Datums
What Are Datums – in Geodesy and Mapping?
A datum is the answer to the practical problem of making an accurate map. If you wish to determine the relative location of a pair of points a few meters apart, the solution is obvious. Just measure the difference with a tape measure. The issue of orientation still exists though, but this can be solved using two "known" points to measure a third. Or observations of the stars can be used to define north.
In effect this defines a local datum. The known point, together with some method for determining the direction of north, define the location of points measured from it. If the reference point is in error by 100 m north, so will all the points using it. They move together. This of course assumes these errors are small, at least as compared to the radius of the earth.
If you look at the legend of a topographic map, you will find that it lists the "datum" that is used. In fact there may be several datums, one for horizontal, one for vertical etc. These are important because they define the reference system that is used for the coordinates. If you use a GPS navigation system not set to the map datum, you can be off by 100m (usually) to a kilometer (sometimes).
A datum can be defined by specifying the ellipsoid, the coordinates of a single point and the direction north. The point ties down the ellipsoid to the physical earth and also implicitly defines the placement of the center of the earth. This location is called the primary reference point. For North America, it is in central Kansas at a place called Meade's Ranch.
The practical way to define a datum is with a whole set of reference markers and their associated coordinates. They should be carefully surveyed together. This gives a network that serves as a "realization" of the datum. This provides a practical set of points spread out over the region covered. Surveyors use the closest survey marker that meets the accuracy needs . You don't have to start all surveys in North America in Kansas.
This means that datums are the reference frames used in the construction of maps.
Things are complicated in practice as the same area may have two datums giving each point two different coordinates. In addition datums were often generated by individual countries, and did not match at the boundaries.
While there are about 20 common ellipsoids, most ellipsoids are used in multiple datums. For example the number of datums on some popular ellipsoids are:
| Ellipsoid | Number of Common Datums |
| Airy 1830 | |
| Bessel 1841 | |
| Clarke 1866 | |
| Clarke 1880 | |
| International 1924 | |
This understates the true complexity and confusion. For example the North American Datum of 1927 (NAD 27) is listed once for Clarke 1866 in the above table. But NAD 27 is really about 25 different datums. They are all on the Clarke 1866 ellipsoid, but have different primary reference points. The continental NAD 27 is different than that in the Bahamas or the one in Greenland or the one in Cuba ... This issue has not gone away with the newer NAD of 1983. There are different datums called NAD 83 in different areas.
In general if a datum covers areas not directly connected by a survey, such as over a body of water, there are really different versions of the datum. This is not an academic point. At least one recent ship grounding in the Caribbean was caused by using the wrong "NAD 83" in a GPS receiver. They used a version of "NAD83" that did not match the "NAD 83" on the chart.
|
||||
|
|
|
|
|