search

ydluo / qdyn

A Quasi-DYNamic earthquake simulator
59 stars 28 forks source link

Fault type #10

Closed ydluo closed 6 years ago

ydluo commented 6 years ago

Dear co-developer I noticed that the fault type (strike-slip, thrust, etc) is hard-coded as constant (in constants.f90), i.e. each time users of QDYN want to use a different fault type they will have to recompile the code (and it is not properly documented in the manual). Is there a good reason for doing so (e.g. computation efficiency, etc.)? If not we need to fix that as a run-time input.

Yingdi

jpampuero commented 6 years ago

I see no reason to leave it as a hard-coded parameter. As noted in my comment in the code: "this should be an input in qdyn.in, not a parameter here". Pablo

martijnende commented 6 years ago

Same for the solver flag (Bulirsch-Stoer vs. Runge-Kutta). I was going to fix this over the next few days, I'll include the FAULT_TYPE as well.

martijnende commented 6 years ago

The user parameters SOLVER and FAULT_TYPE, as well as a dilatancy hardening coefficient (used in thermal pressurisation), are now written to qdyn.in by qdyn.m and/or pyqdyn.py.

jpampuero commented 6 years ago

Thanks for fixing that, Martijn. Can you document these new input variables in the user's manual?

martijnende commented 6 years ago

Done, and added CNS model parameters