Global WPS LULC Static Data: ESA WorldCover 2021 Reclassified into the MODIS 21-Category Classification

Updated 2021 Global Land Use Dataset for WRF (MODIS 21-Category, 3 arc-second) This dataset provides an updated, high-resolution land use field for the Weather Research and Forecasting (WRF) model, replacing the default MODIS data distributed with the WRF Preprocessing System (WPS). Key Features: Current to 2021: Built from ESA WorldCover 2021 v200 (Sentinel-1 and Sentinel-2 imagery). It captures two decades of global urbanization, deforestation, and agricultural expansion missing from the default stock data. High Resolution (3 arc-seconds): Aggregated from 10 m to ~90 m resolution. Because the data is significantly finer than typical model grids, geogrid can calculate genuine fractional land use rather than just sampling a single value. Improved Physics: Accurate fractional coverage drastically improves how sensible and latent heat fluxes are partitioned in tiled surface schemes like Noah-MP, improving near-surface temperature, humidity, and boundary layer simulations (especially over cities and coastlines). Drop-in Replacement: Cross-mapped to the modified IGBP MODIS 21-category scheme (num_land_cat = 21, mminlu = 'MODIFIED_IGBP_MODIS_NOAH'). Installs directly into an existing WPS geographical data tree with no recompilation required. Please Note: Lakes are not distinguished from ocean. ESA WorldCover has a single water class, so all water maps to category 17 and category 21 (Lake) is never produced. Retain modis_lakes in the geog_data_res fallback chain if sf_lake_physics is in use. ESA does not resolve forest type, so all forest maps to category 5 (Mixed Forests); categories 1 to 4 are not produced. Specifications: Source: ESA WorldCover 2021 v200 (cross-mapped by majority vote) Resolution: 3 arc-seconds (~90 meters) Coverage: Global (60°S to 84°N) on a 432,000 x 172,800 grid Included: Dataset, modified GEOGRID.TBL, and full setup instructions (README.md). A USGS 24-category version of the same dataset is available at doi:10.5281/zenodo.22674626.