Hi Spencer, It all does move, but I wouldn't call it "degrading", but "Changing over time".
I remember asking a geodesy professor "if everything's moving, including the definition of the satellite systems datums, how can one tell which set of coordinates is correct and which is not". And the answer I got was basically: none is correct, they're all changing! You'll have to ask the geodesists to get a better and more manageable answer than that :-) But it doesn't make a lot of difference 'cause almost all measurments are local to a single techtonic plate, and so are indifferent to large scale movements and rotations. So the claim to high accuracy is true, although implicitly only in a relative sense. There are few noteable exceptions I can think of : Iceland, California and the Himalayas I guess, they will have to deal with time dependent observations to be precise. Maybe someone on this list from these regions can shed some light on the practical issues of this issue ? As for pointing towards North or not, only at a single Easting value does the Northing axis in a projected system point true North anyway. So having the plate rotate with a local coordinate system is not a real serious issue :-) The engineers are now, as I understand from the Discovery Channel, contemplating cross-plates constructions like a Gibraltar bridge and a cross Atlantic "tunnel". They'll certainly have to take inter-plate movements into serious consideration. I'm curious as to exactly how they'll cope with such a seemingly impossible task. The only issue I can see wrt. day-to-day GIS is contemplating that all GPS measurments are drifting over time. It doesn't matter much if you're always doing differential or relative observations, and most surveyors are, which is why e.g. Europe is beginning to use the "frozen" GPS datum of 1989 (named ETRS89) for all GPS measurments. I've been told that ETRS89 and GPS differ by an amount of 2 cm/yr, which today amounts to about a quarter meter, too large not to be taken into account. As for height variations, we're just beginning to use a new vertical reference system here in Denmark, replacing an older system that's more than 50 years old. The reason is that Denmark "tilts", i.e. the north part is lifting whereas the southern part is sinking, almost precisely across the middle. In some local places the old system were beginning to cause problems when measuring drains etc., so it was definitely necessary to update the system. This kind of problem will necessitate implementing new height datums if the land is indeed in a zone of up or downheavel (is there such a word?). As I see it, large scale movements are important to be aware of, but not a problem in normal day-to-day work. I think you're safe to leave it with the geodesists :-) Best regards/Med venlig hilsen Lars V. Nielsen GisPro, Denmark http://www.gispro.dk/ ----- Original Message ----- From: "Spencer Simpson" <[EMAIL PROTECTED]> To: "'MapInfo-L'" <[EMAIL PROTECTED]> Sent: Friday, May 21, 2004 6:06 PM Subject: MI-L Geological processes affect GIS accuracy? The mention of high-accuracy datums reminded me to ask about how large-scale geological processes affect the accuracy of GIS data through time. For example, after a bit of research, I found that the North American Plate (including myself) is moving away from London (and thus the Greenwich Meridian) at about 4.4 cm/yr. The plate is also rotating, so that the Y axis of a projected coordinate system (say, a State Plane Coordinate System) no longer points due north. Local governments in my area have specified that their data have an accuracy of 0.1 ft (about 3 cm). But they're using high-accuracy GPS to capture other data years afterward. If we're going to use space-based reference systems such as GPS and ITRS, this will definitely degrade the accuracy of GIS data over time. I suppose that if you converted data back and forth between different realizations of datums, you could see the movement. Vertical accuracy would also be degraded by things such as isostatic rebound. My knowledge of geodesy is cursory at best. Does anyone know of any research that has been done as how such processes affect GIS data accuracy? Are movement rates really enough to matter? Spencer --------------------------------------------------------------------- List hosting provided by Directions Magazine | www.directionsmag.com | To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] Message number: 11876 --------------------------------------------------------------------- List hosting provided by Directions Magazine | www.directionsmag.com | To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] Message number: 11879
