Open MarkWieczorek opened 8 months ago
The geocentric_radius
is justified since it's calculated differently when using spherical coordinates. The normal gravity is not and passing the keyword will mean running the coordinates conversion under the hood.
I'm OK with this if it makes life easier for users.
What if we noted that its preferable to input geodetic coordinates (which is the default), because the normal gravity routine uses a formalism that requires geodetic coordinates? If this is done repeatedly in a loop, then it wouldn't be very efficient to do the conversion over and over, but if its just a single call, then the computational load would be the same.
I only bring this up because in planetary geophysics, its really rare to use geodetic coordinates, so it would be a convenience for us. :)
What if we noted that its preferable to input geodetic coordinates (which is the default), because the normal gravity routine uses a formalism that requires geodetic coordinates?
That sounds like a good idea! We can add it to the description of the geodetic
argument.
I only bring this up because in planetary geophysics, its really rare to use geodetic coordinates, so it would be a convenience for us. :)
I've been reading more planetary papers and I noticed 🙂
The
Ellipsoid.normal_gravity()
method takes as input the geodetic latitude. I propose to add an optional argumentgeodetic=True/False
that would allow the user to specify if the input latitudes are geodetic or geo/planetocentric. The default option would be to use geodetic. Though this could easily be accomplished using thespherical_to_geodetic
method, this would make things easier for many applications where the geocentric latitude is the more appropriate value to use.I note that that option is already implemented in the method
Ellipsoid.geocentric_radius()