Answers Finally in 2000!
April
1, 2000 - After Perusing Relevant Internet Links For Opinions Of Learned
Scholars,
we have gathered this set of answers to frequently asked GPS-related questions.
Where
Can I Get Maps With GPS Coordinates?
In response to a large number of requests from recreational GPS users, the U.S. Government has issued an entire set of maps that display latitude and longitude tics marks as well as State Plane Coordinates and Universal Transverse Mercator grid lines or tic marks. The maps, called USGS 7.5 Minute Quadrangles, are at a scale of 1:24,000 and the entire set of some 50,000 map sheets is now available from the U.S. Geological Survey, some in both paper and digital form. Other countries and agencies have followed this trend, producing maps with latitude and longitude tic marks as well as a variety of national grid systems on the map "neat lines." Some national mapping agencies have begun to produce maps with one- or ten-kilometer grid squares that appear to be related to GPS coordinates through some form of map projection mathematics. On all sorts of maps from atlases to nautical charts, one can now see latitude and longitude marks on the surface of the paper. Users no longer need to rely solely on an abstract set of coordinates from GPS receivers to sense where they are; they can now use paper or electronic maps to find their position relative to mapped points, lines, and areas in order to check the validity of the numbers displayed on their GPS receivers. As an additional benefit, hand-held GPS users can use the 1:24,000 scale 7.5 minute quadrangles to derive their latitude and longitude to an accuracy of 20 meters or so, reducing their dependence on the 100 meter GPS uncertainty.
When
Will Selective Availability Go Away?
Selective Availability (SA) is the intentional degradation of the civil Standard Position Service (SPS) GPS service that degrades it from a 30-meter accuracy level to a 100-meter accuracy level. When and if SA is removed, civil GPS will revert to the 30 meter system it once was. Differential GPS (DGPS) provides corrections that turn the 100 meter SPS service into a 2-meter service everywhere that DGPS is available (all of North America and much of the rest of the world). When SA is turned off, DGPS will continue to provide 2 meter accuracies. Those particular users who need 30-meter accuracy and choose not to use DGPS want to know when SA will go away. Most researchers do not know - they just use DGPS now and will continue to use DGPS in the future irrespective of the irrelevant issue of SA to precise positioning projects. The “30-meter only” users can now use DGPS and degrade the accuracy of the 2-meter DGPS positions to 100 meters through a variety of numerical techniques.
Can
I Use Precise Military GPS?
The U.S. Department of Transportation and the Department of Defense (DoD) have issued a document that publishes the accuracy of the military's Precise Positioning Service (PPS). The 1999 Federal Radio Navigation Plan (December 1999) reveals that by using encrypted signals from the GPS satellites, users authorized by DoD, with access to secure decryption devices and appropriate cryptographic keys, can achieve 22-meter accuracies, free from the Selective Availability degradation imposed upon the civil SPS service which can only provide 100 meter accuracy. Currently, civil SPS Differential GPS users can achieve 2 meter accuracies using a number of public and private correction services. Specially equipped DoD-authorized PPS users with these same DGPS corrections can also achieve 2 meter accuracies. Users with a need for positioning accuracies worse than 22 meters and better than 100 meters should make every effort to obtain GPS PPS equipment and necessary security clearances and equipment in order to make use of these "precise military GPS" measurements.
Which
Map Datum Should I Use?
