Time Series Ground Motion Monitoring Before 6.24 Maoxian Xinmo Landslide With Sentinel SAR Data

At about 5:38:55 am local time on 24 June 2017,a massive landslide occurred in the back mountain of the Xinmo village,Mao County,Sichuan Province.This langslide continues 120 s and the center of landslide is located at 103.65°E,32.091°N.The Xinmo landslide is the largest landslide after “5·12”Wenchuan Earthquake,which was extremely destructive.It destroyed the Xinmo village rapidly and caused more than 100 people to be buried,and blocked many motorways and the Songping River at the foot of the hill.In this paper,with landslide of Xinmo village as the research object,we use the Time Series DInSAR to detect the deformation information of the mountain before the landslide and study the temporal motion characteristics of the mountain.Considering the special geographical location of Xinmo landslide,this paper analyzes the geometric characteristics and the formation mechanism of the landslide firstly.The landslide has the characteristic of high position,rock,giant,and high speed.The causes of landslide mainly include active faults,historical earthquakes,rock formation lithology,topography,rainfall conditions and so on.The mountain landslide got squeezed by the Songping river fault and the Minjiang river fault continuously as well as the destruction of the rock structure caused by frequent high-intensity historical earthquakes,which caused Xinmo landslide fundamentally.Soft sandstone and cuesta composed of slates are the material condition of triggering landslides.High and steep slopes are potential energy conditions of generating landslides.The continuous rainfall before landslide directly induced the occurrence of this landslide.In order to detect the temporal deformation of landslide before the landslide,we use the PSInSAR technology to monitor the landslide areas with Sentinel-1 SAR images which covers landslide areas and time span 804 days.Finally,we obtain the information of the time series deformation of the study area.The results show before the landslide,the terrain deformation of Xinmo village at the foot of the mountain was not obvious,only had-5~5mm/yr.The center area of the landslide at the top of the mountain shows obvious deformation that the maximum deformation rate along the LOS direction can reach-35mm/yr,and the direction of the deformation region is consistent with the direction of the slide of this landslide.This paper also uses SBAS technology to monitor the time series deformation of the landslide area.Compared with the results obtained by the PSInSAR method,it can be seen that the deformation results obtained by the two methods have higher consistency,and also prove there are obvious deformation in the area of landslide before the landslide occurred.In order to study the characteristics of mountain motion before landslide,this paper extracts the deformation information of the feature points of different locations on the mountain and the deformation rate information on the profile line of the slide source area.The terrain deformation of Xinmo village at the foot of the mountain is not obvious before this landslide.In the sliding source area above the mountain,the higher the settlement rate of the area along the slope,the closer to the ridge is,and the settlement rate of the area far away from the ridge is gradually reduced.The rate of the subsidence area was accelerated gradually,and the average deformation rate was up to-6mm in the first two months of the landslide.Slightly away from the ridge,there is a tendency of uniform subsidence,and the shape variable is larger.In the further distance of landslide center area,with the dense vegetation and the stable rock,the deformation is not big,only in the first two months of the landslide the deformation signal appears.