[CIG-SEISMO] [Specfem3D Cartesian]- Using a CMTSOLUTION source to trigger a critically-stressed fault

Tue Apr 10 07:39:57 PDT 2018

Dear Ge,

You may find in the following paper some insight on the conditions  
required to trigger an earthquake in your simulations:

M. Galis, J. P. Ampuero, P. M. Mai and F. Cappa (2017)
Induced seismicity provides insight into why earthquake ruptures stop
Science Advances, 3 (12), eaap7528, doi:10.1126/sciadv.aap7528
http://dx.doi.org/10.1126/sciadv.aap7528

Static triggering can be directly addressed with the analytical theory  
developed therein. Dynamic triggering maybe not, but at least the  
result can serve as reference.

Best,

JP Ampuero

Quoting Ge Li <ge.li2 at mail.mcgill.ca>:

> Thank you, Surendra!
>
> I figured out how to deal with it after talking to Kangchen.
>
> I put a Mw 6.2 Right-lateral EQ 1km away from a dynamic fault.
>
> This scenario results in a dynamic stress changes of ~0.5MPa
>
> The results look reasonable to me.
>
> Best,
>
> Ge
>
> From: Surendra Nadh Somala <surendra at iith.ac.in>
> Date: Tuesday, April 10, 2018 at 3:01 AM
> To: Ge Li <ge.li2 at mail.mcgill.ca>
> Cc: Kangchen Bai <kbai at caltech.edu>, Dimitri Komatitsch  
> <komatitsch at lma.cnrs-mrs.fr>, <cig-seismo at geodynamics.org>,  
> <ampuero at geoazur.unice.fr>
> Subject: Re: [CIG-SEISMO] [Specfem3D Cartesian]- Using a CMTSOLUTION  
> source to trigger a critically-stressed fault
>
>
>
> Dear Ge,
>
>
>
> Currently, the implementation is such that all the fault in a  
> particular simulation should be either Kinematic or Dynamic.
>
> Nevertheless, you can run a kinematic simulation, observe the stress  
> changes on the other fault and setup a dynamic rupture simulation  
> based on the updated stresses.
>
>
>
> Btw, what is the magnitude of the point source (CMTSOLUTION) you  
> have put in your simulation?
>
>
>
> Best,
>
> Surendra
>
>
>
>
> ___________
>
> Surendra Nadh Somala
>
> Assistant Professor
> Department of Civil Engineering (Block-E, 208)
> Indian Institute of Technology (IIT) Hyderabad
>
> Kandi, Sangareddy
> Telangana, India - 502285
>
> Phone : +91 (0)40 2301-8457
>
> http://civil.iith.ac.in/surendra/
>
>
>
> On Tue, Apr 10, 2018 at 1:08 AM, Ge Li <ge.li2 at mail.mcgill.ca> wrote:
>
> Hi Kangchen and Dimitri,
>
> Thanks for your reply!
>
> Kangchen, can you give me more explanations on ‘fault-node splitting’?
>
> Do you mean the split-node technique used during the mesh process? I  
> followed the
>
> instructions provided in the specfem3d manual to create a Cubit mesh  
> file. The
>
> mesh works well for a ‘step-over’ scenario and I did observe the   
> dynamic stress transfer.
>
> Back to the test mentioned in my previous email, where I put a point source
>
> (Mw~6 Right-lateral strike slip) about only 1km away from the fault  
> surface. I also observed
>
> stress changes, but with a magnitude of ~  0.0682Pa. Considering the  
> initial stress level of
>
> 82.1MPa, I suspected these stress changes are numerical?
>
>
>
> Also is it possible to implement both dynamic and kinematic fault in  
> a model, i.e. one fault is
>
> Kinematic and the other one is dynamic.
>
>
>
> Thanks !
>
> Best regards,
>
> Ge
>
>
>
>
>
> From: Kangchen Bai <kbai at caltech.edu>
> Date: Monday, April 9, 2018 at 1:12 PM
> To: Dimitri Komatitsch <komatitsch at lma.cnrs-mrs.fr>
> Cc: <cig-seismo at geodynamics.org>, Ge Li <ge.li2 at mail.mcgill.ca>,  
> Surendra Nadh Somala <surendra at iith.ac.in>, <ampuero at geoazur.unice.fr>
