[CIG-SHORT] Green's functions for fault opening

[Brad Aagaard]

On 07/03/2017 09:38 AM, Kyle Murray wrote:
> Hello,
> I am working on example 7.8.10.2 and 7.8.11 in the manual. My goal is to 
> invert ground surface displacements for dike/fault opening. The Green's 
> function generation example is done with impulses of slip on the fault 
> rather than impulses of fault opening. If I want to make Green's 
> functions for the dike/fault opening example, do I need to change the 
> impulse_dof value in the greensfns.cfg? Is there a way to know what 
> different values signify for impulse_dof (I can't find it in the 
> manual).  Is there anything else I would need to change to make 
> my Green's functions for dike/fault opening?

In 2-D use impulse_dof = 1 for fault opening impulses and in 3-D use 
impulse_dof = 2 for fault opening impulses. You are correct that Section 
6.4.6 on FaultCohesiveImpulses does not mention the fault opening degree 
of freedom. We will fix that.

> Also for the inversion, is it necessary to sort the impulse, response, 
> and data arrays before solving?  In the 2D example (strike-slip fault) 
> it is straightforward to just sort the arrays based on the y-coordinate, 
> but I'm not sure how to deal with 3 dimensions.

In the 2-D example, we sort the impulses simply to make it easier to 
plot the slip profile. You just need to be careful that you maintain 
consistency between the order in which the Green's functions are 
calculated and their corresponding locations (as given by the fault 
vertices).

Regards,
Brad