Positional Accuracy Checking

> Does anybody really care?

I think that the RTK users generally seem much less likely to care about meeting relative positional accuracy standards. The rest of your questions really boil down to this: "Does it matter if a land surveyor falsely certifies to compliance with a standard?" In Texas, I'd say it matters quite a lot. I'm unsure about other states.

> Yeah, I suppose a land surveyor should be able to make good measurements but being able to do so doesn't make you anything special anymore.

I would certainly have thought that would be true, but there are surprisingly few surveyors who seem to have much of an idea of what the uncertainties in their measurements actually are. Having a handle on uncertainty is pretty much the test of good measurements; without it, a surveyor is essentially guessing about measurement quality.

>So how is a surveyor to make a living if his/her main specialty is making precise measurements.

Actually, making professional-quality measurements is only one of the four main activities of land surveying that are unlikely to disappear anytime soon. None of the four are sufficient in themselves and all are necessary.

The irony is that as technology becomes more capable, the abuse of that technology is ready to make up for the improvements.

I've known from experience for a long time that there is some slop in RTK measurements and there is some in static also. The CORS station coordinates themselves introduce error into OPUS solutions. More so out west where the earth's crust is less stable (they will probably cure this before long).

BUT, there is another aspect to land measurement and that is speed and economy. Many clients are more than willing to give a bit on the accuracy for quicker turn around and lower cost. If push comes to shove the rules will be modified to make the less costly (and a bit less accurate) measurements acceptable. I'm sorry but this being spot on a dime just isn't worth it to most if not all of the clients I do work for. They are not going to pay a bunch more for the last couple of tenths. What needs to be relaxed is the spec to make RTK OK where it is useful and acceptable. Hey, if you got to have it to a hundredth, they pay more and a different sort of equipment should be used. My guess is that the mass use of RTK and other GPS isn't going away anytime soon.

I've found a lot of GLO monuments using RTK just because of the ease to get to the search area. These markers have been sort of lost for the last hundred years because it was just to much trouble to traverse into the area. Once I find a corner then it usually gets the static GPS treatment or a side shot from a GPS suitable area. Hey the marker is there, iffn a surveyor using GPS can't find it when he is within say a foot then there are other issues that should be addressed, not the measurement skills.

I tried to be a bit more honest and only report my measurements to a tenth and the angles to 20 seconds or so (depends on the length) of the line. Seemed to work fine until the survey had to pass some govment checker somewhere and then it gets rejected because the figure doesn't close to 0.01 in the CAD (must be badder than bad). So IF THEY WANT A COOKIE GIVE EM A COOKIE. They don't realize what kind of crap they are eating anyway!

> BUT, there is another aspect to land measurement and that is speed and economy. Many clients are more than willing to give a bit on the accuracy for quicker turn around and lower cost. If push comes to shove the rules will be modified to make the less costly (and a bit less accurate) measurements acceptable. I'm sorry but this being spot on a dime just isn't worth it to most if not all of the clients I do work for. They are not going to pay a bunch more for the last couple of tenths.

Have you told them how much you could save them with just a consumer-grade handheld receiver? I feel certain that they would want that even more than RTK. If a fence corner shows up fairly distinctly in GoogleEarth, is there really any need to spend all the time driving over to it? Once you start letting a client's desire to get things cheap and fast dictate professional standards, where is the bottom?

> I've found a lot of GLO monuments using RTK just because of the ease to get to the search area.

I trust you realize that when you're searching for monuments of surveys 100 years old submeter accuracy is probably exactly as useful as few centimeter accuracy, right?

Does that mean that all rural surveying in your area should be made using submeter positioning methods, then?

If there are monuments, and the monuments are held and respected and sub-meter allows one to locate the monuments, I'd say sub-meter works just fine.

I suspect there is a lot of DIY surveying going one. As long as there are markers that they find it probably doesn't do that much harm. DIY is a lot cheaper than a LS on dime, recording and all of that. And hey, what are landmarks for if not for the landowners to use to locate their boundaries. I really don't care that much, I'm willing to work for those willing to pay. Sometimes the DIYers cause more work for a land surveyor untangling a mess but this has been going on forever.