> Subject: Re: [CIG-SEISMO] [Specfem3D Cartesian]- Using a CMTSOLUTION  
> source to trigger a critically-stressed fault
>
>
>
> Hi Ge,
>
>
>
> Triggering a dynamic rupture with a point source is do-able with the  
> code. You should see dynamic stress changes as the wave goes across  
> the fault.
>
>  I actually did similar things before by mistake --- I forgot to  
> turn off the point source when doing a dynamic rupture simulation  
> and see the circular stress pattern on the fault.
>
>
>
> We also did the work of one fault rupture triggering another using  
> the code (See Bai & Ampuero 2017
>
> https://doi.org/10.1002/2017JB014848
>
>  ), the stress transfer is not a problem as is shown in Fig 6 & 7.
>
>
>
> If you do not see stress changes, there must be something wrong with  
> the settings.
>
> I suspect the most plausible scenario is that the fault node is not  
> split in which case the stress remain constant anyhow.
>
> Let me know if you have more questions.
>
>
>
> Best,
>
> Kangchen
>
>
>
> On Mon, Apr 9, 2018 at 7:21 AM, Dimitri Komatitsch  
> <komatitsch at lma.cnrs-mrs.fr> wrote:
>
>
> Dear Ge Li,
>
> Thank you very much for your email and for your interest in the work  
> of the community around the package.
>
> Yes, if you want to mimic an extended source using a kinematic  
> source model, you can use many CMTSOLUTION sources that are  
> triggered at different times and at slightly different locations  
> (see e.g. the discussion about extended sources around equation  
> (A12) of  
> http://komatitsch.free.fr/preprints/GJI_1999_figure16_error_fixed_equation3_typo_fixed.pdf, and the way we mimic a Hamming quasi-plane wave using 1000 such sources in http://komatitsch.free.fr/preprints/Ultrasonics_Moysan_2016.pdf figure  
> 5).
>
> If you want a dynamic rupture source model, i.e. something more  
> sophisticated, the package can do that based on the great work of  
> Jean-Paul Ampuero, Surendra Nadh Somala, Kangchen Bai and other  
> people in the community, they can give you more details (they will  
> know much more about this than I do :-)
> Let me thus cc them.
>
> Best regards,
> Dimitri.
>
> On 04/05/2018 03:00 AM, Ge Li wrote:
>
> To Whom It May Concern:
>
> I’m a PhD Candidate in geophysics at McGill University. Currently I’m using
>
> Specfem3D Cartesian for my research.
>
> I’m wondering is it possible to trigger an earthquake on a fault  
> using a CMTSOLUTION source?
>
> I tried several simulations by setting a critically stressed fault  
> and a CMTSOLUTION point source.
>
> The source (Mw~6 Right-lateral strike slip) was located about only  
> 1km away from the fault surface.
>
> But it turned out the stresses on the fault surface(dynamic  
> slip-weakening) remain unchanged.
>
> Based on these tests, I supposed that a CMTSOLUTION won’t change the  
> stress level on a fault surface. Is it correct?
>
> Or did I miss something here?
>
> Thanks! I’m looking forward to your reply!
>
> --
>
> *Ge Li*
>
> /Ph.D. Candidate/
>
> /Department of Earth & Planetary Science, /
>
> /McGill University/
>
> /3450 University Street/
>
> /Montreal, QC, Canada /
>
> /H3A 0E8/
>
> /ge.li2 at mail.mcgill.ca <mailto:ge.li2 at mail.mcgill.ca>/
>
>
>
> _______________________________________________
> CIG-SEISMO mailing list
> CIG-SEISMO at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-seismo

-- 
Jean-Paul Ampuero
Directeur de Recherche
Géoazur, Institut de Recherche pour le Développement / Université Côte d'Azur
http://geoazur.oca.eu
@DocTerremoto