RTK elevation values (local base VS network base) 

Professional Surveyor Boards » Ask an Expert » RTK elevation values (local base VS network base)

Name Message
Thomas LaCorte
Posts: 62

Joined: 10/17/2008
Topic  RTK elevation values (local base VS network base)       Flag »  Reply »
I as well as others are getting better vertical elevations using a local base then when using a network base. After two years of using RTK this seems to always be the case. Not just a once in awhile thing. My assumption is that my local base within 1 to 2 miles of the rover is under the same atmospheric influence as my rover as opposed to a network base being 20 some miles away from it's rover and perhaps not being under the same atmospheric influence and therefor perhaps not giving corrections as well. This theory is beyond my expertise,can someone maybe shed some light on whether this is a viable theory? One thing I know for sure is:Local base = better vertical datum than network base (for myself and others).  
  Tuesday, November 11, 2008 at 8:09:20 PM
BlackWatch
Posts: 14
Location: Orlando, FL USA

Joined: 10/17/2008
Reply  Re: RTK elevation values (local base VS network base) Flag »  Reply »
GPS works by calculating the distance between the satellite and the receiver and your receiver knowing where the satellite is located (or should be).  It takes three satellites to produce a horizontal location estimate and four to get an elevation.  Anything that causes the location of the satellite to be different than expected or for the signal to take longer to reach the receiver will induce error.  You can use realtime differential GPS by receiving a signal from a Continuously Operating Reference Station (CORS) and applying the adjustments required to put the CORS location in the precise position to the location determined by your receiver.  You can also do this after the fact through post processing of the CORS information.  You can instead use your own base station for the correction factors in RTK surveying.  A typical $4,500 differential GPS receiver can routinely compile data with an accuracy of less than a meter in relatively open spaces.  The National Geodetic Survey has online programs to drop the error to a few centimeters with long observations (15 to 60 minutes) and post processing.
    You are correct that ionospheric and tropospheric interference is a primary cause of error in GPS measurements due to varying rates of slowing of the radio signal as it passes through atmospheric layers.  In order to reduce this error, the GPS receiver includes a standard model for compensation; however, what works at the CORS location may not be the same as what works at your distant observation site.  To the extent that atmospheric conditions do differ at the two locations, you will perceive a larger error when using broadcast differential correction compared to your own base station adjustments, which should be close enough to the rover to experience similar conditions.
    Another source of differences in effectiveness between the two correction methods is called ephemeris error, which is due to the perturbations of satellite paths as they move in orbit over different parts of the Earth.  The satellite is actually moving up and down as it orbits in response to small variations in the gravitational field.  This can result in the satellite being at different heights above the Earth when it passes over the CORS base station compared to when it is polled by your rover GPS receiver.
    A third source of error that is method-specific is the difference in satellite locations relative to each other between the CORS location and your own.  Satellite signals can interact if they "shade" each other.  The effect will be different at one location compared to the other.  Just having different satellites overhead can also induce a small error.
    Error sources that may be felt only at the rover location, assuming you put the base station in a good location, are multipath distortions imposed by adjacent objects that cause the signal to travel a greater distance from the satellite to the GPS receiver, poor satellite locations for the required triangulation process, and clock timing errors in the receiver.
  Thursday, November 13, 2008 at 1:00:11 PM
Thomas LaCorte
Posts: 62

Joined: 10/17/2008
Reply  Re: RTK elevation values (local base VS network base) Flag »  Reply »
Thank you for taking the time to respond to my theory. It seems we are in agreeance. I have a 38,000 dollar system the hiper-lite plus, base and rover with GPS and GLONASS. 40 channels with the 2.5 mile spectrum radio. I use the RECON with bluetooth and TDS software. The results are almost unbelievable to me. I shot between two bench marks 2 miles apart with 0.006' error. (I had to change my shorts!) Typically my horizontal positioning is .02 to.05 and my vertical .02 to.08. I realize though that either could be as much as 0.12 on a different day. I always keep my base away from any obstruction and having an open northern sky seems to help with GLONASS. I am very careful for multi path at the rover. These tools as (noted in the manuals) should be used by a professional. In the hands of an inexperienced person could result in disastrous consequences. I noticed that the least experienced people have the most problems because they don't understand the limitations.
Thanks again....Tom
  Thursday, November 13, 2008 at 6:58:32 PM
BlackWatch
Posts: 14
Location: Orlando, FL USA