I went out on one a couple of weeks ago where a guys father-in-law had done his own hand held GPS survey. They had actually found about half the dozen of so corners because of the sentry t posts. They didn't know that there was a separate rebar close by that was the actual corner but in this mountainous area a foot doesn't really matter that much. On one side they got off some (no standing t posts) and where about to set a solar panel 18 feet over the line. So using the RTK GPS and a magnetic detector I was able to find all the corners except one which I don't think was ever set (lot split after original subdivision).

So I had a nice day in the mountains and made a few bucks. Interesting subdivision, nice original survey, probably 200 lots or so. First time I've ever been there, I don't think there has been much retracement by surveyors. That's how it should be, a good original survey with markers that serve the purpose they were set. I'd say the future should be for surveys to be pretty much a one time thing. Set good markers, file the results w/coordinates in a public place and from there other than destruction and some landowner fraud the DIYers should do just fine with the smart phones. If you want to measure and certify to a dime that's fine, a good mark should be findable within a foot or so. Let's just use the mark instead of pincushioning to the dime.

> If there are monuments, and the monuments are held and respected and sub-meter allows one to locate the monuments, I'd say sub-meter works just fine.

Isn't the point of survey accuracy that there won't always be the monuments? The monuments will get destroyed and their former positions as determined by a surveyor will be important to the maintenance of boundaries as they were.

If you got the main point that was made in a thread a couple of days ago, it was that if you locate a monument by a process with a certain standard error and then replace it using a positioning process with that same standard error, then (assuming that the errors in the two positions are independent of each other) the most likely amount by which the actual position of the replaced monument will differ from the position in which it formerly existed will be twice the standard error of the positioning processes. This is for the case when both positioning devices (read RTK controllers) are indicating that the coordinates of the points are exactly identical.

In other words, if you use a process that gives coordinates with standard errors of +/-0.10 ft. in N and in E and you replace the monument by a positioning process with the same s.e. of +/-0.10 ft. in N and E, even though the gadget says that the replacement is at EXACTLY the same coordinates, the most likely reality is that it will be 0.20 ft. away from where the monument really used to be.

I understand what your point is Kent. It just to me and most of the clients I work for a couple of tenths usually doesn't make any difference. There are lost corners I've searched for that a couple of feet or more would have been great (other than some proportion to a hundredth that we all know is not near the original). My point is the law of decreasing returns. The cost of the accuracy may not be worth it to most land boundary clients. All clients shouldn't be required to buy a hundredth when a couple tenths is OK for them.

If you got clients willing to pay for the high accuracy you provide that's great for you. Unless your rates are really low you wouldn't be able to get work much where I'm at. I turn a lot of work away because I won't work for nothing. The accuracy you are promoting would really be costly up in the mountains around here. A survey could cost more than the lot if it all has to be on a LS dime. Wouldn't you rather know where you lot is to within a foot than have to sell the lot to pay for the survey?

And also, most OPUS results I get have 1 to 2 cm spread in the average coordinates to start with, so the error budget is burned before you start. What should one do, traverse back to a main CORS/GPS control station. The ground can be unstable and monuments move (downhill mostly in the mountains). Land surveying should be practical and sensible and not be required to achieve such a high level of accuracy that there is no valuable gain from it.

But hey, we both know there is no end to the argument for arguments sake, eh? So what's the point of continuing this discussion to much. Don't seem to be any other interested posters. Nobody in their right mind is going to get between us, right?

But what about the original question for the thread. Where are the positional checkers. Are there any? Does any state require it or is it just a bunch of words on paper that nobody really does anything about. How would you check it and prove it is out of spec. Just a bunch of math, change a couple of inputs and the answer changes. Measure it again and the answer changes again. Keep messing with the input and sooner or later the answer will work out. I've crunched a few numbers in my life, know all about it!

Gotta go make some money, see ya

> I understand what your point is Kent. It just to me and most of the clients I work for a couple of tenths usually doesn't make any difference.

Well, I think the point is that it really doesn't cost that much to meet most relative positional accuracy standards if a surveyor is aware of what it will take to do it. The problem pops up when one-beep RTK is the standard. It's understood that clients will always complain about how much surveying costs. That's to be expected. You could be charging $20 per hour and you'd still have clients calling explaining that they "only need one line surveyed".

