find_moho fails for fine-grained velocity models

[yetinam]

Hi @dttrugman , I've run into an issue with the find_moho function defined here: https://github.com/dttrugman/GrowClust3D.jl/blob/47743b8a4f1cbb7d792e039eb36350dd583b2711/src/seismotrace.jl#L62-L79 The problem occurs when inputting a fine-grained velocity model that does not have any jumps larger than dalphaM. In this case the function calling find_moho in traveltime.jl will crash. As the parameter dalphaM is not accessible here, this crash can not be avoided with configuration parameters.

Potential solutions that come to my mind are:

• Instead of checking for the size of the jump, detect the moho at the point were a certain velocity gradient is exceeded. Alternatively, one could also combine a gradient condition with an absolute jump condition to be robust to both coarse and fine-grained velocity models.
• Iteratively decrease dalphaM until a moho is found.
• Catch the cases were no moho is detected. From what I can tell, it's only used for an optional filtering of refracted phases that could be skipped.
• Make dalphaM configurable when calling make_trace1D_tables so that users can manually overwrite it if necessary.

[dttrugman]

Thanks for pointing this out. Should be fixed now, using a combination of bullet points (1) and (3) above.

Daniel

[yetinam]

Thanks for the quick fix!