrate for a complex microbial community, including methanogenic
microorganisms. Most studies on the microbiological aspect of CH4 production in flooded rice soil have focused on methanogens (Asakawa and
Hayano, 1995; Asakawa et al., 1998). In addition to methanogens, the
degradation of organic matter to its most reduced status (CH4), however,
involves at least two other kinds of nonmethanogens: the zymogenic bacteria
and the acetic acid- and hydrogen-producing bacteria. Thus, from the point
of view of microbiological ecology, diVerent eVects of various organic fertilizers on CH4 production potential might be closely related to the amount of
easily decomposable organic matter.
In principle, the degradation pattern in soils with and without amended
straw is similar, with acetate, propionate, and H2 as the main intermediates
of anaerobic degradation and CH4 being formed from H2/CO2 (11–27%)
and acetate (84–89%). However, the early phase of straw degradation diVers,
as a large variety of fatty acids accumulate transiently (Glissmann and
Conard, 1999). A study by Weber et al. (2001) indicated that the methanogens colonizing rice straw are less diverse than those inhabiting the soil.
Polysaccharolytic bacteria in rice soils constitute the first step in the degradation process and eventually produce substrates needed for the production
of CH4. Distinct trends of multiple rate patterns for CH4 emission from
waterlogged soils have been shown in laboratory and field studies (Hou et al.,
2000). The first peak, between 20 and 40 days at 25 8C, probably originated
Figure 8 EVect of rice straw application on methane production in a sandy soil (Hou et al.,
from the decomposition of easily decomposable forms of C in the rice straw,
such as microbial products and polysaccharides (Watanabe et al., 1995). The
second change in rate of CH4 emission observed may have been associated
with the decomposition of structural components of the rice straw, such as
cellulose and lignin.
The eVect of rice straw application on CH4 production potential is shown
in Fig. 8. Methane production in the treatment without rice straw supplement occurred at a much lower rate during the whole period of incubation, in
which the highest production rate was less than 40 mg CH4 kg 1 soil day 1.
After the application of rice straw, the CH4 production rate increased
substantially. Both the quantity and the quality of added organic materials
influence CH4 emission from soils. Yagi and Minami (1991) showed that
while rice straw increased CH4 emission by a factor of 3.3, addition of rice
straw compost increased CH4 emission only slightly compared to the application of mineral fertilizers. The extent and variability of observed methane
enhancements by organic additions are governed by several factors, the most
obvious being quantity. Schütz et al. (1989) established that CH4 emissions
from paddy rice progressively increased with increasing rice straw additions
from 3 to 12 t ha 1. Straw levels over 12 t ha 1 did not increase CH4 fluxes
further. Likewise, Wang et al. (1992) found increasing CH4 flux to be
proportional to rice straw input levels. A field study (Yagi and Minami,
1990) also showed that rice straw applied at rates of 6–9 t ha 1 enhanced
CH4 emission rates by 1.8–3.5 times. As reported by Sass et al. (1991) and
Watanabe et al. (1995), CH4 production was enhanced by the addition of
straw in flooded soil only.
Watanabe et al. (1995) proposed a simple straw rate response model to
predict cumulative CH4 emissions from a known rice straw application to
any soil:
Y ¼ k½aðEÞ=ð1 þ bðEÞe
ފ þ Y ð0Þ;
where Y is the fractional increase in CH4 emission relative to a chemical
fertilizer control, and x is the level of incorporated organic matter (t ha 1).
Adjustments to the coeYcients a, b, and c were added to account for
responses to temperature (E) and diVerences of soil type (k). Such modifications reflect observations that daily and seasonal CH4 fluxes are temperature
dependent (Parashar et al., 1991; Schütz et al., 1989; Yagi and Minami,
1991). Incubation studies have shown that large diVerences in CH4
production potential of soils are related to organic C content (Majumdar
et al., 1998).
The extent and rapidity with which added organic materials are decomposed depend greatly on chemical composition, including C:N ratio, lignin
and polyphenol content, and other critical compounds. Yadvinder-Singh
and van Cleemput (1998) reported that maximum methane (9980 mg g 1)
was emitted from soil amended with sugar beet leaves, and emissions of CH4
from wheat and rice straw were 4953 and 5030 mg g 1 in 40 days in a silty
clay soil under flooded conditions. The emissions of CH4 from composted
farmyard manure and poultry manure-amended soils were very low.
From an incubation experiment in a Chinese flooded rice soil, Hou et al.
(2000) reported that organic matter, added as rice straw and organic manure
(pig, chicken, and cattle manure), increased CH4 production rate significantly. The results showed that organic manures had diVerent promoting eVects,
with pig manure increasing the CH4 production rate most, followed by rice
straw, chicken, and cattle manure. The CH4 production potential caused by
organic manures was closely related neither to the total C added to the
system nor to the C:N ratio of the materials. A significant linear relationship
between CH4 production and the logarithm of the number of zymogenic
bacteria was found, with an r value of 0.96. This finding suggests that the
number of zymogenic bacteria may be used as an index to predict CH4
production potential in flooded rice fields.
Bronson et al. (1997a) observed that organic matter additions as rice
straw (5.5 t ha 1, dry) or green manure (Sesbania rostrata, 12 t ha 1, wet)
stimulated methane flux several-fold. Rice straw resulted in higher CH4
emissions than GM. The GM plots showed highest CH4 fluxes in the first
2 weeks, but thereafter straw–amended emitted the most CH4. Green manure has more easily decomposable C than straw, although more C was
added as straw. Sesbania green manure, being easily degradable material,
required the lower activation energy by methanogens to use the substrate as
C source than wheat straw (Bhat and Beri, 2002). Rice straw applied before
the winter fallow period reduced CH4 emission by 11% compared with that
obtained from fields to which the same amount of rice straw was applied
during field preparation. Surface mulching of straw instead of incorporation
into the soil showed 12% less emission.
Composts consistently produced lower CH4 emissions than fresh green
manures or straws. Aerobic composting reduces readily decomposable carbon to CO2 instead of CH4 (Inoko, 1984) and also modifies the original
organic constituents to forms more resistant to subsequent degradation
(Watanabe et al., 1995). Consequently, when compost is incorporated into
anaerobic soils, less available carbon is present for methanogenesis. However, the agricultural benefit derived from compost is maintained, especially if
composts are applied year after year (Inoko, 1984). Thus, composting provides a compatible option for adding organic materials to flooded soils
without substantially enhancing methane emissions. Following the same
principal, Miura (1995) found that fall rice straw incorporation or winter
mulching combined with spring incorporation significantly reduced CH4
emissions during the subsequent summer rice season.
Jain et al. (2000) reported that additions of organic manures and crop
residues enhanced CH4 emissions from rice fields. There were wide variations in CH4 emissions because of the variety of organic amendments.
