Study on the spatial variation of albedo at Urumqi river basin of China, using remote sensing

Using Landsat TM image and DEM, three surface parameters of surface albedo, land cover types and Normalized Difference Vegetation Index (NDVI) at Urumqi river basin are estimated in this research. Then spatial distribution pattern of surface albedo and the relationships between albedo and other two parameters are analyzed. The results show that: 1) in general, surface albedo of the basin appears an obvious zonal distribution, which verifies with the range from 0.005 to 0.708. This pattern maybe caused by the topographic factors such as altitude, slope and aspect, whereas the correlation between surface albedo and altitude appears unobvious; 2) surface albedo verifies greatly with land cover types, albedo of glacier is the largest while that of coniferous forest is the lowest, albedo of glass is larger than that of forest and rock, which is mainly due to the different spectral characteristics of land covers; 3) surface albedo decreases with the increase of NDVI for each of land cover types, the relationship between surface albedo and NDVI for vegetation regions is much closer than that for non-vegetation regions, the closer correlation with the increase of NDVI. The above conclusions could provide the scientific reference to the researches on vegetation transformation, climate and water resource for the basin in spatial scale.