Notifications
Clear all

EGM96 height difference

11 Posts
7 Users
4 Likes
263 Views
mov_hp
(@mov_hp)
Posts: 17
Member Member
Topic starter
 

I asked this question last year and now I was able to test this further. Equipment used is: Trimble R8 and Access, CHCNAV i90 and Landstar 7, Leica FLX100 and Zeno Mobile, Leica FLX100 and TcpGPS.

Interesting findings:

  1. The control points are measured with R8 and Access. Global EGM96 geoid file from Trimble was used.
  2. CHCNAV i90 and Landstar 7 get 25cm higher elevations at this control point using the same geoid file. Landstar 7 can use Trimble GGF files.
  3. FLX100 and Zeno mobile get 15cm higher elevations using a 500m grid geoid and 40cm higher elevations using 1000m grid geoid. Geoid is EGM96.
  4. FLX100 with TcpGPS heights match with CHCNAV i90 and Landstar 7. TcpGPS has its own version of EGM96 which is installed with the software on Android.

Is the EGM96 reliable? Why does it report such differences with different software and hardware?

Sincerely

 

 
Posted : November 9, 2022 10:27 pm
OleManRiver
(@olemanriver)
Posts: 741
Member Member
 

EGM96 is reliable but it is for use with wgs84 realization. Just as geoid03 vs geoid18. Egm96 is not as dense for every day surveying like the hybrid geod models in the states. I believe they have an updated egm model that is to be used with the latest realization of wgs84. Etc.  make sure the horizontal datum is correct realization for use with each geoid model.  

 
Posted : November 10, 2022 4:20 am
Rover83
(@rover83)
Posts: 1601
Member Member
 

Don't compare geoid-derived orthometric elevations until you have confirmed that all units are measuring the same ellipsoid height.

 
Posted : November 10, 2022 5:41 am
toeknee and Bill93 reacted

Get the 2023 SurveyorConnect Wall Calendar

StLSurveyor
(@stlsurveyor)
Posts: 2461
Member Supporter
 

Also be mindful, EGM96 is outdated, available in three different densitiy flavors, Trimble GGF files have user options to make grid spacing an option, WGS 84 (realization) vs. ITRF, ellipsoid parameters...

All of these are handled diffently by different software, manufactors, etc. 

Be sure you know what you don't know. 

 
Posted : November 10, 2022 6:20 am
OleManRiver reacted
MightyMoe
(@mightymoe)
Posts: 8858
Member Supporter
 

Which number is "more" correct to orthometric heights?

 
Posted : November 10, 2022 7:58 am
Bill93
(@bill93)
Posts: 9329
Member Member
 

Posted by: @rover83

Don't compare geoid-derived orthometric elevations until you have confirmed that all units are measuring the same ellipsoid height.

This!

Differences could be in misunderstanding the antenna reference points, or just random error in the solution due to propagation or multipath effects.  Those all affect the ellipsoid height result.

Or it could be differences in the geoid models an software selections. Those affect the difference between ellipsoid height and orthometric estimate.

 

 
Posted : November 10, 2022 8:30 am
toeknee reacted

Get the 2023 SurveyorConnect Wall Calendar

GeeOddMike
(@geeoddmike)
Posts: 1470
Member Member
 

My reading of the OP was that mov_hp was questioning why different software packages would yield different results using the same EGM.

My first thought was that the differences could be the result of different gridding and interpolation choices. 

As I do not have any of your tools, I used the UNAVCO tool specifying the EGM96-15 and EGM96-5 models. Both yield the same geoid height. Both computations used cubic interpolation. No change was noted when unselecting cubic interpolation. 

UNAVCO Geoid Height calculator tool URL:

https://www.unavco.org/software/geodetic-utilities/geoid-height-calculator/geoid-height-calculator.html

As other responses have indicated there is a newer EGM (EGM2008) that should be used (all things being equal). 

Calculator results (for N25.25, E055.25, 3m) were:

EGM96-15 and EGM96-5 = -33.87

EGM2008-5 = -34.16

Pretty big difference IMO.

BTW, there is an overdue update to the NGA EGM in work. I believe it will be named EGM2020.

I tried to find the NGA tool for geoid height computation but had no luck.

On the general issue of EGM’s v Geoid models, here is a screen capture from Dr Yan Ming Wang’s webinar presentation Q&A note:

 

 

 
Posted : November 10, 2022 9:03 am
GeeOddMike
(@geeoddmike)
Posts: 1470
Member Member
 

Some information about NGA EGM products:

https://www.usna.edu/Users/oceano/pguth/md_help/html/egm96.htm

 
Posted : November 10, 2022 9:18 am
StLSurveyor
(@stlsurveyor)
Posts: 2461
Member Supporter
 

@geeoddmike 

here you go

https://earth-info.nga.mil/index.php?dir=wgs84&action=egm96-geoid-calc

 

 
Posted : November 10, 2022 9:24 am

Get the 2023 SurveyorConnect Wall Calendar

GeeOddMike
(@geeoddmike)
Posts: 1470
Member Member
 

@stlsurveyor   Thanks for the link. BTW, the Read Me Document from this URL does NOT work.

Running the NGA tool for the same point used at the UNAVCO site yields large differences.

NGA values.                             UNAVCO.                   EGM

-33.12.                                     -33.87                    EGM96

-33.89.                                     -34.16.                   EGM08

UNAVCO site output sample:

 

 
Posted : November 10, 2022 9:47 am
OleManRiver
(@olemanriver)
Posts: 741
Member Member
 

I believe an i am digging back several years now trying to remember but when EGM08 was being tested they knew that some differences were presented that they were aware of in areas. Also that the wgs84 realization and EGMxx needed to be used together. I will be on a tractor and what I am trying to get out of my brains filing cabinet will emerge but right now I am lost in some storage room. This has my gerbils twitching so I know i have heard the why before. 

As for the OP. How do the ellipsoid heights compare with each on same horizontal datum. That might point to the answer. And check your rod settings for arp or apc.  

 
Posted : November 10, 2022 4:25 pm