[CIG-SHORT] Frictional behavior and imposed episodic slip on a fault

[Brad Aagaard]

Romain,

Matt makes a great point. The new prescribed traction implementation for 
spontaneous rupture allows complex traction time histories to be 
imposed, including using a user-prescribed time history (not just a step 
function). This is definitely the most promising approach and could very 
well produce the behavior you want.

Regards,
Brad


On 08/23/2012 08:40 AM, Matthew Knepley wrote:
> On Thu, Aug 23, 2012 at 10:12 AM, Brad Aagaard <baagaard at usgs.gov> wrote:
>
>> Romain,
>>
>> The current version of PyLith does not support mixing prescribed slip
>> and spontaneous slip on the same fault. I do not know of a workaround
>> for the issue of the single cell impeding slip between the two faults.
>> Oliver Boyd asked for a mixed fault implementation a few years ago and I
>> have some ideas on how it might be possible to extend the current
>> formulation to handle this. I have not tried to flush out the details to
>> see how practical it is because other features have higher priorities.
>>
>> In the current version of PyLith, one possible solution would be to use
>> a single fault controlled by spontaneous rupture (FaultCohesiveDyn) but
>> setup the BC to facilitate episodic slip on the desired portion of the
>> fault. This could be done via spatial variation in the friction model
>> parameters or Dirichlet BC in the vicinity of the fault. The advantage
>> of this approach is that the physics at the intersection between the
>> portion with episodic slip and other portion of the fault will be
>> consistent.
>
>
> You could also think about having prescribed fault tractions generate the
> slip.
>
>     Matt
>
>
>>
>> Brad
>>
>>
>> On 08/22/2012 07:31 PM, Romain Jolivet wrote:
>>> Hello Pylith users and developers,
>>>
>>> Is there a zero slip condition at the tip of a fault (FaultCohesiveKin
>> object) and if yes, is there a way to remove this constraint?
>>>
>>> I explain my problem a little bit more. Here is a little drawing:
>>>
>> #-------------------------------------------------------------------------------------------------#
>>> |
>>                                                              |
>>> |                                                                 Medium
>> 1                                                |
>>> |
>>                                                              |
>>> |
>>                                                              |
>>> |                Fault 1
>>    Fault 2                                       |
>>>
>> #--------------------------------#----------------------------------------------------------------#
>>> |         imposed slip
>>   FaultCohesiveDyn                          |
>>> | (1 earthquake/100 yrs)
>>                                           |
>>> |
>>                                                               |
>>> |
>>   Medium 2                                                |
>>> |
>>                                                               |
>>>
>> #-------------------------------------------------------------------------------------------------#
>>>
>>> I would like to impose slip, episodically, on fault 1 and observe the
>> behavior of fault 2, that has a frictional rheology. I apply dirichlet
>> boundary conditions on the edges of medium 1 (towards the left) and medium
>> 2 (towards the right), so that frictional creep occurs along Fault 2
>> between slip episodes.
>>>
>>> The problem right now is that the element at the junction between the
>> two faults (the one that belongs to fault 2) does not slip. Consequently,
>> creep does not propagate correctly in fault 2 after each event. Then there
>> is a huge stress accumulation at the junction between the two faults (the
>> problem might diverge at some point). One solution would be to have one
>> fault (FaultCohesiveDyn) and to impose slip on part of its elements. Is it
>> possible to impose episodes of slip on elements of a FaultCohesiveDyn
>> object?
>>>
>>> I would like to know if this is possible to do with the latest PyLith
>> version, if I need to implement my own fault module (Would it be
>> easy/complicated/hard/a nightmare/impossible?) or if I should change the
>> design of my problem?
>>>
>>> Thanks,
>>> Romain
>>>
>>> Romain Jolivet
>>> Postdoctoral Scholar
>>> Geological and Planetary Sciences
>>> California Institute of Technology, Pasadena, CA
>>> rjolivet at caltech.edu
>>> +1 (626) 560 6356
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> CIG-SHORT mailing list
>>> CIG-SHORT at geodynamics.org
>>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>
>>
>> _______________________________________________
>> CIG-SHORT mailing list
>> CIG-SHORT at geodynamics.org
>> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>
>
>
>