[CIG-LONG] Faults do not depart away from the axis

[Taichi SATO]

Dear Garrett, Walter

> (1) You should check the rate that the dike zone is widening. If it is
> widening and the fault(s) is on the side of the dike zone then the fault
> should be moving, except if the rate the dike is opening is about half
> the spreading rate (i.e., M=0.5, see Buck et al. 2005 or Behn & Ito
> 2008).
I checked the dike zone is widening(M is not 0.5). 
So I think the fault should be depart away.

> As a minimal hack I made the initialStrainShape be a region where the
> accumulated strain is always zero.  It would be easy to have a maximum
> strain in that region as in the Buck Nature paper.  I am attaching a
> new version of Underworld/Rheology/src/StrainWeakening.c and a
> modified version of the input file I sent you earlier.  Right now the
> zero strain region is the same as the dike region.  You may want to
> make it larger.

I reinstalled gale after replacing with the new 
StrainWeakening.c. I made sure that the 
accumulated strain in the dike zone is always zero .
However, I also need to set initial damage to 
weaken one side of the dike for the initiation of 
one side fault. I think it is difficult to do this 
by a new StrainWeakening.

Sincerely,

Taichi