[CIG-SEISMO] SPECFEM2D code became unstable and blew up

[Dimitri Komatitsch]

Hi Parvathy,

Hard to tell, but this is either the time step that is a bit too high 
(unlikely, since the CFL you used (0.44) seems fine), or if the 
amplitude of the source you used is very high then the code can 
erroneously think that the run became unstable if amplitudes went above 
1.e+25 (that is what we detect, since we have no way of detecting 
instabilities; instead we detect "very high values" that do not seem to 
make sense).

If so, try reducing the amplitude of your source by 1.e+10 or so to see 
if the problem disappears (since the wave equation is linear, doing so 
is safe).

Another thing you can do is configure the code with ./configure 
--enable-debug to see if the run then detects any bug (the 2D code was 
significantly restructured recently and thus we cannot exclude that; if 
so please let us know and we will fix it).

Thanks,
Dimitri.


On 02/11/2016 09:46 PM, Parvathy Sindhu Nair wrote:
> Hi
> I'm a summer research student working with SPECFEM2D software and while
> running the simulation I'm getting this error
>
> /Warning: failed creating color image, maximum value of amplitude in
> image color is invalid
>   amplitude max =    3.7316137641792927E+025  with threshold at
> 1.0000000000000001E+025
>   Please check your simulation setup...
>   code became unstable and blew up (image_color_data)
>   Error detected, aborting MPI... proc            0
> STOP error, program ended in exit_MPI
> /
> I'm not sure why this is happening. I used source wave frequency of
> 100Hz, with mesh file thickness to be 5m. My CFL stability condition was
> 0.44 and my  histogram of min number of points per S wavelength in solid
> regions and P wavelength in fluid region were in the normal region. But
> in the output snapshot my wave suddenly disappears (I've attached my
> snapshots here). Could someone help me with this?
> Thanking you
> Yours truly
> Parvathy Sindhu Nair
> University of Auckland
>
>
> _______________________________________________
> CIG-SEISMO mailing list
> CIG-SEISMO at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-seismo
>

-- 
Dimitri Komatitsch
CNRS Research Director (DR CNRS), Laboratory of Mechanics and Acoustics,
UPR 7051, Marseille, France    http://komatitsch.free.fr