基于SBAS-InSAR技术的川藏铁路澜沧江段滑坡隐患早期识别

早期识别是实现地质灾害防灾减灾的有效途径之一,然而复杂地形区滑坡的早期识别一直是个难题,尤其是位于高山峡谷区的滑坡隐患点。为了全面准确地获取川藏铁路澜沧江段的滑坡隐患,采用SBAS-InSAR技术,通过Sentinel-1(升轨)和RADARSAT-2(降轨)数据结合互补的方式,对川藏铁路澜沧江段进行滑坡隐患早期识别。解译结果显示2018年8月至2020年2月研究区LOS向的形变速率分别为-58~21 mm/a(升轨)和-42~16 mm/a(降轨),转换后的斜坡向最大平均速率达到-128 mm/a。基于升降轨数据的斜坡向形变结果,识别出川藏铁路澜沧江段的113处滑坡隐患点,其中存在4处滑坡隐患密集区以及13处典型滑坡隐患点,进一步分析了两处重点滑坡隐患的形变特征和滑移机制。本次研究结果对于川藏铁路线路选定以及澜沧江大桥上、下游的防灾减灾具有一定指导作用,不同轨道数据结合互补的方式为川藏铁路沿线的高山峡谷地区的滑坡隐患早期识别提供参考。

Early Landslide Detection in the Lancangjiang Region Along the Sichuan-Tibet Railway Based on SBAS-InSAR Technology

Early detection is a key to prevent and reduce geo-hazards.However,it is always problematic in complex topographic region,especially in alpine-valley region.Based on the Sentinel-1(ascending)and RADARSAT-2(descending)radar satellite data obtained from August 2018 to February 2020,quantitative detection and analysis of the potential surface deformation of landslides in the 80 km upstream and downstream of the Lancangjiang region along the Sichuan-Tibet railway was carried out by using the small baseline subsets in terferometric synthetic aperture radar(SBAS-InSAR).The results show that deformation rate in the LOS direction for the study period are from-58 to 21 mm/a(ascending)and from-42 to 16 mm/a(descending),respectively,and the maximum average deformation rate in the slope direction is-128 mm/a.Based on the deformation result in the slope direction,113 potential landslides are detected,which are clustered in four regions.Besides,13 landslides have significant deformation in the 113 potential landslides,deformation characteristics and slipping mechanism of the two critical landslide hazards were further analyzed.The results provide a reference for the route selection of Sichuan-Tibet railway,and for the geohazard prevention in the upstream/downstream of the Lancangjiang bridge.The technique of combining descending and ascending radar data can provide reference for early landslide detection in other alpine-valley regions along the Sichuan-Tibet railway.