Traversing

Maybe a dumb question, but if you were able to shoot the section corner with GPS, then why spend 8 hours cutting line and traversing through muck? I'd rather do an hour or two of static ... or were there other things you were locating along the way?

Anyway, to answer your question, rerunning the traverse forward or backwards, it would not make a difference if you are trying to find a bust. Could be though, that there isn't really one large bust, but maybe several smaller busts (due to the ground instability) that add up and compound into your large misclosure.

Agree either end works and equally distribution of errors.

Curious, when you are about to read next station, set up point etc, do you reobserve backsight?
I also reobserve and record backsight before pulling up and moving on.

Spongy ground can be a pain.
In bad cases I do a few new setup (logging in DC) and record back and fore readings whilst keeping eye on results.
Might take 2-3 minutes extra each setup but can pay (big) dividends.

Doesn't help now I realise.

I believe that if you have known positions at each end of a traverse or are working with a closed traverse, that running the traverse from each end IS a way of finding a single bad angle or a bad distance.
Look for a point where the coordinates suddenly miss each other.

Yes - if it's an "open ended" traverse running between known control computing forward and backwards will allow you to identify a single blunder. Note that this is "in the office" on paper, not in the field.

And you can do something similar with a closed traverse by simply computing it a second time on top of itself and a single blunder can be identified.

Nevermind misread. Open traverse yes you can run it backwards or use a triangle to try to find where the error occurred if there is one error still. If you know the monument you tied to is good then you do have a closing angle to work with I suppose.

You don't have to run it backwards.

Take your two closing points (two each where you tied back in) and draw lines from the good point and the point with error. Then connect the two. If you have one error it will point to where it is pretty close. Then you can go back out and check those traverses and usually find the error. Now if you have more than one then it won't work.

Here is a pic of what I mean. I found many errors in this way over the years. When we used to use field books and write everything down this was the way to do it. Seems like one crew or another made a blunder all the time and this was the quickest way to find it.

Here are a couple pics. It works as long as you have one error not a bunch of errors.

closeup

All you are doing is pointing back to where say your closing angle plays out. If you have 5 minutes error in 6 feet you are just completing the triangle solution to point to where the error is.

I use lines nowadays instead of the triangle solution just because it is quicker.

> Hello,
>
> Has anyone ever heard of running a traverse backwards to find an error? We got to the end of an 1100' (+/-) traverse line to find and locate a section corner. We missed the corner by about 6 feet. Some traverse points were set in muck. I would have used large hubs to stabilize the tripod legs that I had to set in muck. We got the rover out to the corner and missed it by 0.47 feet. On the first setup of the return run I told the party chief the instrument was already sinking after he had given me a rod shot on the backsight. That ended the return run. We spent about 8 hours cutting line. The highlight of that adventure was watching a gray fox as it bounded around in the groundcover.
>
> Regards,
>
> Mark

Mark,

Your story brings back fond memories.

Anyways -

You say "end of a 1100' traverse"
Is this an open ended traverse?

Also, you say that the section corner was "missed"-
Was it a calculated position?
Or is the sect corn part of a close traverse?

Depending upon your field procedures considering the site conditions an error or blunder very well could have creep in.

How many trav pts were there?
What was your shortest trav leg?
Did you read Direct & Reverse on each shot?
Where there multiple sets of angles read?

If this was a closed traverse and the section corner is a traverse point, then the misclosure (imo) is angular and due to the site conditions. Most likly he instrument settling during BS/FS. - I would look at the shortest trav lengths 1st

Let us know how it turns out

Some good advice so far. Pay particular attention to the error isolation methods. Few attempt to use them these days..
I for one like to hear of folks traversing a line. You can't claim you followed the footsteps if you didn't even walk the line. Too much important evidence gets overlooked.
Back to the subject. When forced to perform a linear traverse, use horizon closure to double the angles. That's a fancy way of saying flip the roles of backsight and foresight. It will catch most blunders.

Traversing Backward In CAD

Turn your traverse line into polyline.

Copy that polyline and move the end point to the desired location.

Usually you can see between the 2 polylines the 1 or 2 questionable angle points, that is near where the polylines cross.

It may be necessary to rotate the polylines into endpoint to endpoint alignment, but usually they are close enough that rotation is not required.

If you cannot clearly see the angle points in the polyline, draw faint circles from each endpoint to a traverse point. The circles that are closest together indicate the problem traverse point. That is some extra work but as I said simple inspection is usually all that is necessary.

Sometimes the error can be corrected in the office. Look at the raw data for the suspected bad point, you should have run multiple sets at any suspected soft ground. If the instrument is settling the firs set or two may differ from the last sets. Use the first tightest group and see if that resolves the situation. It is recommended to return to that point and reobserve, if not from a firmer setup, at least using more observations.

Paul in PA

Back in the olden days the best way to locate an error in a traverse was to plot the thing then bisect the line between the ending point and the closing point. Draw a perpendicular offset on the bisected line and it will point to the faulty station or lines.

The muck issue is a real concern. Here in “World World”(3 T shirts for $4.99) if you are on a muck bed you can jump up and down and everything around you moves, tripods included. As a FL boy you should know that.;-)

Have a great weekend! B-)

Yeah, at one traverse point I was standing on a patch of grass. I watched the plate bubble float from side to side as I gently shifted my weight from side to side. Ridiculous!

After cutting line for a few hours I was wondering why we just didn't wander out there with a compass. There was nothing to locate along the way. We had a calculated position for the section corner. We also had some open sky around it. I think you're right on about the compounding errors!

We used the last foresight as a backsight for the return traverse.

Thanks for replying!

The position of the closing monument has not been verified by my company. If I understand you correctly, your method works to detect a blunder at one setup only. How do you use lines instead of the triangle solution?