The current MINIMUM STANDARD DETAIL REQUIREMENTS FOR ALTA/NSPS LAND TITLE SURVEYS (Effective 02/23/2016) specify that :
"The maximum allowable Relative Positional Precision for an ALTA/NSPS Land Title Survey is 2 cm (0.07 feet) plus 50 parts per million (based on the direct distance between the two corners being tested)."
and it defines Relative Positional Precision as:
"the length of the semi-major axis, expressed in feet or meters, of the error ellipse representing the uncertainty due to random errors in measurements in the location of the monument, or witness, marking any corner of the surveyed property relative to the monument, or witness, marking any other corner of the surveyed property at the 95 percent confidence level."
So, how does one compute relative positional precision? The ALTA specification states that:
"Relative Positional Precision is estimated by the results of a correctly weighted least squares adjustment of the survey."
If you are independently determining the coordinates of two points by some method such as, say, RTK, wanting to be certain that the length of the semi-major axis is no more than 2cm at the 95%-confidence level, and are using some positioning technique in a way such that the uncertainties of a pair of points are equal, then what is the maximum uncertainty that is allowable for the position of each point?
The answer is that a DRMS uncertainty greater than 7mm for the points will fail the ALTA/NSPS specification for their relative positional precision when they are close enough together that the magnitude of 50ppm x D is neglible, where D = Distance between points. By DRMS, what is meant is that the uncertainty is such that the coordinates of the point as determined by survey have a roughly 68% likelihood of being within a distance of less than 7mm of the actual position of the point.
In other words, to be certain that one is meeting the ALTA/NSPS specification, the points need to be positioned with an RMS uncertainty significantly less than +/-7mm. If you are using a positioning technique that on a good day, under ideal conditions, just barely has an RMS uncertainty of 7mm, then the odds are very good that the results are substandard.
The cure is simple. When the points positioned are close togther, use a better positioning technique than one that only delivers 7mm DRMS. A total station should be able to easily do this or you can get several repeat measurements of the points positioned by the process with +/-7mm uncertainty, if the repeats actually do give a better answer instead of just one with the same errors as the first.
Kent,
You lost me on your jump from 2cm(0.07') to 7mm(0.02') and the change in confidence levels.
Paul in PA
Paul in PA, post: 383751, member: 236 wrote: Kent,
You lost me on your jump from 2cm(0.07') to 7mm(0.02') and the change in confidence levels.
Paul in PA
You don't understand, 7 is greater than 20
Paul in PA, post: 383751, member: 236 wrote:
You lost me on your jump from 2cm(0.07') to 7mm(0.02') and the change in confidence levels.
If you have two positions, each independently determined (the random errors in the positions aren't correlated) with uncertainties of +/-7mm DRMS (and with nearly equal uncertainties in N and E components of the positions, i.e. approximately circular error ellipses) then the uncertainty in the relative uncertainty of each in relation to the other is :
1.414 x 7mm = 1cm DRMS
and their relative uncertainty at the 95% confidence level is:
1cm x 2 = 2cm 2DRMS.
+/-2cm 2DRMS is the maximum allowable uncertainty per the ALTA/NSPS standard, so one is right at the edge of failure if he is positioning the two points with an uncertainty of +/-7mm DRMS.
For 2D, the multiplier from 1 sigma to 2 sigma is about 1.6 not 2.
I have RTK but I rarely use it because the surveys I do are usually marginal to heavy canopy so trying to use RTK would be difficult at best. I have a base and two rovers. For example, the emergency survey I did over the weekend involved approximately 250' x 150' in steep forest with 9 stations set. We left the receivers running in faststatic base mode overnight (we have overnight security) and the results came out good. One in the open, one mostly open and the third marginal (that one was still running in the morning because I had the cowbell on it). The rest of the points were positioned using a network of total station and scanner measurements to targets on tripod/tribrach. I measure targets in both faces on the scanner. The new scanner is as good as a total station based on the residuals I see.
From my own observations in optimal environments at various ranges, I've found my RTK system provides <1cm+1ppm accuracy (2D) at 95%. Two points positioned by RTK (considering I'm using an on site base which negates ppm error) will have an error estimate of 1.4cm (0.05') at 95%, well inside the 2cm(0.07') allowable error. Repeat observations will reduce this error estimate.
This is exactly what I typically see - Hz error estimates that are below 0.05' at 95% confidence. If I have two or more independent shots on a point that are below 0.05' at 95% confidence, and furthermore they produce a weighted mean that no single measurement deviates from by more than, say, 0.03', any reasonable assessment of that point would determine that it more than meets the standard. However, if I need to produce a least squares adjustment report to make some thick skulled dinosaur happy I can do that too. Whatever it takes.
We had a job recently where on the front end the client said they didn't want / need a proper static network, and then on the back end decided that they wanted a network adjustment report, etc., etc. Fine; we had done the static work, I can kill trees with the best of them.
