The NOAA operational partners in the North American Land Data Assimilation System (NLDAS) include NCEP/EMC and OHD of the NWS and NESDIS/ORA, who have joined with the NLDAS research partners of NASA/GSFC, Princeton and Rutgers University, and the Universities of Maryland, Oklahoma and Washington. These partners have developed, executed, and evaluated a realtime and retrospective uncoupled NLDAS. The NLDAS generates hourly surface forcing (anchored by observation-based solar insolation and precipitation fields) and uses this forcing to drive four LSMs running in parallel to produce hourly output on a common 1/8° grid over a CONUS domain. The paper of Mitchell et al. (2004) contains all the references of other papers published within this project that measure the quality of the forcing data and assess the model results.

The analysis of model simulations of this first phase of NLDAS pointed to areas of potential physics and parameter improvement for all participating models. We present an analysis of reruns of the NLDAS core period (1996 - 1999) that were carried out with the improved Noah model (version 2.9), comparing energy and water fluxes over a broad range of scales, surface skin temperature and snow melt and sublimation results.

We also present the progress within NOAA/NCEP/EMC to develop a land surface data assimilation system. This progress includes the development of the adjoint model / tangent linear model for the Noah LSM and it's incorporation into NLDAS. We demonstrate here the improvement achieved in Noah LSM 1-D column-model simulations of soil moisture by means of an "identical twin" experiment, which assimilates the land surface temperature (LST) produced by a Noah LSM control run. Furthermore, we will show results of the assimilation of hourly satellite-derived surface temperature retrieved from NOAA GOES satellites.

Mitchell, K., D. Lohmann, P. Houser, E. Wood, A. Robock, J. Schaake, B. Cosgrove, J. Sheffield, L. Luo, Q. Duan, D. Lettenmaier, R. T. Pinker, D. Tarpley, W. Higgins, J. Meng, A. Bailey, F. Wen, The Multi-institution North American Land Data Assimilation System (N-LDAS): Leveraging multiple GCIP products in a real-time and retrospective distributed hydrological modeling system at continental scale, J. Geophys. Res. 109,D07S90,doi:10.1029/2003JD003823, 2004.