One of the most mysterious bits of arcana for new GPS users is the selection of datum. Almost as confusing is the choice of traditional latitude/longitude, versus UTM (Universal Transverse Mercator). The native datum for the GPS satellite network is WGS84, but many USGS maps are relative to NAD27. In other parts of the world, there is a multitude of datums from which to choose. One may get stern and emphatic opinions about these choices.

Actually, the choices rarely matter if you are consistent and use a computer-based mapping program with your GPS.

First off, you should read the many good web explanations of the datum-- e.g., check the references at http://www.gpsinformation.net. A working definition of datum -- which you may wish to skip, is: "The datum is the 'name given to a smooth mathematical surface that closely fits the mean sea-level surface throughout the area of interest' (Snyder, 1987). This surface may be defined in two general ways. Older datums use a particular ellipsoid definition, related to an initial surveyed point. NAD 27, the datum employed by the 30m DEMs, uses the Clarke 66 ellipsoid and has an initial point at Meades Ranch, Kansas. Any point may be specified by survey to this location. Newer, earth-centered datums use satellite and terrestrial data to specify an ellipsoid without recourse to a survey origin on the earth's surface. WGS 72, used by the 3 arc second DEMs, is such a datum. Changing datums is not difficult using a computer. The result is a transformation of x and y coordinates; the magnitude of the difference is dependent upon the location of the transformed points."

(http://www.ncgia.ucsb.edu/~ashton/masters/DigTerrain.html for above definition)

 

Second, recognize that if you elect to use UTM, you still need to select a reference datum. UTM is not quite as "Universal" as the name implies -- there are parts of the world where it works only with modification. And, your GPS may not behave as expected when you change UTM zones. Here in Las Vegas, we are in UTM zone 11; but right across the border in Utah and northern Arizona is another zone. When you cross into a new zone, the display changes to give coordinates relative to the new zone. Still, many people swear by UTM, principally those who MUST use a GPS with big maps in the field.

More on UTM: http://www.backpacker.com/article/1,2646,6985,00.html

Before you get hooked on UTM, recognize one of the common complaints about latitude and longitude -- the necessity to use minutes and seconds, and count in base 60 like an ancient Sumerian -- is hooey. You can set a GPS and the mapping software to report your position in decimal degrees.

The datum selection makes a difference only if you try to relate the readouts on your GPS to a list of externally obtained coordinates (that includes a computer file you might download off the web), or to a paper map external to your GPS. By "external", I refer to a map that was created with no awareness of the datum used in your GPS. If you use original paper maps on your treks -- say the official USGS maps -- and you want your GPS coordinates to correspond to those on the paper maps, you must choose the GPS datum that matches the datum printed on the map.

If you upload saved tracks or waypoints to your GPS, from a file on the computer, you must also make sure your GPS and that file have consistent datum before and during the upload. After the upload, you can set the GPS to whatever datum suits your fancy.

If you don't use the coordinates printed on the paper maps, and aren't going to punch in coordinates you got from outside your GPS (e.g., on the web, e-mailed by a friend), or read a computer file, then the datum choice is irrelevant. The GPS is internally consistent, and if you keep a track log or mark your starting point on the GPS, a proper GOTO will still get you back home, regardless of the datum setting (assuming you have good satellite reception). Furthermore, if you use a mapping program with your GPS, and you use that program to upload waypoints and print any maps you will use in the field, in general you need only be sure that the GPS datum matches the datum you've selected for the mapping program. The vast majority of mapping programs will automatically adjust the printout to a particular datum selection. The programs won't change the location of UTM grid lines in bitmaps (scanned from USGS maps). But if you are using the mapping program to its fullest, you are probably ignoring the grid lines (which only exist on newer maps, anyway). In addition, many programs will superpose a correct grid on your printout. However, if you send coordinates to someone else, you must specify the datum -- or he/she might misinterpret your coordinates by several hundred meters.

Now, there are probably some of you who use a GPS in the old fashioned way; you read the coordinates on the GPS, and you look for the position of those coordinates on a true-blue USGS paper map. If you are lucky, the map has a UTM kilometer grid. So you put the map down on a flat dry surface on the mountaintop, pull out your plastic UTM grid tool and mechanical pencil, put on your reading glasses, and plot the coordinates to +/- 10 meters on the big paper map. And then you flap your wings and fly. If your GPS doesn't have a data port for a PC, this may be your only option when in the field, if you really need to relate your position to an external map, and can't use obvious cues to remove ambiguity from your position. Such may be the case if you are in deep, featureless woodland (good luck getting satellite reception, in that case).

But with the "newer" mapping programs (which have been around longer than WAAS and the end of SA), you are more likely to do the following. You set the program on your PC/Mac/LinuxBox to the same datum as on your GPS, once and for all. Before a particular trip, you look at maps (on the computer screen) for your desired goal. You plot routes, tracks, or waypoints on the PC screen; then you upload those data to your GPS. On the trip, you look at your position relative to those tracks, routes or waypoints on the GPS screen, and you blithely ignore the coordinates associated with those points, because ALL that concerns you is your relative position. Occasionally, you may pull out a paper printout map, produced by the same software, and compare your progress relative to the same routes, tracks or waypoints. If you are fortunate enough to have fairly accurate topos loaded into the GPS (as the Garmin Mapsource or the newer Magellan 3D topos), you may not need a paper printout, ever.

I typically record a track when I use my GPS in trailless territory. Yes, that is battery-intensive; but I use rechargeable batteries, principally do day hikes or overnights, and have never had a GPS run out of juice on a trip. My current GPS will go about 25 hours on a pair of AA 2500 mAh NiMH rechargeables. I always carry at least one spare set of batteries (one Li for cold trips), so I can go for nearly a week, recording a track during each 10-hour hiking day. For many of my hikes, there are long class 3 or 4 sections, where it is good to be able to find my return route to +/- 3 meters, not +/- 10 meters. For long backpacks on trails, I wouldn't use a GPS often at all (after all, I did many miles of confusing trails before personal GPS was available). Then I would turn the GPS on occasionally, and look at position relative to waypoints. The maps loaded on my GPS have mountaintops and other features as built-in waypoints, and it is only important that I know my position relative to my goals. Yes, occasionally I might have to resort to plotting my position on a map, which I can still do.

This method may seem like sacrilege to old-timers; but it generally works very well. If you must work with a bunch of folks who gather around a paper map to chart movements -- such as a search and rescue team -- you might change your GPS to that map datum, at least on that day, rather than argue. Search and Rescue teams tend to take their opinions very seriously.

And, to repeat: if you get a list of coordinates from someone, or a track/route to upload, you have to be aware of the datum that person used when generating those lists. You must either convert the list to the same datum as your GPS (before the upload); or set your GPS (and/or mapping program) to that datum and upload the points. After you upload those points, you can reset the datum on the GPS to whatever you want -- they have been converted to the GPS's internally consistent format.*

To convert among datums, consult:

http://jeeep.com/details/coord/

or

http://www.kolumbus.fi/eino.uikkanen/geoconvgb/geobriefgb.htm

 

* I'd like to say that once the file is uploaded into your mapping program, you don't have to worry any more about the datum chosen for GPS or mapping program. Indeed, that is the way ExpertGPS works; once files are in ExpertGPS, the internal data storage is all WGS84, and all communications with the GPS are in WGS84. Unfortunately, some programs use a different internal format for data, and some communicate by NMEA. So to be safe and utterly paranoid, I always convert everything to WGS84.