Abstract:
Eight years (2002-2010) of remotely sensed AMSR-E snow water equivalent (SWE) retrievals and MODIS snow cover fraction (SCF) observations are assimilated separately or jointly into the Noah land surface model. A multiscale ensemble Kalman filter (EnKF) is used, supplemented with a rule-based update. The satellite data are either left unscaled or are scaled for anomaly assimilation. The SWE assimilation estimates (or their anomalies) are validated against in situ observations at 14 high-elevation SNOTEL sites and 4 lower-elevation COOP sites over a domain in Northern Colorado.

Both downscaling of coarse-scale AMSR-E SWE and assimilation of MODIS SCF data result in realistic spatial SWE patterns. At COOP sites with shallow snowpacks, both AMSR-E SWE and MODIS SCF data assimilation are beneficial, and joint SWE and SCF assimilation shows a significantly improved result for both scaled and unscaled data assimilation. However, in deep snowpack areas, AMSR-E retrievals are typically biased low, causing the assimilation without a priori scaling to deteriorate the seasonal SWE variability for the SNOTEL sites. Furthermore, anomaly SWE assimilation could not improve the interannual SWE variations in the assimilation results, because the AMSR-E retrievals lack a realistic interannual variability. SCF assimilation has only a marginal impact on the deep snowpack SNOTEL locations, because these sites experience extended periods of near-complete snow cover. Across all sites, SCF assimilation improves the timing of the onset of the snow season, but without a net improvement of SWE amounts.