Rice fields amended with biogas slurry emitted significantly less CH4 than
those amended with other organic amendments. They further reported that
CH4 emission rates were very low (between 16 and 40 kg CH4 ha 1 season 1)
when the field was flooded permanently. Application of organic manure
(FYM plus wheat straw) in combination with urea (1:1 N basis) enhanced
CH4 emission by 12–20% compared with fields treated with urea only. The
site in New Delhi represents one example of very low CH4 emissions from
rice fields. Emissions from other sites in northern India may be higher than
those in New Delhi, but they are still lower than in other rice growing
regions in India. Jain et al. (2000) reported that organic amendment inputs
promoted CH4 emissions, but total emission remained less than 25 kg CH4
ha 1. This finding contrasts with results from other network stations with
irrigated rice where total emissions generally exceeded 100 kg CH4 ha 1 after
manure application (Wassmann et al., 2000a). The low impact of organic
manure in the experiment in New Delhi could be related to high percolation
rates. Constant inflow of oxygen into the soil and downward discharge of
methanogenic substrate resulted in low CH4 production (Inubushi et al.,
1992; Yagi et al., 1994). Thus, emissions were very low even when organic
matter was applied. In other stations of the network, organic amendments
stimulated emissions during the first half of the season (Wassmann et al.,
Ishibashi et al. (2001) studied the eVect of surface application of rice straw
in no-till rice on methane emission in three soils during rice growing season.
It was found that CH4 emissions from the no-tilled direct-seeded field on the
average were 21, 47, and 91% of that from the tilled transplanted field in
high-percolating site, low-percolating site, and extremely low-percolating
(4.4 mm day 1) site, respectively. Straw incorporation leads to significantly
more methane production than burning or removal. Over the long term,
however, incorporation may provide benefits through the accumulation of
C as soil organic matter.
The biologically mediated reduction processes of nitrification and denitrification are dominant sources of N2O generation in soils (Paul and Clark,
1989). Nitrous oxide is also produced to a much lesser extent by the abiotic
process of chemodenitrification (Bremner, 1997). Denitrification processes
can terminate with N2O, or, more commonly, N2O is further reduced to N2
gas. Conditions that promote N2O emissions over N2 are high NO3 levels
and/or increasing O2 , while increasing organic carbon levels tend to favor
N2 production (Firestone, 1982). Nitrous oxide emissions from rice fields
occur as a result of nitrification–denitrification during periods of alternating
wetting and drying. Emissions are usually small in irrigated rice systems with
good water control and small to moderate inputs of fresh organic material
(OM) (Bronson et al., 1997a,b).
Bronson et al. (1997a) reported that organic amendments, particularly
rice straw, helped in reducing N2O emissions. In the flooded rice soil, straw
addition possibly stimulates O2 consumption in the aerobic soil layer and in
the rhizosphere, resulting in smaller zones in which nitrification can occur.
Enhanced immobilization of fertilizer N with straw would result in less NH4
available for nitrification–denitrification. Additionally, the high CH4 concentration in straw-amended soil could inhibit nitrification (McCarty and
Bremner, 1991). Methane emissions ranged from 3 to 557 kg CH4 ha 1 with
an average of 182 kg CH4 ha 1.
Few measurements have been published for N2O emissions from flooded
rice soils amended with organic materials. The existing information indicates
that N2O emissions from flooded soils with organic additions are similar to
or less than soils receiving chemical fertilizers, indicating that organic
amendments do not appear to influence N2O emissions very much.
Most information on N2O emissions from rice soils focuses on water
management and nitrogen fertilizers as controlling variables (Cai et al.,
1997, 1999). A trade-oV relationship between CH4 and N2O, i.e., conditions
that favor CH4 production suppress N2O and vice versa, is also well
recognized (Mosier et al., 1998a). So, while organic amendments seemingly
have no impact on N2O emissions from flooded soils, management practices
before or after rice may produce a significant eVect.
Aulakh et al. (2001) showed that denitrification is a significant N loss
process under wetland rice amounting to 33% of the recommended dose of
120 kg N ha 1 on a permeable sandy loam soil. Integrated management
of wheat straw (6 t ha 1) and GM (20 t ha 1 supplying 88 kg N ha 1) and
32 kg N ha 1 as urea fertilizer N significantly reduced cumulative gaseous
N losses to 51.6 kg N ha 1 as compared to 58.2 kg N ha 1 for 120 kg N ha 1
alone. The gaseous losses under wheat were 0.6–2% of the applied fertilizer
N. Interplay between the availability of NO3 and organic C largely controlled denitrification and N2O fluxes in flooded summer-grown rice, whereas temperature and soil aeration status were the primary regulators of the
nitrification–denitrification processes and gaseous N losses during winter
grown upland wheat. The irrigated rice–wheat system is a significant source
of N2O, as it emits around 15 kg N2O–N ha 1 year 1.
The quantity of organic additions may also aVect N2O emissions. In one
of the few studies looking at the impact of organic materials on N2O,
Bronson et al. (1997a) suggested that organic additions to flooded soils
stimulated oxygen depletion to the point of inhibiting nitrification and
thereby N2O emissions. From this, one could hypothesize that increased
oxygen depletion with more organic material and consequently N2O emissions would decline even more. Burning of crop residues also contributes to
the global N2O budget. For each ton of crop residue that is burned, 40 g
N2O is emitted (Grace et al., 2003).
The objective of reducing CH4 emissions must be combined with
improvements associated with increased yields and straw recycling; adhering
to CH4 emission quotas might increasingly aVect rice production practices.
Possible mitigation options for reducing methane emission from rice fields
include reduced length of flooding, temporary drainage (Wassmann et al.,
2000b), rice cultivar selection, kind and application mode of mineral fertilizers, and soil and crop management strategies to achieve a high acceptance
(Mosier et al., 1998a,b; Neue, 1993; Yagi et al., 1994). CH4 emission was
reduced significantly by early incorporation of rice straw during the fallow
period, adding to the agronomic benefit of this practice.
Bronson et al. (1997b) recorded seasonal N2O emissions during a fallow
period as high as 172 and 183 mg N m 2, where rice straw and a green
manure had been incorporated the previous season, respectively. Such emissions might be considered maximums because assimilation of nitrogen
mineralized from organic additions by fallow weed species or upland crops
helps to retain N within the system and minimize N2O emissions (Buresh
et al., 1993). Given the influence of soil type, climate, and organic additions
on CH4 and N2O emissions, comprehensive studies are needed to quantify
more thoroughly the trade-oV eVects between CH4 and N2O during an
annual cycle within rice-based cropping systems.
Water management is an important management factor when trying to
minimize CH4 or N2O emissions from rice-based cropping systems. Midseason drainage, which originally was developed in Japan as a means to
supply oxygen to rice roots, is also very eVective in reducing seasonal CH4
emissions from rice (Jain et al., 2000; Yagi and Minami, 1990). Despite
projected decreases in CH4 emission by such methods, aerobic soil conditions during fallow and upland cropping intervals between rice crops
enhance N2O emissions generated by nitrification of mineralized organic
N and subsequent denitrification of NO3 when flooding is reestablished
(Bronson and Mosier, 1993). Unintentional mid-season drainage is possible
in many rice cropping systems of South and Southeast Asia where light
textured soils or water distribution and management problems influence
the ability of farmers to keep their soils flooded (Jain et al., 2000). Sitespecific adaptations will be required for an optimum eVect, considering rice
yields, water consumption, and CH4 emissions.
In summary, methane emissions can be reduced significantly by adopting
the following mitigation practices: water management through intermittent
irrigation or drainage, the use of composted organic manures instead of
fresh manure, allowing pre-decomposition of crop residues under aerobic
conditions before rice planting, and the selection of suitable cultivars that
emit less CH4. It appears that composted organic additions are the best way
to meet sustainable agriculture goals while minimizing greenhouse gas emissions from paddy rice. Adding crop residues or green manures in suYcient
quantities to increase soil organic matter levels or replenish deficient nutrients for flooded rice exacerbates N2O emissions to unacceptable levels. Of
course, it is important to establish that CH4 and N2O emissions arising from
the composting process do not exceed emissions during rice cultivation.
