- Source: /models/lnd/clm/tools/mksurfdata
CLM4 differs from CLM3 in that the surface dataset is created offline, using fortran code included in the CESM release package:mksurfdata. The surface dataset is created from seven mksrf files that contain ‘raw’ data on 0.5 o (e.g., mksrf_glacier.nc) or 1o lat/lon grids (e.g., mksrf_lanwat.nc). The raw data include high-resolution maps of land cover (PFTs or plant functional types), soil color, soil texture, leaf/stem areas and heights (LAI), land water (lakes and wetlands), glaciers, and urban areas (Table 30).
Hint: We recommend identifying your surface dataset by adding your case ID and date. Place the surface dataset in your $CASEROOT/SourceMods/src.clm/ directory for future reference. Then point to your new surface dataset in clm.buildnml.csh:
Example: set fsurdat = ‘surface_data_LGM_64x128.110101.nc’
Quaternary modelers can often use the default CLM4 mksrf files. However, some time periods may require modifications to the present day mksrf files to simulate land cover or vegetation change.
If your simulation requires drastic revisions to the ‘raw’ mksrf files, Deep Time Sections describes a tool for deep time experiments where continental configurations are very different from modern, and/or little is known about land cover and soil distributions.
NOTE: If you have lowered sea level and exposed new land along the continental shelf, CLM4 will automatically define that new land as ‘wetland’. To reassign these points to another PFT you will need to modify your mksrf_pft file using a nearest neighbor algorithm to fill in the new land points. You will also need to define LAI, SAI, and soil color for these cells.
NOTE: Modifying mksrf_pft.nc can be tricky and drastically changing vegetation may result in a climate signal that is larger than the forcing (e.g., solar) that you are trying to simulate.
mksurfdata optimization on bluefire:
% gmake SMP=TRUE OPT=TRUE -j 64
% setenv OMP_NUM_THREADS 64