The issue of terrain processing in the MPAS tutorial practice guide

[jc sun]

Hello，everyone
I am learning MPAS according to MPAS Tutorial, I have successfully compiled initinit_atmosphere_model. When proceeding to the Static, terrestrial field processing chapter, an error occurred when executing ./init_atmosphere_model.I suspect it's the reason it's not running with mpi. Then re-executed mpirun ./init_atmosphere_model and found the same error.

 

[Ming Chen]

Which version of MPAS are you using?

 

[jc sun]

My mpas version is 8.0.1.Download from github and upload server to decompress and compile.

 

[Ming Chen]

Your namelist and streams files look fine. I am not sure what is going on here. Probably there are issues related to the command and the environmental settings. In Cheyenne, the command is:

mpiexec_mpt ./atmosphere_model

I suggest you contact your computer manager and ask what command/environmental settings are appropriate.

 

[jc sun]

Thank you for your suggestion. I checked the relevant process and found that I was not using PIO, I thought it was the cause of the error, so I installed and set up PIO. Then I recompiled and installed again. The result is as follows:

However, when I re-run ./init_atmosphere_model, the error is reported as follows:

Does this mean a failure of the compilation process？

 

[jc sun]

My pnetcdf version is 1.11.2 and PIO version is 2.5.9. I follow the settings of wrf for the rest of the tools, and the settings of wrf can be simulated successfully.

 

[jc sun]

I guess I may have found the answer to my problem. When I replace openmpi with mpich to compile PIO2.5.9, and also use mpich to compile MPAS, the final executable file can be run. My openmpi version is 2.1.2 and mpich version is 3.2. I don't know if it is the reason why MPAS does not support openmpi?

 

[jc sun]

I encountered an error when doing Interpolating real-data initial conditions in the tutorial practice guide. I could successfully obtain the results of terrain processing and generate relevant nc files. However, when doing atmospheric data interpolation, an error occurred. The information is as follows. For this problem How to solve it?

Found 27 levels in the first-guess data

 *************************************************

 Setting pressure field for isobaric data

 Setting SST from SKINTEMP

 Done with soil consistency check

 SEAICE_FRACTIONAL file not found...

 OMLD file not found...

ERROR: extrap_type == 2 not implemented for target_z >= zf(1,nz)

ERROR: *****************************************************************

ERROR: Error in interpolation of t(k,iCell) for k=  39, iCell=         9

CRITICAL ERROR: *****************************************************************

 

 -----------------------------------------

 Total log messages printed:

    Output messages =                  364

    Warning messages =                   0

    Error messages =                     3

    Critical error messages =            1

 -----------------------------------------

 

[jc sun]

Sorry to disturb everyone. I know the reason. The config_extrap_airtemp in namelist needs to be changed to linear. If it is not marked in yellow, it has not been changed, which leads to this reason.

 

[meteowang]

Sorry to disturb everyone. I know the reason. The config_extrap_airtemp in namelist needs to be changed to linear. If it is not marked in yellow, it has not been changed, which leads to this reason.

Hello! I just met the same error as you, I notice that you modified config_extrap_airtemp=linear,then it worked, but I still got the same error. Did you change something else? Thanks a lot!

 

[Ming Chen]

@meteowang,
Would you please send me your namelist.init_atmosphere and streams.init_atmosphere to take a look? Which mesh did you use? Other information such as the MPAS version, the input data, how you compile MPAS and run the model would also be helpful.