I2000Clm50SpRsGs + GLACIER_REGION

[yfang]

I'm running the latest version of CESM with compset I2000Clm50SpRsGs. The model has an error:
ERROR: GLACIER_REGION NOT on surfdata fileERROR in glcBehaviorMod.F90 at line 583
I checked a couple of surfdata and didn't find the variable GLACIER_REGION. Is there a surfdata that has it?

Is there a compset that's similar to ICLM45 as I only need componet DATM and lnd?

Thanks,
Yilin Fang

 

[katec]

Hi Yilin, If you are using CESM 2.1.1, try the compset I2000Clm50Sp . It sounds like you do not need active glaciers, which you will get with compsets that include "G". You can look at all of the supported CESM2 compsets on this website: CESM2 Component Sets Definition

 

[yfang]

Hi Kate, I did try I2000Clm50Sp, but it requires a bunch of coupler files.

 

[sacks]

@yfang - You need the GLACIER_REGION variable regardless of whether or not you are running with CISM. It's true that you can avoid needing some coupler (mapping) files by running without CISM, and you're right that I compsets with Gs in the alias avoid including CISM, and so avoid the need for these mapping files. However, you still need GLACIER_REGION on the surface dataset.

All of our recent surface datasets should have this field on them. It sounds like you may be trying to use a surface dataset from an older version of the model? If so, that's going to cause problems for a variety of reasons. If you describe your full setup in more detail, we might be able to help you.

 

[yfang]

@sacks - Thanks for your response. This is my command to create the new case:
./create_newcase --case ${CASE_NAME} --res f09_g16 --compset ${COMPSET} --run-unsupported
I have to provide a surfdata in user_nl_clm otherwise it complains about no default value for fsurdat:

ERROR: Command software/CESM/components/clm/bld/build-namelist failed rc=255
out=CLM adding use_case 2000_control defaults for var 'sim_year' with val '2000'
CLM adding use_case 2000_control defaults for var 'sim_year_range' with val 'constant'
CLM adding use_case 2000_control defaults for var 'stream_year_first_urbantv' with val '2000'
CLM adding use_case 2000_control defaults for var 'stream_year_last_urbantv' with val '2000'
CLM adding use_case 2000_control defaults for var 'use_case_desc' with val 'Conditions to simulate 2000 land-use'
err=ERROR : CLM build-namelist::CLMBuildNamelist::add_default() : No default value found for fsurdat.
Are defaults provided for this resolution and land mask?

 

[sacks]

Try --res f09_g17 instead of --res f09_g16: The g16 mask is no longer supported, and we now use the g17 mask instead. (The difference is that, with the g17 mask, the Caspian Sea is treated as land rather than ocean.) Hopefully with that it should find an out-of-the-box surface dataset.

 

[yfang]

@sacks - It still can't find the dataset with f09_g17. Could you tell me where I can find the recent surface datasets? Thanks!

 

[yfang]

I found a recent surfdata. Adding it to user_nl_clm works now.

 

[sacks]

@yfang - I think something may be wrong with your setup if you're needing to explicitly set fsurdat in user_nl_clm. I just tried create_newcase from cesm master with --compset I2000Clm50SpRsGs --res f09_g17 and it found a surface dataset. I'm concerned that what you're describing is a sign of a more significant problem. If you give some additional information about what you're doing, including the exact model version (output from running describe_version from the top level of your cesm checkout) and the exact steps you have taken (see https://bb.cgd.ucar.edu/cesm/threads/information-to-include-in-help-requests.4974/ for guidance on information to provide), I can try to help you figure out what's wrong.

 

[yfang]

@sacks - Here are the first few lines. Details are in the attached. Thanks!
./describe_version
------------------------------------------------------------------------
git describe:
cesm2_2_alpha06a-3-gbae1c94
------------------------------------------------------------------------

------------------------------------------------------------------------
git status:
# On branch master
nothing to commit, working directory clean
------------------------------------------------------------------------

 

[sacks]

Thanks for that info. Can you please describe the full set of steps you are taking that is leading to this error? Also, please give the error message you are receiving if it differs from what you posted before.

 

[yfang]

@sacks - I just found out why. It doesn't like I put "irrigate = .false." in user_nl_clm.

 

[sacks]

Great, I'm glad you figured it out. Do you need help with how to turn irrigation off?

 

[yfang]

What's the correct way to turn irrigation off?

 

[sacks]

Hmmm, now that I look, I'm not sure what is the correct thing to do here. I'm pinging @erik for help.

 

[erik]

OK, we currently have things setup so that the two irrigation options are expected to have different datasets. This was a requirement in the model until fairly recently. With the version of the model you are using, this is no longer required. But, it's still setup to expect that. So you have to jump through some hoops to get this to work.

What I recommend doing is setting up an identical case with irrigation on, that you can use as a prototype for the other case. The important thing you'll get from that case is the fsurdat file. In your case with irrigation off, you'll need to set the fsurdat file in user_nl_clm as well as irrigate = .false.. I think that should be all that you need to do. It'll help to have the two cases setup so you can compare between the two. In the end I expect that the only difference between them for the lnd_in file will be the irrigate setting.

 

[yfang]

Thanks!

 

[huazhen]

OK, we currently have things setup so that the two irrigation options are expected to have different datasets. This was a requirement in the model until fairly recently. With the version of the model you are using, this is no longer required. But, it's still setup to expect that. So you have to jump through some hoops to get this to work.

What I recommend doing is setting up an identical case with irrigation on, that you can use as a prototype for the other case. The important thing you'll get from that case is the fsurdat file. In your case with irrigation off, you'll need to set the fsurdat file in user_nl_clm as well as irrigate = .false.. I think that should be all that you need to do. It'll help to have the two cases setup so you can compare between the two. In the end I expect that the only difference between them for the lnd_in file will be the irrigate setting.

Hi @erik , I followed your suggestions to turn off the irrigation by adding "
irrigate = .false.
fsurdat = '/work6/data/cesm/inputdata/lnd/clm2/surfdata_map/surfdata_64x128_16pfts_Irrig_CMIP6_simyr2000_c170824.nc'
" in user_nl_clm. And I also checked the lnd_in file have "irrigate = .false.". But in the CLM outputs, the variable "QIRRIG" still have the same values as the identical case with irrigation on. I think it still not turn irrigation off. Do you have any suggestions? Thanks a lot.

Kind regards,
Huazhen

 

[huazhen]

Hi @erik , I followed your suggestions to turn off the irrigation by adding "
irrigate = .false.
fsurdat = '/work6/data/cesm/inputdata/lnd/clm2/surfdata_map/surfdata_64x128_16pfts_Irrig_CMIP6_simyr2000_c170824.nc'
" in user_nl_clm. And I also checked the lnd_in file have "irrigate = .false.". But in the CLM outputs, the variable "QIRRIG" still have the same values as the identical case with irrigation on. I think it still not turn irrigation off. Do you have any suggestions? Thanks a lot.

Kind regards,
Huazhen

Sorry. I forgot to mention that I'm using CESM2.1.0 version and FSCAM compset.

 

[Wenwen]

Sorry. I forgot to mention that I'm using CESM2.1.0 version and FSCAM compset.

Hi,Im facing the same problem. In the irrigation off case, the CLM output variable "QIRRIG" have the same values as the irrigation on case. Please let me know if you had solved the issue? Thanks!