Running CLM5 with your atmosphere forcing

[wvsi3w]

Hello everyone,
I have a question that might sound trivial:

I have looked into these links link 1 , link 2 , link 3 , link 4 , and I couldn't decide what variables we need to give the CLM5 to run it with our data.
I was thinking of using the last millennial data from another institution, which used high and low variabilities. In order to get their data I should tell them exactly what we need to run CLM5 using their data.
In the mentioned 4 links the first one is for CLM4.5 and I couldn't find something similar to that for CLM5.
The next 3 links are for CLM5 but they are either for Carbon forcing or previously run by CESM or for anomaly forcings.

I am guessing the first link is more related to my answer but that one is unfortunately referring to CLM4.5.
Can you tell me what link to follow, or maybe what specific detailed variables we need in order to run CLM5 (in my case, using last millennial high and low variability dataset).

Thanks a lot in advance

P.S. I also looked at this link in the threads of this forum but I don't think it is similar to my thing.

 

[wvsi3w]

I think this thread also had similar question to mine (I am not sure if what I am planning to do is the same thing this thread did)

Also these links (link 7, link 8 ) are somehow related. but they are referring to problems they had with running it, I am currently at the beginner stage (kind of a noob :) )

 

[wvsi3w]

I think this table (2.2.4) is all we need for this purpose, or am I missing something?

 

[slevis]

An approach that I find helpful is to look at the default atmosphere data that the model reads. This is a quick way of finding out the variables needed and their format.

 

[wvsi3w]

An approach that I find helpful is to look at the default atmosphere data that the model reads. This is a quick way of finding out the variables needed and their format.

Thank you very much,

I inspected the atmospheric forcing files under the path atm_forcing.datm7.GSWP3.0.5d.v1.c170516 (The files are at a 0.5°x0.5° spatial resolution, covering 6-hourly time steps) and found that they include the following:

Precipitation (PRECTmms):
Stored in files under the Precip directory. The data provides total precipitation in mm H2O/sec.

Solar Radiation (FSDS):
Found in the Solar directory, representing total incident solar radiation in W/m².

Surface Pressure (PSRF), Temperature (TBOT), Wind Speed (WIND), Specific Humidity (QBOT), and Longwave Radiation (FLDS):
All these variables are in the TPHWL directory. They provide the necessary atmospheric conditions at the lowest atmospheric level, which I guess is critical for driving CLM5.

But there is an issue which I don't know why these data are only for 1901 till 1920 and I was using the year 1958 for my test runs before. Do you think there is something missing or maybe I am looking into the wrong path?

 

[wvsi3w]

Thank you very much,

I inspected the atmospheric forcing files under the path atm_forcing.datm7.GSWP3.0.5d.v1.c170516 (The files are at a 0.5°x0.5° spatial resolution, covering 6-hourly time steps) and found that they include the following:

Precipitation (PRECTmms):
Stored in files under the Precip directory. The data provides total precipitation in mm H2O/sec.

Solar Radiation (FSDS):
Found in the Solar directory, representing total incident solar radiation in W/m².

Surface Pressure (PSRF), Temperature (TBOT), Wind Speed (WIND), Specific Humidity (QBOT), and Longwave Radiation (FLDS):
All these variables are in the TPHWL directory. They provide the necessary atmospheric conditions at the lowest atmospheric level, which I guess is critical for driving CLM5.

But there is an issue which I don't know why these data are only for 1901 till 1920 and I was using the year 1958 for my test runs before. Do you think there is something missing or maybe I am looking into the wrong path?

There is also a domain file in that path (scratch/cime_INPUT_DATA/cesm2_1_3/inputdata/atm/datm7/atm_forcing.datm7.GSWP3.0.5d.v1.c170516):
domain.lnd.360x720_gswp3.0v1.c170606.nc

Which I don't know if I need to make something similar to this for my own forcing (sorry if it sounds silly).

Also in the main path (scratch/cime_INPUT_DATA/cesm2_1_3/inputdata/atm/datm7) we have these:
atm_forcing.datm7.cruncep_qianFill.0.5d.v7.c160715 atm_forcing.datm7.GSWP3.0.5d.v1.c170516 CO2 NASA_LIS topo_forcing

And I only mentioned the files inside one of the folders above (from GSWP3 folder). So I think I would need to make (have) all of these files from the forcing I want to use (?)

I am asking these because the institute needs to know everything before they give me the forcing data.

Thanks a lot

 

[oleson]

GSWP3 data is available for 1901-2014. It may be that your specific case is only setup to use 1901-1920 and thus that is the only data that was downloaded. You can look at "DATM_CLMNCEP_YR_START" and "DATM_CLMNCEP_YR_END" in your case directory.
Yes, you will need to create a domain file that matches your atmospheric forcing resolution and mask.
You only need files similar to the ones in the GSWP3 directory. atm_forcing.datm7.cruncep_qianFill.0.5d.v7.c160715 is a different atmospheric forcing dataset.

