基于静止气象卫星数据的区域尺度土壤水分反演方法研究

Soil moisture is the link of surface water, ground water and biological geochemistry circulation. It plays a significant role in the formation, transformation and consumption of water resources, and is closely related with the hydrology, climate, ecology, agriculture and some other relative fields. On the basis of the geostationary meteorological satellite and its multi-temporal and multi-spectral characteristics, the project aims to develop a regional soil moisture retrieval model using the FY-2D/E data, which is supposed to make it possible to retrieve soil moisture with part-time cloudy in the whole day by the optical remote sensing. Based on the daily evolution of LST (land surface temperature) and NSSR (net surface shortwave radiation), the relationship between LST and NSSR would be acquired theoretically by the mathematical deduction. In addition to the theoretical deduction, a strong physics-based land surface model was conducted to investigate the inner relationship between LST and NSSR with different soil moisture status. Further with the validation by field measurement data, soil moisture retrieval model will be acquired. Finally, with the new generation of geostationary meteorological satellite (FY-2D/E), the proposed soil moisture retrieval model will be applied to estimate regional soil moisture with the support of scaling transformation. Meanwhile, in order to enrich and improve the theory and method of soil moisture retrieval, the demo application will be also conducted to monitor regional soil moisture using remotely sensed data.

土壤水是水循环过程中的一个重要环节。本项目从深层挖掘静止气象卫星多时相、多光谱遥感数据的角度出发，构建利用静止气象卫星定量反演区域尺度土壤水分的模型，以完善国内外光学遥感在全天部分有云条件下土壤水分反演方法。首先基于地表温度日变化模型和地表短波净辐射日变化模型，通过数学推导，从理论上找出地表温度与地表短波净辐射之间的关系模型；然后基于地表能量平衡原理和热传导方程，采用具有较强物理机制的陆面模型来探求不同土壤水分条件下地表温度与地表短波净辐射之间的内在关系，并利用地面观测数据对模型进行真实性检验，以确定土壤水分与地表温度-地表短波净辐射之间的关系模型；最后利用新型静止气象卫星数据（FY-2D/E），结合尺度转换，将已建模型拓展到卫星层次，实现区域尺度土壤水分全遥感定量反演，并开展相应的遥感动态监测区域土壤水分的应用示范，进一步丰富和完善遥感定量反演土壤水分的理论和方法。

本研究首先对晴天条件下观测的白天地表温度与地表短波净辐射的散点呈现近似椭圆这一现象从理论上进行了推导，建立了白天地表温度-地表短波净辐射的椭圆关系模型；在此理论基础上，利用陆面过程模型的模拟数据建立了裸土表层土壤水分定量反演模型；为了使发展的表层土壤水分反演模型更具实用性，本研究还进一步分析了自然地表下垫面植被与土壤质地对裸土表层土壤水分反演模型的影响。基于陆面过程模型的模拟数据，分别分析了裸土表层土壤水分反演模型与模型系数在低植被覆盖(植被覆盖度范围0-0.3)和中-高植被覆盖(植被覆盖度范围0-0.7)条件下的可行性。初步实现了利用具有统一的模型形式和物理基础的模型直接定量反演自然地表条件下表层土壤水分的目标；最后利用发展的表层土壤水分反演模型，分别基于欧洲第二代静止气象卫星数据和我国风云静止气象卫星数据进行了初步的表层土壤水分反演的应用研究。结果表明：基于静止气象卫星数据反演的表层土壤水分表现出较高的精度，其与土壤水分实测值以及微波土壤水分产品之间在时间序列上均具有较好的相关性。. 本研究首次将时间序列的遥感信息用于表层土壤水的反演，实现了利用地表温度与地表短波净辐射来直接反演表层土壤水分的目标，具有重要的科学意义和实际意义。