通过涡动相关方法观测的全球通量网可以获取较准确的地面通量观测，但是稀疏的站点分布难以满足区域应用需求，由单源卫星遥感数据反演得到的地表蒸散发难以同时满足高空间分辨率和高时间分辨率的田间尺度水分平衡研究需求。针对这些问题，本项目拟联合Landsat8的高空间分辨率和MODIS的高时间分辨率，以及地面通量观测的高精度，开展基于集成学习和时空融合的高时空分辨率日蒸散发遥感反演方法研究。首先，利用集成学习方法构建考虑观测源区的日蒸散发遥感反演模型，提升地表日蒸散发估算精度。其次，通过建立基于长时间序列的蒸散发时空融合模型，开展Landsat8和MODIS的地表蒸散发时空融合，实现高时空分辨率地表日蒸散发的遥感反演。最后，结合地面通量观测数据和流域水量平衡蒸散发，开展精度验证和评价研究。本项目的实施有助于提高田间尺度的地表蒸散发遥感反演水平，对于农田精准灌溉和提高农业水分利用效率具有重要意义。

Accurate ground-based flux observations could be obtained from the global networks of fluxes measured by the eddy covariance method, but data from sparse distribution of measured sites is difficult to meet the needs of regional applications. Remote sensing data from single-source satellite unable to meet the needs of field-scale water balance research which requiring land surface evapotranspiration with high spatial resolution and high temporal resolution. To solve these problems, this project intends to combine Landsat8 with high spatial resolution and MODIS with high temporal resolution, as well as ground flux observations with high precision, and aims to perform the study of methodologies on remote sensing retrieval of daily evapotranspiration with high spatiotemporal resolution based on ensemble learning and spatiotemporal fusion. Firstly, the project will construct a remote sensing retrieval model of daily evapotranspiration that considering the observation source area by using ensemble learning to improve the estimation accuracy of land surface daily evapotranspiration. Then, a spatiotemporal fusion model based on long time series for evapotranspiration will be constructed to carry out the spatiotemporal fusion of Landsat8 and MODIS, achieving remote sensing retrieval of land surface daily evapotranspiration with high spatiotemporal resolution. Lastly, the estimated daily evapotranspiration will be validated and evaluated using ground-based flux measurements and evapotranspiration from watershed water balance. The implementation of this project will help to improve the study on the remote sensing retrieval of land surface evapotranspiration in the field-scale, which is of great significance for precision irrigation of farmland and the improvement of agricultural water use efficiency.