This page borrows liberally from Gautam's excellent CLM4.5 dataset for year 1850 at ne120np4 resolution. It adds some more information on how to incorporate
...
cd $SCRATCH/ACME/components/clm/tools/clm4_5/mksurfdata_map/
For 1950 high res:
./mksurfdata.pl -res ne30np4 ne120np4 -d -y 1850-2015 -dinlc /global/project/projectdirs/acme/inputdata -usr_mapdir /global/project/projectdirs/acme/inputdata/lnd/clm2/mappingdata/maps/ne120np4
For 1850 - 2015 low-res:
./mksurfdata.pl -res ne30np4 -years 1850,2015 -d -dinlc /global/project/projectdirs/acme/inputdata -usr_mapdir /global/project/projectdirs/acme/inputdata/lnd/clm2/mappingdata/maps/ne30np4
4. Modify namelist file if land cover information should be used
...
mksrf_fvegtyp = '/scratch2/scratchdirs/ritvik/ACME/components/clm/tools/clm4_5/mksurfdata_map/luh2/mksrf_landuse_rc_1850_10202016.nc'
5. Modify hardcoded reading of text in mksurfdat.F90Make
cd $SCRATCH/ACME/components/clm/tools/clm4_5/mksurfdata_map/src
vi mksurfdat.F90
Modify this line: read(nfdyn, '(A195,1x,I4)', iostat=ier) string, year
This line splits the text read in from mksrf_fdynuse. Currently the A195 value is hard-coded and needs to be changed any time the path lengths in mksrf_fvegtyp change
>> gmake clean
>> gmake
6. Log onto a debug or regular queue at EDISON
...
7. Create surface dataset
srun ./mksurfdat_map < namelist