Retrieval of the surface evapotranspiration patterns in the alpine grassland–wetland ecosystem applying SEBAL model in the source region of the Yellow River, China

The source region of the Yellow River is located in the eastern of Tibetan Plateau at an altitude of 3900–5300m. Applying the method of unsupervised classification, ISODATA, we defined the land covers of the region into five types, snow cover, water body, bare soil, low coverage grassland, and middle-and-high coverage grassland. To learn the pattern of evapotranspiration (ET) at large scale is essential for the local water use and management. With surface energy balance algorithm for land (SEBAL), we retrieved ET and the energy fluxes with Landsat 5 remote sensing images of the area in the summer of 2006. It was estimated that the instantaneous evapotranspiration was highest in middle-and-high coverage grassland of the southern region at 0.31mm/h on average level, and lowest in bare soil of the northern region at 0.13mm/h on average level. The key variables in SEBAL model, including the normalized difference vegetation index, surface albedo, the surface radiation temperature, net radiation, and sensible heat fluxes were calculated and illustrated to see their patterns in the region. At final, based on the correlation analysis among the key variables, we conducted a conceptual framework of cause–effect loops involving the main factors in the process of ET. Under the framework, we described the typical situations of hydro-heat in the source region.