Failure using predefined RRFS_SUBCONUS_3km grid

I am running SRW v2 that I compiled using intel on a Boise State HPC.

I successfully ran the FV3_RRFS_v1alpha (and GFS_v16) suite using the predefined RRFS_CONUS_3km grid. When I switched to use the RRFS_SUBCONUS_3km grid, it failed after running for ~1.5 mins with the error:

'Error in subr radiation_aerosols: unrealistic surface pressure =           1                     NaN'

Is this predefined grid supported yet for this suite? or at all? It wasn't listed in the UG, but I saw it was an option in valid_param_vals and thought to try it out.

Thanks,

Tracy

Hi, Tracy,

I just tested  RRFS_SUBCONUS_3km on NOAA HPC Hera, it works fine with the default case (2019061518). Are you using the default case?

Thanks,

Linlin

Hi Linlin - I am using the default case. Thanks for testing on Hera. (Note: my SRW v2 build actually failed on Hera using the 'devbuild.sh --platform=hera --compiler=intel' options, which should be a straightforward build, but maybe I need to build it manually; error happened in building UFS_UTILS/sorc/emcsfc_snow2mdl.fd/snow2mdl.F90)

Anyways, do you have any suggestions based on the error for running on Borah? Link to log: https://drive.google.com/file/d/1B7lgB2xStpfuFtRl8eh7v8SwB-sXlHwE/view?usp=sharing

Thanks,

Tracy

Hi Linlin,

Thanks - The develop branch builds on Hera for me, but since we are using the public released SRW V2 on the Boise State HPC, that is the preferred version to be testing on Hera.

Are you able to build the v2 public release on Hera?

What version did you use when you had a successful SUBCONUS domain run? develop or v2?

Thanks,

Tracy