Mapping Evapotranspiration from ASTER Data through GIS Spatial Integration of Vegetation and Terrain Features

Accurate estimation of evapotranspiration (ET) is difficult to obtain over heterogeneous landscapes presenting spectrally diverse land covers and topographic terrains. The goal of this study is to build advanced remote sensing and surface energy balance algorithms to map ET in a heterogeneous semi-arid area. ET of 12 different land covers is computed by applying the Surface Energy Balance Algorithm for Land (SEBAL) modified for roughness, vegetation index, and topographic-corrected reflectance, with comparison to the JIC model. A GIS raster/vector platform is used to integrate multispectral thermal and reflectance imagery with meteorological, terrain, land-cover, and astronomical data. SEBAL computed with all the modifications showed the best agreement with the ground measurements, compared to the SEBAL versions without any single modification, and it could significantly discriminate ET among 75.8 percent of vegetation types (at threshold differences in ET >0.5 mm/day). SEBAL without any modifications could not discriminate any vegetation types.