Impact of new land boundary conditions from MODIS data on the climatology of land surface variables

This study uses the Community Land Model (CLM2) to investigate the improvements of a new land surface data set, created from multiple high-quality collection 4 Moderate Resolution Imaging Spectroradiometer data of leaf area index (LAI), plant functional type, and vegetation continuous fields, for modeled land surface variables. The previous land surface data in CLM2 underestimate LAI and overestimate the percent cover of grass/crop over most of the global area. For snow-covered regions with abundant solar energy the increased LAI and percent cover of tree/shrub in the new data set decreases the percent cover of surface snow and increases net radiation and thus increases ground and surface (2-m) air temperature, which reduces most of the model cold bias. For snow-free regions the increased LAI and changes in the percent cover from grass/crop to tree or shrub decrease ground and surface air temperature by converting most of the increased net radiation to latent heat flux, which decreases the model warm bias. Furthermore, the new data set greatly decreases ground evaporation and increases canopy evapotranspiration over tropical forests, especially during the wet season, owing to the higher LAI and more trees in the new data set. It makes the simulated ground evaporation and canopy evapotranspiration closer to reality and also reduces the warm biases over tropical regions.