Hi, I asked this question earlier. Now I tested more and I am unable to resolve this issue. With CHCNAV i73 I use Landstar7 software. I always get higher height values than R8. I use exact same geoid (universal EGM96) and same CORS corrections. LS7 help says "Best Practice" height fitting method uses Trimble algorithms. But with Best Practice > Bilinear it is of by 30cm. It comes closer using Best Practice > Spline but still 10cm off. I don't see any such settings in Trimble Access.
Sincerely
The naming of those functions is highly suspicious - "Best Practice", "Height Fitting", "Bilinear" all seem to indicate that some sort of transformation or modification is taking place.
When using a geoid, there's no "algorithm" that needs to be calculated, Trimble or otherwise. The software should simply apply the geoid separations to the ellipsoid height at the appropriate position, interpolated from the geoid grid files.
As Rover says, geoid heights are fixed numbers tagged to a Lat, Long. From there it's a simple addition or subtraction from the Ellipsoid Height to the Orthometric Height (Elevation). What you're seeing is a difference in Ellipsoid height in the different systems you're applying the geoid height to, or the programs have different geoid heights and the same ellipsoid heights.
You need to "first principal" your data and figure it out yourself.?ÿ
Either your data has different geoid heights or different ellipsoid heights. It's up to you to find out. Time to dig into the data.?ÿ
Best Practice > Bilinear it is of by 30cm. It comes closer using Best Practice > Spline but still 10cm off.
Please clarify if that is comparing two different observations or the same observation processed two different ways.
If you are getting that much difference with the same observed data processed by two different algorithms, there is either a mistake in the algorithms, or more likely insufficient or inconsistent data in the geoid model it is operating from.?ÿ
The fancier the fitting algorithm is, the more sensitive it is to missing, too widely spaced, or inconsistent geoid data to interpolate from.
I agree that for help you need to be more clear about what information you have and what you are trying to determine.
The title of your post "Different Geoid Heights with CORS Trimble and CHCNAV" implies that the two packages yield different geoid heights for the same point/s. You further state that the geoid heights were determined with respect to the Earth Gravity Model 96 (EGM96).
The geoid heights from EGM96 are determined by specifying WGS84 values: latitude and longitude (and sometimes) ellipsoid height). It has nothing to do with your GPS processing.
I hope that the two software packages include information about how the model data is stored and used. I expect they both use the EGM96 15 Minute Interpolation Grid code downloadable from the NGA site: https://earth-info.nga.mil/index.php?dir=wgs84&action=wgs84
Note that there are on-line tools to compute geoid heights from EGM96 (NGA also provides values wrt EGM08).
NGA tool site:
https://earth-info.nga.mil/index.php?dir=wgs84&action=egm96-geoid-calc
UNAVCO tool site:
Unfortunately, entering the same geodetic coordinates into the two tools will yield different results!
There are a number of issues explaining why this is the case among the better explanations are from here: https://www.ngs.noaa.gov/PUBS_LIB/EGM96_GEOID_PAPER/egm96_geoid_paper.html
Unfortunately, my hope that the tool's "Read Me document" would answer some of my questions were dashed by the fact that the file does not exist on the site.
?ÿ
?ÿ
?ÿ?ÿ
