A Global Snowmelt Product Using Visible, Passive Microwave and Scatterometer Satellite Data

Seasonal snow cover is a key component of the Earth’s energy balance and a key storage mechanism for water. In many areas of the world, people rely on snowmelt runoff for their water resources. For example, melting snow contributes upwards of 70% of the total annual water supply in the western U.S., and in India, Pakistan, Afghanistan, and Nepal snow and ice melt from the Hindu Kush and Himalayan ranges is a vital resource for nearly 1 billion people. The ability to characterize snow storage more accurately at the drainage basin scale is crucial for improved water resource management. Snow-water equivalent (SWE), snow cover, and melt onset are critically-needed parameters for climate modeling and the initialization of forecasts at weather and seasonal-time scales. Snowmelt data are also needed in hydrological models to improve flood control and irrigation. In addition, knowledge of snowpack ripening is essential for natural-hazards applications such as flood prediction. This purpose of this paper is to show results of a blended, global snowmelt product utilizing Moderate Resolution Imaging Spectrometer (MODIS), Advanced Microwave Scanning Radiometer for NASA’s Earth Observing System (AMSR-E) passive microwave data and QuikSCAT scatterometer data. These data are being blended into a single, global, daily, user-friendly product. Though snow-cover extent is currently a product available from various sensors, including MODIS, and snow water equivalent (SWE) is an available product from AMSR-E, snowpack ripening and snowmelt are currently unavailable as satellite-derived products. Nonetheless, there is a significant body of literature showing that AMSR-E radiometric data and QuikSCAT scatterometer data are highly suited to these tasks. AMSR-E data will be used to estimate the onset of snowmelt by looking at increases in brightness temperatures. QuikSCAT will be used to detect melting snow at 14 GHz Ku-band (13.4 GHz) backscatter is sensitive to snow wetness properties. The detection of wet snow from space-borne microwave data is possible because a strong and sudden increase of the imaginary part of snow permittivity (and as a consequence of brightness temperature) occurs when liquid water particles appear within the snowpack, even in small amount. If snow is dry during the night (either frozen or refrozen) and it becomes wet during the day, then daytime brightness temperature will be higher than those acquired during nighttime. Also, if melting persists during the night, then the brightness temperature will remain high. At the wavelength of 2.2 cm (Ku-band, 13.4 GHz), QSCAT backscatter is highly sensitive to snow wetness. In wet snow, liquid water (no salinity) has an imaginary part of about 19,000 times larger than that of dry ice. Thus, a small amount of wetness can significantly change the imaginary part of the snow effective permittivity, and consequently decrease the backscatter. The initial blended-snowmelt product will be 25 km resolution and validated (to begin with) using data from the lower Great Lakes region of the U.S. and from data gathered in Colorado at Cold Lands Project Experiment (CLPX) sites in 2002 and 2003.