Can't read the emission data from use_nl_cam

[Sherry Hou]

Dear all,
New species with emissions are added in CESM2 with CAM-chem in my researches. When I have created NO bio, bmb and other emission datasets (using ncl program), there are some problem in running simulation. Given NO_BIO as a example. The problem as follows:

Opened existing file /nuist/u/home/zhubin/zb_houxw/scratch/tagged_emissions/emissions.NO.surface.1.9x2.5_bio.nc 327680
Opened existing file /nuist/u/home/zhubin/zb_houxw/scratch/tagged_emissions/emissions.NO.surface.1.9x2.5_bio.nc 327680
NetCDF: Variable not found
NetCDF: Variable not found
NetCDF: Invalid dimension ID or name
NetCDF: Invalid dimension ID or name

The information of "emissions.NO.surface.1.9x2.5_bio.nc" are as follows:
"

netcdf emissions.NO.surface.1.9x2.5_bio {
dimensions:
time = UNLIMITED ; // (230 currently)
lat = 96 ;
lon = 144 ;
variables:
float time(time) ;
time:long_name = "Time" ;
time:units = "days since 1850-01-01 00:00:00" ;
time:calendar = "Gregorian" ;
float lat(lat) ;
lat:units = "degrees_north" ;
lat:long_name = "Latitude" ;
float lon(lon) ;
lon:units = "degrees_east" ;
lon:long_name = "Longitude" ;
int date(time) ;
date:units = "YYYYMMDD" ;
date:long_name = "Date" ;
float NO_BIO(time, lat, lon) ;
NO_BIO:units = "molecules/cm2/s" ;
NO_BIO:long_name = "Soil emissions" ;
NO_BIO:history = "POET" ;

I added the new species to the namelist and the path in user_nl_cam, as:

srf_emis_specifier = 'NO_BIO -> /nuist/u/home/zhubin/zb_houxw/scratch/tagged_emissions/emissions.NO.surface.1.9x2.5_bio.nc'

The user_nl_cam and mo_srf_emissions.F90 files are in the attachments.

Could you give me some suggestions?
Does the emission data have problems?
Or can't the new species be read from emission data in some Fortran program?
If the data have problem, what's kind of data should I created?
If the new species can't be figured out, which programs should I modify?

Looking forward to your reply.
Best,
Sherry

 

[fvitt]

Which specific version of CESM2 are you using? There seems to be a problem reading one of the netcdf input files. I am not so sure the NO_BIO emissions file is the culprit. Do all of those emissions files look okay?
Have you tried running without NO_BIO in the srf_emis_specifier, or pointing NO_BIO some other emissions file?

 

[Sherry Hou]

Hi,
I use CESM1.2.2, sorry for not CESM2. I have tried without srf_emis_specifier. It is running without any problem. I have added a "write" command in mo_srf_emiss.F90 to check the "non-zero emission". It seems the read of emission data has no problem. But the output data of the new species in history, such as NO_BIO concentration in air, is NAN. Could you give me some suggestions about the NAN output? Thank you so much!

Another question is about FSDCHM in CESM1.2.2. Is there any problem about the code of FSDCHM?
When I run it, the result always shown:

Fatal error in PMPI_Allgatherv: Invalid buffer pointer, error stack:
PMPI_Allgatherv(1210): MPI_Allgatherv(sbuf=0x1b0cdd40, scount=3, MPI_DOUBLE_PRECISION, rbuf=0x1b0cdd40, rcounts=0x1b0d3b10, displs=0x1b0d15e0, MPI_DOUBLE_PRECISION, comm=0xc4000006) failed
PMPI_Allgatherv(1158): Buffers must not be aliased

I have searched the problem in CESM Forums, other people also meet the problem. See CESM ‘’PMPI_ALLGATHERV(1072): BUFFERS MUST NOT BE ALIASED‘’ ERROR | DiscussCESM Forums (ucar.edu). The problem seems not been solved. I also have tried to set I_MPI_COMPATIBILITY=4 in the environment and to change version of MPI. They all don't work. Could you give me some suggestions about the problem, too?
Thanks!
Best,
Sherry