It is estimated that the annual value of water from the major watersheds being forecasted in California is ~$3.3 billion, and the corresponding value of hydropower is equal to ~$1.3 billion, with the specific value of the forecasting program estimated at $140-230 million. The proposed project aims to assess the potential benefits of using remote sensing and earth science datasets to improve these forecasts, with a special emphasis on drought years where the value of forecasts is highest and current accuracy is lowest. The activities in Stage 1 (Year 1) will focus on clearly establishing whether remote sensing data can enhance the existing forecast system via in-depth examination of two test basins. Our Stage 1 analysis will span 30 years of available remote sensing and earth science datasets. We will analyze the ability of the remote sensing measurements to quantitatively improve the drought forecasts in water years 1982 - present. We will perform focused analysis of the use of the remote sensing measurements during the droughts of 1987-1992 and 2007-2009. If approved for Stage 2 (Year 2-4) we will move to extend the approach to other basins across the Sierra Nevada. We hypothesize that the proposed work will enhance the existing systems, which are based primarily on historical regression relationships. Also as part of Stage 2 we propose to investigate whether a remote sensing data assimilation system could prove to be a useful new framework for use in the future by the partner agencies, where it is hypothesized that less reliance on regression relationships and more reliance on physically-based methods will provide further enhancements to water supply and drought forecasts. The region and system chosen is an excellent testbed due to its scale and complexity, and lessons learned in this project are expected to not only benefit CA, but could be extended to other regions that rely heavily on snowmelt for their water resource supplies.