Guidance on Running with LME datasets with CESM2.1.3 (CLM5)

[slevis]

The path of the file
inputdata/atm/cam/solar/SolarForcing1995-2005avg_c160929.nc
and the values that you quoted
1361.1
suggest to me that this file contains the top-of-atmosphere solar radiation used by an active atmosphere model (CAM). If you do not use an active atmosphere model, then I suspect that this file and its values are ignored.

I do not have answers about volcanic forcing. I hope somebody else will answer.

 

[wvsi3w]

The path of the file
inputdata/atm/cam/solar/SolarForcing1995-2005avg_c160929.nc
and the values that you quoted
1361.1
suggest to me that this file contains the top-of-atmosphere solar radiation used by an active atmosphere model (CAM). If you do not use an active atmosphere model, then I suspect that this file and its values are ignored.

I do not have answers about volcanic forcing. I hope somebody else will answer.

Dear Sam, Thank you so much for your answer.

I understand. So in a fully coupled simulation or a CLM-CAM simulation, I could use my tsi instead of the default one (?)

 

[slevis]

I understand. So in a fully coupled simulation or a CLM-CAM simulation, I could use my tsi instead of the default one (?)

I have never tried this myself, but I do not see why not.

 

[wvsi3w]

I have never tried this myself, but I do not see why not.

And do you know if we can assign its path to a different directory in our case, like we do for other forcings?
I looked at the CaseDocs and user_namelists, but I dont see where we can do that.

Because if I need to just remove that file from inputdata/atm/cam/solar/ and replace it with my file then I should wait until first member is finished then replace it with another member solar file which is not rational.

 

[wvsi3w]

And do you know if we can assign its path to a different directory in our case, like we do for other forcings?
I looked at the CaseDocs and user_namelists, but I dont see where we can do that.

Because if I need to just remove that file from inputdata/atm/cam/solar/ and replace it with my file then I should wait until first member is finished then replace it with another member solar file which is not rational.

And another question would be why we have this path inputdata/atm/cam/solar/ in IHIST compsets, but we dont have it in I1850 compsets? The first one is historical, the second one is used when we want constant values of urban, landuse, etc.

I personally cant see the reason why we dont have that path in I1850 compset inputdata. In I1850, we only see cam/chem path and nothing else.

 

[slevis]

And do you know if we can assign its path to a different directory in our case, like we do for other forcings?
I looked at the CaseDocs and user_namelists, but I dont see where we can do that.

Because if I need to just remove that file from inputdata/atm/cam/solar/ and replace it with my file then I should wait until first member is finished then replace it with another member solar file which is not rational.

I have not tried this myself but here are my thoughts:

I suspect that the solar file of your choice could end up in atm_in via user_nl_cam or maybe it would be specified in a "streams" file. Overwriting the default file with your file seems the easiest immediate solution, though it could confuse you in the future, because you could forget which data you actually used for your various simulations. In any case, you could keep trying these ideas until something works.

You could also post in the CAM Forum where people are more likely to know about this file.

About your last question, I am surprised in the opposite sense than you are:
I would expect no such file in I1850, I2000, IHist or any I-compset because, from my understanding, this is top-of-atmosphere radiation that matters only if you have an active atmosphere model in your compset. If such a file is specified in I-cases, then I would expect no effect from changing it because I would not expect top-of-atmosphere data to be used in an I-case.

Again, you may get more conclusive responses in the CAM forum about this file.

 

[wvsi3w]

Hello again,
I would like to ask another question related to my experiment. I would be glad to know your thoughts on it dear @slevis and dear @oleson

First, to remind us of my simulation conditions:
Starts at year 500CE, ends in 2000. Has 43m bottom boundary in the shallow version (using the 23SL_3.5m soil layer structure) and 500m in deep version (modified 23SL_3.5m), uses two members of monthly 1500 years of forcings (started from an initial condition (Spun-Down) that picked first 30years of forcings from each member to better capture the climatology of that time (500-529CE)), my members of 1500y forcings have the variables required for land-only simulations (TPHWL, Prec, Solar) and are monthly temporal resolution, 1deg spatial. I have used (I1850Clm50BgcCrop --res f09_g17) for my 500 to 1850 period (that has constant CO2 and aerosol of 1849, and has no landuse change), and I have used (IHistClm50BgcCrop --res f09_g17) for my 1850-2000 period that has transient CO2, etc for that period.

So, the simulations are done and I have compared my results with the observation, and I see that there are some differences between my deep model and the shallow (default) model in regards to the end year of the simulation (year 2000) for permafrost extent, but it is not very significant. Therefore, based on several studies that extended their simulation onwards to the future, I was wondering how I could do the same thing here?

