ioshchepkov / pygeoid Goto Github PK

I'm starting to think about how to integrate this into pyshtools, and I have a couple minor comments:

1. In LevelEllipsoid could you give the option to input geocentric latitude for the various functions like normal_gravity()? The way I would like to use this would be to pass a vector of geocentric latitudes that corresponds to one the shtools grids, and output vectors of the normal gravity and geocentric radius. You probably have the conversion functions written somewhere, but it would be a bit more convenient to specify something like latitude='geocentric'.

2. It is not clear which function gives the geocentric radius to a specific latitude: Is this circle_radius()?, or is that something else, like the distance to the rotation axis?

3. The pyproj package doesn't define ellipsoids for the other planets. One way to do so would be to create a new pyproj ellipsoid using pyproj.pj_ellps.update, the other way would be to pass the required keyword args directly to Ellipsoid. Could you provide minimal documentation as to what keyword args are necessary for Ellipsoid to function correctly? I am guessing a, and then either rf or b.

4. Similar remark for LevelEllipsoid: To use an ellipsoid that is not defined, it is necessary to either update LEVEL_ELLIPSOIDS, or input all the necessary keyword args. Could you add minimal documentation as to what arguments are necessary for this to work? I'm guessing here you only need gm, omega, a, and then either rf or b.

5. Would you be interested in adding both ellipsoids and level ellipsoids for the other planets? If not, could you perhaps create some kind of method that allows you to update both pyproj.pj_ellps.update and LEVEL_ELLIPSOID externaly? I'm not sure what the best way to do this is. I'm not sure if anyone has defined such an ellipsoid for Venus and Mercury given that they rotate so slowly. For the Moon it is more complicated, because a level ellipsoid should also include the c22 tidal potential. For Mars, I know that there is a paper by Ardalan et al. (2010). I'd be happy to help out with finding the numbers later.

This is useful for planar approximations and formulas in geoid computation.

• free air anomaly

• simple Bouguer anomaly

• spherical Bouguer anomaly

• complete (refined) Bouguer anomaly (planar and spherical)

• other...

• Hirt C, Featherstone WE, Claessens SJ (2011) On the accurate numerical evaluation of geodetic convolution integrals. J Geod 85:519–538. https://doi.org/10.1007/s00190-011-0451-5

• Hirt C (2011) Mean kernels to improve gravimetric geoid determination based on modified Stokes’s integration. Computers & Geosciences 37:1836–1842. https://doi.org/10.1016/j.cageo.2011.01.005

Support initializing pygeoid.reduction.normal.LevelEllipsoid from (GM, J2, W0, omega).