I reprocessed a file through OPUS. First 3/23/13 then today.
OPUS used different CORS and gave a slightly different answer. Today's solution looks better.
Anybody know why. I don't think I've seen this before, always got the same answer from the same file before (I think).
Here are the two solutions:
FILE: 54411251.DAT OP1364007400419
NGS OPUS SOLUTION REPORT
========================
All computed coordinate accuracies are listed as peak-to-peak values.
For additional information: http://www.ngs.noaa.gov/OPUS/about.jsp#accuracy
USER: DATE: March 23, 2013
RINEX FILE: 5441125a.12o TIME: 02:59:30 UTC
SOFTWARE: page5 1209.04 master50.pl 082112 START: 2012/05/04 00:44:00
EPHEMERIS: igs16865.eph [precise] STOP: 2012/05/04 13:57:00
NAV FILE: brdc1250.12n OBS USED: 33381 / 33476 : 100%
ANT NAME: TRM33429.00+GP NONE # FIXED AMB: 96 / 101 : 95%
ARP HEIGHT: 1.700 OVERALL RMS: 0.010(m)
REF FRAME: NAD_83(2011)(EPOCH:2010.0000) IGS08 (EPOCH:2012.3396)
X: -1799127.113(m) 0.013(m) -1799127.917(m) 0.013(m)
Y: -4578500.021(m) 0.021(m) -4578498.727(m) 0.021(m)
Z: 4048923.190(m) 0.015(m) 4048923.121(m) 0.015(m)
LAT: 39 38 44.27482 0.007(m) 39 38 44.29193 0.007(m)
E LON: 248 32 51.34160 0.011(m) 248 32 51.29038 0.011(m)
W LON: 111 27 8.65840 0.011(m) 111 27 8.70962 0.011(m)
EL HGT: 1812.377(m) 0.028(m) 1811.632(m) 0.028(m)
ORTHO HGT: 1829.200(m) 0.050(m) [NAVD88 (Computed using GEOID12A)]
UTM COORDINATES STATE PLANE COORDINATES
UTM (Zone 12) SPC (4302 UT C)
Northing (Y) [meters] 4388524.808 2145690.421
Easting (X) [meters] 461183.879 504084.851
Convergence [degrees] -0.28865500 0.03048827
Point Scale 0.99961855 0.99990425
Combined Factor 0.99933441 0.99962002
US NATIONAL GRID DESIGNATOR: 12SVJ6118388524(NAD 83)
BASE STATIONS USED
PID DESIGNATION LATITUDE LONGITUDE DISTANCE(m)
AJ2124 PUC1 CARBON COUNTY COU CORS ARP N393557.066 W1104831.329 55524.9
AJ8478 MYT2 MYTON 2 CORS ARP N400609.568 W1100254.602 130415.4
DH3864 P105 DELTAMUNI_UT2004 CORS ARP N392315.140 W1123014.669 94890.6
NEAREST NGS PUBLISHED CONTROL POINT
KN0180 SS 42 N393757. W1112634. 1678.7
This position and the above vector components were computed without any
knowledge by the National Geodetic Survey regarding the equipment or
field operating procedures used.
FILE: 54411251.DAT OP1380168239167
NGS OPUS SOLUTION REPORT
========================
All computed coordinate accuracies are listed as peak-to-peak values.