What compset should I use from 2000 to 2100? I mentioned that I used IHistClm50BgcCrop for 1850-2000 (with my own forcing scenarios), but with SSP scenarios, how should I do it, especially because my forcings were monthly, and I think the future scenarios available as compset in CLM5 are 6-hourly or finer resolution?

I see that in the compset list, there are 8 types of SSP compset there, but I am not sure if I could just pick one (ISSP245Clm50BgcCrop). Does it start from 2000, or would I miss 20 years, as they probably start from 2020?

I would like to start this future scenario from my own restart file of my already finished simulation that ended in 2000 (from 1500 years of simulating monthly), so that I could see what would happen to my experiment in the future using different future scenarios if we start them with my own initial conditions.

P.S. Also, any Idea if there are developed future scenarios available for 2100 onwards (lets say till 2300?)

 

[slevis]

In CLM5 I think that you will find that IHist simulations can continue to 2014. Both the landuse and the datm files include data for these years. And ISSP simulations start in 2015.

Running out to 2300 sounds familiar, but I am not finding data readily available. I searched "2300" in the Forums and found a few mentions. This is the first, and you may find interesting information in this one and/or the others:

How to do a BSSP126 run with cesm2.1.3 and resolution f09_g17 with the extended SSP126 scenario for years 2100-2300

Dear all, I did a BSSP126 run with CESM version 2.1.3. and with resolution f09_g17 run for years 2015-2100. It was setup with command: ./create_newcase --run-unsupported --case b.e21.BSSP126_BPRPcmip6_hosing.f09_g17.001 --compset...

bb.cgd.ucar.edu

 

[wvsi3w]

In CLM5 I think that you will find that IHist simulations can continue to 2014. Both the landuse and the datm files include data for these years. And ISSP simulations start in 2015.

Running out to 2300 sounds familiar, but I am not finding data readily available. I searched "2300" in the Forums and found a few mentions. This is the first, and you may find interesting information in this one and/or the others:

How to do a BSSP126 run with cesm2.1.3 and resolution f09_g17 with the extended SSP126 scenario for years 2100-2300

Dear all, I did a BSSP126 run with CESM version 2.1.3. and with resolution f09_g17 run for years 2015-2100. It was setup with command: ./create_newcase --run-unsupported --case b.e21.BSSP126_BPRPcmip6_hosing.f09_g17.001 --compset...

bb.cgd.ucar.edu

Dear Sam, thank you so much for your response and the link you provided.

About the gap that I mentioned, the problem was that I am using two different forcings (my own forcings) that end in 2000, and they have different TSI, so I can not simply pick an IHist compset and fill the gap of 2000 to 2015 (because IHist has one TSI in it). I have asked my source to extend their run so that I could have this different TSI output for the mentioned period, as my forcings, and they agreed. I will provide the results here when it is finished.

I have a question: You mentioned that SSPs start in 2015, I checked the input data link (- Revision 70792: /trunk/inputdata/atm/datm7) and couldn't find anything related to these SSP compsets. Do you know if there is a link for these SSP forcings that the model reads and downloads from
? or there is no such thing?

 

[slevis]

If I understand your question, I think you need to read this section of the user's guide:

1.5.9. Running with anomaly forcing — ctsm release-clm5.0 documentation

 

[wvsi3w]

If I understand your question, I think you need to read this section of the user's guide:

1.5.9. Running with anomaly forcing — ctsm release-clm5.0 documentation

Oh, ok, I thought there were different sets of datasets for future scenarios. Cool, I get it now. Thanks.

So, unfortunately, right now my cluster system is down and I can't run any simulation, but later when it is up and running, I only need to create a new case using ISSP compsets that starts from 2015, then give my finidat restart file to the namelist file to start from, then I should follow the steps in the link (anomaly forcing) you provided for stream files and datm namelist file? But, when we start a new case with ISSP compsets, wouldn't it automatically set those files for us?

 

[oleson]

It doesn't come out of the box with release-cesm2.1.3, which is what I assume you are still using...

 

[wvsi3w]

Dear Keith @oleson and Dear Sam @slevis
Thanks for your responses,

1) I have access to new forcing data from 2000 to 2021 (for Precip, Solar, TPHWL), and I was wondering what I should set DATM_PRESAERO to?

Because I think RCPs (e.g., RCP4.5 or RCP8.5) could be used for the years 2006-2021, but what to do for 2000 to 2006?

