How Accurate are Adjusted Benchmark Coordinates?

[+Thot]

A graduate student contacted me to help him locate a couple of benchmarks he can use for a study he is doing. He needs locations that are accurate to less than a centimeter and more accurate if possible. I've read that adjusted marks are accurate to less then one centimeter. Are all adjusted marks this accurate? Many in our area were done sixty to 75 years ago. I don't understand how they could have gotten this accuracy back then and surely some errors were made.

[kayakbird]

'--- couple of benchmarks ----'.

 

Do you mean 'Triangulation Station'? Very unlikely that a true 'Bench Mark' will have a Lat/Long that accurate. kayakbird

[+Thot]

Do you mean 'Triangulation Station'? Very unlikely that a true 'Bench Mark' will have a Lat/Long that accurate.

Are you saying Triangulation Stations have this kind of accuracy, but ordinary adjusted benchmarks don't?

[+Ernmark]

Do you mean 'Triangulation Station'? Very unlikely that a true 'Bench Mark' will have a Lat/Long that accurate.

Are you saying Triangulation Stations have this kind of accuracy, but ordinary adjusted benchmarks don't?

 

Any horizontally adjusted mark will be very accurate. True 'benchmarks' usually have very accurate vertical information, but were 'scaled' horizontally (i.e., with a ruler on a USGS 7.5' topo map)..

 

It is amazing how accurate most of these horizontal control marks are considering many go back to the 30s or earlier!

[kayakbird]

'------True 'benchmarks' usually have very accurate vertical information,------'

 

It has been quite awhile since I did any 'three-wire' differential level runs (reading to 0.001 ft); but I believe that this over 100 year old method is still more accurate that the best Survey Grade GPS on the market. MEL

[Bill93]

The key is the Order of the stations, as shown on the NGS data sheets. For older measurements, this will be the relative accuracy compared to other stations in the area. For newer, GPS derived, data they may give both the local accuracy and the accuracy relative to the world as measured in the NAD83 latest refinement. Try to use First Order stations.

 

Is the student just needing a distance, or is lat-lon also critical? How far apart should they be? The most accurate distances, maybe down to a millimeter, are for calibration baselines, of which most states have at least one, although some of them (like the one nearest me) have been allowed to deteriorate. Baselines often go out to near a mile.

 

On the sidetrack of vertical accuracy, there are two problems with GPS elevations. The first is that since you don't have satellites both above and below the accuracy is poorer than for horizontal where you can have satellites in any direction. The second and more fundamental reason is that GPS measures a different kind of elevation than does optical leveling. Optical leveling follows the surfaces of constant gravity and measures with respect to the theoretical "sea level" geoid underground. GPS has no knowledge of gravity and computes the elevation above the mathematically perfect ellipsoid model. NGS is working on the conversion between systems by getting more gravity data in the 10-year GRAV-D program.

Edited September 27, 2014 by Bill93

[+Thot]

Is the student just needing a distance, or is lat-lon also critical?

I believe he's only interested in lat-lon

 

The most accurate distances, maybe down to a millimeter, are for calibration baselines,

That's the kind of lat-lon accuracy he's looking for. How would I find this mark in Texas?

 

<musing>I can't see how the scribe mark on the disk can be located with millimeter accuracy.</musing>

[Bill93]

Cal baselines are ONLY for distance, and have no better lat-lon than is needed to find them.

http://geodesy.noaa.gov/CBLINES/calibration.html

 

The best available for absolute lat-lon are HARN or CBN stations with GPS coordinates. First order stations may be as good relative to other nearby ones, but not relative to the world. If he's interested in that level of accuracy, he must develop a very good understanding of datums, the flavors thereof, and what those imply.

 

If this is somehow astronomically related, he needs to know that astronomical latitude and longitude will probably differ by up to hundreds of feet from GPS lat-lon due to variations in gravity, which pull the local vertical away from the normal to the ellipsoid, so it will in general not point to the axis of the earth.

[DaveD]

The control stations with the most accurate coordinates are the network of Continuously Operating Reference Stations (CORS) managed by the National Geodetic Survey (NGS) - http://www.ngs.noaa.gov/CORS/. In general these stations have horizontal accuracies of right around 1 cm and ellipsoid heights at around 2 cm. As for 1st-, 2nd- or 3rd-Order horizontal stations these are now at the lower end of the accuracy chain in the National Spatial Reference System and should seldom be used for anything that requires high accuracy. With the completion of the national adjustment of all the GPS stations in the NSRS by NGS in 2011 (NAD 83 (2011)) those stations are no longer classified with the old proportional accuracy statements. Rather they are given their positional accuracy in horizontal and ellipsoid relative to the entire network. For examples station CS2783 (NEWTON) has a network accuracy of 0.9 cm in the horizontal and 1.4 cm in the ellipsoid height. BILL93 is correct that this student needs to determine if the accuracy is required to be local (all of the U.S.) in which case he should use the NAD 83 (2011) coordinates, or global in which case he should use ITRF coordinates. He can access these data by going to the NGS home page www.geodesy.noaa.gov and then selecting "Survey Mark Datasheets" on the left side. There are several ways to access control. Once he's got an idea of the area he wants to search he should limit the search under "Data Type Desired" to "GPS Sites Only." If he has any problem accessing these data and/or understanding the geodetic elements on the datasheet I would be happy to discuss these issues with him -- base9geodesy@gmail.com

 

CS2783 PACS - This is a Primary Airport Control Station.

CS2783 DESIGNATION - NEWTON

CS2783 PID - CS2783

CS2783 STATE/COUNTY- TX/DALLAS

CS2783 COUNTRY - US

CS2783 USGS QUAD - FERRIS (1968)

CS2783

CS2783 *CURRENT SURVEY CONTROL

CS2783 ______________________________________________________________________

CS2783* NAD 83(2011) POSITION- 32 37 06.48374(N) 096 44 09.28934(W) ADJUSTED

CS2783* NAD 83(2011) ELLIP HT- 148.888 (meters) (06/27/12) ADJUSTED

CS2783* NAD 83(2011) EPOCH - 2010.00

CS2783* NAVD 88 ORTHO HEIGHT - 175.23 (meters) 574.9 (feet) GPS OBS

CS2783 ______________________________________________________________________

CS2783 NAVD 88 orthometric height was determined with an earlier geoid model

CS2783 NAD 83(2011) X - -630,747.576 (meters) COMP

CS2783 NAD 83(2011) Y - -5,340,407.498 (meters) COMP

CS2783 NAD 83(2011) Z - 3,418,476.319 (meters) COMP

CS2783 LAPLACE CORR - 4.04 (seconds) DEFLEC12A

CS2783 GEOID HEIGHT - -26.33 (meters) GEOID12A

CS2783

CS2783 FGDC Geospatial Positioning Accuracy Standards (95% confidence, cm)

CS2783 Type Horiz Ellip Dist(km)

CS2783 -------------------------------------------------------------------

CS2783 NETWORK 0.90 1.41

CS2783 -------------------------------------------------------------------

CS2783 MEDIAN LOCAL ACCURACY AND DIST (008 points) 0.97 1.39 30.51

CS2783 -------------------------------------------------------------------

[Bill93]

I'm curious how this project went, as we never heard any more back about what the data was to be used for.

[+Thot]

I gave him the information I had and the what I got here (except the post about the CORS stations because I didn't get an email notice of that post). I got him a list of all adjusted benchmarks in this(Harris) county. I haven't heard from him since. So, I don't know what happened with his project.