Direct dry seeding of rice as well as other crops following rice into surface
residues will reduce N2O and CH4 emissions. Grace et al. (2003) suggested
three feasible, cost-eVective agronomic interventions that would have an
immediate eVect by reducing greenhouse gases production in the rice–
wheat cropping systems and that will no doubt be applicable to other ricebased cropping systems in the tropics: (1) a reduction in residue burning, (2) a
reduction in flood irrigation frequency for rice, and (3) the use of minimum
or no tillage for upland crops following rice (e.g., wheat or maize). It was
estimated that Adopting these measures would result in total savings in CO2
equivalent emissions of 1680 kg ha 1 year 1.
Recycling of crop residues is an essential component in achieving sustainability in crop production systems. Since crops respond diVerently to the
application of diVerent organic materials to soil, evaluation of crop residues
in terms of fertilizer eVect is complicated by the variable nutrient contents of
the materials and the host of other eVects (as already discussed) these may
have on crops and soils. In some cases, straw incorporation can actually
lead to a reduction in crop yields due to release of phytotoxic compounds
during decomposition and immobilization of soil and fertilizer N, causing
N deficiency in the crop planted immediately after straw incorporation. In
many studies in which crop residues proved to be superior to inorganic
fertilizers, the eVect may be due only to better supply of nutrients from
organic matter. It is, in fact, impossible to monitor all the eVects of organic
matter on nutrient availability. Evaluation of the fertilizer eVects of an
organic resource (e.g., for N) requires that the material be assessed both at
an equal N application and on equal mass (or carbon) application, preferably in each case over a range of application rates. The eVect of residue
incorporation on succeeding crops depends on the amount of residues and
the time and method of incorporation. Though the long-term eVects of crop
residue incorporation are generally expected to be beneficial in terms of
increasing soil organic matter content, availability of nutrients, cation exchange capacity, and microbial, the time scale for these improvements is
generally long (e.g., >5 years). However, improvements in soil conditions do
not always flow to yields. Thus, despite the very large body of literature on
the recycling of crop residues, there exists very little information that enables
proper evaluation of organic residues for their fertilizer value.
Effect on Crop Yields
Crop residue management as practiced in the rice–wheat cropping system
is of three types (1) wheat straw management in rice and its residual eVect in
following wheat, (2) rice straw management in wheat and its residual eVect
in following rice, and (3) wheat straw management in rice and rice straw
management in wheat (cumulative eVect). In several studies, incorporation
of wheat straw into the soil had pronounced but variable eVects on the
growth and yield of subsequent rice (Table XV). For example, in a field
experiment on clay loam soil, rice yields under removal or incorporation of
Table XV
EVect of Wheat Straw Management on Grain Yield (t ha 1) of Rice and Its Residual EVect on the
Grain Yield of the Following Wheat in Rice–Wheat Cropping System in India
Wheat straw management in rice
Haryana, 3-year
study, clay loam
Punjab, 12-year
study, loamy
sand soil
Madhya Pradesh,
2-year study
West Bengal, acid
silty clay loam
soil, 2-year
study, Wheat
incorporated 10
days before rice
Uttar Pradesh, clay
loam soil (pH
8.6), wheat straw
(10 t ha 1)
incorporated 30
days before rice
30 kg ha
removed burned incorporated
Straw þ
(a) 100%
(b) 50%
Agrawal et al.
et al.
Pandey et al.
Sharma and
extra fertilizer N.
wheat straw were similar (Agrawal et al., 1995). On a loamy sand soil,
incorporation of wheat straw reduced rice yield by 7% (average for 12
years) compared to when it was removed (Yadvinder-Singh et al., 2004a).
Incorporation of wheat straw into an acidic clay loam soil significantly
increased the grain yield of rice, with significant residual eVect in the succeeding wheat crop (Sharma and Mitra, 1992). Similar observations were
also made by Pandey et al. (1985) and Rajput (1995). A beneficial eVect of
wheat straw on the grain yield of rice even in the first year of study has been
reported by many workers (Alam and Azmi, 1989; Zia et al., 1992).
Table XVI
EVect of Time of Rice Straw and Fertilizer N (120 kg N ha 1) Management in Wheat and Its
Residual EVect on the Following Rice in Rice–Wheat Cropping System in Indiaa
Experiment 2:
et al. (2004b)
Experiment 1:
Bijay-Singh et al. (2001)
Grain yield (t ha 1)
Grain yield (t ha 1)
Straw removed
Straw burned
Straw incorporated
(40 DBSb)
Straw incorporated
(20 DBS)
Straw incorporated
(20 DBS) and
25% N applied at
Straw incorporated
(10 DBS)
In a column, figures followed by a common letter are not significantly different.
DBS, days before sowing of wheat.
In a long-term experiment (1984–94) in the Indo-Gangetic plains of India
(Rattan et al., 1996), both rice and wheat had a higher yield with inorganic
fertilizers than any of the crop residue management treatment in the first
year. After 2 to 3 years, the combination of wheat straw and inorganic
fertilizers produced yields similar to those with inorganic fertilizers. It was
after 3 to 4 years that the combined use of inorganic fertilizers and wheat
straw started giving higher yields than inorganic fertilizer treatment.
Interestingly, green manuring in conjunction with wheat straw helped to
mitigate the adverse eVect of wheat straw in rice (Yadvinder-Singh et al.,
2004a). Similarly, Aulakh et al. (2001) reported that compared to application
of 120 kg N ha 1 through urea alone, rice production was greater with wheat
straw incorporation when an average of 86 kg N ha 1 of a prescribed 120 kg
N ha 1 dose was applied as green manure and the balance as urea N. Green
manure and incorporation of wheat straw in rice–wheat cropping systems has
the potential to increase soil organic matter while maintaining high yields.
In a field experiment conducted using 15N-labeled urea, grain yields of
wheat and the following rice were not adversely aVected by incorporation
of rice straw at least 20 days before sowing of wheat (Table XVI). In
another 7-year study, compared with residue removal or residue burning,
incorporation of rice residue 10 to 40 days before seeding wheat did not
show any adverse eVect on wheat yield (Table XVI). The application of 25%
of fertilizer N as starter N at the time of residue incorporation showed some
depression in wheat yield in all years compared with no starter N under
20-day incorporation treatment, although the diVerences were not significant.
It was suggested that N applied concurrently with straw incorporation gets
immobilized and does not remineralize easily. In this study, annual additions
of 40–50 kg N ha 1 through rice residue for 7 years did not influence grain
yield of wheat, as the recommended split application of 120 kg N ha 1 (one
half drilled at sowing and the remaining half top dressed at 21–25 days after
sowing) was already applied to all the treatments in wheat.
There also exist several other reports showing similar rice and wheat
yields under diVerent residual management practices (burning, removal, or
incorporation) (Singh et al., 1996; Walia et al., 1995). Kavinandan et al.
(1987) reported that incorporation of wheat straw 10 days before rice
transplanting and rice straw 3 weeks before wheat sowing gave 0.25 and
0.42 t ha 1 higher yields of rice and wheat, respectively, over incorporation
of wheat straw at 3 days before rice transplanting and rice straw at 2 weeks
before wheat sowing; the diVerences, however, were not significant.
