GPS Control

What was the question?

I think you swapped a couple digits. Is it 1022 or 1202?

Are your numbers the mark-to-mark 3-D distance, the horizontal distance, or the State Plane grid distance? Or have you mixed them?

hmmm...6.31 feet. Were you using a fixed 2M rod? When you submitted to OPUS, did you check that your antenna heights were the same for both points?

I think it's fine. I have a good friend and surveyor who started using GPS this way. As a matter of efficiency, he bought a second OPUS receiver a few months later. With four hours of data (at least in my neck of the woods) I think you can achieve the results you mention consistently. You could even use the results in a least squares adjustment, so long as the coordinates are properly weighted (or even better, use the vectors from the CORS in the adjustment).

There was a typographical error in my original post. Thank you Bill for pointing it out.

The vector is 1202.228 feet between the two GPS points and the vector computed by the traverse 1202.266 feet. The first GPS point is 1022.703 feet elevation and 1029.012 feet elevation at the GPS point 2 for a difference in elevation of approximately 6.31 feet.

The difference in elevation is between two points approximately 1200 feet distant. The antennae height was 2.000 meters for both observations. I included the information about the relative elevations to demonstrate that the site is relatively mild slope.

The distances are not mark to mark. The geodetic calculator I use is computing the grid and ground distances.

Think in terms of a traverse from COORS station number one to your control number one, to COORS 2.

Then, COORS 1 to CONTROL 2 to COORS 2.

Think of them as individual traverses.

This will give you an idea of what you did.

Nate

> The difference in elevation is between two points approximately 1200 feet distant. The antennae height was 2.000 meters for both observations. I included the information about the relative elevations to demonstrate that the site is relatively mild slope.
>
>

I initially read that as 6.31 feet difference from two observations on the same point. :-$

I use a similar procedure. At a minimum I always try to include a 3rd point as a check for any sort of rotational problem. Also, I've broken the rinex file into smaller files and submitted those to OPUS-RS just to run a check using the different algorithm. How did your elevation difference between the GPS observations compare with the trigonometric leveling elevations? (Again, I like to have at least one other point as a check for elevation, horizontal, and just as a redundancy.)

I am getting a level instrument to get the elevation difference on my traverse. There is a first order Bench Mark nearby. I did not collect elevation data on the traverse with the survey instrument. It will be interesting to see how the orthometric heights from OPUS compare.

"I did not collect elevation data on the traverse with the survey instrument."

Any time you combine GPS with a field traverse the field traverse should include vertical observations. The vertical agreement is one of the better checks in comparing your results.

How close were your various OPUS Observations on each point?

It appears that your accuracy is within the OPUS points precision.

However the difference you see could even be the result of failing to have the traverse instrument set for the particular atmospheric observation conditions.

Paul in PA

I surveyed a traverse between the two OPUS positions using ground measurements in an arbitrary coordinate system. Then I calculated a transformation into the State Plane Coordinate System with an average elevation for the project.

The elevation used in the transformation was the average of the OPUS orthometric heights on the two points which varied in elevation approximately 6.3 feet.

The calculated grid coordinates were compared to the OPUS results within approximately 0.04 feet.

Atmospheric corrections were used in the traverse and amounted to only 1 ppm that day. Before I complete the project I plan to perform more redundant observations.

Not that there is anything wrong with it; but that's a lot of GPS time on two control points 1200' apart, I wouldn't charge a client for that unless it was something really special. And there is no redundancy in the GPS vectors.

So to get that would need another 8 hours minimum. Whew!!!

Typically we run control for short legs like that using much shorter time frames, and multiple occupations, I would suggest computing to a local cors station with 1/2 hour locations (at the most) and at least twice on each point, if the station is near then maybe only 10-15 minutes will be necessary.

OPUS is nice as long as it's within the work flow, but GPS can eat up charges like crazy sitting on site points for 4 hours at a time.

I am working on other parts of the boundary and road layout while the receiver collects data. Besides it's only money, right? Actually I am working pro bono for the Historical Society.

Perhaps I will try some shorter sessions on the control to compare the results. The CORS stations I am using to process are approximately 30 miles away. It is a very rural area.

30 miles is really close.......at least for my area.

Actually, if you are that near some Cors, I would take a bit of time and do some testing, try 15 minute sessions and you may be really surprised just how tight your points will process.:-)

The recommendation has always been about a minute per mile, plus 10 minutes, so 40 minutes, but practically 15-20 minutes will return really good results.

I'm assuming you can process your own static

ok, thank you.