Joined: 10/17/2008
Reply  Re: RTK elevation values (local base VS network base) Flag »  Reply »
I think you have hit on one of the big elements of surveying art: being able to distinguish accuracy from the appearance of accuracy.  Knowing how to use high-accuracy GPS equipment requires much more than learning how to press the buttons.  The problem is always what you don't know you don't know.
  Thursday, November 13, 2008 at 8:46:01 PM
Amberftw
Posts: 11

Joined: 1/2/2009
Reply  Re: RTK elevation values (local base VS network base) Flag »  Reply »
I am not an expert on GPS, but I have used trimble GPS with the network base and whenever I measured a point it was correct within the thousandths. The only time I have been working with someone and I got an incorrect reading was because it was set up incorrectly. I love the network base. The one I used gave really good readings and I couldnt imagine having to set up another base for my GPS shots.
  Friday, January 02, 2009 at 11:02:19 PM
Jack Chiles
Posts: 3

Joined: 2/13/2009
Reply  Re: RTK elevation values (local base VS network base) Flag »  Reply »
Amber, if you return to those points on a different day and with at least 4 hours of difference in time of observation, you will observe a tenth of difference in the positions of the point. RTK cannot and will not guarantee you any better than that. You might think it is better than that, but it isn't. Listen to the manufacturers - they will tell you the same thing, that is, a tenth in the horizontal and more than that in the vertical difference. RTK will give you thousanths if you reshoot that point (you know, lose lock, reshoot and check) within a short timeframe. That is because the mathematical soultions for that position are based upon measurements to satellites that have not changed their positions enough to allow for a different solution. RTK is great, but don't ever think it is going to give you a reliable position with less than a tenth of error in precision unless you remeasure it at different times of the day (at least a 4-hour difference) and on different days and then average the results.
  Friday, February 13, 2009 at 3:13:46 PM
Amberftw
Posts: 11

Joined: 1/2/2009
Reply  Re: RTK elevation values (local base VS network base) Flag »  Reply »
I disagree with you Jack. I looked up the monuments and they have a stated elevation that is set when they put the monument in. So couple years later I come back and measure the exact same elevation. The elevations that I compare are not two of my own shots on the same point but the given elevation from the monument website and one I shoot my self.
I dont think I was too clear on my post before... I am not a GPS expert.. I have not examined the postioning of the point, my observations where dealling with just elevations.. not positioning.
But when I get the chance I will experment with the positioning.. just to see if what you say is true.
  Sunday, May 31, 2009 at 12:27:25 PM
BlackWatch
Posts: 14
Location: Orlando, FL USA

Joined: 10/17/2008
Reply  Re: RTK elevation values (local base VS network base) Flag »  Reply »
Amber & Jack, it is entirely possible that you are both right.  The error surface for GPS is not continuous; i.e., the amount of system error is not constant across the earth.  The plus or minus a tenth is a general average, but if you are taking a measurement where the geoid surface matches the real surface, then the error can be much less.  GPS is tied to a gravitational geoid rather than the real surface elevation.  When the two are the same, then you are left to deal only with atmospheric and other errors, but they have less impact.  In your case, Amber, both the network base station and the rover are likely in very favorable locations relative to the geoid.  (That's certain where I would put a base station.)  Measure somewhere else, and you may get a much bigger error.  Measuring some other time at the same location will not result in much of a change, as long as PDP is fairly consistent with various satellite constellations.  (BTW, those satellites are whizzing by.  It doesn't take but a moment for there to be a different solution.)
  Sunday, May 31, 2009 at 1:34:58 PM
Amberftw
Posts: 11

Joined: 1/2/2009
Reply  Re: RTK elevation values (local base VS network base) Flag »  Reply »
Jack excuse me for my earlier post. I had partial information on GPS at the time of those post. Since then I have taken a GPS course at my school. I was not working with RTK GPS I was working with VRS which is .003 accurate for horizontal and .006 accuate for vertical. RTK is working with a base you set up at the site and VRS is with the network bases. My instructor also told me my readings may have been so good because I was close to the equator. But the shot I did take on the monuments where set like 10 years before I measured them with my VRS GPS and got the same exact reading to the tenth so I did have quite a bit of time in between shots.
  Sunday, July 19, 2009 at 9:33:17 PM

Professional Surveyor Boards » Ask an Expert » RTK elevation values (local base VS network base)

 
» Reply to RTK elevation values (local base VS network base) Topic

In order to post a new message, you must first be logged into an account.

Website design and hosting provided by 270net Technologies in Frederick, Maryland.