In a field experiment in Faislabad (Pakistan), incorporation of rice straw
into the soil produced significantly higher yields of wheat (3.51 t ha 1)
compared to when rice straw was removed (2.91 t ha 1) (Salim, 1995).
Singh et al. (1996) reported that in Pantnagar (India), incorporation of rice
straw 3 weeks before wheat sowing significantly increased wheat yields on a
clay loam but not on a sandy loam soil. In the Himachal Pradesh state of
India, however, incorporation of rice straw at 30 days before wheat sowing
produced significantly lower wheat yields than removal or burning of straw
in the first 2 years, remained at par in the third year, and produced a
significantly higher yield and N uptake from the fourth crop onward
(Table XVII) (Verma and Bhagat, 1992). The causes of lower yields with
straw incorporation, particularly in the initial period of study, were immobilization of N and slow decomposition of rice straw at low temperatures
during wheat growth. However, with the advancement of time (fourth crop
and onward), the previously added rice straw might have decomposed,
resulting in significantly higher wheat yield and N uptake under this treatment. Straw mulch increased the wheat yield and N uptake significantly over
straw incorporation during the 2 years, which might be due to more favorable soil moisture regime, regulation of soil temperature, control of weeds,
and an increase in the microbiological activity. Yield and N uptake of
following rice under straw burn treatment did not vary significantly from
the straw removal during the entire study period. Straw mulch produced the
lower yield and N uptake of rice as compared to other straw management
treatments without N application during the first two rice crops but had a
Table XVII
EVect of Rice Straw Management on Grain Yield (t ha 1) of Wheat and Its Residual EVect on
Grain Yield of the Following Rice in Rice–Wheat Cropping System in India
Crop residue management
Pradesh, data
averaged for 4
years, acidic
clay loam soil,
rice straw
chopped and
incorporated 4
weeks before
wheat sowing
Pradesh, 5year study,
acidic clay
loam soil, rice
straw chopped
incorporated 4
weeks before
wheat sowing
Punjab, sandy
loam soil. Data
are reported
for the fourth
cropping cycle
Sharma et al.
(1985, 1987)
Verma and
V. Beri and B. S.
Sidhu (personal
significant residual eVect from the third crop onward. Straw incorporation in
wheat increased rice yield by 38% in the third crop and 45% in the fourth
crop over straw removal.
Sharma et al. (1985, 1987) observed no significant eVect of straw incorporation on the grain yield of wheat and on the following rice (Table XVII).
Pathak and Sarkar (1997) observed that at recommended fertilizer N (120
kg ha 1), rice straw incorporation produced lower rice yields than straw
In a long-term field experiment in Ludhiana in northwestern India, Beri
et al. (1995) found that incorporation of rice and wheat residues into soil
resulted in significantly lower yields than removal or burning of residues
(Table XVIII). It was suggested that the depression in rice yield was not due
EVect of Wheat Straw Incorporation in Rice and Rice Straw Incorporation in Wheat on Crop
Yields in Rice–Wheat Cropping System
and country
Beri et al.
Brar et al.
Type of
residue and
soil type
Rice straw
in wheat and
wheat straw
in rice,
sandy loam
Rice straw
in wheat and
wheat straw
in rice,
loamy sand
Rice straw
in wheat and
wheat straw
in rice,
sandy clay loam
Grain yield
(t ha 1)
Half residues
Full residues
Table XIX
EVect of Wheat Straw Incorporation in Rice and Rice Straw Incorporation in Wheat on Crop
Yields in Rice–Wheat Cropping System in India
Grain yield of
wheat (t ha 1)
LSD ( p ¼ 0.05)
Grain yield of
rice (t ha 1)
From Dhiman et al. (2000).
to N immobilization. Dhiman et al. (2000) reported that on a clay loam
soil, rice yields increased significantly with the incorporation of residues of
both rice and wheat as compared to burning or removal (Table XIX),
but wheat yields decreased with residue incorporation, particularly in the
initial 2 years of the study. In the fourth cropping season, wheat yield was
higher in the residue-incorporated treatment than residue burning or removal treatments. The average productivity during the 4-year period was
11.5 t ha 1 year 1 when residues of both the crops were incorporated, and it
was higher by about 0.61 t ha 1 year 1 than from burning and removal of
residues. In a calcareous sandy loam soil in Bihar (India), Prasad and Sinha
(1995b) studied the eVect of incorporation of crop residues after chopping (2 cm size), soaking in 2% urea solution, and then inoculating with
cellulytic culture (Aspergillus spp.) to hasten the decomposition. At recommended fertilizer levels, incorporation of crop residues compared to removal
increased mean yields of wheat and rice by 7.2 and 8.5%, respectively.
Prasad et al. (1999) concluded that residues of both rice and wheat
can safely be incorporated without any detrimental eVects on the crops of
rice and wheat grown immediately after incorporation. Rice straw was incorporated 32–42 days before sowing of wheat, and wheat straw was
incorporated 65–76 days before rice planting. Wheat yields were slightly
reduced with rice straw incorporation in the first year of study (3.7 versus
4.1 t ha 1).
In microplot experiments with early rice-late rice–wheat rotation in
China, fertilizer utilization by rice was 82.6 and 47.7% on a sandy loam
soil and 75.5 and 51.8% on a light clay soil for rice straw (C:N ¼ 89:1) plus
N fertilizer and N fertilizer alone, respectively. The grain yield from the total
rotation was also higher under rice straw plus fertilizer N than under
fertilizer N alone treatment (Xu, 1984).
In a field experiment at Yanco (Australia), Bacon et al. (1989) observed
that increasing quantities of rice stubble retained on the soil surface
increased soil NO3 –N concentrations by 46% and wheat on these plots
had a 37% increase in grain yield and 29% increase in N uptake. Bacon
and Cooper (1985) obtained higher yields from wheat direct-drilled into
undisturbed rice stubble plots over where stubble was incorporated at sowing. The high yields were due to increased availability of both soil and
fertilizer N. Delayed stubble incorporation until wheat sowing caused greater yield depression due to N immobilization than when stubble was
incorporated early after rice harvest. It is also possible that the wide range
of phytotoxins released during stubble decomposition directly inhibited
plant growth.
Fertilizer Management in Straw-Amended Soils
Incorporation of cereal residues into the soil generally causes rapid
immobilization of soil and fertilizer N during the early stages of decomposition, resulting in N deficiency in the succeeding crop. Proper management
of fertilizer N may lead to reduced rates of N immobilization by crop
residues, thus increasing the eYciency of N usage. The improved fertilizer
management practices may include optimum method, time, and rate of
fertilizer N application, which may diVer from that when residues are
removed or burned. One obvious solution to the N immobilization problem
would be to place the fertilizer below the C-enriched surface soil layer formed due to surface placement of crop residues (Doran and Smith, 1987).
Yadvinder-Singh et al. (1994b) concluded that on soils amended with crop
residues, band placement of urea prills and deep placement of large urea
granules would lead to significantly lower amounts of fertilizer N immobilization than mixed application of commercial urea granules. The limited
contact between fertilizer N and the decomposing microorganisms was
the main reason for the low rates of N immobilization with large urea
granules. The adverse eVect of N immobilization on crop growth can also
be avoided by applying additional fertilizer N or by delaying planting.
