Against irrigation_on, evolving land cover lowers the median NSE from 0.106 to 0.097 even though more gauges improve than degrade (547 better, 414 worse in NSE). Details: Results (Configuration sensitivities).
Case card¶
| Item | Value |
|---|---|
| Baseline | Because evolving_landcover also has irrigation on, its land-cover effect is measured against irrigation_on, not control (Results). |
| Configuration | Common setup (Configuration summary) |
| Setup script | setup_landcover.sh, see Code and data (Scripts) |
| Spinup start | 1980-10-01 |
| Spinup length | 22 yr |
| Spinup final restart | 2002-10-01 |
| Evaluation run | evolving_landcover |
| RUN_REFCASE / RUN_REFDATE | evolving_landcover / 2002-10-01 |
| CTSM finidat | evolving_landcover.clm2.r.2002-10-01 |
| mizuRoute fname_state_in | evolving_landcover.mizuroute.r.2002-10-01-00000.nc |
Evaluation run¶
evolving_landcover: check_finidat_year_consistency=.false., because the restart year does not match the land-use year at the evaluation start. David Lawrence expected this to work with SP.
Issues and fixes¶
Check build-namelist defaults. Supplying flanduse_timeseries with a matching fsurdat turns irrigation on by default; preview_namelists shows it before the run.
Regional grids: no default land-use time series exists for the NLDAS2 grid, and mksurfdata_esmf warns against regional grids (CTSM issue #2430). The evolving land-cover files were generated by Sean Swenson.