I am searching for a simplified model of EGM2008 that can be used for rapid, repetitive computations. It needs to be valid only up to GPS altitudes. I'm thinking such a model already exists but cannot find it and would prefer not to reinvent the wheel. I'm aware of the math, so I could make such a model if I had to do so but would prefer to use a community model. Suggestions?

  • 3
    $\begingroup$ interesting question! The farther you get from Earth, the more accurate any truncated multipole model will be, so it's more a question of what level of accuracy you need at the lowest altitude you'll use, not the highest. So for example, it might be 1 part per million at 400 km above a reference sphere of 6378 km (just an example). If you'll only be working in the MEO neighborhood of GPS then all you'd need to say is roughly how accurate. Also you might state which software tools are you using so someone doesn't write a lengthy answer for package X and then find out you use Y. $\endgroup$ – uhoh Sep 27 '18 at 15:48
  • 1
    $\begingroup$ Do you have the ability to just truncate EGM2008 to a lower number of coefficients, i.e. lower degree and order? If you are using a package, there might even be an option to do this automatically. $\endgroup$ – uhoh Sep 27 '18 at 15:56
  • 1
    $\begingroup$ I truncated the SGG-UGM-1 model: xb.sinomaps.com/EN/10.11947/j.AGCS.2018.20170269 to degree and order 15 to be used with sourceforge.net/projects/geographiclib. I can send it over. $\endgroup$ – Cristiano Sep 27 '18 at 16:24
  • $\begingroup$ Yes, I looked at geographiclib and thought about making a wrapper for it. However, for my immediate purposes geographiclib as is would be overkill. (Nevertheless, I have to admire the breadth of geographiclib.) As for the question of SW platform, for now I'm working in MATLAB and will probably duplicate it in Python when I get to that stage. Currently, my shop is in the shift from MATLAB ($) to Python (free and more widely used); the world has changed a lot since we were a Fortran shop decades ago. So, I'm open to suggestions. At the moment I'm doing lots of reading on gravity models. $\endgroup$ – Christopher Ison Sep 27 '18 at 20:17
  • $\begingroup$ I'm not sure of exactly what you're looking for.... Are you looking for code that implements the spherical harmonics? Are you looking for an analytic approximation of a low degree/order field? Or are you simply just looking for a truncated set of coefficients, as @uhoh asked? There are options for all of these. $\endgroup$ – DuffBeerBaron Oct 3 '18 at 14:06

We recently developed a web API that provides access to the EGM2008 model, as implemented by the open-source GeographicLib Gravity library.

Geoid undulation values calculated using the API agree with corresponding values generated by the official EGM2008 harmonic synthesis program to within 1 mm. The API is open and available here in case that satisfies your needs.

  • $\begingroup$ Cool! fyi How can I verify my reconstructed gravity field of Ceres from spherical harmonics? still needs a better answer. Also I'm curious how does one calculate a gravitational potential "to within 1 mm"? $\endgroup$ – uhoh Jul 16 '20 at 8:19
  • 1
    $\begingroup$ Thanks! The "to within 1mm" refers to the calculated geoid undulation values; i.e., the undulation values calculated by the GeographicLib Gravity library and the EGM2008 harmonic synthesis program agree to that level of precision. It might be worth contacting the author of GeographicLib regarding the other question. $\endgroup$ – Team Amentum Jul 16 '20 at 21:55
  • $\begingroup$ Okay got it; thanks × 2! $\endgroup$ – uhoh Jul 17 '20 at 0:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.