期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2007
卷号：XXXVI-7/C50
出版社：Copernicus Publications
摘要：The Moderate-Resolution Imaging Spectroradiometer flies on two NASA/EOS satellites, , each imaging most of the Earth every day, Terra in the morning, Aqua in the afternoon. MODIS has 36 spectral bands covering wavelengths from 0.4 to 14.4μm, 2 at 250m spatial resolution, 5 at 500m, and 29 at 1km. Using reflectance values from the 7 "land" bands with 250 or 500m resolution, along with a NASA- identified cloud flag that does not always distinguish cloud from snow, we estimate the fraction of each 500m pixel that snow covers along with the albedo (reflectance) of that snow. The model MODIS Snow Covered Area, Grain Size, and Albedo (MODSCAG) is a multiple endmember spectral mixture model coupled with a radiative transfer model of snow directional reflectance. Its products are then used in hydrologic models in several mountainous basins. The daily products have glitches. Sometimes the sensor cannot view the surface because of cloud cover, and even in the absence of clouds, an off-nadir view in a vegetated area "sees" less ground area than a nadir view. Therefore, we must use the daily time series in an intelligent way to improve the estimate of the measured snow properties for a particular day. We consider two scenarios: one is the "forecast" mode, whereby we use the past, but not the future, to estimate the snow-covered area and albedo on
