SSP1 and SSP3 land use timeseries datasets are the same

[James King]

Hi all,

I came across an issue a few months ago which I discussed with NCAR colleagues who resolved it as a science issue on their end, but it looks like the problematic data are still in the public areas of Glade. If you go to the directory containing the default land surface files for CLM:

/glade/p/cesmdata/cseg/inputdata/lnd/clm2/surfdata_map

the datasets

landuse.timeseries_0.9x1.25_SSP1-2.6_78pfts_CMIP6_simyr1850-2100_c181220.nc

and

landuse.timeseries_0.9x1.25_SSP3-7.0_78pfts_CMIP6_simyr1850-2100_c181220.nc

are identical, i.e. the differences between all the variables contained in each file are 0. Just wanted to flag this for the admins - I got confused when plotting land surface change from these files and want to avoid others having the same problem!

Best,

James

 

[oleson]

Thanks, I'll file an issue and we'll see if we can get this fixed.

 

[oleson]

I filed an issue for this:

c181220 SSP126 and SSP370 landuse files are identical and are still in cesmdata · Issue #1634 · ESCOMP/CTSM

Brief summary of bug A user reported these SSP126 and SSP370 landuse files are identical and are still present in cesmdata: /glade/p/cesmdata/cseg/inputdata/lnd/clm2/surfdata_map/landuse.timeseries...

github.com

I've been told that the files can't be removed permanently from both the cesmdata directory and the svn inputdata repository because they are needed to maintain a history for simulations that might have used these files. They are going to remove it from the cesmdata directory on glade only however.

In general, the choice of files used in a simulation should be extracted from the CLM namelist xml files by the case setup, not selected based on looking at the inputdata directories.
We were wondering if there was a tag you or others used that had those particular files listed in the CM namelist xml files, or were they chosen manually based on looking at the inputdata directories?
Thanks for your input.

 

[James King]

Hi Keith,

Thanks for the update. I found the files by looking in the input directories for CLM while setting up some experiments using a user-defined compset within CESM2.2.0 - I was interested in how land use changed in different SSP scenarios so these files seemed like the right place to look.

James

 

[oleson]

Ok, thanks for the info.