Multiscale study on the spatial heterogeneity of remotely-sensed evapotranspiration in the typical Oasis of Tarim Basin

Abstract

Time series of regional evapotranspiration (ET) of the typical oasis in the Tarim Basin were investigated and calculated by integrating remotely sensed surface parameters into the improved Surface Energy Balance Algorithm for Land (SEBAL) model. The distributed average and maximum daily actual ET was calculated based on the instantaneous ET, corresponding to oasis vegetation types for investigating the relationships between spatial difference of actual ET and land cover types in arid environment. Different land cover types and surface water conditions in the Keriya Oasis can be distinguished obviously in NDVI-LST feature space and characterized by the Temperature Vegetation Dryness Index (TVDI). Lacunarity index was employed to determine heterogeneity and spatial patterns of ET landscape in the Keriya Oasis by using RS and GIS techniques. Spatially distributed scale-dependent regularities of land cover types across a range of patch sizes were revealed by changes of lacunarity curves. It is explicitly suggested that lacunarity index is quite effective to quantify the scale-dependent ET spatial heterogeneity and useful to figure out the water use pattern of oasis-desert landscape.