青藏高原土壤未冻水的微波遥感监测和数据同化方法研究

Currently, soil moisture retrievals based on the satellite measurements are only developed for thawed soil conditions, and efforts are still needed to develop algorithms for retrieving unfrozen water in frozen soils. This project will first focus on the understanding of microwave emission of soil freeze–thaw process via analysis of ground-based microwave radiometry measurements, and a recently developed integrated land surface process-microwave radiative transfer model will be further validated. In addition, the zero-order approximation of the radiative transfer equations will be updated by the integrated model, and the algorithm for retrieving the unfrozen water in frozen soil will be developed. Furthermore, currently operational satellite measurements of brightness temperature will be evaluated using the ground-based microwave radiometry measurements, and the retrievals of unfrozen water will be developed for the whole Tibetan Plateau, which will be further validated using measurements collected from currently available in-situ soil moisture monitoring networks. Finally, a hydrologic data assimilation system will be developed based on the Noah land surface model and Ensemble Kalman Filter method to investigate the impact of assimilating unfrozen water retrievals on the simulations of surface water and energy budgets across different scales, e.g., site, catchment and regional scales. The findings of this project could contribute to the application of passive microwave remote sensing to monitor the hydrologic processes on the Tibetan Plateau and other cold regions.

针对当前冻土中未冻水动态变化遥感监测产品缺失、反演算法欠完善等问题，本项目将基于近期构建的青藏高原L波段地面微波辐射计观测实验平台，研究土壤未冻水迁移转化对微波辐射过程影响的机理，验证和完善近期发展的陆面-微波辐射过程综合模拟系统，评估和改进SMAP参数化微波辐射模型，并发展地表土壤未冻水的被动微波反演算法。进一步的，利用地面微波辐射计观测数据评估和选取最优卫星亮温观测资料，发展青藏高原地表土壤未冻水动态变化遥感监测产品，并结合地面卫星尺度土壤湿度观测网络验证产品在不同典型区域的反演精度。依据微波遥感反演研究的成果，基于改进的Noah陆面模式和集合卡尔曼滤波方法发展土壤未冻水的陆面同化系统，探索同化土壤未冻水遥感监测产品对改进流域及区域水分循环和地表能量平衡模拟精度的贡献。本项目研究成果对拓展被动微波遥感应用以及解析高原陆面水文过程对土壤冻融变化的响应特征具有现实意义和科学价值。

气候变暖导致青藏高原冻土快速退化、冻融变化加剧、植被可利用的土壤水分变少，加剧了亚洲水塔水资源危机和生态环境问题，青藏高原地表土壤水分以及冻融变化的准确监测与模拟对解析冻土退化的生态水文效应具有重要科学意义。针对当前冻土中土壤未冻水动态变化遥感监测产品缺失、反演算法欠完善等问题，本项目以青藏高原土壤未冻水的微波遥感监测和数据同化为研究目标，围绕地基L波段被动微波遥感观测、冻土分层介质辐射传输模型研发、SMAP卫星算法改进和土壤未冻水反演算法发展、数据同化系统构建等开展研究。项目取得的主要成果包括：（1）建立了青藏高原首个陆面-主被动微波辐射传输过程的综合观测平台，获取了包含完整冻融循环周期的多频主被动微波遥感协同观测数据，发布了青藏高原首套10年尺度15分钟分辨率的土壤温湿度廓线观测数据集；（2）改进了青藏高原水热传输与冻融过程的模式模拟，发展了多频率、主被动一体化的陆面-微波辐射传输耦合模型，揭示了土壤冻融循环对微波辐散射日变化特征的影响机制和L波段辐射计在冻结和融化土壤的可感深度；（3）发展了土壤未冻水微波遥感反演算法，解决了青藏高原冻土中未冻水的遥感反演难题，改进了SMAP卫星算法反演高原土壤水分的精度；（4）构建了土壤未冻水的数据同化系统，揭示了土壤水储量及冻融变化对江河源区降水-径流时滞特征的影响机制。在项目支持下，发表论文10篇，主要发表在Remote Sensing of Environment、Earth System Science Data等遥感和地学领域著名期刊；发布数据集3个，累计被下载2650余次，支撑和推动了高原遥感和再分析产品的评估和应用。项目负责人晋升研究员，1名骨干晋升副研究员，培养硕士研究生3名。本项目的研究成果促进了冻土微波遥感机理的发展与应用，为卫星算法评估与改进提供了重要参考，可为准确认识冻土退化过程及其水文效应提供数据与理论支撑。