Another fertilizer management option may be to apply a part N fertilizer
at the time of straw incorporation to enhance decomposition of residues or
to allow suYcient time for the decomposition of crop residues before the
planting of next crop. Thakur and Pandya (1997) reported that preconditioning urea with rice straw and soil in the ratio 1:3:1 (urea:straw:soil) was
significantly superior to urea alone in respect of grain yield and N uptake of
a. Fertilizer N Rate. The target of eYcient nutrient management is to
maintain stable nutrient cycling in the long term while supplying suYcient
nutrients to crops in the short term. From a 3-year study on a rice–wheat
cropping system in Uttar Pradesh (India), Misra et al. (1996) reported that
total grain yields of rice and wheat increased due to the incorporation of
both the straws, with an extra dose of 20 kg N ha 1 applied at straw
incorporation over burning and straw incorporation without an extra
N dose. Singh and Sharma (2000) showed that application of 20 kg extra
fertilizer N as compared to recommended N levels of 120 kg N ha 1 to wheat
on straw removal plots gave a significantly higher grain yield and nutrient
uptake on straw-amended plots. In another study, the application of 30 kg
extra N ha 1 compared to the recommended fertilizer increased the rice yield
only slightly (Table XV) (Agrawal et al., 1995).
Sharma and Mitra (1992) found that incorporation of rice straw 15–20
days before wheat sowing decreased the grain yield, while incorporation
of wheat straw 15–20 days before transplanting increased rice yields. However, application of 15 kg N ha 1 as a starter dose with straw application
increased the yields of both rice and wheat crops. From a 2-year study, Brar
et al. (2000) reported that application of 40 kg N ha 1 at rice straw incorporation in addition to the recommended N fertilizer dose (120 kg ha 1) in
two equal splits at sowing and 3 weeks after sowing produced a significantly
higher wheat yield (4.94 versus 5.31 t ha 1) and N uptake (101 versus 116 kg
N ha 1) than application of recommended N fertilizer. Application of
irrigation at straw incorporation to enhance straw decomposition further
increased the wheat yield by 0.2 t ha 1 compared to no irrigation.
Narang et al. (1999) reported that wheat responded significantly to
the application of 160 kg N ha 1 during the first 2 years of straw incorporation (both rice and wheat) as compared to the recommended N rate of
120 kg N ha 1 when residues are removed. In the third year of the study, a
significant response to fertilizer N was observed up to 120 kg N ha 1 in
straw-amended wheat plots. Irrespective of the residue load, response of rice
to fertilizer N was also observed up to 160 kg N ha 1 in the first year of
study, but the grain yield increased significantly up to 120 kg N ha 1 in the
second year of study. Results from this study suggested the application of
25–30 kg ha 1 higher fertilizer N doses to rice in wheat on straw-amended
fields during initial 1–2 years after residue incorporation compared to the
rates recommended for straw removal fields. Later on, recommended fertilizers may be needed to achieve higher yield productivity of rice–wheat
Singh and Sharma (2000) reported that incorporation of wheat residue
(40–50 days before rice transplanting) with no N or at low N rates resulted
in an adverse eVect on crop yields of rice and wheat. When adequate
N (180 kg N ha 1) was applied, residue incorporation increased productivity by 0.4–0.7 t ha 1 and nutrient uptake by 40–65 kg ha 1 over removal or burning of residues. Residue incorporation increased eYciency of
applied fertilizer N in rice and had a significant residual eVect in following
Thakur and Singh (1987) estimated optimum N rates of 115 and 140 kg
ha 1 for rice on fields without and with wheat straw (5 t ha 1) incorporation.
Thus, higher fertilizer N may be required for crops grown on soils amended
with crop residues to get maximum benefits. Jha et al. (1992) obtained the
highest grain yield of rice when rice straw and green manure (mungbean)
along with 60 kg N ha 1 was applied in three equal splits compared to rice
straw or green manure applied alone. Grain yield of following wheat was
also higher when rice straw and green manure were incorporated.
In a 2-year field experiment in Modipuram (northwestern India), incorporation of half of the crop residues along with recommended fertilizers
consistently produced higher yields of both rice and wheat than incorporation of full residues or removal of residues (Sarkar, 1997). The rice grain
yield with half residue was 5.80 t ha 1 and wheat yield was 4.38 t ha 1
compared to 4.70 t ha 1 of rice and 3.71 t ha 1 of wheat for no straw
In a rice–wheat rotation on a calcareous soil, application of crop residues along with FYM gave the highest yield followed by FYM, crop
residues, and no amendment (Prasad and Sinha, 1995b). The grain yield
recorded with 50% NPK plus FYM plus crop residues was higher than that
with the 100% recommended dose of NPK alone, indicating that FYM
plus crop residues substituted 50% of NPK in each of wheat and rice
Malik and Jaiswal (1993) reported that application of 58 kg N ha 1
as urea super granules plus 28 kg N ha 1 as wheat straw produced a
significantly higher rice yield than the recommended practice of applying
87 kg N ha 1 as commercial urea granules. The grain yield of the following
wheat was not aVected by the previous residue management practices in the
rice crop. Under dryland conditions in Uttar Pradesh (India), Singh and
Singh (1995) found that incorporating rice straw (10 t ha 1; C:N ¼ 75.5) 3
weeks before planting rice integrated with 50% of recommended NPK
fertilizers produced the highest rice yield and improved the profitability
of the system. Rajput (1995) found that incorporation of wheat straw
(10 t ha 1) resulted in up to 50% savings in the recommended NPK fertilizers
(60 kg N þ 13.1 kg P þ 25 kg K ha 1). A higher yield potential of rice was
achieved when wheat straw was applied along with recommended NPK
fertilizers (Table XV). The residual eVects of wheat straw on the following
wheat were also substantial. In another study (Kundu et al., 1994), however, wheat straw applied to rice had little eVect on the grain yield of the
succeeding wheat crop. In fact, several other long-term studies also showed
that it is not possible to substitute a part of fertilizer N requirement of rice
with N added through wheat straw (Table XX).
Katyal et al. (1998) reported results from long-term field experiments
conducted at five sites in India during 1983–1991. At Kanpur (U.P.,
India), 50% recommended NPK fertilizers plus 50% N through wheat straw
in rice followed by 100% recommended NPK fertilizers in wheat stabilized
the yields of rice and wheat. However, at the three other locations it was not
possible to substitute a part of fertilizer N (25% N) with crop residues
(Table XX). In West Bengal (India), the application of semi-decomposed
wheat straw (0.78%N, dry weight basis) at 3 t ha 1 along with 25 or 50% of
the recommended NPK fertilizers (80 kg N þ 26 kg P þ 33 kg K ha 1) 30
days before sowing of rainfed rice resulted in the highest yields. In a rice–
wheat cropping system, 5 t wheat straw ha 1 was incorporated 3–10 days
before rice transplanting and 5 t rice straw ha 1 was incorporated 2–3 weeks
before seeding wheat (Kavinandan et al., 1987). Most of the studies on rice
straw management were conducted at recommended N rates, and thus it is
diYcult to quantify the contribution of rice straw in supplying N to plants in
the cropping system due to the fact that the amount of N fertilizer applied
exceeds that for optimum yields.
