Spin up for the cold start in land-only runs

[Yuan Sun]

Hi all,

I am designing my land-only simulations for the project where I change 'numurbl' from 3 to 10 in the land surface file.

For spinup from a cold start to generate initial data, in my understanding, maybe it should start with I1850, followed by IHIST(1850-2000), and then I2000, and I2010.

According to 1.5.4. Spinning up the Satellite Phenology Model — ctsm CTSM master documentation, I should run 50-year spinup.

in the I1850 run, I set RUN_STARTDATE=1850, DATM_YR_START=1850, CLM_FORCE_COLDSTART= on,
STOP_OPTION=nyears, STOP_N=49,
in the IHIST run, I set RUN_STARTDATE=1850, DATM_YR_START=1850, STOP_OPTION=nyears, STOP_N=149

So the simulation journey is I1850 (1850-1900) -> IHIST (1900-2000 or 1850-2000 ?) Could i use the 50year I1850 output for the IHIST with
RUN_STARTDATE=1850-01-01.

Thanks for any comments.

Best,
Yuan

 

[slevis]

The I1850 and I2000 are time-slice simulations and IHist is transient. As such, think of the simulation years as 1-50 in the first time-slice case and as 1850-2000 (or whatever it is) in the transient.

 

[slevis]

I would certainly use the last restart file from the I1850 case to initialize the transient.

 

[Yuan Sun]

Dear Slevis,

Thank you very much. I will have a try.

Best,
Yuan