Impact of different surface layer schemes on surface fluxes in the Noah land model used at NCEP

Tests with the Noah land-surface model at times show large biases in surface fluxes when compared with observations from several flux sites. For example, these biases can be in the form of significant negative sensible heat fluxes in stable conditions over snow, but are not confined to the cold-season. Such biases may be due in part to the treatment of the atmospheric surface-layer. We begin with Noah land-model-only (off-line) runs using different surface layer schemes (both explicit and implicit forms), with atmospheric forcing and validation data from several surface flux sites in the U.S., and examine terms in the surface energy budget, plus skin temperature, aerodynamic conductance, and snow-water-equivalent during the cold-season. This is followed with "2-D" off-line runs using the North-American Land Data Assimilation System (NLDAS) 2005-2006 annual forcing data for the NLDAS domain (continental U.S.) as a testbed. Finally, Noah-model/surface-layer tests are extended to coupled land-atmosphere runs in the North American Mesoscale (NAM/WRF-NMM) and Global Forecast System (GFS) models at NCEP.