andreww
commented
8 years ago Thanks - looks good. I've added a handful of in-line comments. If you could change things and push to your branch it will show up here.
Also, any chance of a test case and do we need to update the main documentation?
alanfbaird
I've updated MS_Phasevels to add optional outputs (P, S1 and S2 eigenvectors, as well as phase propagation vector) and added MS_groupvels.
MS_groupvels outputs the group velocity in vector format corresponding to a given phase angle (Note that these are generally not parallel to each other). It also optionally outputs phase polarizations (eigenvectors) and the slowness vectors.
A potential issue with this is that S1 and S2 are currently sorted by phase velocity magnitudes, but since one phase vector corresponds to separate group vectors for P, S1 and S2, this doesn't translate to group velocities very well (see the attached figure for an example: blue is "S1" and red is "S2"). They should probably be sorted by which S velocity sheet they belong to. This is simple enough to do for VTI media (by sorting based on S polarisation), but I'm not sure how to do this for the general case.