Should I create a case where I start from 2000 to 2006 and use aerosol forcing of clim_2000, and then a new case with aerosol forcing of RCPs for 2006-2021?

2) How about "datm.streams.txt.co2tseries.20tr" or CLM_CO2_TYPE? Or even landuse timeseries file?

Thanks again for your support.

 

[wvsi3w]

Dear Keith @oleson and Dear Sam @slevis
Thanks for your responses,

1) I have access to new forcing data from 2000 to 2021 (for Precip, Solar, TPHWL), and I was wondering what I should set DATM_PRESAERO to?

Because I think RCPs (e.g., RCP4.5 or RCP8.5) could be used for the years 2006-2021, but what to do for 2000 to 2006?

Should I create a case where I start from 2000 to 2006 and use aerosol forcing of clim_2000, and then a new case with aerosol forcing of RCPs for 2006-2021?

2) How about "datm.streams.txt.co2tseries.20tr" or CLM_CO2_TYPE? Or even landuse timeseries file?

Thanks again for your support.

Please ignore this message because I think I realized how I should do it (correct me if I am wrong):

The run from 2000 to the end of 2014 should be using these:
- datm.streams.txt.presaero.trans_1850-2000 (because the file has it until 2015: aerosoldep_WACCM.ensmean_monthly_hist_1849-2015_0.9x1.25_CMIP6_c180926.nc)

- datm.streams.txt.co2tseries.20tr (fco2_datm_global_simyr_1750-2014_CMIP6_c180929.nc)

Then I create a new case from 2015 to the end of 2020 using these:
co2tseries.SSP1-2.6
DATM_PRESAERO => SSP1-2.6

I still don't know what to do with landuse file in these.

 

[slevis]

Seems reasonable. For many such files, the SSP version starts in 1850 (or 1849 or 1750 or similar). So you may find that the co2 and aerosol files for your SSP1-2.6 simulations start all the way back, in which case you could point to the same files before and after 2015.

When you set up the SSP case, I think it will pick up a corresponding landuse file, as well. Only the anomaly forcing will not be available by default if I understood @oleson's comment correctly, so you will still need to look at the User's Guide, as recommended above.

 

[wvsi3w]

Seems reasonable. For many such files, the SSP version starts in 1850 (or 1849 or 1750 or similar). So you may find that the co2 and aerosol files for your SSP1-2.6 simulations start all the way back, in which case you could point to the same files before and after 2015.

When you set up the SSP case, I think it will pick up a corresponding landuse file, as well. Only the anomaly forcing will not be available by default if I understood @oleson's comment correctly, so you will still need to look at the User's Guide, as recommended above.

Dear Sam, thank you for the response.
Ok, good.
And I guess I don’t need to overcomplicate things, as my supervisor mentioned in our study using clim_2000, or clim_2010 for aerosol would not change the output much for this 2000-2020 period (therefore suggesting to have one case for 2000 to 2020 with aerosol (and CO2) starting from 2000 and ending in 2015).

But I wanted to try being strict and having rational aerosol for each section (the aerosol of 2000 to 2015 for the first case, then for the second case from 2015 to 2020, using SSP aerosol).

About Keith's comment, I assume it is for when we use ISSP compset, but here I am only changing the aerosol and CO2 in the namelist and stream. Do I need to follow the anomaly forcing you mentioned earlier for this case too (when our compset is IHist but the aerosol and CO2 are changed to be ssp)? Because I think since I already have those TPHWL, Solar, Precip forcings for 2000-2021, I don't need to go through the steps of that anomaly link, right?

 

[slevis]

You are correct, the anomaly forcing is for later, i.e. the ssp phase of the simulation.

 

[wvsi3w]

Dear Sam @slevis and Dear Keith @oleson

I have finished my simulations from the year 500 CE to 2020 successfully and I am going to use anomaly forcing (two SSP scenarios, 4.5 and 8.5) until 2100.

