A key indicator of water availability, and the primary input to streamflow models, is April 1 snow water equivalent (SWE), which has historically been monitored from a network of in-situ SNOTEL observing sites across the West, but two remote-sensing-based approaches have recently been developed to complement and extend the SNOTEL network. In the first approach, used by Molotch and group, MODIS satellite snow-cover measurements along with a regression from historic SNOTEL data are combined to reconstruct SWE. In the second, used by Deems and group, airborne LIDAR measurements of snow depth are used to estimate SWE. This project will assess the usability of these spatial SWE products by water managers, and their potential to improve runoff forecasts. In summer 2016, phone interviews were conducted with water managers in the Uncompahgre River and Rio Grande basins to better understand how they have used snowpack data and runoff forecasts in their decision-making, especially managing for drought years and very wet years. Focusing on those recent years in which there were unusual snowpack and runoff conditions that were difficult to monitor, forecast, and prepare for, we will use the spatial SWE products and hydrologic model simulations to assess whether the new SWE products could have provided better information to prepare for those events. PI Molotch and K. Jennings have prepared MODIS data for the two basins to be used in hydrologic models. PI Deems has prepared ASO data from the Uncompahgre Basin for use in the models. K. Jennings leveraged high-resolution snow depth observations and snow water equivalent (SWE) estimates from NASA’s Airborne Snow Observatory (ASO) in order to quantify errors in current snow products. He found the large spatial scale of NLDAS-2 grid cells made the snow product unsuitable for use. Only 8 grid cells completely or partially overlapped the Uncompahgre River Basin and they often showed snow-free conditions when snow was still reported at the two aforementioned SNOTEL sites. Similarly, the SNODAS SWE product performed poorly relative to ASO data (Fig. 1). SNODAS typically overestimated subalpine SWE while significantly underestimating alpine SWE. We have also taken initial steps towards modeling SWE in the Uncompahgre using the Distributed Hydrology Vegetation Soil Model (DHSVM).