CESM2 with ic's from different runs?

[tomas]

Hi

I would like to start CESM2 using land/atm ic's from one run, and other component ic's from another. I couldn't find any guidance/advise on how to go about doing this, so I'm posting here in hopes that someone can provide some.

Thanks.

Bob

 

[nusbaume]

Hi Bob,

I am not sure exactly what kind of simulation you are doing, but you should be able to specify CAM's initial conditions by setting the ncdata namelist variable, and CLM's initial conditions via the finidat namelist variable, both of which you should be able to copy from your original run (while leaving everything else the same).

Of course if that doesn't work just let me know.

Thanks, and have a great day!

Jesse

 

[tomas]

Hi Bob,

I am not sure exactly what kind of simulation you are doing, but you should be able to specify CAM's initial conditions by setting the ncdata namelist variable, and CLM's initial conditions via the finidat namelist variable, both of which you should be able to copy from your original run (while leaving everything else the same).

Of course if that doesn't work just let me know.

Thanks, and have a great day!

Jesse

That is along the lines of what I came up with. Set the run up as an initial run? I believe trying a hybrid or branch would cause the scripts to balk. As far as specify ic's via user_nl_* files, here is what I came up with:

atm - ncdata
lnd - finidat
mosart - finidat_rtm
ice - ice_ic
pop - maybe via the restart file?
marbl - ???
cism - ???
wav - ???
anything else???

 

[tomas]

That is along the lines of what I came up with. Set the run up as an initial run? I believe trying a hybrid or branch would cause the scripts to balk. As far as specify ic's via user_nl_* files, here is what I came up with:

atm - ncdata
lnd - finidat
mosart - finidat_rtm
ice - ice_ic
pop - maybe via the restart file?
marbl - ???
cism - ???
wav - ???
anything else???

I forgot to answer your question. But from the namelist parameters in my list, you can figure it is a fully coupled simulation. I am starting a b.f19_g17 run. I want to use ocean/ice ic's from a b.f09_g17 run. For atmosphere and land, I want to use a different b.f19_g17 run,

 

[nusbaume]

Hi Bob,

I am also not sure exactly how the initial conditions are set for POP, MARBL, CISM, and/or WW3. @mlevy, @katec, @altuntas do any of you by chance know?

Thanks!

Jesse

 

[katec]

Hi Bob,

The initial conditions for CISM are defined in the user_nl_cism namelist with the variable "cisminputfile" . Also, it might be helpful to use the namelist definitions websites from the CESM2 online documentation: CESM2 Component Configuration Settings

 

[lvxin]

Hi Bob,

I am not sure exactly what kind of simulation you are doing, but you should be able to specify CAM's initial conditions by setting the ncdata namelist variable, and CLM's initial conditions via the finidat namelist variable, both of which you should be able to copy from your original run (while leaving everything else the same).

Of course if that doesn't work just let me know.

Thanks, and have a great day!

Jesse

Hi Jesse,
If my simulation starts in year 2010, what 'ncdata' variable should I set? Is the default file '******0009-01-01-00000_c130329.nc' ok? If not, where can I download the ncdata?
Thanks!

 

[nusbaume]

Hi Ivxin,

The ncdatachoice will depend on your particular model set-up, such as what grid you are using, and whether you are using CAM, CAM-Chem, or WACCM. If you are using the CESM2 code-base, then I would recommend just creating a new case with your grid of interest and compset F2010climo if using CAM, FC2010climo if using CAM-Chem, or FW2010climo if using WACCM.

You can then run ./case.setup && ./preview_namelists in your new case to generate the name of the correct ncdata file (which will be listed in atm_in in your run directory), and ./check_input_data to automatically download the file to your machine if it doesn't exist already.

Of course if you run into any problems using those compsets to acquire the correct file(s) please let us know.

Thanks, and have a great day!

Jesse

 

[lvxin]

Hi Ivxin,

The ncdatachoice will depend on your particular model set-up, such as what grid you are using, and whether you are using CAM, CAM-Chem, or WACCM. If you are using the CESM2 code-base, then I would recommend just creating a new case with your grid of interest and compset F2010climo if using CAM, FC2010climo if using CAM-Chem, or FW2010climo if using WACCM.

You can then run ./case.setup && ./preview_namelists in your new case to generate the name of the correct ncdata file (which will be listed in atm_in in your run directory), and ./check_input_data to automatically download the file to your machine if it doesn't exist already.

Of course if you run into any problems using those compsets to acquire the correct file(s) please let us know.

Thanks, and have a great day!

Jesse

Hi Jesse,
Thank you very much for your reply. I will create cases following your suggestion.