How should I replace the default raw data with my own data?

[LiwenWu]

Dear Scientists,
I want to run a regional case using my own landuse_timeseries data and surface data. Now I have made my raw datasets into nc format that meets the requirements. According to CLM user guide, I should create mapping files from my output grid to my raw datasets. Am I right? I saw many mapping files in the CTSM/tools/mksurafdata_map/mksurfdata_map.namelist, but they are for the default raw data instead of my own raw data. So if I replace the default data with my own raw datasets，how can I create the mapping files that correspond to my raw datasets (the resolution of my raw datasets are all 0.01°x0.01°)?
What's more, I saw many mksrf_data files in /mksurfdata_map.namelist such as mksrf_fglacier and mksrf_fwetland. These data are included in my landuse_timeseries data that I prepared. Should I extract them separately from the landuse_timeseries files to create new original files? Or is there any other way for CLM to directly read from landuse_timeseries data?
Lastly, I have prepared a list of LAI datasets for the years 2000-2020, and I would like to know how to input dynamic LAI data into CLM? It seems that this section is not mentioned in the user guide.
Any advice would be greatly appreciated. Thank you!

 

[oleson]

If you have all of the data required by the CLM surface and landuse timeseries datasets at 0.01deg resolution, then it seems like you could simply create surface and landuse timeseries datasets manually without having to create mapping files. If some of your raw data is at resolutions other than 0.01deg, then you would have to create mapping files first and then use mksurfdata_map. The sequence for creating mapping files for a new grid is here:

1.3.4. Creating input for surface dataset generation — ctsm release-clm5.0 documentation

Alternatively, we have a new tool, which has not been officially released but is close, that eliminates the need for mapping files. In particular, this may be useful for you to generate regional datasets.

git clone GitHub - ESCOMP/CTSM: Community Terrestrial Systems Model (includes the Community Land Model of CESM)
cd CTSM
git checkout ctsm5.2.mksurfdata
./manage_externals/checkout_externals

Further instructions can be found in the README in the checked out directory under tools/mksurfdata_esmf.
You would have to port it and build it on your machine however You could then generate a namelist that you modify to point to your new raw datasets. You'd also need to create a mesh file for your 0.01deg grid.

More detailed discussion/instructions are here:

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

If you create a new example SP case and then set use_lai_streams = .true. in your user_nl_clm and run ./preview_namelists you should see something like this in your lnd_in:

&lai_streams
lai_mapalgo = 'bilinear'
model_year_align_lai = 2001
stream_fldfilename_lai = '/glade/p/cesmdata/cseg/inputdata/lnd/clm2/lai_streams/MODISPFTLAI_0.5x0.5_c140711.nc'
stream_year_first_lai = 2001
stream_year_last_lai = 2013
/

It seems like you could use that method to get the model to read in your own LAI file.

 

[LiwenWu]

Thanks for your kind reply Oleson, as your said, "If some of your raw data is at resolutions other than 0.01deg, then you would have to create mapping files first and then use mksurfdata_map." I can understand what you mean but I still wonder if some of the raw data, such as mksrf_fglacier and mksrf_fwetlnd, are replaced by my own 0.1 ° raw data, how should their corresponding mapping (such as map_fwetlnd) be modified?

 

[oleson]

I assume that you would just use the mapping files that you generated as described above.