典型山区InSAR地表形变监测与滑坡灾害识别

InSAR Surface Deformation Monitoring and Landslide Hazard Detectionin Typical MountainousAreas

云南省复杂山地和脆弱的地质条件,导致滑坡、崩塌、泥石流等地质灾害频发,迫切需要高精度的广域地表形变监测方法.利用多源SAR卫星数据和多种InSAR处理技术,在云南省典型山区开展了地表形变监测研究.针对雷达卫星在复杂山地成像时的几何畸变问题,提出耦合叠掩阴影绘图法(Layover and Shadow Map,LSM)和R指数模型(R-index)的几何畸变精细识别与可视性分析方法,提高InSAR监测效率.同时,结合地表相干性时序变化特征提出一种SAR数据时空适用性评估方法,降低时空失相干的影响.此外,基于ERA-5数据构建了大气延迟分析模型,量化了不同海拔下对流层延迟对形变时间序列的影响,有效缓解了复杂山区的对流层带来的时间振荡偏差,提升了时序InSAR反演地表变形的精度.通过对多时相、多源SAR数据的综合分析与优化利用,实现了对云南典型复杂山区的滑坡灾害监测与识别.结果表明:优化后的InSAR方法不仅在宏观尺度上可有效捕捉复杂山地的微弱形变信号,识别潜在滑坡灾害,还能针对局部重点区域做出更加精细的活动性评估.研究成果为云南省地质灾害的监测和防治提供了重要的技术支撑.

Yunnan Province's unique geological environment and the characteristics of its mountainous gorge terrain contribute to the susceptibility and frequency of geological disasters such as landslides,collapses,and debris flows,underscoring the urgent need for high-precision and wide-area ground deformation monitoring methods.To address the geometric distortion issues in radar satellite imaging of complex mountainous terrain,we propose a technique that couples the LSM(Layover and Shadow Map)approach with the R-index model for fine identification of geometric distortions and visibility analysis,thereby enhancing the efficiency of InSAR monitoring.Concurrently,we introduce a spatiotemporal suitability assessment method for SAR data based on the temporal variations in surface coherence,which reduces the impact of spatiotemporal decorrelation.Furthermore,leveraging ERA-5 data,we developed the PZTD-NEF model to quantitatively analyze the effects of tropospheric delay on deformation time series at different altitudes,effectively mitigating the temporal oscillation biases caused by tropospheric effects in complex mountainous areas,and improving the accuracy of time-series InSAR surface deformation inversion.Through comprehensive analysis and optimized utilization of multi-temporal and multi-source SAR data,we achieved significant results in landslide disaster monitoring and identification in Yunnan's typical complex mountainous regions.The study demonstrates that the optimized InSAR method not only effectively captures subtle deformation signals in complex mountainous terrain on a macro scale and identifies potential landslide hazards but also provides more detailed activity assessments for key local areas.This research offers significant technical support for the monitoring and prevention of geological disasters in Yunnan Province.