Using GLDAS/LIS to derive global land climatology for the NOAA Climate Test Bed

Accurate assessment of land surface states, namely, soil moisture, soil temperature, vegetation, and snowpack, is critical in numerical environmental prediction systems because of their regulation of surface water and energy fluxes between the surface and atmosphere over a variety of spatial and temporal scales. To provide the NOAA Climate Test Bed (CTB) optimal land climatology, an uncoupled, land-component only, Global Land Data Assimilation System (GLDAS) will be implemented. GLDAS, developed jointly by NASA Goddard Space Flight Center (GSFC) and NOAA National Centers for Environmental Prediction (NCEP), aims to perform high-quality land surface simulation using state-of-art land surface models and further minimizes the errors of simulation by constraining the models with observation-based precipitation and radiation, and satellite land data assimilation techniques. The GLDAS-based Land Information System (LIS) infrastructure has been successfully ported to the NCEP supercomputer that serves the CTB framework. In this implementation, the NCEP Global Reanalysis-2 and the CPC Merged Analysis of Precipitation (CMAP) will be used to drive the Noah land surface model. The latest version of Noah has been coupled to the NCEP Global Forecast System (GFS) for operational weather prediction and test bed versions of the NCEP Climate Forecast System (CFS) for seasonal prediction. It is crucial that the uncoupled GLDAS/Noah use exactly the same Noah code (and soil and vegetation parameters therein), and execute on the same horizontal grid, landmask, terrain field, soil and vegetation types, seasonal cycle of green vegetation fraction and surface albedo as in the coupled GFS/Noah and CFS/Noah. The GLDAS/Noah will be executed for a period from 1979-2004, allowing the first two years to serve as a spin-up period, thus providing a 24-year retrospective of GLDAS land states. This GLDAS retrospective will be used for both climate variability assessment and as a source of land-state initial conditions for ensemble CFS/Noah seasonal hindcast experiments. Finally, this GLDAS retrospective will provide the GLDAS climatology needed as the foundation for a GLDAS-based global drought/flood monitoring system that includes realtime daily updates of the GLDAS land states.