So my view is that it's a losing game to try to cater to the clients's expectation of Walmart prices. Naturally, there is such a thing as surveying economy, but the truly economical solution is seldom identified by the cheapest invoice since there is always a hidden cost to low quality.

In Texas, I'd say it matters quite a lot. I'm unsure about other states.

So does Texas have and agency that checks positional accuracy of surveyors? Does the board have a checking department that reviews and checks say, just 10% of the certified measurements? What equipment do they use that is so much better than survey grade stuff so that the results are truly reliable? What's the sense of having a spec that nobody checks for enforcement purposes? What says the LS results from the second set of measurements are better than the first?

Always unanswered questions. My guess is that Texas doesn't do any checking or if it does only very rarely (like a lawsuit maybe but I've never read it in an opinion). If you are going to get nailed for bad measurements its more likely in some construction layout than a boundary being a bit out of a LS report. The good construction layout surveyors are better at measurements than most land surveyors and pay a much higher price for messing up.

> So does Texas have and agency that checks positional accuracy of surveyors? Does the board have a checking department that reviews and checks say, just 10% of the certified measurements?

Actually, the question of competent practices usually arises when things go wrong. A surveyor who adopts practices that obviously don't meet the standards he or she is obligated to meet just raises their liability for negligence.

As a general rule, if you have to defend your work, you don't want to admit that you routinely neglect to meet common professional standards. A pattern of negligent practice is bad. If you certify that you meet a standard when it's obvious that a competent surveyor would recognize that you most likely did not, it even calls your truthfulness into question as well as your competence.

Generally, a surveyor wants to be able to demonstrate that he or she followed a more than acceptable standard of care and does not want to concede that virtually none of his or her work would meet the standard that it ought to.

I think that the RTK users generally seem much less likely to care about meeting relative positional accuracy standards.

I think surveyors in Texas generally drink too much at lunch.
😀

> I think that the RTK users generally seem much less likely to care about meeting relative positional accuracy standards.

One need look no further than the recent threads in which virtually no RTK users have wanted to discuss methods to overcome some obvious problems with RTK accuracy. Instead, the trend has been to argue that the standards are too restrictive or somehow would be ridiculously expensive to meet.

Using RTK for control or boundary work shows that you don't know which tool to use for the job.

I've never had them come and check my positional accuracy, doesn't mean someday they won't.

Kent, have they ever checked your positional accuracy? If so, how did they do it?

I'd like to hear about someone who has had there positional accuracy checked and better yet someone who failed and what was the consequences.

If there is a problem it's usually something bigger than a little out of the positional accuracy spec (like measuring very preciously from the wrong point or staking out when one should be holding a previous marker or established boundary).

RTK has some slop in it, I know been using it for almost 15 years. Some conditions degrade it and a bad fix will really send it out. Funny thing is the worst problem I ever had with a survey was when a total station went wacko and measured ever shot 0.11 feet bad. I sensed something was wrong, had to go back and redo a whole days work.

Considering some of the overall advantages to RTK it shouldn't be denied use on boundary just because it doesn't meet the tighter specs of other measuring devices. There are better tools than survey grade total stations available. Maybe we should upgrade the specs to a couple millimeters. Not because it's needed but just because it’s possible.

exellent point, Kent.

> Does any state have an agency that checks the positional accuracy of reported survey measurements? Do they randomly check say 10% of all land surveys. If so what kind of measurement equipment do they use? What kind of fines are imposed for failures? Are there many big judgments for being a bit out of spec.
In an around about way, any recording state has a form of accuracy checking when another surveyor reruns a line of record and then files his own. And a state, like Oregon, which has County Surveyors or other agencies checking Records of Survey prior to recording has an extra layer in the system.

> I suppose you could properly tape anything under a hundred feet but how would you check say a 25 mile baseline and do better than GPS without GPS.
There was a once upon a time not so long ago when a 1:5000 closure was considered pretty good. I don't think any surveyor would be very impressed with your 1:5000 closure today. Expectations are much higher. Sure, you can push button a position to any number of decimal points, but it is easy to push button such a position with poorer precision that a 1:5000 tape and transit job.

