J.D. and I can speculate endlessly about this, but here's the question: when will the next three to six hour period arrive when the K index will be 0? This is when space weather is optimal for long L1 vectors, the ionosphere is still and "clear".
Being of a sporting inclination, and using the 27 day rule, I'll say that it will arrive between about 11:00 PM and 5:00 AM CST on 2/02/2002. Any other guesses?
Best regards,
Kent McMillan, RPLS Austin TX
Modified By Kent McMillan on 1/21/2002 at 6:49 PM
I guess I should have read this challenge before I set the date this afternoon for O.R.G.I. 2 Interstate.
You may be right about 02/02, but I'll be looking at the conditions on 01/31 for my own assessment.
J.D.
It's a moral matter. ;>} (Just kidding, ---Brian)
We can change the color of anyone's receivers (for a price).
Modified By Brian D. Ewing, PLS on 1/21/2002 at 9:02 PM
I'm willing to wager that you're not apt to devote the time and/or energy to get three of your own units scattered around the countryside for this little experiment. Or am I wrong.
I'm planning to have one on POST, one some 100 miles South to position towards "the Crawfish Kitchen" and one up to the North (position to be determined later). Two on unknowns for densification of network.
Wanta' play?????
J.D.
J.D.:
It sounds lazy, I know, but I'm only a few miles away from the Austin CORS, AUS5. The RINEX file from it should be available. We can just process the L1 observables from it.
If the space weather is right, there's a chance that 300-500km baselines from AUS5 would be resolvable into fixed integer solutions with the right ephemeris and software. Don't tell Brian Ewing about the non-blue software angle, though.
Best regards,
Kent McMillan, RPLS Austin TX
Kent,
You aren't going to run a unit of your own? I know AUS5 is running, but you don't have a position or two to run a unit on overnight. Maybe between Austin and Houston somewhere. I'm looking for a convenient location, state or national park, to overnight around the Red River so as to tie Indian Territory to the POST. That's if Indian Territory responds. My wife is all up for a weekend out next week. I just haven't defined the term "out" yet.
J.D.
p.s. I don't think the non-blue is a threat. At least the numbers I've seen so far. But, this ain't a manufacturer competition and I've tried to not make it one.
J.D.:
I'm probably missing the point, but isn't the idea behind the test to see what sort of a large L1-only network can be solved and to see what sort of accuracy is possible with minimum constraint?
Why aren't the L1 parts of the CORS records just fine for that?
I've been thinking that the general direction of the project was to demonstrate the feasibility of getting a good connection to NSRS via the CORS network and using only L1 receivers for the unknown points. Your experiments demonstrated the sort of precision that could be accomplished under favorable space weather conditions at night.
On othe lines, I've been exploring the outer limits (sounds spooky) of baseline length and absolute single vector accuracy using only L1 data. That is how the IGS post-fit orbits and ITRF97 entered the mix, since the first directly effect the accuracy of long vectors and session length for integer fixing, and the second is necessary to deal with systematic errors that become significant as vectors stretch out to 300km.
The original design for L1-OPUS that we briefly fantasized about months ago (before Dave Doyle put on the green eyeshade and reminded us how little money is available to NGS to fund these sorts of things) contemplated that detailed ionosphere models derived from the CORS network could be used to improve L1 positioning to make connections to NSRS at close to the 2cm level (2-sigma) routinely possible. Showing what is possible with L1 in nighttime work seemed to me to be a contribution toward that end.
Along the way, what Mr. Geodesist had been mentioning (fairly regularly) about the effect of the ITRF97/NAD83 coordinate system mixing also entered the picture. It also strikes me that it effects the problem of estimating the uncertainty of points positioned since the increase in residuals in a mixed adjustment of ITRF97 vectors in NAD83 suggests that the uncertainties in the adjusted positions are significantly larger than your night-to-night repeats indicate can be had.
From a technical standpoint, what part of the movie have I missed?
Best regards,
Kent McMillan, RPLS Austin TX
I guess it's not all just the "technical" side I'm concerned with. The whole idea is a "user" network. Not a just "user process CORS" network. I think your work on the ITRF side, as well as precise ephemeris is great. I would like to participate in that phase as well. Certain constraints currently prohibit that. For now, densification between distant CORS sites may have great importance to some. That's in part what we are going to find out.
Sooner or later you may have to get your hands dirty. It's not all theory you know.
J.D.
J.D.:
You say that it is not all theory?
I use L1 CORS ties all of the time. They work great. That is how virtually all of my land surveys are connected to NSRS, either directly from the CORS or from points that I have surveyed on a network connected to the Austin CORS.
In the hands-on experience department, I offer this as either evidence of bona fides or of non compos mentis, you be the judge.
I'm working on surveying out a tract from a ranch that I resurveyed about ten years ago. The work involves getting some GPS ties to old control points so that the old conventional work can be readjusted in Star*Net-GPS in combination with the GPS ties that connect it to NSRS. This is a superior method to a simple coordinate transformation. Unless there is some problem with the adjustment, it is unlikely that much of the conventional work will get re-run. That part is pretty clean office work. I suppose that theoretical understanding is useful.
But the piece I'm cutting out is to be bounded by a creek. The owner and I have discussed the work off and on for several years and each time I end with "well, just be sure to get us to do the work when the creek is down." For most of the year, the creek is dry and it has a rocky bed that would be ideal for setting markers. (The creek probably hasn't moved in 500 years and isn't likely to start. Getting permanent markers in place to define the line for the purposes of exact location would be good practice given land values. It is navigable, but is crossed by an original survey with a huge shortage (this is Texas, after all), so the bed is in private ownership.)
So years pass and now there is a big rush to do the survey. The job was estimated assuming that we could GPS the centerline. Only now the creek is up and flowing pretty well. The banks are overgrown. The sky is out in the creek, so to speak.
Who was out wading waist deep in that 50 degree water with non-blue waterproof GPS equipment, trying to locate the meanders of the creek without sliding into the numerous holes and pools? And who after about a chill mile of that action was starting to wonder how in the heck he'd found himself in that particular creek on that particular day, good space weather or no?
I didn't get my hands dirty, but I don't think that we can write that off to theoretical understanding.
Best regards,
Kent McMillan, RPLS Austin TX
Modified By Kent McMillan on 1/22/2002 at 3:08 AM
Wasn't talking about daily dirt. We all have that under our fingernails.
J.D.
J.D.:
Well, if you're just talking about pushing the button to turn a GPS receiver on or playing chopsticks on a data collector, I'm afraid that both lost their novelty for me years ago. Beyond that "hands on" part, GPSing is mostly that bothersome "theory".
Good luck with the ORGIing,
Kent McMillan, RPLS Austin TX