For additional information: http://www.ngs.noaa.gov/OPUS/about.jsp#accuracy
USER: DATE: September 26, 2013
RINEX FILE: 5441125a.12o TIME: 04:13:29 UTC
SOFTWARE: page5 1209.04 master63.pl 072313 START: 2012/05/04 00:44:00
EPHEMERIS: igs16865.eph [precise] STOP: 2012/05/04 13:57:00
NAV FILE: brdc1250.12n OBS USED: 31194 / 31478 : 99%
ANT NAME: TRM33429.00+GP NONE # FIXED AMB: 100 / 105 : 95%
ARP HEIGHT: 1.700 OVERALL RMS: 0.013(m)
REF FRAME: NAD_83(2011)(EPOCH:2010.0000) IGS08 (EPOCH:2012.3396)
X: -1799127.128(m) 0.003(m) -1799127.932(m) 0.003(m)
Y: -4578500.019(m) 0.003(m) -4578498.725(m) 0.003(m)
Z: 4048923.195(m) 0.010(m) 4048923.126(m) 0.010(m)
LAT: 39 38 44.27487 0.006(m) 39 38 44.29198 0.006(m)
E LON: 248 32 51.34098 0.003(m) 248 32 51.28976 0.003(m)
W LON: 111 27 8.65902 0.003(m) 111 27 8.71024 0.003(m)
EL HGT: 1812.383(m) 0.009(m) 1811.638(m) 0.009(m)
ORTHO HGT: 1829.206(m) 0.022(m) [NAVD88 (Computed using GEOID12A)]
UTM COORDINATES STATE PLANE COORDINATES
UTM (Zone 12) SPC (4302 UT C)
Northing (Y) [meters] 4388524.809 2145690.422
Easting (X) [meters] 461183.865 504084.836
Convergence [degrees] -0.28865511 0.03048816
Point Scale 0.99961855 0.99990425
Combined Factor 0.99933440 0.99962002
US NATIONAL GRID DESIGNATOR: 12SVJ6118388524(NAD 83)
BASE STATIONS USED
PID DESIGNATION LATITUDE LONGITUDE DISTANCE(m)
AF9633 RBUT RED BUTTE CORS ARP N404651.807 W1114831.490 129707.2
AJ2124 PUC1 CARBON COUNTY COU CORS ARP N393557.066 W1104831.329 55524.9
CQ6018 MIDV MIDVALE CORS ARP N403716.045 W1115426.030 115070.5
NEAREST NGS PUBLISHED CONTROL POINT
KN0180 SS 42 N393757. W1112634. 1678.7
This position and the above vector components were computed without any
knowledge by the National Geodetic Survey regarding the equipment or
field operating procedures used.
Different base stations used.
OPUS - Resubmit after 6 months - Case closed
> Different base stations used.
Case closed. (Unless the fence corner post actually moved, that is.)
Not knowing much about OPUS, am I right in reading that there are only three base stations used in the processing?
I took a screenshot from the last Auspos (our version of OPUS) that I did.
My point is the yellow dot but the processing used fifteen different base stations which are the red triangles.
Resubmit To Same CORS
Then compare.
What you have is an E-W difference, check the layout of the CORS and see which one is pulling harder.
Paul in PA
Resubmit To Same CORS
> Then compare.
He may still get slightly different coordinates because the a-priori assumptions could be different. Although that difference should be very slight.
> Not knowing much about OPUS, am I right in reading that there are only three base stations used in the processing?
Yes. Three bases used with OPUS-Static, which is for sessions over 2 hrs. OPUS-RS, which is for sessions under 2hrs, will use several bases.
Paradoxically, and for that reason, I think, the positions OPUS-RS produces with shorter data sets sometimes seem to be superior to those from OPUS-S. So there is that opportunity for OPUS to improve.
OPUS - Resubmit after 6 months - Case closed
It's not a different observation, same file different submission dates. I let OPUS choose the base stations, so why did it select different CORS?
There must have been some change to the processing program.
It's not a fence post and I'm not going back to punch the marker again!
That's what I think.
OPUS - Resubmit after 6 months - Case closed
> .. so why did it select different CORS?
I can't really answer that. But the selection of stations is based on a statistical analysis, and the outcome of a statistical analysis is dependent upon, in part, the a-priori assumptions that are loaded in the front end. Some of these assumptions are chosen at random. So there need not have been a change in the program for there to have been a different outcome.
When using OPUS-S I like to choose the stations I use. I also like to get redundant sessions, and I get the extended solution. Then I use StarNet to LS the results. I'd rather do that with a pair of 2 hour sessions than to rely on a single 13 hour session.
OPUS - Resubmit after 6 months - Case closed
Utah doesn't have enough CORS to make OPUS-RS useful (at least where I'm at).
I'm not upset with the results just curious as to why. Both answers are fine for me and would land on the 3-1/4" section corner marker. For the work I'm doing they will accept an OPUS solution for documentation (single) but it must be at least 2 hours.
I ran the second OPUS to get the xml extended output which makes it easy to insert the point into TBC.
It takes two trips to set up and retrieve a GPS receiver. So the only difference is how long you let it cook. So lately I set them out in the afternoon and pick them up in the morning. Long session overnight. Seems to work good for me. Actually I don't really care about the last 5-10 millimeters. Hey, if you get to the cap with my coordinates job accomplished in my book.
