Hello,
I'm giving my first steps on hysplit starting by compiling hysplit version 3.9.1 along with wrf-chem 3.9.1 (making the required modifications on registry). I use the option 32(serial) and 34(dmpar) GNU for configuration, getting the same error in the same step while compiling em_real:
hyconc.f90:262:32:
mpi_comm_here = MPI_COMM_WORLD
1
Error: Symbol 'mpi_comm_world' at (1) has no IMPLICIT type
Command exited with non-zero status 1
I would like to know how can I solve this problem. Than you very much!
Christian
Compiling problem
-
- Posts: 16
- Joined: January 25th, 2018, 1:28 pm
- Registered HYSPLIT User: Yes
Re: Compiling problem
To be clear, you were trying to compiler the inline WRF-HYSPLIT.
"MPI_COMM_WORLD" is one of the functions in MPI library. Have you checked the MPI library to be installed properly on your machine with the gfortran complier? And you should be using option 34 for gfortran compiler.
"MPI_COMM_WORLD" is one of the functions in MPI library. Have you checked the MPI library to be installed properly on your machine with the gfortran complier? And you should be using option 34 for gfortran compiler.
-
- Posts: 16
- Joined: January 25th, 2018, 1:28 pm
- Registered HYSPLIT User: Yes
Re: Compiling problem
Hi, thank you very much for the advice.
I reinstalled libraries and it solve the mpi error. Unfortunately there are many new errors starting with this one:
mediation_integrate.f90:72.18:
grid%hycsum=0.0
1
Error: 'hycsum' at (1) is not a member of the 'domain' structure
mediation_integrate.f90:173.31:
IF (config_flags%run_hysplit .EQ. 1) THEN
1
Error: 'run_hysplit' at (1) is not a member of the 'grid_config_rec_type' structure
mediation_integrate.f90:182.6:
ENDIF
1
Error: Expecting END SUBROUTINE statement at (1)
I would attach my log.compile, but there is a problem with the attachment.
Thank you very much for your help.
Christian
I reinstalled libraries and it solve the mpi error. Unfortunately there are many new errors starting with this one:
mediation_integrate.f90:72.18:
grid%hycsum=0.0
1
Error: 'hycsum' at (1) is not a member of the 'domain' structure
mediation_integrate.f90:173.31:
IF (config_flags%run_hysplit .EQ. 1) THEN
1
Error: 'run_hysplit' at (1) is not a member of the 'grid_config_rec_type' structure
mediation_integrate.f90:182.6:
ENDIF
1
Error: Expecting END SUBROUTINE statement at (1)
I would attach my log.compile, but there is a problem with the attachment.
Thank you very much for your help.
Christian
Re: Compiling problem
The error is due to newly added HYSPLIT variables are not defined in the Registry file. The inline HYSPLIT package has two files (registry.hysplit and Registry.EM) that should be copied to your Registry directory. Please check if it was done properly.
The inline HYSPLIT was implemented in a standard version of WRF (README file in the package). If you are using it with WRF-CHEM, you will need to put registry.hysplit in the registry file used for WRF-CHEM. Again, the inline HYSPLIT has not been tested with the WRF-CHEM code.
The inline HYSPLIT was implemented in a standard version of WRF (README file in the package). If you are using it with WRF-CHEM, you will need to put registry.hysplit in the registry file used for WRF-CHEM. Again, the inline HYSPLIT has not been tested with the WRF-CHEM code.
-
- Posts: 16
- Joined: January 25th, 2018, 1:28 pm
- Registered HYSPLIT User: Yes
Re: Compiling problem
Thank you very much, the problem was not adding the new Registry.EM as you said. Its solved now, thanks!