The question of which geodetic datum the user should use has finally been settled. The National Imagery and Mapping Agency (NIMA) has published a document (NIMA TR8350.2) that lists the parameters for many geodetic datums. Some have exotic sounding names such as "Astro Tern Island (FRIG) 1961" and "Pico de las Nieves" while others carry more mundane names like "Geodetic Datum 1949" or "Arc 1950." Setting the GPS receiver to one of these datums has the result of shifting the measured coordinates from between a few meters to as much as 1,000 meters from the nominal WGS-84 datum of the GPS system. Recent research has found that the best results are obtained by setting the receiver to the geodetic datum used to produce the paper or electronic map on which GPS positions are being charted. When a map legend states that the map datum is the North American Datum 1927, the user can use "NAD27" or a locally more appropriate datum such as "NAD27 Alaska" when in Alaska. In foreign countries such as Ecuador, one might find the phrase "Dato Horizontal: El Provisional de 1.956 Para America del Sur" on the legend of the map. Experimentation and careful analysis will show that for such a map the "Provisional South American Datum of 1956" seems to work the best. Similar results can be obtained by reading the information on other maps and setting the GPS to that datum.
How
Much Does Averaging Improve GPS Accuracy?
Averaging over a few hundred seconds is an effective way to estimate the average GPS error over that period by comparison of the averaged GPS readings and the known geodetic position of the antenna. Averaging of SPS GPS measurements is a good way to convert what might be a set of independent measurements with 100 meter accuracy to an ensemble measurement with a 100 meter accuracy. At unknown locations it is possible that the ensemble average might be better than 100 meters but that is impossible to determine. Users who do not need to actually know whether their position estimates are better than 100 meters can make good use of ensemble techniques by averaging for a few hundred seconds and hoping it helps. With averaged readings taken over many thousands of seconds (3600-7200) the position error of the ensemble measurement will be reduced somewhat as the biases from SA and the atmosphere begin to average to 30 meters or so around a zero mean. When only a few hundred seconds are available for averaging, users should check the averaged value against a map or other accurate reference to pick the averaged or ensemble set of measurements that best represent the actual position.
What
Is The Algorithm For Distance Between Two Points?
It seems that distance between two points depends on what is meant by "points" and "distance." Points can be described on the surface of a sphere, an ellipsoid, a projection surface, the topographic surface of the earth, or within in a three dimensional coordinate system. Points can be dimensionless intersections of defined axes, the approximate position of cultural centers, or one of various geometric centroids of regions such as cities, counties, states, or nations. Distance too can be defined in many ways. There is distance over the surface of a map projection converted to distance on the earth, distance over the surface of a sphere on a great circle course, distance over the surface of a sphere over a constant compass course, distance over the surface of an ellipsoid along an ellipsoidal geodesic, distance along a series of local level traverses at a constant heading with respect to true north, distance over the topographic surface of the earth, distance along a three-dimensional vector from one point to another, distance along an ellipsoidal path with varying heights above the ellipsoid from one point to another, distance along a series of line segments called a route and many others. Current wisdom suggests then that there is no single "algorithm" to compute distance from one point to another. It has been found far better to figure out what one means, how one intends to use the results, and what the required accuracy is and then find an appropriate method from the extensive literature on projections, navigation, and geodesy.
Why
Does "Z" Not Vary Correctly With Height?
Studies indicate that height (or elevation or altitude) may not always be the same as "Z." Recent research indicates that while Earth Centered, Earth Fixed (ECEF) X, Y, and Z coordinates vary linearly in "Z" with changes in height above the reference ellipsoid (HAE) at the north and south poles, Z may not represent height at other locations around the globe. African and South American research indicates that along the Equator changes in latitude are more apt to result in changes in ECEF Z, while HAE seems to cause changes in X or Y or both X and Y. The mechanism for this may lie in the mathematical relationships between spherical and rectangular coordinates. In most cases it is best to find out what is meant by “Z” and expect to see only effects that are likely within that definition.
What
Is The Best Place To Find GPS Information?
Books and published papers (particularly papers published in refereed journals) may be better places to obtain reliable GPS information than Internet News-Group discussions and personal World-Wide-Web pages. Research indicates a higher correlation between presentations of GPS-related details and the actual, often complex, real-world situations in responsible books, articles, and papers written by authoritative sources than in the easier to read and more convenient materials provided by folks who willingly express their personal (and sometimes un-informed) opinions on the Internet.
AF2000 – Peter H. Dana - April 1, 2000