Shawn Billings, post: 383770, member: 6521 wrote: For 2D, the multiplier from 1 sigma to 2 sigma is about 1.6 not 2.
Yes, sorry.
7mm would be the maximum allowable uncertainty at 50% confidence level for each point in order for the relative error not to exceed 2cm. At the 68% confidence level, the point uncertainty would have to be certainly less than 9mm, which is hardly fatal to the point that RTK probably won't meet the ALTA/NSPS specification except under either ideal conditions or after a number of remeasurements that would have to be verified.
Shawn Billings, post: 383779, member: 6521 wrote: From my own observations in optimal environments at various ranges ...
Doesn't pretty much everything work in optimal environments, with longer sessions, and over very short baseline lengths?
Should one assume that there probably is a good reason why RTK accuracy isn't quoted in real world environments ?
Kent McMillan, post: 383785, member: 3 wrote: ...or after a number of remeasurements that would have to be verified.
Have you been reading our sales brochures? That's practically plagiarism.
I think the main takeaway for readers of this thread is that the premise is simply incorrect. RTK can provide positions that satisfy ALTA/NSPS requirements. As with any surveying methodology and equipment there are notable conditions, however, any absolute claims that RTK cannot provide results that exceed this minimum requirement are simply not true.
Shawn Billings, post: 383792, member: 6521 wrote: I think the main takeaway for readers of this thread is that the premise is simply incorrect.
Sorry, the premise (as amended) was:
"The cure is simple. When the points positioned are close togther, use a better positioning technique than one that only delivers 9mm DRMS. A total station should be able to easily do this or you can get several repeat measurements of the points positioned by the process with +/-9mm uncertainty, if the repeats actually do give a better answer instead of just one with the same errors as the first."
If you don't think that is correct, you're free to say why you think otherwise.
In the real world, however, the obvious problem is how to demonstrate what the uncertainties are in the points positioned and to show that the relative positional uncertainties between adjacent points positioned by a survey don't exceed maximum allowable values. That is a problem that can't be wished away.
Your approach of testing an RTK receiver under optimal conditions may not have much to recommend it for real world use of RTK. It's understood that salespeople are going to present their wares in the most favorable light possible.
Shawn Billings, post: 383789, member: 6521 wrote: Have you been reading our sales brochures? That's practically plagiarism.
LOL! I'm looking forward to anyone actually doing any rigorous testing in real world environments of the equipment you sell. The test on a 2.93 ft. baseline was something that I don't qualifies as other than a novelty item that identifies the maximum performance that can ever be possibly expected under conditions so ideal as to be completely unrealistic as to expect.
Kent McMillan, post: 383785, member: 3 wrote: Yes, sorry.
Kent, this is progress. I'm impressed!
Nate The Surveyor, post: 383801, member: 291 wrote: I'm impressed!
I'll bet that you don't use your RTK in optimal environments. Maybe you could help Shawn out by describing the sort of uncertainties in positions that you are getting under the trees in Arkansas and how you are demonstrating that the uncertainties you think are present in your work are realistic.
Maybe some day I'll get around to testing my RTK system in canopy.
https://surveyorconnect.com/community/threads/testing-rtk-and-post-processing-in-canopy.326194/
Shawn Billings, post: 383808, member: 6521 wrote: Maybe some day I'll get around to testing my RTK system in canopy.
https://surveyorconnect.com/community/threads/testing-rtk-and-post-processing-in-canopy.326194/
I recently had the good fortune to do some testing under canopy with the R10's, it was surprising, just under the 7mm stated and under heavy canopy. I was not expecting those results.
I'm beginning to revise my error allowances and maybe will dip my toe into using the R10's where I never would have used GPS.
I've not done an ALTA survey with one (the boundary anyway), the combination of the R10 and S6 is difficult to beat for accuracy, most closures with the S6 checking the R10's are <0.02'.
From our real world testing it's very tight, much better than the old days, the micro centered antennas, error trapping and processing, there is a marked improvement.
But as always check everything, never assume anything.
Shawn Billings, post: 383808, member: 6521 wrote: Maybe some day I'll get around to testing my RTK system in canopy.
RTK Distribution from RTK average:
100% H 0.161 V 0.323
99% H 0.100 V 0.225
95% H 0.080 V 0.162
68% H 0.050 V 0.075
50% H 0.038 V 0.056
So, basically, you were finding such a high failure rate that only 50% of your positions had horizontal errors of less than 1.1cm and you didn't even bother to note the even smaller fraction of positions that contained horizontal errors less than 9mm?
Okay, so far you've determined that your RTK works well in wide open settings with long sessions and does not work well in environments with high multipath and rates of cycle slips. Was either a surprise?
MightyMoe, post: 383812, member: 700 wrote: I recently had the good fortune to do some testing under canopy with the R10's, it was surprising, just under the 7mm stated and under heavy canopy. I was not expecting those results.
I'll bite. How did you determine that 7mm [something, DRMS?] was a realistic estimate of [something]?