Now if what yer doing is going to some over evaluated, least squared and super massaged coordinate instead of looking for the in the ground position of the marker, well then by all means, punch the cap and prove it wrong. You might impress someone else but not me.
OPUS - Resubmit after 6 months - Case closed
> ... I set them out in the afternoon and pick them up in the morning. Long session overnight....
Use your TBC to export an interval of about 4 hours worth of data during darkness. You will probably get better results than with the whole 13 hrs.
And now you know why OPUS isn't a good tool for lot surveys.
I NEVER let OPUS decide which CORS to use, GRANTED...it's does a better job that it used to...BUT all CORS are NOT created equal (and even those that are [PBO sites] are subject to Murphy's Law.
I know (pretty much) which CORS in my area are winners, and (more importantly) which are losers. BUT I ALWAY look at the Short Term (90 Day) Plots, to be sure that "MY" CORS are behaving themselves.
PLUS...I ALWAYS get the Extended Output, and extract the G-FILE vectors (and data) so that I cam evaluate EACH Baseline.
Loyal
OPUS - Resubmit after 6 months - Case closed
If you really want a solution it's two separate setups with different HIs and different tribracks or the same tribrack rotated. Then you know.
Nothing is to stop you from taking a 15 minute session and tie it to one of your own control points or to a near cors station. This will give you more info and check your HI measurements and your tribrack.
Setting up a receiver only locates the receiver somewhere above the point. Doing it twice and having it check locates the point on the ground.
OPUS - Resubmit after 6 months - Case closed
:good:
OPUS - Resubmit after 6 months - Case closed
My solutions are just fine. Using them a monument can be found. I could set up without a tribrach and get within a foot. So in the future if a surveyor gets within a foot and can't see the marker maybe he should take up a new occupation.
I have rotating optical plumets and tribrachs. With them you can rotate over the point and see if they stay on the point or trace a circle around them. So centering error is reduced. I'm careful about heights but a mistake is always possible.
So I'm documenting PLSS corners that currently have error in location of up to 300 feet in the GCDB. Some I've found nobody knew where they were. Reducing that error in geodetic location down to under a foot is a big improvement. Under a tenth even better. The extra effort to get it under 5mm (in the math exercise with multiple observations and time in the field) at a doubling of cost I'm not sure is cost effective. Might be better to spend the few dollars to locate more corners. Let the surveyors of the future wiggle out the last few centimeters or millimeters when the high rise buildings go up.
OPUS - Resubmit after 6 months - Case closed
I'm not beating you up at all. When I first was trained with GPS and control work it was beat into my head that each point gets at least two setups.
If you can use two different tribracks and two different receivers so much the better. Often I will use a zephyr and a 5800 on the same point.
If I'm locating a brass cap I do two sessions, gives you proof that you have the correct number. You don't need 4 hours on each point often a short session as you set up the tripod then break the setup once you get everything in the notes and pictures taken, remeasure and do the long one.
I never worried about the perfect number either. Mainly what I'm looking for is a reduntant measurement within a tolerance set for what I'm doing. Out there in the big country 0.04' is nothing.
Either way I always locate a section corner at least twice. Even if it's RTK.;-)
OPUS - Resubmit after 6 months - Case closed
Yes redundancy is the best way to sure up the measurements.
I been using CORS and OPUS for about 15 years now. I'm not an expert like say, Loyal Olsen, but I've done a bunch of them. I've control points with up to 20 or more long static sessions. I've certainly seen some ratty sessions but on the other hand ratty is something like 10 centimeters (horizontal - vertical is subject to HI errors), that's a lot for an OPUS session to be off the mark. To be more than that off you'd probably need to be set up over the wrong mark. Maybe a real bad environment under trees or something but I personally don't recall ever having multiple OPUS sessions on the same mark come in several feet different. I've seen it with bad fix's with RTK. That doesn't mean it can't happen but I've never seen it.
So I suppose it depends on what you're doing and why. If it's a control point that is going to be used to locate other points I'd say that multiple long sessions and careful reducing the data to get it right on should be done. For other points like PLSS corners in the hills with solid markers not likely to move isn't an improvement from 300 feet to under a foot in geodetic coords at minimal cost an acceptable thing. The goal is to preserve as many as we can, not locate a few to under 5mm. The dollars are limited or non existent for this work.
OPUS - Resubmit after 6 months - Case closed
:good:
