Accuracy of triangulation stations

[Bill93]

You may want to read the related post about datum realizations to help this make more sense.

 

I have a project that I've worked with occasionally over several years to play with measurements, and get a better understanding of the C&GS triangulations. I observe angles to tall objects such as the red lights on radio towers, serving as intersection stations. I have a few points with known coordinates as a basis and do least squares fitting of the measurements to estimate coordinates for everything.

 

A church spire NJ0770 didn't seem to fit very well when I used the latest data sheet NAD83(96) coordinates and converting to NAD83(2011) with GEOCON didn't help much.

 

Then last fall I acquired an "antique" professional GPS, a Trimble 4000sst. Anybody who used one when they came out in the early 1990's would consider them junk today because they only process L1 and sort of process L2, but no P code. Newer receivers process all of those signals. I figured out how to manipulate the data files so the NGS OPUS service would process them and return NAD83(2011) coordinates.

 

I got a couple OPUS sessions on the ground near the church and did triangulation to tie in the spire, and find it to be about 1.6 foot east of where the data sheet plus GEOCON would put it, and much closer to my older measurements.

 

So I went to first-order triangulation station NJ0775 a few miles away, dug down to the disk, and got a 4-hour GPS session that came up 1.0 ft mostly east from the 96+GEOCON coordinates.

 

The nearest tri-station NJ0769 is deep in someone's yard and doesn't have enough sky anyway. Its RM3 is also somewhat obscured and gave me a file that was so weak in L2 OPUS wouldn't give useful results. I got someone to run the L1-only portion of the data and it came back 1.0 ft mostly east of prediction.

 

My conclusion is that the old triangulation network in this area was very good in relation between nearby tri-stations, and good to about a foot with respect to stations that are tens of miles away, but the intersection stations (church spire) can be off more than that even with respect to nearer triangulation stations. The triangulation network had some warpage in this area for stations without GPS data, relative to the stations used in the '96 adjustmen,t because the nearest stations with both triangulation and GPS data are tens of miles away. That is well within the goals of the old triangulations.

 

This has been an interesting and educational exercise. I'd like to eventually check more tri-stations around here to better characterize the shift.

 

(Corrections welcome)

Edited May 14, 2016 by Bill93