An application of evapotranspiration partitioning for monitoring drought from Sentinel-2 data

Abstract

<p>Drought is one of the globally extreme events and is physically defined as an extended imbalance between moisture supply and demand, which might severely affect crop growth and water preservation. More specifically, agricultural and meteorological droughts are manifest as the deficits in actual evapotranspiration (ET) and a surplus in atmospheric evaporative demand ET<sub>0 </sub>(sometimes referred to as potential ET). Therefore, the accurate estimations and reliable information regarding to ET might enhance the drought monitoring and provide better agricultural management policies. However, ET estimations and its components remain many challenges. For example, as three ET contributors are soil evaporation (ET<sub>soil</sub>), plant transpiration (ET<sub>veg</sub>), and vegetation interception evaporation (ET<sub>ic</sub>), current ET models tend to ignore the ET<sub>ic </sub>and consider it as residual of two others. This leads to the uncertainties and incomprehensive reflection of ET. Additionally, ET models-based flux measurement might produce good accurate results, but they have limitations of spatial coverage. With the rapid development of remote sensing platforms, the models-based remote sensing are able to cover a large and regional scale, but they remain higher uncertainties due to the low spatial resolution, complexities in processing, requirements of many input data. Currently, the optical Sentinel-2 is a newly launched product with the Multi Spectral Instruments that might provide 10, 20, and 60-m spatial resolution, which potentially supports to design and improve ET models with superior performance. To overcome these mentioned disadvantages of ET models, the main objective of this study are to propose a simple and objective method using only optical Sentinel-2 dataset to improve the accuracy of ET estimation; and to project the enhanced ET partitioning as feasible method for further monitoring the agricultural drought. This study might bridge the gap in knowledge and applicability of ET in monitoring the hydrological disasters under severe climate change context.</p>