Well, here's the OPUS results to add to the position on station POST. Two seperate sessions of 9 hours or so each were run on station POST and submitted to OPUS. Results of both sessions are based on "Rapid Orbit", not "Precise Orbit". Plans are to resubmit at a later date for "precise". I will just cut and paste the data as returned to me by the independent observer that put 2 days of L1/L2 observations on the POST. Thanks to Joey Stanger of Tyler Texas for participating in this project. I just wouldn't have felt right about doing the L2 obs. "in house". I hope that if there are any questions pertaining to the dual frequency side of this project that Joey will answer them as needed.
Data received from Joey Stanger:
TEST PROJECT ON J. D. BILLINGS POST MONUMENT - KILGORE, TEXAS
FT/MET 3.28083333
STATION NORTHING EASTING ORTHO FILE DATUM
POST 2083571.680 945090.913 122.775 61343610 OPUS
POST 2083571.681 945090.915 122.771 61670030 OPUS
AVG 2083571.681 945090.914 122.773
POST 6835851.422 3100685.774 402.798 AVG FEET
POST 6835851.505 3100685.732 402.979 HARN
DIFF OPUS-HARN -0.083 0.042 -0.181 FEET
POST 6835851.427 3100685.739 402.873 JD TEST 1
POST 6835851.422 3100685.757 402.956 JD TEST 2
POST 6835851.415 3100685.742 402.728 JD TEST 3
AVG 6835851.421 3100685.746 402.852 AVG JD
DIFF OPUS-JD 0.00044 0.028 -0.055 FEET
Interesting numbers to say the least, especially when you consider the end user actually had gps units up and running on the actual test site.
J.D. Billings
Stumpwater R&D
p.s. Joey also recommended purchase of dual frequency. Said something about how amazing the OPUS system was with regards to processing time, repeatability, simplicity...blah, blah
(sorry Joey, must be the nicotene withdrawals talking....)
J.D.:
Those are interesting results. Just out of curiosity, have you ever adjusted POST in ITRF97 to see what further improvements are possible?
Thanks again for the report,
Kent McMillan, RPLS Austin TX
Modified By Kent McMillan on 1/8/2002 at 10:23 PM
Gotta check my numbers against yours tonite when I come back home.
Is the Harn coordinate shown yours or from L1/L2 and a shot on those airport PACS monuments you were speaking of earlier ?
I think I've figured out a way to check the location of my aerial point....."offset plumb bob holder", plumb bob, post construction ground monument, long string.....not really hard but those particular brain cells haven't been used in ages.
Jimbo
Jimbo,
The coordinates noted as "HARN" are my original "in house" coordinates used for the POST. These are the values we determined from the various PACS in the area, strictly from L1 positioning. I have always tried to keep that data set in it's place by stating as being HARN "derived", so as to denote the datum basis.
Kent,
Does OPUS take ITRF into account when processing? I have both nad83 coord's and itrf coord's for station POST as returned by OPUS for each of two 9 hour dual frequency sessions. What else can I tell you?
J.D.
J.D.,
Processing is done in WGS84 (that's the system the orbits are referenced to). Any other system (including ITRF) is a post-computed transformation.
Regards,
Brian
It appears from the post that I was;
Opus
Northing 0.093
Easting 0.210
Elev 0.192
And from the HARN;
Northing 0.010
Easting 0.172
Elev 0.011
Still thinking about the ramifications of all this.....(now my head hurts even more and the voices are getting louder)....
TM
J.D.:
If you got ITRF97 coordinates back from the OPUS runs, you've got excellent control values for POST to compare to.
All you need to do is to substitute the ITRF97 coordinates of the CORS stations for their NAD83 coordinates and run the least squares adjustment of the vectors to POST. The adjustment will then give you the ORGI-derived ITRF97 position of POST to compare to control.
I'll bet you a Mexican lunch (in Austin, naturally) that the Up residuals will be significantly smaller in the ITRF97 adjustment.
By the way, pay no attention to Brian's insistence that the satellite ephemerides are referenced to WGS84. While WGS84 and ITRFxx are essentially identical, I believe that ITRFxx is actually the reference frame that the international fiducial stations are positioned in.
And if I'm wrong about that, Brian, I'll buy you a Mexican lunch, also. It will have to be in Kyle, Texas, however, since they are used to seeing foreigners at my regular cafe there and so the sight of a Californian will not scare them quite so much. :)
Best regards,
Kent McMillan, RPLS Austin TX
Modified By Kent McMillan on 1/9/2002 at 7:48 PM
I agree...The SV's have been refrenced to the ITFR for quite a while...It's not a major difference from WGS, but it's a difference none the less...
The debate I have ,is that in the LS adjustment (while the statistics may prove smaller), the position will not change in the least.. They should remain coorelated (about the same in any direction), which should produce the same station coordinate...
I welcome the data to prove this wrong Kent...Just don't see it in my software...
Kindly Asking,
TM
You owe me lunch. You are correct that the fiducials are published in ITRF, but I'm correct that the broadcast orbits (ephemerides) are referenced to WGS84. I'm particularly fond of good chile verde and rellenos (not the omelet sort, but big green chiles stuffed with cheese, breaded and fried).
Regards,
Brian
Modified By Brian D. Ewing, PLS on 1/9/2002 at 9:34 PM
Brian:
But what you wrote was:
Processing is done in WGS84 (that's the system the orbits are referenced to). Any other system (including ITRF) is a post-computed transformation.
I'm afraid that the folks who actually compute and distribute the best orbital data (the IGS Ultra-rapid, Rapid, and Final Products)aren't aware that they're computing them in WGS84.
IGS Orbit Information
You can still come to lunch on Friday, but the special is carne guisada. The waitresses are cute, but you'll need to habla some Espanol.
Best regards,
Kent McMillan, RPLS North Kyle TX
Deral:
You're right for short vectors, but as vector length increases, the weighting of the vectors gets thrown off by the non-identity of NAD83 and ITRF, in particular in the vertical components.
Would you like a numerical example to demonstrate the principle in two dimensions? You can construct it yourself this way:
(a) take a pair of "known" coordinates of control points on opposite sides of an unknown point, one at a distance of 50km, the other at a distance of 300km.
(b) choose coordinates to represent the unknown point.
(c) calculate the bearings and distances (by inverse from the exact coordinate values) from the unknown point to the fixed points.
(d) rotate the bearings of those inverses by some constant amount (and scale the distances also if you want) and treat those as the actual observations that were taken to the unknown point from the fixed points.
(e) find a least squares solution for the unknown point without modelling the azimuth and scale bias of the ties.
(f) find a solution that also models the azimuth and scale bias.
(g) compare the results and note that one will be inferior when the azimuth and scale bias passes a threshhold value.
Best regards,
Kent McMillan, RPLS Austin TX
Your starting the OPIE/ORGI activity has resulted in important understanding of what can be done using single frequency GPS in conjunction with CORS, both through your work and that of others who have joined in. It is always great to me to see people get answers to questions based on actual experience with real data --- something called the scientific method.