I have done this so far, and it worked ok until now:

cime/scripts/create_newcase --case /home/USER/links/projects/rrg-XXX/USER/MYCASES/SSP45_Mem1_Deep --compset ISSP245Clm50BgcCrop --res f09_g17 --run-unsupported

cd /home/USER/links/projects/rrg-XXX/USER/MYCASES/SSP45_Mem1_Deep

./case.setup

./xmlchange STOP_OPTION=nyears,STOP_N=5,JOB_WALLCLOCK_TIME=03:59:00,RUN_STARTDATE=2021-01-01,STOP_DATE=21000101,RESUBMIT=15
./xmlchange DATM_CLMNCEP_YR_END=2014,DATM_CLMNCEP_YR_START=2000,DATM_CLMNCEP_YR_ALIGN=2000


emacs user_nl_clm

use_init_interp = .true.
soil_layerstruct= '23SL_3.5m_D500'

finidat ='/home/USER/links/projects/rrg-XXX/USER/MYCASES/SSP45_Mem1_Deep/DeepMem1_2015to2021_IHist.clm2.r.2020-01-01-00000.nc'




emacs user_nl_datm

anomaly_forcing = "Anomaly.Forcing.Precip","Anomaly.Forcing.Temperature","Anomaly.Forcing.Pressure","Anomaly.Forcing.Humidity","Anomaly.Forcing.Uwind","Anomal
y.Forcing.Vwind","Anomaly.Forcing.Shortwave","Anomaly.Forcing.Longwave"


fillalgo = "nn", "nn", "nn", "nn", "nn", "nn", "nn", "nn", "nn", "nn", "nn", "nn", "nn"

streams = "datm.streams.txt.CLMGSWP3v1.Solar 2000 2000 2014",
"datm.streams.txt.CLMGSWP3v1.Precip 2000 2000 2014",
"datm.streams.txt.CLMGSWP3v1.TPQW 2000 2000 2014",
"datm.streams.txt.presaero.SSP2-4.5 2015 2015 2101",
"datm.streams.txt.topo.observed 1 1 1",
"datm.streams.txt.co2tseries.SSP2-4.5 2015 2015 2500",
"datm.streams.txt.Anomaly.Forcing.Precip 2015 2015 2100",
"datm.streams.txt.Anomaly.Forcing.Temperature 2015 2015 2100",
"datm.streams.txt.Anomaly.Forcing.Pressure 2015 2015 2100",
"datm.streams.txt.Anomaly.Forcing.Humidity 2015 2015 2100",
"datm.streams.txt.Anomaly.Forcing.Uwind 2015 2015 2100",
"datm.streams.txt.Anomaly.Forcing.Vwind 2015 2015 2100",
"datm.streams.txt.Anomaly.Forcing.Shortwave 2015 2015 2100",
"datm.streams.txt.Anomaly.Forcing.Longwave 2015 2015 2100"



./case.build

question:
These base files (e.g., datm.streams.txt.CLMGSWP3v1.Precip) that I have set to 2000-2015, what exactly do they do here? I know the anomaly files are in regard to the base, but how should we set these start and end years of datm.streams.txt.CLMGSWP3v1.Precip or solar or TPQW file? Would it be different if I set it to 1900-2015? I mean, for what purpose someone sets it to 2000-2015, and for what purpose someone else sets this to a different period??

If my study is focused on permafrost and it started from year 500CE to 2020, and the anomaly files are until 2100, then what should I set the start and end date for CLMGSWP3 stream files? Does it even matter what year we set these GSWP3 stream files?

 

[oleson]

You've set it to 2000-2014, not 2000-2015 right?
Regardless, it probably doesn't matter too much what you set it to as long as it is some "present-day" period compatible with the GSWP3V1 forcing.
There are several versions of the anomaly forcing files that are available, not sure which ones you are using. But the earlier ones had anomaly/scaling factors calculated relative to 2000-2014, hence that's why the GSWP3V1 forcing was set to 2000-2014. Later ones had anomaly/scaling factors calculated relative to 2005-2025. GSWP3V1 was only available through 2014, so the forcing was set to 2005-2014.

 

[wvsi3w]

You've set it to 2000-2014, not 2000-2015 right?
Regardless, it probably doesn't matter too much what you set it to as long as it is some "present-day" period compatible with the GSWP3V1 forcing.
There are several versions of the anomaly forcing files that are available, not sure which ones you are using. But the earlier ones had anomaly/scaling factors calculated relative to 2000-2014, hence that's why the GSWP3V1 forcing was set to 2000-2014. Later ones had anomaly/scaling factors calculated relative to 2005-2025. GSWP3V1 was only available through 2014, so the forcing was set to 2005-2014.

Hello Keith,
Thank you so much for your response.
Yes, I have set them to 2000-2014 (DATM_CLMNCEP_YR_END=2014,DATM_CLMNCEP_YR_START=2000)

Ok, great to know that.

Yes I am trying to use SSP4.5 and 8.5, which I guess I should download manually first from - Revision 70792: /trunk/inputdata/atm/datm7/anomaly_forcing and set the downloaded path in the user_datm.stream files.

Thanks again for your help