复杂山区滑坡时序InSAR多维动态监测关键技术研究

滑坡是一种极具破坏性的地质灾害，快速准确获取滑坡形变信息对于机理分析与早期预警具有重要指导意义。目前，在复杂山区滑坡InSAR形变监测中存在植被茂密、冰雪覆盖等低相干区有效监测点数量不足、单一平台SAR影像无法准确揭示滑坡真实三维形变、海量数据下InSAR滑坡监测信息难以准动态更新等难题，本项目将研究融合永久散射体与分布式散射体的时序InSAR技术，增加低相干滑坡区的监测点密度，提高滑坡形变监测的有效性；提出融合升降轨SAR数据与先验约束条件构建适合不同量级滑坡的三维时序形变解算模型，获取高精度、高可靠性的三维形变序列，为滑坡机理分析提供数据支撑；研究基于抗差序贯平差的滑坡灾害多维时序形变动态处理方法，提升监测对象的空间维度和动态分析能力。本项目研究成果有望拓展时序InSAR滑坡灾害监测的理论与方法以及适应InSAR灾害准动态监测应用需求。

Landslide is a very destructive geological disaster. It is of great significance to obtain landslide deformation quickly and accurately for mechanism analysis and early warning. At present, the following problems exist in the monitoring of landslide deformation in complex mountain areas with InSAR technique. First, the coherent points are seriously insufficient in low-coherence area caused by such as dense vegetation or ice and snow. Second, it cannot accurately reveal true three-dimensional (3D) deformation just from the single-platform SAR datasets. Third, it is difficult to realize quasi-dynamic monitoring in the era of big data with the conventional InSAR technique. Therefore, this project will propose a new time-series InSAR analysis method by integrating permanent scatterers and distributed scatterers to increase the density of coherent points and improve the effectiveness of landslide deformation monitoring in the low-coherence areas. Then, we study the 3D time-series deformation model by combining the SAR datasets from ascending and descending tracks with the prior constraints to obtain accurate and reliable deformation and provide data support for the analysis of landslide mechanism. Last, we study a multi-dimensional time-series dynamic processing method based on robust sequential adjustment to improve the spatial dimension and dynamic analysis in the deformation monitoring of landslides. The research results of this project are expected to expand the theory and method of time-series InSAR in the deformation monitoring of landslides and meet the application requirements of quasi-dynamic disaster monitoring.