Yes, pfts will change with time as prescribed by flanduse_timeseries in the namelist. Dynamic global vegetation model is generally a term that refers to a different type of model, one in which vegetation types change in response to climate (not prescribed). We used to have a DGVM but don't anymore.
Problem about run time
- Thread starter MikeYuan
- Start date
- Status
Thanks for your help, Mr. Oleson.
Now, I have a confusion about spin-up. If I wanna use the compset I1850Clm50BgcCrop to spin-up 100years, and I found that the default set is "DATM_CLMNCEP_YR_START=1991, DATM_CLMNCEP_YR_END=2010", shoud l change it as "the DATM_CLMNCEP_YR_END =2090"? ,if I wanna run 100years.
I appreciate it very much,
Mike
The I1850Clm50BgcCrop compset will come out of the box with:
"DATM_CLMNCEP_YR_START" value="1901"
"DATM_CLMNCEP_YR_END" value="1920"
not 1991 and 2010.
The I2000Clm50BgcCrop compset will come out of the box with:
"DATM_CLMNCEP_YR_START" value="1991"
"DATM_CLMNCEP_YR_END" value="2010"
Either way, the idea of a spinup is to repeatedly loop over a number of years of atmospheric forcing, in this case either 1901-1920 or 1991-2010.
You shouldn't need to change DATM_CLMNCEP_YR_END.
These compsets come out of the box with this setting in datm_in:
taxmode = "cycle", "cycle", "cycle", "cycle", "cycle"
This tells the datm to cycle or loop through the specified years.
So if you wanted to spinup for 100 years, the datm will repeat the forcing 100/20 = 5 times.
"DATM_CLMNCEP_YR_START" value="1901"
"DATM_CLMNCEP_YR_END" value="1920"
not 1991 and 2010.
The I2000Clm50BgcCrop compset will come out of the box with:
"DATM_CLMNCEP_YR_START" value="1991"
"DATM_CLMNCEP_YR_END" value="2010"
Either way, the idea of a spinup is to repeatedly loop over a number of years of atmospheric forcing, in this case either 1901-1920 or 1991-2010.
You shouldn't need to change DATM_CLMNCEP_YR_END.
These compsets come out of the box with this setting in datm_in:
taxmode = "cycle", "cycle", "cycle", "cycle", "cycle"
This tells the datm to cycle or loop through the specified years.
So if you wanted to spinup for 100 years, the datm will repeat the forcing 100/20 = 5 times.
Well, I got it, thank you very much , Mr.Oleson. By the way, I wonder another question. If I wanna simulate permafrost soil, should I change the compset? And how to check the region of permafrost soil, is there an individual module about simulating permafrost in CLM5?
I appreciate it very much,
Mike
There is some discussion about permafrost in the CLM5 paper (section 4.5):
David M. Lawrence, et al., 2019: The Community Land Model version 5: Description of new features, benchmarking, and impact of forcing uncertainty. Journal of Advances in Modeling Earth Systems, 11, 4245-4287. https://doi-org.cuucar.idm.oclc.org/10.1029/2018MS001583.
and in the CLM4 paper (section 4.6):
In the CLM4 paper, near-surface permafrost extent was defined as the integrated area in which at least one soil layer within the uppermost 3.8m remains below 0C through the year. So one could use the history variable TSOI to diagnose this.
Another variable of interest is "ALT" (active layer thickness; the depth to which permafrost soils thaw each summer). See src/biogeophys/ActiveLayerMod.F90.
TSOI should be available in any configuration. ALT is available in BGC mode and can be turned on in SP mode as well.
David M. Lawrence, et al., 2019: The Community Land Model version 5: Description of new features, benchmarking, and impact of forcing uncertainty. Journal of Advances in Modeling Earth Systems, 11, 4245-4287. https://doi-org.cuucar.idm.oclc.org/10.1029/2018MS001583.
and in the CLM4 paper (section 4.6):
In the CLM4 paper, near-surface permafrost extent was defined as the integrated area in which at least one soil layer within the uppermost 3.8m remains below 0C through the year. So one could use the history variable TSOI to diagnose this.
Another variable of interest is "ALT" (active layer thickness; the depth to which permafrost soils thaw each summer). See src/biogeophys/ActiveLayerMod.F90.
TSOI should be available in any configuration. ALT is available in BGC mode and can be turned on in SP mode as well.
Dear Mr.Oleson
I am using the compset "IHistClm50BgcCrop", now, I wanna do a sensitivity analysis experiment. The control factor is temperature and CO2, If I wanna explore the impact of temperature and CO2 on GPP/NPP, Where to modify the code to increase or decrease the temperature and CO2?
Looking forward to your reply~
I appreciate it very much
Mike
To modify the CO2 in that compset, I would look at the CO2 stream which should be something like the file pointed to by:
datm.streams.txt.co2tseries.20tr
In release-cesm2.1.3, the file is:
/glade/p/cesmdata/cseg/inputdata/atm/datm7/CO2/fco2_datm_global_simyr_1750-2014_CMIP6_c180929.nc
So you could modify that file accordingly.