b. Time and Method of Fertilizer Application. Sharma and Bali
(1998) showed that application of 30 kg N ha 1 at straw incorporation
and remaining 90 kg N ha 1 top dressed during wheat growth soil produced
Table XX
EVect of Wheat Straw and Fertilizer Management in Rice and Their Residual EVect on Wheat in Rice–Wheat Cropping System in India
Uttar Pradesh (Faizabad)
Madhya Pradesh, 3-year study
Punjab (Ludhiana)
8-year study
Uttar Pradesh (Kanpur)
8-year study
Madhya Pradesh (Jabalpur)
8-year study
West Bengal (Kalyani)
8-year study
100% NPK
100% NPK
50% NPK þ
50% N as WSa
100% NPK
75% NPK þ
25% N as WS
75% NPK
( p ¼ 0.05)
Kumar and Yadav (1995)
Dubey et al. (1997)
Katyal et al. (1998)
Katyal et al. (1998)
Katyal et al. (1998)
Katyal et al. (1998)
WS, wheat straw.
a significantly higher yield (2.0 versus 3.12 t ha 1) than that from applying
120 kg N ha 1 in two equal split doses (half drilled at sowing and half top
dressed at 1 month after sowing) on a silty clay loam. Incorporation of rice
straw reduced the wheat yield over straw removal at recommended N level
through its beneficial eVect on residue decomposition. In field experiments
using N-labeled urea, application of a part of the recommended N (25%) at
the time of straw incorporation (to hasten decomposition of straw) led to
large N losses and low wheat yield (Bijay-Singh et al., 2001). Recovery
of 15N by wheat was maximum (41.8%) when rice straw was removed or
burned and the minimum (30.4%) when 30 kg of 120 kg N ha 1 fertilizer was
applied along with straw incorporation at 20 days before wheat sowing
(Table XVI).
From long-term field experiments (1990–96) in three locations in India,
Yadav (1997) reported that when 20 kg N ha 1 was applied at the time of
incorporation and the remainder of the recommended N (100 kg or 120 kg
N ha 1 for rice and 120 kg N ha 1 for wheat) during the growth period,
grain yields of rice and wheat were significantly lower than those obtained
with other N scheduling practices included in the study. However, when an
extra 20 kg N ha 1 was applied at the time of residue incorporation over and
above the recommended N dose at one location (Jammu and Kashmir) or
when N levels were enhanced by 20 kg N ha 1 over the recommended rates
at the other site (Uttar Pradesh), the crop yields were the highest. On average
of six crop cycles, these practices produced an additional 150 and 510 kg
grain ha 1 at the first site and 570 and 810 kg grain ha 1 at the second site in
rice and wheat crops, respectively. Jiang et al. (1998) recommended the
application of 105 kg N ha 1 in three equal splits (sowing, tillering, and
stem elongation) on plots amended with wheat straw (3 t ha 1). Split
application of fertilizer N increased wheat yield by 43% compared to fertilizer N applied in single or two splits. Bacon and Cooper (1985) found that
application of N to wheat at tillering or stem elongation, compared to at
sowing, significantly increased soil mineral N content at least until anthesis.
Wheat on the stubble-incorporated plots did not respond significantly to
N application at sowing or stem elongation, while N application at any time
more than doubled wheat productivity on the stubble-retention plots. Delaying N application until tillering significantly increased yields on stubble
incorporation, stubble retention and burned plus till plots. While only 70
kg N ha 1 was required for maximum yield at stem elongation, 140 kg
N ha 1 was necessary at sowing. It was concluded that stubble and fertilizer
management techniques could be manipulated in order to regulate soil
mineral N status, which in turn determined plant N uptake and yield of
wheat. In Kanto (Japan), with continuous application of rice and wheat
straw, rice yields were low during the initial 3–4 years, but the yield of rice
increased dramatically by continuous application of rice straw (Roppongi,
1987). The adverse eVects were alleviated when the basal application of
N was increased.
Effect on Crop Yields
Rice–rice is a dominant cropping system in Bangladesh, China, Philippines,
Korea, Japan, Indonesia, and eastern and southern parts of India. Studies
with rice straw in rice–rice systems in a number of countries demonstrate
widely varying response to straw incorporation. In sharp contrast to rice–
wheat cropping systems, the majority of the studies on rice–rice cropping
systems show that incorporation of rice residues enhances rice yield and
N use eYciency. Ismunadji (1978) from Indonesia reported that incorporation of 10 t rice straw ha 1 increased grain yield of rice to 2.6 t ha 1 from
2.2 t ha 1 in the untreated control. Burning of rice straw produced a rice
yield almost similar to that obtained with its incorporation. Experiments in
the Philippines showed that straw incorporation for more than 6 years
increased the rice yield by 0.4 to 0.7 t ha 1 compared with fields that used
to receive chemical fertilizers and where rice straw was either removed or
burned (Table XXI) (Ponnamperuma, 1984). On the unfertilized plot, when
the soil was a P-deficient acid clay, the extra yield amounted to 23%
(Ponnamperuma, 1984). The beneficial eVects of rice straw incorporation
were, however, small during the initial 2–6 years of the study. Sharma et al.
(1989), however, observed no significant eVect of rice straw incorporation
(6 t ha 1) on grain yield of rice in a short-term study in the Philippines. In a
long-term study in Japan, Beaton et al. (1992) noted no beneficial eVects of
Table XXI
EVect of Long-Term Management of Rice Straw on Crop Yields (t ha 1) in Rice–Rice
Cropping System
Experiment 1
Experiment 2
Experiment 3
Experiment 1 and Experiment 2 from Ponnamperuma (1984); Experiment 3 from Beaton et al.
incorporation of rice straw over straw removal in the initial 15 years of the
study (Table XXI), but in the final 4 years (20–23 years), an average increase
of 0.70 t ha 1 in rice grain yield was observed with rice straw incorporation
over removal.
The length of the period allowed for decomposition of crop residues
before the sowing/planting of the next crop aVects the agronomic response to applied residues. Houng and Hwa (1975) found that when rice
straw was allowed to decompose for 4 or more weeks before sowing, there
was no adverse eVect on germination of rice seeds. In many other studies,
crop residues were allowed to decompose for 2 or more weeks before
rice transplanting to avoid the adverse eVects of phytotoxicity and
N immobilization on crop growth (Ali et al., 1995; Lanjewar et al., 1992;
Wu et al., 1997). Sharma and Mitra (1990) observed that rice yields were
increased significantly when rice straw was applied 30 days before transplanting, and rice straw also exhibited a favorable residual eVect on the yield
of the second rice crop.
Witt et al. (2000) reported that early residue incorporation improved the
congruence between soil N supply and crop N demand by wet season rice,
especially during the vegetative stage of crop growth. This has resulted in
13–20% greater rice yields with early (60–63 days before transplanting)
compared to late (14–15 days prior to transplanting) residue incorporation
in rice–rice systems without applied N or with moderate rates of applied N.
From South Korea, Han et al. (1991) reported that the application of
7.5 t rice straw ha 1 along with the recommended dose of fertilizers produced a rice yield of 4.8 t ha 1, which was significantly higher than the
4.3 t ha 1 obtained with the application of recommended fertilizers alone.
Similarly, Sistani et al. (1998) from Malawi, Lee et al. (1995) from South
Korea, Beye (1977) from Senegal, Ali et al. (1995) from India, and Gotoh
et al. (1984) from Japan also observed beneficial eVects of rice straw incorporation on rice yield. From a 6-year field experiment, Finassi (1976)
observed that incorporation of rice straw caused a significant increase in rice
yield at the highest rate of N (120 kg ha 1) application only.
In a field experiment over four cropping seasons in Andhra Pradesh
(India), Vamadevan et al. (1975) obtained the highest rice yield and
N uptake in all the seasons when the rice straw was incorporated into the
soil, but there was a small eVect at high levels of applied N (100 or 150 kg
N ha 1). Houng and Lin (1976) and Oh (1979) observed that incorporating
rice root residues into the soil increased rice yield on a poor soil rather than
on a fertile soil. In a pot study, incorporation of 10 t rice straw ha 1 into
0–6 cm soil layer without fertilizer increased rice grain yield by 21.2%, while
application of fertilizer N along with rice straw increased yield by 52.5%
(Rao, 1973). The residual eVect of rice straw in the following rice crop was
equal to 42% increase in grain yield. The beneficial eVect of rice straw
was explained by the increased biological N2 fixation by the free-living
microorganisms in the flooded soil amended with rice straw.
Adverse as well as no eVects of incorporation of crop residues into soil
on rice yield have also been documented. In poorly drained paddy fields,
incorporation of rice straw adversely aVects the rice growth due to the
presence of strong reducing conditions in the soil (Kuboto, 1984). On such
fields, it is recommended to incorporate the rice straw shortly after harvest.
Allowing decomposition of the straw over longer period alleviates the
injury occurring in the initial growth of the rice. In paddy fields with heavy
clay soils, rice yield decreased in the first and second year after straw
incorporation; however, after 3–4 years of continuous application, as the
amount of soil mineralized N increased and reduction in soil becomes less
pronounced, plant growth and the yield increased (Kuboto, 1984). Proper
water management, which includes drainage, is also important in such soils.
In a lysimeter experiment, Kondo et al. (1980) observed that rice straw
tended to decrease the rice yield in the presence of fertilizer N compared
with no straw. Corft et al. (1985), however, reported that 6 t rice straw ha 1
along with recommended N, P, K, and S fertilizers showed no eVect on rice
yield over N, P, K, and S fertilizers alone.
Integrated Management of Fertilizers and Crop Residues
Crop residues incorporated during fallow periods will cause immobilization of soil N, but net N mineralization occurring during the
following cropping season would need to be accounted for when evaluating N requirements of the following crop. To determine the amount
of N fertilizer that can be reduced with annual straw incorporation, a
N fertilizer response study was conducted by Eagle et al. (2001). As the
level of N fertilizer applied increased, grain yield increased when straw was
burned or incorporated. However, grain yields when straw was incorporated
continuously for 5 years were higher than when straw was burned. These
trials indicated that N fertilizer application can be decreased when straw is
incorporated, because no yield response was further observed when more
than 115 kg of N ha 1 was applied. It was recommended that N rates can be
decreased by at least 30 kg N ha 1 after 5 years of straw incorporation.
Clearly, an active, labile N pool was formed when straw was incorporated
that led to a reduction in fertilizer N dependency for rice.
In long-term experiments carried out at four locations in India, Hegde
(1996) observed that at three locations it was not possible to substitute a part
of N (25% of the recommended N) needs through rice straw without adversely aVecting crop yields (Table XXII). In Kerala, located in deep South
of India, however, rice straw incorporation could substitute for 25% of
Table XXII
EVect of Integrated Management of Inorganic Fertilizers and Rice Straw on Crop Yields (t ha 1)
in Rice–Rice Cropping System at Four DiVerent Locations in India
Rainy season
100% NPK
75% NPK
50% NPK
75% NPK þ
25% N as RSe
50% NPK þ
50% N as RS
LSD ( p ¼ 0.05)
Rainy season rice
Winter season rice
Winter season Site 1 Site 2 Site 3 Site 4 Site 1 Site 2b Site 3c Site 4d
100% NPK
75% NPK
100% NPK
100% NPK
100% NPK
Site 1: Kharagpur (West Bengal), data averaged for 5 years, Udic Ustochrepts sandy clay loam
soil (pH 5.4).
Site 2: Bhubneswar (Orissa), data averaged for 10 years, Haplustalts sandy loam soil (pH 5.9).
Site 3: Maruteru (Andhra Pradesh), data averaged for 4 years, Chromustrets clayey soil
(pH 7.0).
Site 4: Karemane (Kerala), data averaged for 8 years, Typic Tropfluvent sandy loam soil
(pH 5.2).
RS, rice straw.
From Hegde (1996).
fertilizer N needs of kharif rice. The relatively high temperatures during both
kharif and winter seasons in Kerala might have helped for quick decomposition of rice straw. At low-fertility levels (50% N, P, K), application of rice
straw significantly increased the grain yield of rainy season rice over no
straw treatment at all the four sites. Raju et al. (1987) and Elankumaran
and Thengamuthu (1986) also concluded that it is not possible to substitute a
part of N through rice straw due to its high C:N ratio, which causes
immobilization of soil and fertilizer N. In a field trial in eastern India,
Bhattacharyya et al. (1996) recorded N substitution of up to 50% of the
recommended fertilizers from the incorporation of 5 t rice straw ha 1 in an
acidic red soil. Russo (1974) reported that incorporating 3–6 t chopped rice
straw ha 1 with 65 kg N plus 28 kg P ha 1 produced a slightly higher rice
yield than that obtained with the application of 120 kg N plus 43.1 kg P plus
110 kg K ha 1. Kamalan et al. (1989) obtained 0.3 t ha 1 of additional rice
yield with the application of urea super granules (USG) combined with rice
straw over USG alone. In another study in Malawi, Sistani et al. (1998)
observed that on rice straw-amended plots, application of urea in briquette
form compared with prilled urea significantly increased rice grain yield in
two of the three experiments.
Raju and Reddy (2000) conducted a 6-year study on a clay loam soil to
investigate the eVect of rice residue management on crop yields in rice–rice
system. Rice straw equivalent to supply 25 and 50% fertilizer N requirement
of the rainy season crop was incorporated before rice transplanting, and
the residual eVect of rice straw was studied in the succeeding winter rice.
The incorporation of rice straw along with recommended fertilizers proved
superior to inorganic fertilizers alone in increasing rice yield, soil organic
matter, and available K contents in soil. This study showed that it is possible
to reduce the total fertilizer needs of both rainy and winter season rice by
25% without any adverse eVect on system productivity. The N balance was
negative in all the treatments, but the P balance was positive. The K balance
was positive when 50% of the fertilizer N was applied as rice straw.
In a greenhouse experiment, Shen et al. (1993) obtained a higher rice grain
yield by adding 10 g wheat straw plus 350 mg urea N kg 1 soil, but the 10 g
straw plus 150 mg urea N treatment registered higher fertilizer use eYciency. It
was suggested that an adequate N supply to rice plants could be maintained by
applying suYcient N fertilizer with straw to have a C:N ratio equal to 20.
Huang and Lu (1996) observed no adverse eVect of rice straw on plant growth
and total 15N recovery when rice straw was incorporated along with (NH4)2SO4
at a C:N ratio of <25:1. The application of rice straw helped alleviate the
adverse eVects of excessive N rates and thus increased the rice growth and
yield response to high N rates. Patel et al. (1997) found that preconditioning
of urea with rice straw plus soil in the ratio of 1:3:1 was significantly superior
to urea alone with respect to grain yield and N uptake of rice.
