[CIG-SEISMO] SPECFEM2D output files

[Alexandre KAZANTSEV]

Good afternoon, 

I write you by e-mail to avoid a potentially useless issue on github. 
I have encountered two problems with specfem2D outputs. 

1. I wanted to sample the wavefield densly in space. I first tried to use the following options in Par_file : 
output_wavefield_dumps = .true. 
use_binary_for_wavefield_dumps = .true. 

This resulted into a crash. 

2. I bypassed this issue by generating a STATIONS file with several thousands stations and using the options : 
save_ASCII_seismograms = .false. 
save_binary_seismograms_single = .true. 
use_existing_STATIONS = .true. 

This worked, but generated heavy files. To avoid this, I tried to use 
subsamp_seismos = 50 , 

But it looks like it works well only with ascii seismograms . The size of the binary file 'Uz_file_single.bin' remains the same (huge) as it was without downsampling. The import of the "downsampled" files with Matlab "fread" produced strange results, while it worked well if no downsampling was used. 

If you have an idea on why these problems occur and how to handle them, please let me know. 

You will find enclosed an example of simulation parameters. The executable is ./main.sh, which must be modified to account for the true location of the scripts xspecfem2D, xmeshfem2D and gmsh. 

Best regards, 
Alexandre (PhD student in Geophysics at Mines-Paristech, France) 





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.geodynamics.org/pipermail/cig-seismo/attachments/20170502/44b29fe9/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: export_simu_params.zip
Type: application/zip
Size: 1057649 bytes
Desc: not available
URL: <http://lists.geodynamics.org/pipermail/cig-seismo/attachments/20170502/44b29fe9/attachment-0001.zip>