Guidance on Running with LME datasets with CESM2.1.3 (CLM5)

[oleson]

I have a tar file of the anomaly forcing files here:

ftp://ftp.cgd.ucar.edu/pub/oleson/anomaly_forcing_mct_for_ftp.tar

I would use it. The files at, e.g., /glade/campaign/cesm/cesmdata/cseg/inputdata/atm/datm7/anomaly_forcing/CMIP6-SSP3-7.0 were created from CMIP6 simulations that had an emissions error. The files I'm providing are from the corrected simulations.
Alternatively, you could try the file that has all of the variables in one file, e.g.,

/glade/campaign/cesm/cesmdata/cseg/inputdata/atm/datm7/anomaly_forcing/CMIP6-SSP3-7.0/af.allvars.CESM.SSP3-7.0.2015-2100_c20220628.nc

That file was also created from the corrected simulations.

 

[wvsi3w]

I have a tar file of the anomaly forcing files here:

ftp://ftp.cgd.ucar.edu/pub/oleson/anomaly_forcing_mct_for_ftp.tar

I would use it. The files at, e.g., /glade/campaign/cesm/cesmdata/cseg/inputdata/atm/datm7/anomaly_forcing/CMIP6-SSP3-7.0 were created from CMIP6 simulations that had an emissions error. The files I'm providing are from the corrected simulations.
Alternatively, you could try the file that has all of the variables in one file, e.g.,

/glade/campaign/cesm/cesmdata/cseg/inputdata/atm/datm7/anomaly_forcing/CMIP6-SSP3-7.0/af.allvars.CESM.SSP3-7.0.2015-2100_c20220628.nc

That file was also created from the corrected simulations.

So if I understood your message clearly, I could either download the .tar file you mentioned,

or

I could use the files (e.g., http://af.allvars.cesm.ssp2-4.5.2015-2100_c20220628.nc) which are in this link: - Revision 70792: /trunk/inputdata/atm/datm7/anomaly_forcing

I don't have access to glade (which I guess is an internal server for NCAR).

 

[oleson]

Correct.

 

[Zh Chen]

Correct.

Hi, Oleson. Thanks for providing the anomaly forcing files (e.g., af.allvars.cesm.ssp2-4.5.2015-2100_c20220628.nc). I am currently running SSP5-8.5 using CTSM5.3.026 with the NUOPC. I noticed that the repository currently seems to only contain the land use file (flanduse_timeseries) that is necessary for CTSM for SSP2-4.5 (link:- Revision 70792: /trunk/inputdata/lnd/clm2/surfdata_esmf/ctsm5.3.0) compatible with CTSM5.3.
Could you please advise on how to obtain the flanduse_timeseries files for other SSP scenarios (e.g., SSP3-7.0, SSP5-8.5)? Should I generate them using the CTSM surface dataset tools, or are files from previous CLM/CTSM versions compatible with CTSM5.3.026? Thanks for your help.

 

[Zh Chen]

Hi, Oleson. Thanks for providing the anomaly forcing files (e.g., af.allvars.cesm.ssp2-4.5.2015-2100_c20220628.nc). I am currently running SSP5-8.5 using CTSM5.3.026 with the NUOPC. I noticed that the repository currently seems to only contain the land use file (flanduse_timeseries) that is necessary for CTSM for SSP2-4.5 (link:- Revision 70792: /trunk/inputdata/lnd/clm2/surfdata_esmf/ctsm5.3.0) compatible with CTSM5.3.
Could you please advise on how to obtain the flanduse_timeseries files for other SSP scenarios (e.g., SSP3-7.0, SSP5-8.5)? Should I generate them using the CTSM surface dataset tools, or are files from previous CLM/CTSM versions compatible with CTSM5.3.026? Thanks for your help.

or the SSP landuse of CTSM5.2 (- Revision 70792: /trunk/inputdata/lnd/clm2/surfdata_esmf/ctsm5.2.0). Thanks

 

[Zh Chen]

Additionally, I noticed that the presets for various SSP scenarios (such as SSP3-7.0 and SSP5-8.5) with path of repository in namelist_defaults_ctsm.xml

e.g.,

<flanduse_timeseries hgrid="0.9x1.25" sim_year_range="1850-2100" ssp_rcp="SSP5-8.5"
>lnd/clm2/surfdata_esmf/ctsm5.3.0/landuse.timeseries_0.9x1.25_SSP5-8.5_1850-2100_78pfts_c240908.nc</flanduse_timeseries>

yet these files appear to be missing from the repository link. Thanks for your help.

 

[oleson]

I am not sure about compatibility between different versions, @slevis may be able to provide some information. The ctsm5.2 and ctsm5.3 are not official releases so that is likely why we are not providing a full set of SSP landuse timeseries files in the public repository. However, if they are available on our internal server (i.e., they've been created using the mksurfdata_esmf tool) then I could put them on our ftp site, or, it might be easier for you to create them using the mksurfdata_esmf tool.
What specific SSP(s) and resolution(s) are you looking for?

 

[Zh Chen]

I am not sure about compatibility between different versions, @slevis may be able to provide some information. The ctsm5.2 and ctsm5.3 are not official releases so that is likely why we are not providing a full set of SSP landuse timeseries files in the public repository. However, if they are available on our internal server (i.e., they've been created using the mksurfdata_esmf tool) then I could put them on our ftp site, or, it might be easier for you to create them using the mksurfdata_esmf tool.
What specific SSP(s) and resolution(s) are you looking for?