Thanks Kent

I appreciate the effort. I certainly learn new things almost every time you post these types of posts. My take away from this one IS, the issues relating to the EVALUATION OF MEASUREMENT EVIDENCE. This is a proper discussion. We are,after all, discussing the basis of professional opinion and how that professional opinion is presented to the consumer of professional surveying services and not Granny's cornbread recipe. Thanks

2cm +/- OPUS ? Reply to LR Day

OPUS accuracy of 2cm +/- plus residuals is entirely sufficient when combined with a static GPS network and/or a standard traverse to meet 0.10' positional tolerances. OPUS was never intended to be the "do all, end all" single solution. It is a single tool in a rather large toolbox. Can you to build a house with just a hammer?

To claim that surveying to meet minimal standards is in fact a premium product that the client should pay extra for is a sad comment on your overall professional outlook.

Kent has in essence made a blanket criticism of general surveyor capabilities to meet minimum standards. The sum of the replies tells me that overall they cannot.

As to my question, "Can you to build a house with just a hammer?" The answer is YES!

Before WWII my father was contacted by a gentleman who had purchased a Sears home kit, Maplewood Model 3302. The man felt incapable of building it so he hired my dad. My dad says it was the easiest work he ever did. Everything was precut, even the short headers and fillers for windows, labeled plus included pilot holes at every nail location. Simply take a hammer, follow the instructions and fill every hole with a nail. Until it was time to hang the doors, when a screwdriver was recommended, this house was built on site with only a hammer. As I travel around I see the exact same model and voila they all look exactly alike.

So go ahead and use just only OPUS and you to will have an empty shell, without doors.

If you can make your living selling that, more power to you.

It seems that every house built now has multiple inspections, inspections which a Sears kit house could readily have met.

Alas states do not inspect your surveys. It seems the state assumes that a licensed professional would meet professional standards. Silly states!

Paul in PA

2cm +/- OPUS ? Reply to LR Day

I use OPUS to establish my surveys on world wide control.

I use OPUS to document PLSS corner records as that is an acceptable documentation in my state (I submit the OPUS solution with the other docs).

As far as only using OPUS for all my work that would be difficult. One project going on is the topo of a 80 acre area in a reservoir. Going to take thousands of shots. So I'd either need thousands of receivers or take a few years to complete, not possible under the contract, so I'll need to stoop to the use of RTK (they have given me a tolerance of 1 foot H and V so I’ll probably meet the spec).

I'm not against specs for survey work. I'm reluctant to require specs to build a Swiss watch when a Boy Scout sundial will be sufficient. So boundary work is somewhere in between and it depends where you may be working (zero setback in NYC or 100 foot set back in very rural area).

Actually if you are getting good RTK shots the 0.1 tenth (horizontal) spec is not that hard to meet as that is under the 1-2cm claimed precision. But bad conditions for observation can cause problems which I dully note and have experienced many times.

I'm somewhat perplexed by the idea that LS can improve the claimed precision or prove that things are good with a procedure that only averages two RTK measurements and massages other black box data that comes with an RTK observation. Maybe I'm just not all that up on it for GPS but LS from my view and experience depends on redundancy and many measurements from different control points and vectors between the points to bring it in. So to really tighten the network you should have multiple measurements from multiple base stations or a big static network to get vectors between your points. And I have done all this before and compared to RTK. Usually the RTK positions have been within a cm or less of the LS adjusted network. So there is not all that much to gain other than the claim that you are a better operator and got a few more millimeters. Multiple shots will catch a blunder and probably show you are working in a bad observation area so I'm all for that. A 3 minute observation is better than a beep but not all that much better a 5 epoch but I always use the 3 minute on important points usually with a tripod set up to stabilize the antenna.

One possibility for short distances only using GPS would be fast static. At least you would have a vector between the points other than two longer vectors from a base. So then the least squares would be able to actually better tune your network. Running two static session less than 50 feet apart might seem strange but would help one get within the spec with GPS. Might be easier to break out a total station if one is handy and add the vector that way (or even tape it).

I suppose maybe I'm relying upon my long time experience and developed judgement to decide just how much risk I take with my work.