干旱区绿洲与荒漠植被蒸散发及其组分定量遥感反演研究

Evapotranspiration (ET), which includes evaporation from soil and transpiration from vegetation, is a major hydrological variable that links water, energy and carbon cycles. Therefore, accurate estimations of ET are required for understanding local to global water and energy cycles and for improving hydrological applications, water resources management, drought monitoring and assessment, irrigation scheduling and climatological studies. Although substantial progress has been made in ET studies, accurately estimating ETs based on remote sensing is a challenge, particularly in arid regions with large barren or sparsely vegetated areas over heterogeneous landscapes. In addition, separating ET into soil evaporation (Es) and vegetation transpiration (Ec) is difficult. To address these challenges, objectives of this proposed project are to: (1) assess the impact of soil moisture and temperature on ET for oasis and desert regions in northwestern China based on field experiments, and then to improve the three-temperatures model accordingly, which has been tested to be accurate in our previous project supported by NSFC, with relative errors generally within ~20%; (2) analyze the uncertainty of air temperature of the proposed algorithm for modeling ET; (3) analyze the uncertainty of scaling remotely sensed instantaneous ET to daily or longer values; (4) test and validate the developed method using ground datasets observed in arid northwestern China (mainly the Heihe River and Shiyanghe River Basins); (5) generate continuous ET (including Es and Ec) using the proposed method and Moderate Resolution Imaging Spectroradiometer (MODIS) data; and (6) quantify the spatiotemporal distribution of the ET in the arid northwestern China. If this project is founded, this study will develop a new scientific method and an optional ET algorithm for remotely estimating ET for arid regions. In addition, the product for the soil evaporation and vegetation transpiration components of a given pixel can be generated to obtain detailed information for practical applications, such as water resources management.

蒸散发及其组分（土壤蒸发和植被蒸腾）的准确定量反演是水热研究领域的热点与难点。干旱区植被稀疏、地表异质性高，蒸散发受温度与水分胁迫严重，使其反演面临极大挑战。在青年基金资助下，申请者发展了独具特色、较简单、可反演蒸散发及其组分的三温模型，在干旱区初步验证表明反演误差较小（约20％），但模型参数化对温度与水分胁迫考虑不足、蒸散发组分缺少观测数据验证等亟需系统研究。为此，本项目拟在中国西北干旱区开展典型样地水热通量、氢氧同位素等观测，揭示温度与水分对蒸散发的胁迫机理，依此改进三温模型，定量评估瞬时气温空间参数化对蒸散发反演的影响、正弦曲线法关键变量物理边界对蒸散发时间尺度扩展的影响等共性问题，借助通量数据，检验并完善考虑水、温胁迫参数化方案的三温模型，生成时空连续的高精度蒸散发及其组分产品。研究成果可望揭示水温胁迫对蒸散发建模的影响、提高蒸散发及其组分反演精度，为水热研究提供理论和应用参考。

围绕蒸散发及其组分（土壤蒸发和植被蒸腾）准确定量反演的难题，项目组利用我国西北黑河流域HiWATER共享数据，通过在宁夏、内蒙古等区域开展定位监测、样地调查、氢氧同位素采样等方法，在获取大量蒸散发及其组分资料基础上，开展蒸散发过程与土壤水分等要素的关系研究，以独具特色、输入参数较少、具有自主产权的三温模型为研究对象，从模型关键要素参考温度参数化角度改进和完善了三温模型遥感应用，多角度、系统地定量评估了三温模型反演的蒸散发及其组分结果，发展了适用于中国西北干旱地区、可精确反演蒸散发及其组分的三温模型；解明了气象站点空间组合方式及高程对气温空间插值结果的影响，提出了考虑地形校正的气温空间插值方法；在此基础上，利用完善后的三温模型初步研发了全球尺度0.25°(25 km)的蒸散发月尺度产品。研究成果丰富了蒸散发及其组分测算与反演方法，为干旱区研究提供了蒸散发产品。截至2020年12月，项目组已发表论文11篇（含接收待刊4篇），其中SCI论文5篇、EI论文1篇（第一资助），发表在《Journal of Geophysical Research: Atmospheres》、《Journal of Hydrometeorology》、《Environmental Research Letters》、《IEEE Geoscience and Remote Sensing Letters》、《Sustainability》等相关领域权威期刊；另有一项发明专利申请已通过国家专利局初步审查，部分成果仍在整理撰写中（至少2篇论文）；项目负责人在内的3名固定参与人员从讲师晋升为副教授；项目负责人联合培养博士研究生2名（已毕业1名）、独立培养硕士研究生4名（已毕业2名）。成果超出预期目标。