Global Land Surface Temperature Retrieval with AMSR-E Brightness Temperature and MODIS Land Cover Type Products

Abstract

AMSR-E brightness temperature data and MODIS land cover type data are combined to retrieval land surface temperature(LST).Using the MODIS data,we classified the global land surface into 16 types,and make assumption that in each land surface type,the emissivity of each channel is constant separately.Based on this assumption and the land surface types,we build land surface retrieval algorithms for each land surface types.Before the algorithms were built,we chose the pixels that with above 90% single land type area and pixels that the MODIS land surface temperature exceed 273 K.This is because the emissivity is dubious in snow cover,ice cover,and mixed-pixels.Coequally,considering the precipitation effect,we winkled the rainfall pixel from our dataset.Finally,we employed the algorithms to retrieval global land surface temperature of Jan to Oct,2004.Compared with the MODIS land surface temperature product,the result is fine,with RMSE of 2~4 K.