I could sure use some help with my gps processing. I'm working on an elevation certificate in the middle of nowhere. I only have L1 gps and no benchmarks nearby, so I set my 3 units in an open field and let them cook for about 5 hours. A friend helped me process the data in GNSS Solutions using the nearest CORS stations, and the points checked very well vertically between each other according to what I shot with the total station.
My problem is, the numbers say the existing building needs to come up about 1.7 feet. I was expecting 6 feet or more, according to the depth of the previous flood. I asked another friend, who told me that I can't use the CORS sites for good elevations, only for horizontal work. I trust both of the guys I asked, so I don't know who to believe.
What's the story, can I use the CORS data for this, or do I need to find a benchmark?:-S
Thanks in advance for any input!
I haven't done any work with elevation certificates, but the OPUS solutions I've done don't agree with local benchmarks. Sometimes the difference is as much as 6 or 7 feet. So I'd try to find a local benchmark to calibrate to.
Dave
Bring in more CORS, reprocess and check the derived elevations vs the published. Then check the CORS result against against OPUS for each individual station. This should give you enough redundancy or prove that you are correct or that something is wrong. If the results are good but dont agree with the local datum to which the flood level is related then you will need to tie into a something locally to determine the shift.
OPUS and a level
> OPUS and a level
It's been a while since I checked, but does OPUS process single frequency data?
Surely there is a benchmark somewhere within five miles of your site. Forget that magic answer from the sky foolishness and get to surveying.
forget the opus because of the L1.
you should have placed one unit on a known BM (or an eccentric mark if sky visibility was poor) and constrained to the published height.
other factors to consider are BL distances for the length of occupation.
toss in the CORS data for QC check.
L1 is going to be limited in baseline length. How far were the CORS?
L1 will not work with OPUS, however you should be able to use the CORS data in your PP software with your L1 units to obtain a decent solution as long as your baseline length is around 20 km or less, five hours of clean data should be more than enough.
Tying to a couple of BM's is going to maybe be a better way to go OR in addition to the CORS for a reality check.
CORS + Geoid model with correct procedures should get you far better than "feet", so I suspect something is wrong and until you add some checks you don't really know what could be wrong.
SHG
Someone correct me if I'm wrong here, but if you know the local flood height elevation, that is the datum you should be relating your elevations to. The ellipsoid/ortho height of a CORS station a hundred miles away is academic.
OPUS requires L1/L2.
Do two sets (break a leg) of simultaneous observations on (local flood height datum) BM & your TBM so errors cancel out. If two solutions don't agree within a couple of cm., rinse and repeat.
That might be due to subsidence, I know there are areas around Tucson that have a measureable difference in just a few years, in the case of subsidence of course the local BM's sink too whereas the CORS may not...
SHG
> ...the points checked very well vertically between each other according to what I shot with the total station.
The local points checked but how did the various CORS check one to another? In other words, what were the residuals CORs v. your 5 hour point?
> .... I asked another friend, who told me that I can't use the CORS sites for good elevations, only for horizontal work.
Your friend is mistaken. While it is true that levelling is more precise, you can certainly use CORs to determine elevations to a fraction of a foot.
> What's the story, can I use the CORS data for this.....
Yes you can. Properly done, you should be able to determine a NAVD88 elevation to +/-0.25 ft or so without any heroic effort.
>or do I need to find a benchmark?:-S
Always prudent to check into some brass...
Also, just because an elevation is published on some physical benchmark doesn't mean that said elevation is correct. There may be datum issues, which can account for a few feet (or more). I've known local benchmark networks with huge errors in them. So consider the source.
Why don't you drive to the nearest benchmarks and process a CORS elevation on them as well? You can do it without doing a differential level survey and build some confidence in what you are getting for an elevation. I would do that kind of checking until I built a confidence level as to how many I should check to and if I have some consistency in precision.
If you can do some differential leveling, get a feel for what kind of vertical differences you get between benchmarks from both GPS and levelling. You can listen to everyone tell you what kind of precisions you have, but I would think you would want to build your own level of confidence. ("level" of confidence. LoL)
Send Me Your Files
Send them as RINEX "O" and "N" files.
Mississippi and Arkansas are rather light on CORS sites, making L1 only harder.
Point to point in the field is alway good L1 only.
From remote CORS it is important to have good orbit data, my minimum is ultra-rapid.
In GNSS Solutions it is better to use the VRS options for L1 data.
I began using 3 L1 receivers and will let you know what is the best you can get from that data. Let me give it a try. Be sure I have your antenna heights, and let me know what 3 CORS your friend used.
Reply to my email to you.
Paul in PA
Thanks again for the help, everyone!