In Maharashtra state of India, Bulbule et al. (1996) observed that management practices consisting of basal incorporation of rice straw (2 t ha 1)
integrated with deep placement of urea briquette (2.1 g per four hills) at
transplanting using a modified 20 20 cm spacing produced significantly
higher grain yield (average increase of 1.3 t ha 1) of rainfed transplanted rice
than did the rice straw incorporation integrated with two equal splits of
prilled urea at the same N rate. The additional yields obtained with deep
placement of urea briquettes were possibly due to reduction in immobilization of fertilizer N over top dressing of prilled urea. Similarly, Elankumaran
and Thandamuthu (1986) reported that USG combined with rice straw
produced about 10% higher rice grain than USG alone at the same rate of
N application in south Indian state of Kerala.
Pandey and Tripathi (1992) reported that application of 33% of recommended fertilizer N (87 kg ha 1) through rice straw plus 66% as prilled urea
produced a rice yield of 4.7 t ha 1, which was higher than that obtained with
the application of prilled urea in three equal splits (4.0 t ha 1). In South
India, Ramaswami (1979) observed that incorporation of 10 t rice straw
ha 1 along with 39 kg P ha 1 produced a rice yield that was similar to that
obtained with the application of 180 kg N plus 39 kg P plus 75 kg K ha 1.
The straw application increased biological N2 fixation and P availability in
soil. Rutu and Widjaja (1994) reported that application of crop residues
increased N and P use eYciency from <20 in control to 40 kg grain kg 1
N and from 30 kg in control to 60 kg grain kg 1 P. Similarly, Sudjadi et al.
(1989) observed a twofold increase in N use eYciency in rice after 4 years of
incorporation of rice straw in Indonesia. In another field experiment in
Indonesia, incorporation of 3 t rice straw ha 1 increased grain yields of 13
rice cultivars by up to 1.3 t ha 1 and significantly increased the K content of
shoot compared with untreated control (Ismunadji et al., 1973). It was
suggested that rice straw could be incorporated into soil as a substitute for
K fertilizer. Kwak et al. (1990) reported that rice straw application increased
rice yield in the no-P plots.
Becker et al. (1994a) reported that co-incorporation of green manure and
straw from previous rice depressed yield and N use eYciency in the second
rice dry season when yield potential is high. A residual eVect equivalent to
10% increase in grain yield however, was, observed in the third rice crop.
Synchronizing soil N supply with N demand by incorporating residues with
suitable chemical fertilizers may not immediately increase rice grain yield but
will improve long-term soil fertility.
In pot and laboratory studies using 15N-labeled rice straw and
N-labeled (NH4)2SO4, Hwang et al. (1993) found that 17.5 to 23.5% of
straw N was mineralized during rice growth, of which 30–50% was subsequently absorbed by rice plants. Rice straw inhibited plant growth in its
early stages, but application of fertilizer N with rice straw stimulated its
decomposition, thereby increasing the mineralization of straw N and
subsequent recovery by rice plants and reducing levels of residual N in
soils. He et al. (1994) measured N recovery by rice grain from rice straw
N incorporated in the field from 9.8 to 14.5%. At IRRI, Philippines, Broadbent and Reyes (1972) obtained a high recovery of fertilizer N (>72%) by
rice in two soils even when chopped rice straw was incorporated under
greenhouse conditions. In a pot study in Japan, Shiota et al. (1984) found
that recoveries of fertilizer 15N and rice straw compost N were 58–61% and
13–15%, respectively. Total N uptake at harvest was the highest when rice
straw compost was applied along with fertilizer. Li et al. (1981) from China
reported that rice straw incorpotaton increased N uptake by rice compared
to straw removal.
In Bangladesh, continuous cropping with rice–rice–rice has led to destruction of soil clay through the process known as ferrolysis This is the major
degradation process in the fine-textured soils, which may be one of the causes
of yield decline/stagnation of rice, including organic matter depletion (Farid
et al., 1998). Rice straw incorporation after each rice harvest continuously for
12 years resulted in average yield of 10.99 t ha 1 (for three rice crops in a year)
compared with 6.48 t ha 1 in the traditional practice of growing rice. The
sustainability of rice yields was also achieved with straw incorporation, as rice
grain yields declined from 7.15 t ha 1 in 1984–85 to 6.15 t ha 1 in 1994–95
under the farmers’ practice, but grain yields increased from 9.50 t ha 1 in
1984–85 to 12.20 t ha 1 in 1994–95 under straw incorporation treatment. The
yield increase and sustainability in the straw-incorporated treatment was
ascribed to improvement in the physical and chemical properties of soil
and the release of several plant nutrients such as K, S, and micronutrients
from the decomposing straw, thereby increasing soil fertility. On average, 12
t ha 1 of rice straw were incorporated each year, which recycled 70 kg N, 12
kg P, and 166 kg K ha 1. The nutrient balance for N, P, and K was positive
in straw-incorporated plots, but large negative nutrient balances were
recorded when rice residues were removed from the fields.
Green gram (Vigna radiata), black gram (Vigna mungo), cowpea
(V. unguicuata), soybean (Glycine max), and groundnut (Arachis hypogae)
are the important summer food legumes that fit into the rice-based multiple
cropping systems; they leave substantial amount of residues. Legume residues provide biological fixed N to the next crop in addition to the benefits
oVered by non-legume crops. Food legumes are capable of fixing large
amounts of N, but removal of seed or green pods can constitute an export
of considerable N. Accumulation of N by legumes in tropical rice-based
cropping systems is influenced by water regimes, inoculation, soil fertility,
nutrient supply, and soil and crop management (Buresh and De Datta, 1991;
Yadvinder-Singh et al., 1994a).
The quantity of N in above-ground residues remaining after grain harvest
ranges from 17 kg to 101 kg N ha 1, and N fertilizer equivalent ranges from
37 to 100 kg N ha 1, with a mean value of 40–45 kg N ha 1 (Buresh and De
Datta, 1991). In addition to above-ground residues, roots of food legumes
contain up to 40 kg N ha 1 at final harvest, which is progressively released
during crop growth. This may explain in part the reported considerable
benefits of legumes to a subsequent rice crop despite a modest return of
N in residues (Kulkarni and Pandey, 1988) or even when all the aboveground residues had been removed prior to cultivation (De et al., 1983).
Residues of grain legumes after harvesting pods have a lower N content than
those of green manure but rapidly release N in tropical flooded soils. The
potential benefits of legume residues and legume green manures in rice other
than fixed N are described by Buresh and De Datta (1991) and YadvinderSingh et al. (1991). Studies with 15N-labeled legume residues indicate that
N recovered by subsequent rice ranges from 25 to 45% of the N originally
contained within the legume residues (Yadvinder-Singh et al., 1991).
In a rice–wheat-mungbean system, Rekhi and Meelu (1983) incorporated
straw of mungbean grown after wheat just before rice planting and observed
an increase in rice yield equal to the application of 60 kg fertilizer N ha 1.
Similar results were also obtained by Bhandari et al. (1992). Incorporation
of legume residues after harvesting grains exhibited high agronomic
eYciency and apparent N recovery compared to cowpea used as green
manure under both lowland and upland conditions in Philippines (John
et al., 1989). Prasad and Palaniappan (1987) reported than incorporation
of mungbean and soybean residues along with recommended fertilizers
produced the highest rice yield. In a 2-year field experiment on lentil-rice
crop rotation, Prasad et al. (1990) found that incorporation of lentil residues
(2.7 to 5.6 t ha 1) in rice exhibited no additional benefit over lentil root
biomass. Incorporation of lentil residues,