For a constant CO2 compset, e.g., an I2000 one, I think you could modify the CO2 value using CCSM_CO2_PPMV in env_run.xml
To modify temperature, you could manipulate the forcing temperature in the forcing file you are using, for example, the GSWP3 field TBOT.
Or modify the temperature after it has been read in by the datm. cime/src/components/data_comps/datm/datm_comp_mod.F90 might be one place to do that.
datm.streams.txt.co2tseries.20tr
In release-cesm2.1.3, the file is:
/glade/p/cesmdata/cseg/inputdata/atm/datm7/CO2/fco2_datm_global_simyr_1750-2014_CMIP6_c180929.nc
So you could modify that file accordingly.
For a constant CO2 compset, e.g., an I2000 one, I think you could modify the CO2 value using CCSM_CO2_PPMV in env_run.xml
To modify temperature, you could manipulate the forcing temperature in the forcing file you are using, for example, the GSWP3 field TBOT.
Or modify the temperature after it has been read in by the datm. cime/src/components/data_comps/datm/datm_comp_mod.F90 might be one place to do that.
Hello, Mr.Oleson.
In terms of the DGVM you mentioned, I found the compset "I2000Clm50BgcDvCrop" from CESM2.1.2, does it mean there is a DGVM?
Besides, I wanna ask a question about the simulation for the future scenario, if I want to simualte the carbon cycle of year 2100, which compsets could I choose?
View attachment 2948
Looking forward to your reply~
Thank you very much.
Sincerely,
Mike
There is, but it is likely not fully functional and we don't support it.
You have a choice of compsets for future scenarios, the SSPs. They are setup to run from 2015-2100. The "I" versions of these compsets use 2015-2100 forcings except the atmospheric forcing is present day by default.
Alternatively, you could force the model in "CPLHIST" mode in which the atmospheric forcing is from output of the fully coupled model in a given SSP scenario. This requires a lot of disc space to store the forcing data however and the model speed is slower.
Also available is "anomaly forcing" (1.5.9. Running with anomaly forcing — ctsm release-clm5.0 documentation).
Another alternative is to run the coupled model, e.g., the "B" compsets, to fully simulate future climate.
You have a choice of compsets for future scenarios, the SSPs. They are setup to run from 2015-2100. The "I" versions of these compsets use 2015-2100 forcings except the atmospheric forcing is present day by default.
Alternatively, you could force the model in "CPLHIST" mode in which the atmospheric forcing is from output of the fully coupled model in a given SSP scenario. This requires a lot of disc space to store the forcing data however and the model speed is slower.
Also available is "anomaly forcing" (1.5.9. Running with anomaly forcing — ctsm release-clm5.0 documentation).
Another alternative is to run the coupled model, e.g., the "B" compsets, to fully simulate future climate.
Dear Mr.Oleson,
I found that there is no flanduse_timeseries in the I2000Clm50BgcCrop. So, If I am using the ISSP585Clm50BgcCrop and I wanna fix the land use condition with the same as the present-day condition(I2000Clm50BgcCrop), how should I deal with it?
Looking forward to your reply.
Thank you very much
Mike
You could do this in your user_nl_clm:
do_harvest = .false.
do_transient_crops = .false.
do_transient_pfts = .false.
flanduse_timeseries = ' '
Simply setting flanduse_timeseries to a blank string might be sufficient but it doesn't hurt to include the other settings.
Depending on the model version, the fsurdat might be set to 1850 by default. So then set fsurdat in user_nl_clm to the surface dataset used in I2000Clm50BgcCrop.
do_harvest = .false.
do_transient_crops = .false.
do_transient_pfts = .false.
flanduse_timeseries = ' '
Simply setting flanduse_timeseries to a blank string might be sufficient but it doesn't hurt to include the other settings.
Depending on the model version, the fsurdat might be set to 1850 by default. So then set fsurdat in user_nl_clm to the surface dataset used in I2000Clm50BgcCrop.
Dear Mr.Oleson,
You said the there is not DGVM any more. If I wanna the vegetation types change in response to climate, how to implement it by using the "I" compenset (e.g. I1850..or ISPP...)
Looking forward to your reply.
Thank you very much
Mike
The old CLM-dgvm was last documented in this paper using clm4/cesm1:
Evaluation of the new CNDV option of the Community Land Model: Effects of dynamic vegetation and interactive nitrogen on CLM4 means and variability
CK Gotangco Castillo, S Levis, P Thornton - Journal of Climate, 2012
I don't think that the CLM-dgvm worked very well (or at all) in later versions of the model.
The new CLM-dgvm is FATES. You can find corresponding documentation and articles online.
Evaluation of the new CNDV option of the Community Land Model: Effects of dynamic vegetation and interactive nitrogen on CLM4 means and variability
CK Gotangco Castillo, S Levis, P Thornton - Journal of Climate, 2012
I don't think that the CLM-dgvm worked very well (or at all) in later versions of the model.
The new CLM-dgvm is FATES. You can find corresponding documentation and articles online.
- Status