Hello all,
I am wondering if there are any other users out there with a Spectra SP80 GNSS receiver having issues with elevations in their survey data. We are adjusting to a published benchmark elevation after running our Static over said benchmark for 20-30 minutes as a standalone survey and are tying down project control with both our Total Station and GPS receiver as well as running level loops over our initial control to establish elevations for the total station points. Overall, our projects look fine but in some sections that have clear visibility overhead and are fairly open spaces, we are getting shots that will show up to a 0.30' elevation difference from shots taken with the total station. I am stumped as to what the issue may be as it is not a consistent problem on every job and we are careful with the equipment ensuring we take a few seconds at each shot to allow the GPS to establish good location. Any help would be appreciated! We are located in Southern California and work mostly in the LA and Orange County area for reference. We are also connecting to base stations with a MiFi jetpack and the CRTN service through UCSD.
Replies
It may be the position of the satellites at the time of observation. If the satellites orientation are near the horizon you will get higher error for elevations. Check your PDOP (position dilution of precision)
Another issue you may be having is simply relying on a RTN to be able to duplicate work tighter than most are capable. RTN's are great but I have found them to be inconsistent at best. I think a lot of people on here, whether they like RTNs or not, can attest that getting a perfect matching pair and perfectly duplicating (as well as GPS can anyway) data is not the strong suit of an RTN.
RTNs provide almost immediate access to a correction that can be used, however, do you know where the actual reference station you are using is? Let's say your reference station is 5 miles away...think of the vectors involved, the horizon being shared, and all of the objects in between the rover and base....what if it is 13 miles away...22 miles away, etc?
I have routinely seen errors when setting pair points from various RTNs. You can shoot each one for 300 epochs, walk away feeling great, and then set your instrument up and all of a sudden it's 0.11' H and 0.14' V. Additionally, you can go back to the same point later in the day w/ the rover on the RTN and get a totally different solution. I am not a Network hater by any means...I actually think it is an amazing tool to originate site control, but with a single point.
Some other questions that could affect the site, outside of the RTN itself and some of the other items mentioned above (measure ups, geoid models) can be scale factors as well as what are you comparing your results to? If you provided a site control based off of a network and they localized to it, there could be an error on a point affecting the plane created when compared to a localization/site calibration. If one of your points was off by 0.3' and was then used to localize control, the plane created would accept that 0.3' error and adjust to match, whereas if you are going out there still on a RTN and relying on a geoid alone, the error would be there.
Also...could this be a Grid/Ground issue? Do you know how you are completing this transformation...combined scale factor, avg job height, etc. These can all have an effect on vertical. The last thing would be the equipment...the receiver could be out of, or going out of, phase center and need maintenance. This usually effects vertical the most drastically. You could be seeing longer float solutions before fixing and inconsistent results (especially vertical), which could signify some service is needed on the receiver.
It could all be wrong for your situation, but I hope some part of this helps!!!
Yes sir...all the factors mentioned are all valid hence prepping the job in whatever program you used to calculate your points need to be the same datum and projection in the field process. Details like your HI and BS heights have to be addressed with extra care until it becomes routine and later procedures to follow as part of your workflow. Start with the basics in tracking your errors and systematically move towards more complex errors that could be generated by human interaction...if you know what I mean.
Soon you will start seeing in your minds eye what and who was directly involved in the final conclusions created by your investigation...hope that helps you as you start attacking your problem
So I would start with making sure your measuring your Hi's to the correct mark on the rod. Then I would see if any adjustments were done for those points directly from caltrans or whomever if you know what I mean when you get a chance get off site and go to a job you already have good elevations and check those...if you still have .30 tenths out ...you will know that its not the points but rather settings in your GPS base setup or maybe your not using the correct geoid model...or ?? Ellipsoidal..?
Try a diferential measuring beginning at an good benchmark, and apply the best geoid model that you can afford, things will be better¡¡¡
You might also be adding to errors by using multiple base stations / reference systems.