run with fixed SST and sea ice, active CAM

[littledddna@gmail_com]

Dear all,

I am using iCESM to perform a fixed SST and sea ice run with active atmosphere and land.
I think this is similar to F (or AMIP) run except that AMIP run has updated observed SST. Following the instruction of CESM1.2 userguide, it seems that I need to add time and datesec to the SST_ICE file. So, my questions are,
1. since I only need a fixed SST and sea ice, do I need consider these time variables in the input file? Should I set time to 1 in the file?
2. The new generated SST and ICE file should be in the CAM grid, right ? Not the 2D POP grid.

Thanks very much and best regards,

Xiao

 

[dbailey]

When you set up your F compset, you should see a variable SSTICE_DATA_FILENAME in env_run.xml. You can have a look at:

$DIN_LOC_ROOT/atm/cam/sst/sst_HadOIBl_bc_0.9x1.25_1850_2017_c180507.nc

as an example. This is on the CAM FV 1-degree grid. Ultimately, if you are doing an FHIST or historical run, then you would expect *real* years in the dataset. If you are just doing a repeat climatological year then the dates would simply be 0001-01 to 0001-12. Or some arbitrary year. When you specifiy the grid for an F compset it will be f09_f09. You can always do:

./create_newcase --list grids

 

[littledddna@gmail_com]

When you set up your F compset, you should see a variable SSTICE_DATA_FILENAME in env_run.xml. You can have a look at:

$DIN_LOC_ROOT/atm/cam/sst/sst_HadOIBl_bc_0.9x1.25_1850_2017_c180507.nc

as an example. This is on the CAM FV 1-degree grid. Ultimately, if you are doing an FHIST or historical run, then you would expect *real* years in the dataset. If you are just doing a repeat climatological year then the dates would simply be 0001-01 to 0001-12. Or some arbitrary year. When you specifiy the grid for an F compset it will be f09_f09. You can always do:

./create_newcase --list grids

Thanks very much for the reply, David. As far as I understand, repeat climatological year simulation can only performed use f09_f09 grid. Is that correct ?
I have T31 and gx3v5 for my old run which i would like to continue with fixed repeat climatological year SST.

 

[littledddna@gmail_com]

I see there are T31 file there. I will try again :)

 

[littledddna@gmail_com]

So i tried to generated new SST and ice file for the new run of T31 resolution. But the run failed. The model is stating that there is a mismatch between land fraction and sea ice fraction. But i checked the domain file for CICE and input sea ice fraction. The sum of land fraction from domain file and sea ice from new data is 0.99999 something. There is also no sea ice on land (i checked the land-sea mask in the domain file). But the model still shows the problem. Should I check other things? Thanks very much . Attached the SST and domain file I am using.

Here is the error message: (if fraci is 1, it is missing, right?)

inconsistency between land fraction and sea ice fraction
n= 1 fracl= 1.00000000000000 fraci=
inconsistency between land fraction and sea ice fraction
n= 1 fracl= 0.811322998149999 fraci=
0.811322998149999 sum= 1.62264599630000
1.00000000000000 sum= 2.00000000000000