spin-up setting for a restart file

[Fuhow]

Hi, I am using ctsm5.3 now, I am ready to do a regional and self defined resolution run from 1980 to 2024 using IHISTBgcCrop case. Thus, I need a restart file in 1850 generated by spin-up from 1850. Therefore, I want to know if my Settings are correct, please help me check my Settings file.

I want to know if this restart file generated after this run has achieved the goal of spinning up 130 years from 1850 to achieve equilibrium, what if I wanna a 500 years spin-up?

In env_run.xml, setting:
RUN_STARTDATE=1850
STOP_OPTION="nyears"
STOP_N=130
ATM_DOMAIN_MESH="/home/user/ctsm5.3/tools/mksurfdata_esmf/lnd_mesh_TP_5km_hist.nc"
LND_DOMAIN_MESH="/home/user/ctsm5.3/tools/mksurfdata_esmf/lnd_mesh_TP_5km_hist.nc"
MASK_MESH="/home/user/ctsm5.3/tools/mksurfdata_esmf/mask_mesh_TP_5km_hist.nc"
CLM_ACCELERATED_SPINUP=on
CLM_FORCE_COLDSTART=on
DATM_YR_START=1901
DATM_YR_END=1980

In user_nl_clm, setting:
fsurdat='/home/user/ctsm5.3/tools/mksurfdata_esmf/surfdata_TP_5km_hist_1850_78pfts_c250218.nc'
hist_fincl1 = 'NEE:A','GPP:A','NEP:A','NPP:A','TSOI:A','TSOI_COL1:A','TSOI_COL2:A','TSOI_COL3:A','TSOI_COL4:A','TSOI_COL5:A','TSOI_COL6:A','TSOI_COL7:A','TSOI_COL8:A','TSOI_COL9:A','TSOI_COL10:A','TSOI_COL11:A','TSOI_COL12:A','ALT:A','TLAI:A','SOILICEWAT_LAY1:A','SOILICEWAT_LAY2:A','SOILICEWAT_LAY3:A','SOILICEWAT_LAY4:A','SOILICEWAT_LAY5:A','SOILICEWAT_LAY6:A','SOILICEWAT_LAY7:A','SOILICEWAT_LAY8:A','SOILICEWAT_LAY9:A','SOILICEWAT_LAY10:A','SOILICEWAT_LAY11:A','SOILICEWAT_LAY12:A','SOILLIQWAT_LAY1:A','SOILLIQWAT_LAY2:A','SOILLIQWAT_LAY3:A','SOILLIQWAT_LAY4:A','SOILLIQWAT_LAY5:A','SOILLIQWAT_LAY6:A','SOILLIQWAT_LAY7:A','SOILLIQWAT_LAY8:A','SOILLIQWAT_LAY9:A','SOILLIQWAT_LAY10:A','SOILLIQWAT_LAY11:A','SOILLIQWAT_LAY12:A','SOIL_LIQICEWAT_LAY1:A','SOIL_LIQICEWAT_LAY2:A','SOIL_LIQICEWAT_LAY3:A','SOIL_LIQICEWAT_LAY4:A','SOIL_LIQICEWAT_LAY5:A','SOIL_LIQICEWAT_LAY6:A','SOIL_LIQICEWAT_LAY7:A','SOIL_LIQICEWAT_LAY8:A','SOIL_LIQICEWAT_LAY9:A','SOIL_LIQICEWAT_LAY10:A','SOIL_LIQICEWAT_LAY11:A','SOIL_LIQICEWAT_LAY12:A'
hist_mfilt = 1
hist_nhtfrq = -8760

Thanks for any advice!

 

[Fuhow]

Hi, here is an update on my thoughts.

Firstly, the setting seems right for the 130-year spin-up. It might be necessary to set RESUBMIT to 4 for the period from 1850 to 1980. The total spin-up years would be 130 * 4 = 520 years.

Secondly, I found out that most say it should be at least a 400-year spin-up for Bgc to reach the equilibrium state. However, it costs much time on my server. As tested, it takes about 19 hours per spin-up year. Since I am running a simulation for a regional area (Tibet plateau) with a self-defined resolution, I am wondering if it can reach the equilibrium state after a 20-year spin-up.

Finally, I found out that I can test the output with the SpinupStability_BGC_v10.ncl tool, but I haven't found specific instructions about this tool yet.

Here are my thoughts for today. Thanks for any comments!

 

[Fuhow]

The reason why I want to try a 20-year spin-up is that I found the instruction about spin-up Bgc said that generally, the criteria that less than 3% of the land surface is in total ecosystem carbon disequilibrium takes the longest to satisfy due to slow soil carbon (TOTSOMC) turnover times in the Arctic. So, I think the soil biogeochemistry might be more active for my regional latitude between 25 and 40, although there is still a lot of seasonal frozen soil.

My setting is:

env_run.xml:
RUN_STARTDATE=1960
STOP_OPTION=nyears
STOP_N=20
DATM_YR_START=1960
DATM_YR_END=1980
CLM_ACCELERATED_SPINUP=on
CLM_FORCE_COLDSTART=on
MOSART_SIM_YEAR=1960

user_nl_clm: (This setting was guided by Oleson, Accelerated spinup mode)
fsurdat='/home/user/ctsm5.3/tools/mksurfdata_esmf/surfdata_TP_5km_hist_1960_78pfts_c250304.nc'
hist_empty_htapes = .true.
hist_fields_list_file = .false.
hist_fincl1 = 'TOTECOSYSC', 'TOTECOSYSN', 'TOTSOMC', 'TOTSOMN', 'TOTVEGC', 'TOTVEGN', 'TLAI', 'GPP', 'CPOOL', 'NPP', 'TWS'
hist_mfilt = 20
hist_nhtfrq = -8760

 

[slevis]

From a quick look at your posts, it seems that you are running successfully and are requesting feedback on issues that I would classify as scientific choices or decisions around your simulations. If so, I will let others comment. If I have misunderstood, please summarize your questions in a few short bullet-points, to help me understand what you are asking.

 

[Fuhow]

Hi, @slevis , thanks for your comment. You're right about my question.

I didn't know whether my spin-up was set up correctly earlier, even though it started to work properly. But now it seems to be set up correctly when compared to others' settings.

In addition, due to the limitation of computing power and regional simulation, I want to try to reduce the time of spin-up to 20 years. So I don't know whether my spin-up strategy is correct. I hope you scientists can help me evaluate whether this strategy is scientific.

Thanks for your reply again.