Blow-up at Initial time when using coupled ROMS and SWAN

[okami_zxh]

Hi all,
I am trying to run a case with coupled ROMS and swan (COAWST) to do sediment transport. Before I add sediment part I did a case purely by running coupled ROMS and SWAN with the same input files, which runs well. But when I modified initial file by adding sediment part with matlab_script named 'create_roms_init.m' which is provided as a part of COAWST tools. The model blows up at initial time due to NAN for Potential Energy and Net volume:

Code: Select all

 STEP   Day HH:MM:SS  KINETIC_ENRG   POTEN_ENRG    TOTAL_ENRG    NET_VOLUME
          C => (i,j,k)       Cu            Cv            Cw         Max Speed

       0   250 00:00:00           NaN           NaN           NaN           NaN
          (000,000,00)  0.000000E+00  0.000000E+00  0.000000E+00           NaN

But I did check the temperature and salinity which is reasonable. I totally don't know why it happened. I attached the ocean.in file, header file, output file.
Due to the big initial file, I uploaded on google drive which is open to everyone, here is the link:https://drive.google.com/file/d/1Zf-h3_ ... sp=sharing
If any friend has suggestions I would be very appreciate it.

[noah.eckman]

Looks like one of your input has NaN values somewhere. ROMS doesn't allow this even on masked cells. See here: viewtopic.php?f=14&t=4454&view=next