I agree with everything Amir said, make sure the data is clean before the adjustment. Topology issues can be exacerbated by an adjustment. Definitely run the consistency check, it will find data entry blunders. Here is the best advice I can give when adjusting the PLSS:
- A priori error estimates on the lines are pretty key. Direction and distance accuracies have to be realistic for the surveys. If you are adjusting using survey measurements from the 1860’s, chances are good the 30” and 15 cm defaults will not be loose enough to adjust the measurements between GPS points. If you have new survey measurements, make sure they are held tighter than the stuff from the 19th century - you want to avoid having older and likely poorer quality measurements degrade the accuracy of the new measurements. Don’t let the tail wag the dog.
- Azimuth Types – If you are using GLO/BLM measurements, remember they report bearings at the midpoint of the line (mean geodetic). You can use the Azimuth types in the adjustments if you are using 3.2. If you aren’t in 3.2 yet, you may need to loosen the direction accuracies to accommodate the rotation between ground and grid.
- Sometimes it may be a good idea to constrain the adjustment area by setting points on the edges as fixed and constrained temporarily, especially if the movement would be insignificant on those margins and there are other datasets lined up to them already. It's not fun having to re-align vertices in associated feature classes by a few centimeters because of an adjustment. If the movement is that small, how much improvement have you really gained on a line a mile long? You can also exclude movement less than a certain tolerance when you apply the LSA.
- Control – sounds like you have good GPS measurements to use, but you should still assign realistic XYAccuracies to them. For example, you wouldn’t want to hold a resource grade GPS point like it was survey grade. If you have all 8 monumented points on a section boundary, you may be able to just conflate and re-subdivide. Remember, unless the plats say otherwise, 16th corners are usually at midpoint between section and quarters. If you do not include them in the adjustment, you should re-midpoint them afterwards.
- Re-subdivide after the adjustment – the positions of the aliquots and lots in the interior of a section are determined by the positions of the points on the section boundaries. You can’t just divide a section up into 16 equal area parts and call it good. The record may say 40 acres, but the survey measurements can stretch and shrink those subdivisions. Refer to chapter 3.99 of Manual of Surveying Instructions 2009 (blm.gov) to see how these should get drafted.
- LSA's are an iterative process, you will end up tweaking things as you go. I find selecting the lines works best for me, but as I am working through adjustments, I end up unselecting and reselecting lines frequently. Rather than manually reselecting, create a layer from your selected features to make it less tedious to select the lines you are adjusting.
Good luck and I hope your adjustments go well.
