Satellite-derived reflectance of snow-covered surfaces in Northern Minnesota

Abstract

Its high albedo and low thermal conductivity make snow a key component of the global energy balance. Snow depth and reflectance are crucial inputs to General Circulation Models (GCMs). Satellite data from the Landsat Thematic Mapper (TM) have been used to estimate the reflectance of snow and snow-covered surfaces integrated over part (0.45–0.90 μm) of the reflective range of the electromagnetic spectrum. The integrated reflectance, R 1, has been computed for snow-covered agricultural and forested areas, and a frozen lake in northern Minnesota using Landsat TM scenes acquired in November 1984 and January 1985. TM-derived reflectances, corrected for atmospheric effects, are mapped for subscenes within the Minnesota TM scenes and reflectances have been color coded. The average R 1 within the November 1984 subscene was 0.429 ± 0.176 whereas the average R 1 within the January 1985 subscene was 0.669 ± 0.236. Snow reflectance is known to increase with solar and sensor zenith angle. The solar zenith angle was 70° at the scene center of the 17 November 1984 scene, and 75° at the scene center of the 04 January 1985 scene. Both areas were predominately snow covered. Lowest reflectances (near 0%) were found in the non-snow-covered agricultural fields. Reflectances of lake ice in the January scene averaged 0.957 ± 0.0008. In the future, the Moderate Resolution Imaging Spectrometer (MODIS), as part of the Earth Observing System, will enable calculation of R 1 over most of the reflective part of the electromagnetic spectrum. R 1 may then be used, along with bidirectional reflectance data, to calculate snow albedo, an important input to GCMs and regional energy balance models.