Analyzing influence of spatial resolution on the estimated evapotranspiration by using remote sensing data over an oasis area in Northwestern China

Abstract. In this paper, the influence of spatial resolution on the precision of estimates was analyzed through evapotranspiration (ET hereafter) modeling over a typical oasis in northwestern China by using the Landsat-TM and MODIS data. A relatively high consistency was observed between the TM-based latent heat flux and daily ET estimates and in-situ measurements, with relative errors of 9.7% and 8.8%, respectively. Despite lower precision of the relative errors of 22.4% and 17.0%, respectively, the MODIS-based latent heat flux and ET estimates can effectively depict the basic trend of the spatial distribution of the land surface processes. When the visible and near-infrared information of 250 m resolution was syncretized into MODIS LST retrieval algorithm, the precision of latent heat flux prediction was improved evidently. Additionally, the diurnal variation of the reference ET fraction shows that the temporal upscaling method of ET is suitable for the study area. In spite of suffering the influence of the heterogeneity of land surface, the moderate resolution MODIS data, combined with the parameterization model of land surface energy flux applied in this investigation, are suitable for the ET mapping at large scale while high-resolution data can serve as an important supplement.