Thank you for your reply. I am looking for three files for SSP3-7.0, SSP4-6.0, and SSP5-8.5 at 0.9x1.25.These are as follows, and they are from namelist_defaults_ctsm.xml of CTSM5.3.026:

<!-- SSP3-RCP7.0 - relative to {csmdata}) -->
<flanduse_timeseries hgrid="0.9x1.25" sim_year_range="1850-2100" ssp_rcp="SSP3-7.0"
>lnd/clm2/surfdata_esmf/ctsm5.3.0/landuse.timeseries_0.9x1.25_SSP3-7.0_1850-2100_78pfts_c240908.nc</flanduse_timeseries>

<!-- SSP4-RCP6.0 - relative to {csmdata}) -->
<flanduse_timeseries hgrid="0.9x1.25" sim_year_range="1850-2100" ssp_rcp="SSP4-6.0"
>lnd/clm2/surfdata_esmf/ctsm5.3.0/landuse.timeseries_0.9x1.25_SSP4-6.0_1850-2100_78pfts_c240908.nc</flanduse_timeseries>

<!-- SSP5-RCP8.5 - relative to {csmdata}) -->
<flanduse_timeseries hgrid="0.9x1.25" sim_year_range="1850-2100" ssp_rcp="SSP5-8.5"
>lnd/clm2/surfdata_esmf/ctsm5.3.0/landuse.timeseries_0.9x1.25_SSP5-8.5_1850-2100_78pfts_c240908.nc</flanduse_timeseries>

If these files are available on your internal server, would it be possible to upload them to the FTP site? It would be greatly appreciated, as running the mksurfdata_esmf tool myself seems challenging due to the large amount of raw SSP dependency files required. Thanks for your help.

 

[oleson]

The first two files are there:

ftp://ftp.cgd.ucar.edu/pub/oleson/ctsm5.3_landusetimeseries/landuse.timeseries_0.9x1.25_SSP3-7.0_1850-2100_78pfts_c240908.nc
ftp://ftp.cgd.ucar.edu/pub/oleson/ctsm5.3_landusetimeseries/landuse.timeseries_0.9x1.25_SSP4-6.0_1850-2100_78pfts_c240908.nc

The disc space is limited, so if you grab those files I can start putting the last one on.

 

[Zh Chen]

The first two files are there:

ftp://ftp.cgd.ucar.edu/pub/oleson/ctsm5.3_landusetimeseries/landuse.timeseries_0.9x1.25_SSP3-7.0_1850-2100_78pfts_c240908.nc
ftp://ftp.cgd.ucar.edu/pub/oleson/ctsm5.3_landusetimeseries/landuse.timeseries_0.9x1.25_SSP4-6.0_1850-2100_78pfts_c240908.nc

The disc space is limited, so if you grab those files I can start putting the last one on.

Thanks a lot! I've grabbed the two files, so you can delete them now. I'm ready for the last one whenever you are.

 

[oleson]

Ok, the last one is there now.

 

[wvsi3w]

Hello Keith @oleson and Sam @slevis

Following my case (which I mentioned here), my simulations started running, and all of a sudden failed in the middle of the second resubmit. I did try to redo the whole thing from the beginning and they all failed at the same year (2033).

I have four cases, and all of them crashed with the same error:

ERROR: ERROR in BalanceCheckMod.F90

I have attached the cesm log file (had to reduce its size by turning it into gz file).

Please let me know what you think the root cause of this issue is. Thanks.

P.S. a reminder to specify my cases: I had two verson of the depth (shallow and deep), and two members of solar variability (mem1, mem2), hence I had 4 conditions to start my SSP simulations which are SSP45_Mem1_Deep, SSP45_Mem1_Shallow, SSP45_Mem2_Deep, SSP45_Mem2_Shallow (started the SSP from 2021 and all failed in year 2033, twice!).

P.P.S. If you want, I can share my user_datm.streams.txt.Anomaly* files and the rest of the setup.

 

[Zh Chen]

Ok, the last one is there now.

Thanks! Download complete, so please feel free to remove it. I realized I still need the SSP1-2.6 file listed below. If you could upload that one, I'll grab it instantly. Thanks for your patience!

<!-- SSP1-RCP2.6 - relative to {csmdata}) -->
<flanduse_timeseries hgrid="0.9x1.25" sim_year_range="1850-2100" ssp_rcp="SSP1-2.6"
>lnd/clm2/surfdata_esmf/ctsm5.3.0/landuse.timeseries_0.9x1.25_SSP1-2.6_1850-2100_78pfts_c240908.nc</flanduse_timeseries>

 

[wvsi3w]

Hello Keith @oleson and Sam @slevis

Following my case (which I mentioned here), my simulations started running, and all of a sudden failed in the middle of the second resubmit. I did try to redo the whole thing from the beginning and they all failed at the same year (2033).

I have four cases, and all of them crashed with the same error:

ERROR: ERROR in BalanceCheckMod.F90

I have attached the cesm log file (had to reduce its size by turning it into gz file).

Please let me know what you think the root cause of this issue is. Thanks.

P.S. a reminder to specify my cases: I had two verson of the depth (shallow and deep), and two members of solar variability (mem1, mem2), hence I had 4 conditions to start my SSP simulations which are SSP45_Mem1_Deep, SSP45_Mem1_Shallow, SSP45_Mem2_Deep, SSP45_Mem2_Shallow (started the SSP from 2021 and all failed in year 2033, twice!).

P.P.S. If you want, I can share my user_datm.streams.txt.Anomaly* files and the rest of the setup.

Hello again,
Would this part of my previous run which is being used as finidat in my current SSP2-4.5 run impact the results and create that BalanceCheckMod Error?
In that I used SSP1-2.6 for 2015-2020, and here from 2021 I am using SSP2-4.5, do you think that is the cause?

 

[oleson]

Ok, that one is there now.

 

[Zh Chen]

Ok, that one is there now.

Download complete, so please feel free to remove it. Thanks for your patience!