using user datasets to run a simulation for a specific region

[Xueli Huo]

Hi

I am reading the CLM5.0 Documentation 1.6.3. Running Single Point Configurations — ctsm release-clm5.0 documentation
1.6.3.4. Example: Using CLM_USRDAT_NAME to run a simulation using user datasets for a specific region

In the regional example, only domain and surfdata were replaced with the user datasets. I am wondering to run a regional simulation using the user datasets whether the other inputdata files (the path for default files is /glade/p/cesmdata/cseg/inputdata) such as PFT physiology and parameter file, snow optical properties and aging parameter files, urban parameter, fire parameter and nitrogen decomposition files should also be replaced ? If so, is there any documentation that guides the generation of these files for a regional run ?

Sincerely,

 

[oleson]

The other datasets are either independent of resolution/domain or will be interpolated to your domain, so you shouldn't have to replace any of them.

 

[Xueli Huo]

Thanks, Keith. That really helps.

Another question about preparing the domain and surfdata is as I know there are two ways to prepare the two files. One is using the CLM tools such as mksurfdata_map illustrated in the CLM5 documentation. Another way is to use the python script under the $CTSMROOT/tools/contrib/subset_surfdata. Which one do you think is a better way to do it ?

Sincerely,

 

[oleson]

If you are just going do a regional run at the same spatial resolution as the global dataset that you specify as input to subset_surfdata, then subset_surfdata is a reasonable choice. Keep in mind that it is a contributed script and as such is not really supported if you run into trouble. The mksurfdata_map tool is supported and gives you the flexibility to create different grids. It is then of course more complicated.

 

[Xueli Huo]

Thanks, Keith.

So the subset_surfdata works in the way that extracting the surface data in the region from the global surface dataset. If the resolution for the region is not supported, i.e., no supported global dataset with the intended resolution, the way to create the surfdata for the region is to use mksurfdata_map, right ?

Another question is that what does the simyr represent in naming the surfdata, say,
surfdata_4x5_simyr1850_c090928.nc
surfdata_4x5_simyr2000_c090928.nc

Sincerely,

 

[oleson]

Right.

The simyr refers to the year of the data that is year-dependent on the surface dataset. E.g., simyr1850 means that the pft distribution and lai are representative of year 1850 conditions.