CTSM resolution problem

[maruf2812]

I was trying to run CTSM cases and looking for 27 km resolution model. I tried this one (./create_newcase --case /glade/u/home/maruf/CLM_test4 --compset I2000Ctsm50NwpBgcCropGswp --res f10_f10_musgs) and found the resolution is very low. Since f10_f10 is not working for me, what can I change for 27 km resolution model run.

Thank you!!

 

[aherring]

Your best bet is to try ne120pg3_ne120pg3_mt13, but I can't be certain that it will work. This grid alias pertains to the 1/4 degree spectral-element unstructured grid. What code base are you running?

 

[maruf2812]

Thank you for your reply. I am running it on WRF-CTSM on Cheynne machine

 

[oleson]

For WRF-CTSM, please see these instructions:

3.4.1. Using CTSM with WRF — ctsm CTSM master documentation

 

[maruf2812]

I was trying to run the cases offline. In that case which resolution should I use to get 27 km domain?

 

[oleson]

I don't think that 27 km grid is supported in offline mode, but I'll ask someone to check on that.

 

[negins]

I don't think that 27 km grid is supported in offline mode, but I'll ask someone to check on that.

I'd like to confirm @oleson : we don't currently have the same 27 km grid supported in CLM standaone.

 

[maruf2812]

Hi @negins thank you for your reply. In that case can we use some other grid options close to 27 km. Our goal is to build high resolution model for whole USA. Can we use NLDAS-2 grid over the U.S which is 0.125 degree resolution. I have attached the screenshot.

 

[oleson]

I tried this with cesm2.2.0:

--compset I2000Ctsm50NwpSpGswp --res nldas2_rnldas2_mnldas2

and that ran for 1 month.
There doesn't appear to be an nldas2 surface dataset for the BgcCrop version of this compset (I2000Ctsm50NwpBgcCropGswp) so you'd have to create one.
You can create one using:

./mksurfdata.pl -res 0.125nldas2 -y 2000

 

[maruf2812]

@oleson thank your kind help. After executing the command ./mksurfdata.pl -res 0.125nldas2 -y 2000 the following error showed up. Could you please enlighten me with some ideas what I need to change?

 

[oleson]

I'm guessing you haven't built the executable (mksurfdata_map). See the README in that directory.

 

[maruf2812]

Hi @oleson thank you for the help. It's working now!! Can we use NLDAS forcing and BGC to create this following case: --compset I2000Ctsm50NwpBgcCropNldas --res nldas2_rnldas2_mnldas2

 

[oleson]

That compset shortname doesn't exist.
You can see available clm compsets using:
./query_config --compsets clm

However, you could pick a compset that is close to what you want (I2000Ctsm50NwpBgcCropGswp) and modify its longname:

2000_DATM%NLDAS2_CLM50%NWP-SP_SICE_SOCN_MOSART_SGLC_SWAV_SIAC_SESP

in your create_newcase command:

--compset 2000_DATM%NLDAS2_CLM50%NWP-BGC-CROP_SICE_SOCN_MOSART_SGLC_SWAV_SIAC_SESP

 

[maruf2812]

Hi @oleson, I appreciate your kind support. I tried with the newcase (--compset 2000_DATM%NLDAS2_CLM50%NWP-BGC-CROP_SICE_SOCN_MOSART_SGLC_SWAV_SIAC_SESP) and this error showed up during case build. Do I need to again create the surface data?

 

[oleson]

You need to tell the model where to find the surface dataset you made and what the name is of that dataset.
Add this into your user_nl_clm in your case directory:

fsurdat = 'YOUR_SURFACE_DATASET'

 

[maruf2812]

Hi @oleson I have added that link in user_nl_clm but same error showed up. not sure if I added it correctly.

 

[oleson]

The exclamation point in front of the statement means that it will be interpreted as a comment.
Put it below this lower line without an exclamation point in front of it:

!-------------------------------------------------------------------------------------------

If you haven't already, I recommend you take a look at the CTSM Tutorial material:

2019 CTSM Tutorial

The National Center for Atmospheric Research is offering a week-long tutorial on the Community Terrestrial Systems Model hosted by the Climate & Global Dynamics laboratory.

www.cgd.ucar.edu

And more broadly the CESM tutorial material and the CESM Distance Learning Course:

Tutorials | Community Earth System Model

www.cesm.ucar.edu

 

[maruf2812]

Hi @oleson thank you very much for the tutorials and suggestions. It is working now.

 

[maruf2812]

Hi @oleson, I am trying to run the same case and forcing for 1 deg resolution. I run this case (2000_DATM%NLDAS2_CLM50%SP_SICE_SOCN_MOSART_SGLC_SWAV_SIAC_SESP --res f09_g17) but this is 1 deg. for global. If I want it for only USA, do I need to follow the same procedure mentioned above, making surface data ./mksurfdata.pl -res 1nldas2 -y 2000?

 

[oleson]

If you want to create a 1 deg surface dataset over just the US then I would probably use the getregional_datasets.ncl script. For release-cesm2.2.0, it is in components/clm/tools/ncl_scripts.
In newer versions of ctsm, there is a tool called subset_data.py in tools/site_and_regional that you could try.