jscheps explained it very well. The critical thing is to properly set up your base station with the geodetic coords. of the occupied station correctly keyed in. Be careful not to key in the Orthometric height after the Lat. Long. It needs to be the ellipsoidal height.
Not all NGS points have the same level of accuracy either. Some are tight horizontally and not so much vertically and others are leveled to vertically, but the horizontal may just be scaled. So be careful selecting a starting point. The NGS website lets you filter your search by a certain order of accuracy and not just what happens to be the closest to your project. Once you're tied in to a good point at your base, everything you shoot will be in NAD83.
If you choose to use a geoid model for orthometric heights, be aware that it is more accurate in some areas than others. It is a model based on observed gravitational data in the area. In between points it is just a straight line interpolation. The more data points used in your area the more accurate it is. But some areas are few and far between. If you find good vertical control encompassing your project, you might be better off tying in to it and using an inclined plane projection.
Too many surveyors get into GPS without bothering to learn some basics of geodesy which are important to know. GPS is a dangerous tool when someone is worried only about the coordinate geometry and ignores the rest. It can be wrong if not set up properly, expecially with horizontal distances. It doesn't measure distances, it calculates them based on your geodetic position. If your grid is on the ellipsoid and your at 6000 feet, it will tell you two points are 1000 feet apart when they are really more like 1000.3 feet apart. In a mile you'll be over a foot and a half longer than it tells you. Make sure your grid is at the average ellipsoidal height of your project.
Edited by: firstname.lastname@example.org on Aug 10, 2009 1:42 PM