Note to self
Lesson learned, double or triple shoot all your boundary points if they are under canopy.
Tie/dump/tie has been our mantra for any boundary or control points for as long as I can remember, regardless of conditions.
Sounds like the mysterious 8-foot problem I encountered a few weeks back. In my case the data did not match my own previous data from a few years earlier. Without that doublecheck it is extremely easy to assume perfection.
With an EDM it was possible to have what appeared to be a 10 meter error in the distance reading. Hard to know at a half mile but very easy to tell at 50 feet.
If I am running full GNSS multi-constellation RTK/VRS...
Observe 1 minute, turn rod 180 degrees, observe another minute, then average.
Wait 15-20 minutes (30 is better), dump init, and repeat. If those observations agree, I'm likely in the clear.