基于SBAS技术的岷江流域潜在滑坡识别

摘要

Landslide often lead to property damages and severe casualties.Potential landslide monitoring and identifying are crucial apporoach to reduce or avoid serious geohazards.Previous studies has indicated that the Interferometric Synthetic Aperture Radar (InSAR) serve as an effective technique for landslide identification and monitoring.However,the application of the InSAR is restricted due to the issues of atmosphere delay,temporal and spatial decorrelation,and so on.In recent years,Small Baseline Subsets (SBAS) technique with prospective applications in surface deformation monitoring has been widely used to obtain tiny and long-time serial change in topographical deformation.In order to identify potential landslide sites in typical steep terrain,in this research SBAS technology was applied to monitor the surface deformation in the Minjiang watershed,the southeast of Longnan City,Gansu Province,where landslides has been prevailing.At first,the deformation rate along the radar sight was obtained based on the SBAS technique but it could not effectively express the true deformation rate along the slope.This is because landslide is usually slid along a slope,and it is transformed into deformation rate along the slope.Nuclear density analysis can transform the deformed area with significant motion into the potential landslide area,so that the potential landslide area in Minjiang watershed was identified by the nuclear density analysis.According to historical landslide data of Minjiang watershed,the identification accuracy of the potential landslide area was verified.The results showed that among the 30 historic landslide points in the Minjiang watershed,18 points were located in the potential landslide region,and 11 points were located in the deformed zones.This study suggests that the potential landslide identification method based on SBAS technique is a feasible and reliable method to monitor and identify potential landslide.%小基线(Small Baseline Subsets,SBAS)技术可以获取微小的形变信息和长时间序列的缓慢地表形变场,在地表形变监测中具有较广的应用前景.本文以地形陡峭的岷江流域为研究区,基于小基线技术获取沿雷达视线方向的形变速率,将其转化为沿坡面方向的形变速率,利用核密度分析对潜在滑坡区域进行提取.研究发现,岷江流域的30处历史滑坡中,有18处位于提取的潜在滑坡区中,有11处位于发生形变的地区范围内,仅有1处位于发生形变的地区之外.这表明,SBAS技术可以较好地监测地势陡峭地区的地表形变,在潜在滑坡识别方面具有较高的可行性.