多时相InSAR技术及其在滑坡监测中的关键问题分析

Multi-Temporal InSAR Techniques and Their Challenges in the Landslides Monitoring

多时相InSAR(Multi-Temporal InSAR,MT-InSAR)技术可以提取毫米级、大范围的地表形变结果,因此在滑坡监测领域引起国内外的研究热潮。但是,由于滑坡所处地理环境的特殊性,如地形崎岖起伏,植被覆盖率高,大气环境复杂多变等,使得InSAR观测值中的固源误差相比城市地区更加难以消除。该文首先针对MT-InSAR在滑坡监测中的关键技术(即相干点目标选取、形变参数估计和大气误差改正),阐述了目前国际上的主流方法,并对比和分析了它们在滑坡监测中的优劣势和适用条件;进而总结了目前MT-InSAR在滑坡监测中面临的主要挑战,包括由山区地形导致的几何畸变、植被覆盖引发的失相关和无法监测真实三维形变等;最后指出MT-InSAR在滑坡监测领域的未来发展方向,包括面向滑坡监测的MT-InSAR技术、星载和地基MT-InSAR联合监测以及建立MT-InSAR形变监测与滑坡触发因子之间的时空响应模式。

Multi-temporal InSAR(MT-InSAR) techniques had been widely exploited in the investigation of landslides due to their unique abilities in the monitoring of ground deformations with millimeter precision and large scale.However,since the slopes are generally located in the environments with steep terrain,heavy vegetation and complicated atmosphere,it is very difficult to mitigate the inherit errors in the InSAR measurements.The key techniques of MT-InSAR for the monitoring of landslides are coherent point selection,deformation parameter estimation and atmospheric artifact correction.In this paper,the current classical methods for these key techniques are firstly introduced and compared in the view of their advantages and applicability in the monitoring of landslides.Then,the main challenges of MT-InSAR are summarized for the monitoring of landslides,including geometric distortion induced by topographic relief,decorrelation associated with vegetation cover and unavailability of three-dimension deformations.Finally,the future developments of MT-InSAR are provided for the monitoring of landslides,such as MT-InSAR algorithms against landslide monitoring,combination of spaceborne and ground-based MT-InSAR,and establishment of spatio-temporal response model between MT-InSAR observations and landslide triggers.