Alright guys
I've been asked to carry out a topographical survey which is an add-on to someone else's work.
I've got a couple of his stations I can set up on in the southern part of the site. There are a couple of his stations in the north part of the site too which I can end up on.
If I did this without closing the traversed loop is there any way of assessing traverse accuracy (1 in XX,XXX, or is it just a case of "well I started on two of his known points and ended up within 20mm of another of his known points so I must be good".
I'm not using GPS, it's a local arbitrary grid.
I know some folk when doing long linear surveys start on a known GPS point and finish on a know GPS point. How is accuracy determined on this type of scenario - its similar to mine.
Cheers, Andrew
?ÿ
One could look at your example as a closed loop where the predecessor did half the loop and you did the other half. Then the error of closure is 20 mm.
Some surveyors would not be comfortable without checking the whole loop themselves. Perhaps some would accept it.
You are starting on two known stations, and closing on two known stations? That's a closed traverse, of the link variety (as opposed to the polygon variety that most think of when they consider closed traverses). Closure does not require a literal/physical loop on the ground.
Angular closure is computed by comparing your final raw azimuth against the record azimuth between the two end stations, and precision is the raw linear misclosure to the final control point, divided by total traverse length. Same goes for starting and ending on GNSS-derived points.
The most rigorous way to do this would be to traverse from one pair to the other and then back, so you have redundant measurements and can check for systematic/compensating error.
?ÿ
You are starting on two known stations, and closing on two known stations? That's a closed traverse ...
The difference is that you won't have any ability to access the internal accuracy of your own work vs. the other guys. ie/ there will be some misclosure and you won't know how much of that is attributable to your work, and how much is in the other guys work. But it is still a closed traverse, and adjustable.?ÿ
?ÿ
If you traverse from 1 known pair and close on the 2nd known pair, you??ll only be seeing a closure based on the original control, which may or may not be Good. Do you have any idea of the fit of the original control? If you fit him within a tolerance you??re comfortable with, and you use error trapping methods to ensure your work, you can have some measure of comfort in that closure and it might be acceptable based on the needed accuracy of your pick up topo.?ÿ
you should be able to figured out an expected closure based on your methods.
if you don??t hit the 2nd known pair within a comfort level, you??ll need to decided on the value of the return loop. If you do that, you can do a minimally constrained adjustment of your own work to validate your own goodness of fit - then evaluate the project including all known points for an overall fit.?ÿ
As others have said, it's closed, closed-connecting is what I've always heard and used.
20mm error seems pretty good.?ÿ Either your work and the original are pretty good, or you have compensating errors that make it seem so.
Hi guys
Thanks for the replies. I can't be as clever/experienced as you all because I just don't see how I can adjust or quote standard 1 in XX,000 accuracy figures as you normally would for a Bowditch (or similar) traverse adjustment. Surely you can only do this when you have all the angles within the traverse. As the first traverse (assuming a traverse was carried out) was done by a different surveyor I don't have access to his angles between stations.
If you're saying I can work out, please point me in the right direction. This wasn't covered on my surveying course.
Cheers, Andrew
As others indicated, you first would have to be assuming that the knowns you have are good. So you would be holding 4 coordinates as being without error and adjusting to them. You should be able to determine if you are comfortable doing that based on the results of traversing between the points. However, any error you find will be attributed to and adjusted in your observations without knowing the quality of the provided coordinates.
Occupy the provided pair at the beginning (holding calculated azimuth from coordinate values), turn through your new traverse points until you close by observing the angle created by backsighting your last new point - occuping one of the provided end points - foresighting the other provided end point.
Using the coordinates provided for the end points, calc azimuth between ending pair. Holding beginning azimuth, calc azimuth of your lines based on your observed angles. Your ending azimuth observed between the provided pair and the calc'ed ending azimuth from the held pair of points will indicate angular error. Distribute and adjust in the same manner as a closed polygon. Latitudes and departures calc'ed per usual and added to beginning provided coordinate. difference between ending coordinates calc'e d per your observed azimuths and distances and the provided ending coordinate is linear error. adjust based on total error and distance of each line per usual, adjust latitiude and departures per usual which should provide matching calculated and held ending coordinates.
Quick google search reveals this website (don't know how well presented or accurate the information is, just found it as one of the first links that provided sketches and explanations) - http://www.jerrymahun.com/index.php/open-access/topic-trav-comps/47-travcomps-chap-h?showall=1
Cheers guys?ÿ - many thanks for explaining. As you have said I'll end up adjusting my traverse not knowing if the errors/misclosure is due to my traverse or the stations already in the ground.
I think for belt and braces I'll survey some common points such as corners of buildings, inspection chamber corners etc that he has surveyed and see how my results match up. Either that or I traverse back to my starting stations and close my traverse.
Regards, Andrew
?ÿ
I??d recommend closing yourself out, with a few ground truthing shots inside- especially since the tendency for projects is suffer scope creep. Remember- not counting your accumulated errors, the difference you observe between your survey and the original is likely only about half of the error adjusted thru the original network, assuming the originals was a compass adjustment. I??ve been on jobs where the original control was dang sketchy and tptb didn??t want to ?ÿspend any time/money on sorting it out, but they wanted good pickup info. It sucked- when every backsight ?ÿshot results in questioning whether or not you??re jacked up or the control is jacked up - the days are long with a capital L. ?ÿ
