Bidirectional reflectance for multiple snow-covered land types from MISR products

Abstract

Bidirectional reflectance factors (BRFs) play a key role in land surface studies. Snow has a significant influence on vegetative surface BRF. To evaluate the surface reflectance behaviors of snow-covered regions, a surface BRF database has been constructed from Multi-angle Imaging SpectroRadiometer BRF products for five biomes in the mid-high latitude regions of the U.S. (evergreen needleleaf forests, shrublands, grasslands, croplands, and urban areas). Using corresponding surface snow depth data from 26 meteorological stations, BRF signatures with snow cover are derived from the database to show the effect of snow on the BRF of vegetation. Five bidirectional reflectance distribution function models' abilities of capturing vegetation-snow mixed BRF shape are evaluated by fitting all the BRF data with snow. The results show that the Rahman model, Ross-Li model, and Walthall model perform well in fitting forest, grassland, and cropland BRFs when the surface is covered by snow. The Rahman model, Ross-Li model, and Roujean model fit visible reflectance well for mixed surfaces. The Rahman model best captures the BRF shapes, followed by the Ross-Li model. © 2012 IEEE.