在全球环境变化的背景下，区域及全球尺度地表参数和能量循环的研究中迫切需要高分辨率、长时序的土壤水分数据。本项目拟通过协同新一代多观测模式的主-被动微波和光学遥感数据，改进传统的变化检测方法，评估其与裸土/植被微波散射模型在田间尺度的土壤水分反演方面的精度，研究微波反演土壤水分方法对植被含水量的不同参数化方案的敏感性，以准确估算田间尺度的表层土壤水分。在此基础上，拟对申请者在青年基金项目中提出的流域尺度同化系统的诊断指标进行多尺度特性验证，将其推广到田间尺度，以解决模型背景场与观测值自相关差异所引起的同化效果降低问题，进而提高田间尺度的根层土壤水分预报精度。本研究将有效扩充土壤水分的微波遥感反演方法和陆面同化系统的理论体系，并为农业干旱监测预警、农田用水管理提供科学保障和技术支持。

In the context of global environment change, high-resolution and long-term soil moisture retrieval is crucial for the study of water and energy cycles at regional and global scales. This research proposes to synergistically utilize the new generation SAR (Synthetic Aperture Radar) system, i.e., Sentinel-1 and optical remote sensing data sets for accurate estimation of high-resolution soil moisture. Due to the high revisit frequency of Sentinel satellite, we proposed an improved change detection method which applies its core hypothesis of near-constant surface roughness and vegetation cover in the consecutive SAR signals. To assess the applicability of proposed method, intercomparison will be conducted between retrievals from conventional/improved change detection methods and those from microwave scattering model. In addition, the sensitivity of microwave-based soil moisture retrieval methods to vegetation index products from various satellite platforms will be analyzed. As data assimilation (DA) is the most effective approach for acquiring root-zone soil moisture information, it is employed in this research to estimate field-scale root-zone soil moisture. In the previous project supported by NSFC, the applicant proposed a diagnostic index for the coarse-resolution DA system to avoid degradation in the assimilation performance caused by model/observation auto-correlation mismatch. In this research, the scalability of the previously proposed diagnostic index will be carefully examined, and adjustment will be made to the diagnostic index according to model operators of different spatial scales, in order to make it properly function in this high-resolution DA system and retrieve accurate root-zone soil moisture. This research will supplement the theoretical framework of microwave-based soil moisture retrieval methods and land surface data assimilation system, and research results will be valuable in agricultural drought monitoring practice at various scales.