Geographic Coordinate Systems
Datums are one component of a geographic coordinate system (GCS). A GCS includes a datum, angular unit of measure, and a prime meridian. Datums are often thought of as geographic coordinate systems because it is very rare that the units are not degrees (versus grads) and often utilize the same prime meridians.
Geographic coordinate systems use latitude and longitude values to describe point locations on earth. Latitude and longitude are angle measurements from the earth’s center to locations on the surface. For example, in Figure 4 there is a point location that can be described as 60 degrees east of the prime meridian and 40 degrees north of the equator. The units are degrees because it is difficult to make surface measurements on a curve. We use geometry and describe the distance with the angle from the center of the earth to the surface. If we had a second point, say at 75 degrees east and 15 degrees north, we wouldn’t describe the distance between the two points as 15 degrees latitude and 25 degrees longitude. This is one of the main reasons we flatten the curved models of the earth and deal with spatial data on Cartesian or planar grids.
Geographic coordinate systems are spherical models of the earth. They are curved surfaces measured in divisions of latitude and longitude with typically units of degrees (decimal degrees or degrees minutes seconds which divide a full circle into 360 parts). Latitude lines are horizontal parallel lines (parallels), and longitude lines are vertical lines that converge at the poles (meridians). The distance between lines of latitude is consistent. The distance between longitude lines varies from the equator to the poles. Latitude and longitude are inconsistent units of measure relative to each other, which becomes important when we start dealing with continuous surface data (rasters). The grid of parallels and meridians is called a graticule.
Again, Geographic coordinate systems are curved coordinate systems. All cartographic representations must address the shape of the earth first. Thus, we define our location first relative to a coordinate system that models the curved shape of the earth’s surface. Geographic coordinate systems form the first level of defining where we are on the earth. All spatial data must first address the curved surface of the earth, thus all coordinate systems will have a geographic coordinate system “core” that defines the spheroid, geoid, units of measure, and prime meridian on which all further measurements and calculations will be based.
When we display geographic coordinate systems in ArcGIS, degrees are treated as linear or planar units of measure. That’s how we end up with the severe east/west stretch in the northern and southern latitudes. One degree at the pole narrows to 0 km wide. But at the equator, a degree is approximately 111 km wide. The Plate Carrée projection is very similar to the display of degrees of lat/lon in ArcGIS.
So we have spent a lot of brain power trying to define a great set of models we can use as proxies for the earth’s curved surface in order to accurately define where things are. Why do more? Why not just work in Geographic Coordinate Systems?
Area and distances are much easier to measure on a flat surface than on a curved surface, to be sure. But computers should be able to handle that kind of math without frying any circuits. However, GIS is a complicated, hot mess. And we are often using data structures that simply depend on X and Y being consistent units of measure relative to each other. When a raster grid is displayed in a geographic coordinate system, cells or pixels that should be square display as rectangles or trapezoids. As powerful as computers are, the complexity of GIS software still demands that we convert our curved models to flat planes, just as conventional maps are forced to be drawn on flat media (paper maps).