Running CLM5 and DATM in different resolutions

[ChrisPp]

Dear colleagues

Reposting from the CLM forum.

I am running CLM5 regionally in a 3 km resolution over the EU-CORDEX domain. Thus far, our model has worked with forcings that we downscaled to this 3km EU-CORDEX grid without problems. Now, I need to run the same clm5 model but with 12 km resolved forcings (over the same domain). I created a new domain file that matches the 12 km grid of the forcings, passed it to the model as ATM_DOMAIN_FILE, and specified it also in the stream files. I thought that CLM5 could interpolate the forcings (with the "mapalgo"). Unfortunately, I get an error:

(prep_atm_init) : Initializing mapper_Fl2a
(seq_map_init_rcfile) called for mapper_Fl2a initialization
(seq_map_gsmapcheck)gsmap global sizes different 2458048 174688
ERROR: (seq_map_gsmapcheck) different gsmap size

Is there any way to make this work?
Best regards
Christian

 

[oleson]

I think this file in your datm_in:

domainfile = "/p/scratch/cjibg31/jibg3105/projects/SCENIC/forcing_domain/domain.lnd.EUR-11_focus_EUR-11_focus.241213_final_pos_pos.nc"

should actually be:

domainfile = "/p/scratch/cjibg31/jibg3105/CESMDataRoot/InputData/share/domains/EUR-0275/domain/domain.lnd.EUR-0275_final.nc"

 

[ChrisPp]

Dear Keith Oleson
Thanks a lot for your response. Yes, I tried that, too, but I got an error "sooner" in the initialization of atm, so I assumed the other setting was correct. Maybe it worked for the wrong reasons. When I change the atm domain file to the same as lnd (as you suggested) while keeping the right atm domain file in the stream files, I get this error:

(shr_strdata_init)  calling shr_dmodel_mapSet for remap
(shr_map_checkGrids_global) ERROR Xsrc not monotonic        1  350.0361137390      349.9361200333   
(shr_map_checkGrids_global)  ERROR Xsrc not regular lat,lon        1       2  349.8890037537      349.8890037537   
(shr_map_checkGrids_global)  ERROR Ysrc not regular lat,lon        2       1  22.02783775330      21.98782920837   
 ERROR: ('shr_map_checkGrids_global')  ERROR

As is seen in the attached atm log file. I understand the problem: Our domain spans from the left to the right of the prime meridian, so we have longitudes 350..360 and then 0 to 20..
A monotonic increase of lat and lons is also not necessarily true for the CORDEX (curvilinear) grids.

When the forcings align in resolution with the CLM5 simulation (both 3km), I do not get this error; it just works, even though the problem should be the same. I think some map checks in the beginning are then left out.

Do you know of any way I could solve or bypass this? The other solution would be to manually regrid the forcings again to 3km, but I would like to avoid this.

 

[oleson]

Unfortunately, I'm not personally aware of a way to solve or workaround this.
Based on this post maybe @xiulingao would be able to help since it looks like they have used curvilinear grid forcing (WRF) before?

Run CLM5 with downscaled 9km WRF forcing

Hi there, I'm trying to create a customiazed regional simulation in California annual grasslands using 9km forcing data from WRF. I never did this before, so my questions might be very broad, and I'm trying to get some suggestions to start with. So basically my quesitons are: 1) WRF outputs...

bb.cgd.ucar.edu

I think this was part of a broader discussion here (maybe):

Setting up (high-res sparse) regional-grid CTSM simulations · ESCOMP CTSM · Discussion #1919

Context This started as issue #1773 from which I linked a google doc (now out of date and obsolete) with step-by-step instructions intended initially for a case that @XiulinGao needed to run. To th...

github.com

 

[ChrisPp]

Thank you for your reply and the resources. If I find any solution, I will post it here.