Operational snow cover estimation at subpixel scale using NOAA-AVHRR data

摘要

Snow and ice play an important role in the earth's radiation balance because of the high albedo in comparison to other natural surfaces. Furthermore ice and snow is the largest contributor to rivers and ground water over major parts of the middle and high altitudes. These are reasons why hydrological and climatological studies require estimates of snow covered areas. Most of such snow cover maps generated from satellite data include information of snow or not snow for each image pixel. In this study a linear spectral unmixing algorithm is used to calculate snow cover portions within each data cell. We examine the ability of this algorithm for operational and near-real time snow cover estimation at subpixel scale using medium spatial resolution satellite data from NOAA-AVHRR. The automated methodology is presented which produces snow cover fraction maps showing plausible distribution of snow in comparison to TERRA-ASTER data. The qualitative analysis of the results present how suitable the approach implemented in the preliminary processing chain is. Simplifying assumptions are made to the procedure which explains some variation between derived snow cover fraction map and reference data. Further work should include an accurate quantification of areal snow coverage comparison to traditional approaches.