 

[wvsi3w]

GSWP3 data is available for 1901-2014. It may be that your specific case is only setup to use 1901-1920 and thus that is the only data that was downloaded. You can look at "DATM_CLMNCEP_YR_START" and "DATM_CLMNCEP_YR_END" in your case directory.
Yes, you will need to create a domain file that matches your atmospheric forcing resolution and mask.
You only need files similar to the ones in the GSWP3 directory. atm_forcing.datm7.cruncep_qianFill.0.5d.v7.c160715 is a different atmospheric forcing dataset.

Thank you very much for your response,

I checked every case I have created recently and Yes they are using 1901 to 1920 for the atm forcing.
I think this 1901-1920 is the default and I think I should have changed these dates before running the test (I mean as far as I know each test must be using the same period in atm, therefore if I am testing it for 1958 the atm files should start with 1958 too, unless we are going to do some "spinup to reach equilibrium" runs and that's a different subject).

So I need to do these:
- make the domain file as you mentioned (which I think it should be done using the mksurfdata_map tool) that matches my atmospheric forcing resolution and mask.
- and have the rest of the files (PRECT, FSDS, PSRF, TBOT, WIND, QBOT, FLDS) that are similar to what is inside the GSWP3 directory (and it should be 0.5°x0.5° spatial resolution?) (I will find a way for the temporal resolution conversion from monthly to 6hourly)

By the way, I think I should have these directories as well in my own forcing "CO2 NASA_LIS topo_forcing", right? these are inside the "inputdata/atm/datm7" path, and I don't think without them the model will run. So I need to ask for these from the institute (or maybe I personally can generate them with the tools in the model?).

Thanks again.

 

[oleson]

Right, changing the DATM dates would provide you with the appropriate forcing for 1958.
Doesn't necessarily have to be 0.5deg forcing, could be some other resolution as long as your datm domain file describes the grid.
Other datm streams include co2 (if transient), aerosol deposition (transient in a transient simulation), and topo (static even if a transient simulation). However, it seems to me as if you could use the model default for those.
There are other transient files that are specified in lnd_in, including NASA_LIS for lightning, but again it seems like you could just use the defaults.
I assume you really only need to replace only the meteorological forcing (e.g, GSWP3).

 

[wvsi3w]

Hello again,

Please ignore the message in this thread (CLM output restart file missing) because my question there was not related and a bit wrong.

I have some questions regarding the last millennial simulation how they do it and how I can use the data for my purpose.

From this thread, I have found that the finidat file in lnd_in is spunup already. And from the LME archive and related papers (like these: paper1, paper2, paper3, paper4, etc) I found that with CESM they have used data from 850 to 1850 (or even 850 until 2100).

Q1- Let's say I want to do a simulation from 850 to 2000 (or 2100). How can I use the spunup file from one of these datasets, and how can I ask for it? I want to save time for the spinup and with having that spunup file that was done for 850 at 1 or 2 degree resolution I could save a lot of time.
In the link of LME, there is no restart file from the 850 spinup simulation. Like for example here it says "forcings fixed at 850CE conditions" which means they used 850 and did the spinup with that year for several years of spinup time (I don't know maybe it took 200y or 500y or ...), so there must be a restart file for that.

The reason I want that is to start my transient run from 850 and continue until let's say 2100 or 2300 with my modified soil layer structure that has deeper values and to give land enough time to capture surface temperature signals to reach that depth. So using spunup finidat files that are from 1850 is not enough because in 150 years or so it (temperature) wouldn't reach that depth so we need at least 1000y (even from the year 1000 is good for us). So it would be really helpful to know how I can have that restart file and more importantly what the process is (explained in Q2 below)

Q2- So lets say we have the restart file from 850 and I can start my run (transient) from that point, then we need all the required forcing data from that point of time like "PRECT, FSDS, PSRF, TBOT, WIND, QBOT, FLDS" and "CO2 NASA_LIS topo_forcing" that are from 850 till now. So how can I get these? rationally speaking the team who has the restart file should also be able to give us these files too, or maybe I am wrong. Is this email the one that I should contact "esg-support@earthsystemgrid.org"? The link says dataset available to community but I see that these are simulation results and not forcings or restart files. or maybe I should contact the author of the paper for LME project?

forgive my numerous questions because I really don't know if I can have these data or not and how I should obtain them.

Q3- what is the process of implementing these forcings? would it be the same as "using your own forcing to run", which is the topic that is discussed in the forum a lot? Or are there other precautions to consider? For example, the case to create should use a compset, and all compsets are 1850 onward so I assume that wouldn't be an issue, and later on we can change that right? because our restart is for 850 and our transient is from 850 onward so the initial compset doesn't apply here.

I really need to run this and our purpose is also to use NorESM's last 1000y data later. Hence figuring out this first could provide us a lot of insight for the future NorESM simulation.

Thank you for your support and consideration @oleson

 

[oleson]

I'm going to move this to the paleoclimate forum. I don't know about the availability of restart and forcing files for LME simulations. Presumably, if you want to run an LME in land-only mode, then you would run with the datm in CPLHIST mode using coupler history files generated from the LME. However, I don't know if these would be available.