[CIG-SHORT] slip orientation for a nonplanar fault surface

[Brad Aagaard]

Jiangzhi,

All fault slip and tractions are specified (and output) in the local 
fault coordinate system (left-lateral, reverse, and opening). If you 
know the slip in a global XYZ coordinate frame, you will need to compute 
the fault slip in the local coordinate system. If you have the fault 
slip in the fault slip coordinate system (left-lateral, reverse, 
opening) no conversion is necessary.

If you need to convert from a global XYZ coordinate system, then I 
suggest running a simulation that only initializes

[pylithapp]
initialize_only = True

and request output of the fault orientation information

[pylithapp.timedependent.interfaces.fault.output]
vertex_info_fields = [strike_dir,dip_dir,normal_dir]

so that PyLith will give you the local fault coordinate orientation to 
use in the conversion.

Regards,
Brad



On 10/31/2014 02:11 PM, Jiangzhi Chen wrote:
> Hi Brad,
>
>      Just one small question. When specifying the fault parameters in
> terms of lateral motion and reverse motion for a 3d nonplanar fault, are
> the motions evaluated along the fault surface locally? To prescribe
> left-lateral-slip and reverse-slip for a specific point on the surface,
> I still need to calculate the strike, dip and rake angle at that point,
> is that correct?
>
> Jiangzhi
>
> On 10/20/2014 03:57 PM, Brad Aagaard wrote:
>> On 10/20/2014 03:55 PM, Jiangzhi Chen wrote:
>>> Oh I understand what you mean, since the strain is not finite, the
>>> deformation should not change the slip direction great enough, so the
>>> prescribed slip should work.
>>
>> Correct.
>>
>> Regards,
>> Brad
>>
>>> On 10/20/2014 03:50 PM, Brad Aagaard wrote:
>>>> Jiangzhi,
>>>>
>>>> I don't understand why you think prescribed slip won't work for a
>>>> kinematic fault interface?
>>>>
>>>> Regards,
>>>> Brad
>>>>
>>>>
>>>> On 10/20/2014 03:44 PM, Jiangzhi Chen wrote:
>>>>> The time scale is several hundreds of years, about one earthquake
>>>>> cycle.
>>>>>
>>>>> On 10/20/2014 03:34 PM, Brad Aagaard wrote:
>>>>>> Jiangzhi,
>>>>>>
>>>>>> What time scale are you modeling? PyLith should be appropriate if the
>>>>>> problem can be solved with infinitesimal or small strains.
>>>>>> Additionally, the fault formulation in PyLith breaks down when the
>>>>>> offset become large enough so that the slip vector would need to be
>>>>>> changing. We expect that a finite-strain formulation would be needed
>>>>>> as well for those types of problems.
>>>>>>
>>>>>> Regards,
>>>>>> Brad
>>>>>>
>>>>>> On 10/20/2014 03:24 PM, Jiangzhi Chen wrote:
>>>>>>> Thanks for the quick response. The region is part of the Cascadia
>>>>>>> subduction zone, where the Juan de Fuca plate moves northeastward
>>>>>>> and
>>>>>>> subducts. The aim is to understand how the subduction influences the
>>>>>>> surface displacement field, and explains the geodetic
>>>>>>> observation. The
>>>>>>> upper slab interface is supposed to be a dynamic fault surface, and
>>>>>>> the
>>>>>>> lower interface is a kinematic surface. For now, I am using two
>>>>>>> kinematic surface just to make the slab go under.
>>>>>>>
>>>>>>> Jiangzhi
>>>>>>>
>>>>>>> On 10/20/2014 03:08 PM, Brad Aagaard wrote:
>>>>>>>> Jiangzhi,
>>>>>>>>
>>>>>>>> I don't understand what you are trying to do. What research
>>>>>>>> question
>>>>>>>> are you trying to answer? What do you want to learn from the
>>>>>>>> simulation?
>>>>>>>> What are your observations/constraints?
>>>>>>>>
>>>>>>>> Regards,
>>>>>>>> Brad
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> On 10/20/2014 03:02 PM, Jiangzhi Chen wrote:
>>>>>>>>> Hi everyone,
>>>>>>>>>
>>>>>>>>>      I am working on a subduction simulation, where the fault
>>>>>>>>> interface
>>>>>>>>> is nonplanar, and the subduction slab moves in northeast direction
>>>>>>>>> at 4
>>>>>>>>> cm/year before it meets the overriding plate. My question is how
>>>>>>>>> I can
>>>>>>>>> specify the kinematic fault interface. It is not appropriate to
>>>>>>>>> use
>>>>>>>>> predefined slip functions, because the slip rates depend on the
>>>>>>>>> location
>>>>>>>>> which is changing with time. Is there a way to make this work? Any
>>>>>>>>> help
>>>>>>>>> is appreciated.
>>>>>>>>>
>>>>>>>>> cheers,
>>>>>>>>> Jiangzhi
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> CIG-SHORT mailing list
>>>>>>>>> CIG-SHORT at geodynamics.org
>>>>>>>>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> CIG-SHORT mailing list
>>>>>>>> CIG-SHORT at geodynamics.org
>>>>>>>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> CIG-SHORT mailing list
>>>>>>> CIG-SHORT at geodynamics.org
>>>>>>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>>>>
>>>>>> _______________________________________________
>>>>>> CIG-SHORT mailing list
>>>>>> CIG-SHORT at geodynamics.org
>>>>>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>>>
>>>>> _______________________________________________
>>>>> CIG-SHORT mailing list
>>>>> CIG-SHORT at geodynamics.org
>>>>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>>
>>>> _______________________________________________
>>>> CIG-SHORT mailing list
>>>> CIG-SHORT at geodynamics.org
>>>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>
>>> _______________________________________________
>>> CIG-SHORT mailing list
>>> CIG-SHORT at geodynamics.org
>>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>
>> _______________________________________________
>> CIG-SHORT mailing list
>> CIG-SHORT at geodynamics.org
>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short