Loess Landslide High-precision Identification And Monitoring In Heifangtai,Gansu Province With InSAR

Loess covers a very large area where has a large population and abundant resources in China.However,geological disasters in the area are serious and frequent.Every year,one-third of the geological disasters occur in the loess area.Landslides are one of the most destructive geological disasters,which poses a serious threat to the lives and economic development of the local people.Therefore,it is very important and necessary to monitor loess disasters.InSAR is a satellite remote sensing technology developed in recent decades which has been widely used in landslide monitoring due to its high spatial resolution and accuracy,wide coverage,all-days.The landslides of Heifangtai have a long history and are still very active.They have already created a major threat to the lives and property security of the local people.This paper aims at the landslide of Heifangtai,uses InSAR technology,conbines various types of SAR data,catalogs landslides in time,analyzes the characteristics of landslide which is large and dangerous.The main work and achievements are as follows:(1)The process and main errors of loess landslide monitoring using InSAR technology are analyzed.The SBAS-InSAR method is selected for the monitoring of landslides,and the errors of DEM and unwrapping errors are mainly corrected.(2)This paper counts the landslides that have occurred in the Heifangtai,analyzes the relationship between the number of landslides,annual irrigation,monthly irrigation,and rainfall.The results show that landslides in the area mainly occur in March and July of each year and water is the main cause of landslides in the area.(3)The SAR data covering the Heifangtai was investigated,including ALOS data,TerraSAR data,and Sentinel data.Based on these data,the inventory of landslides at the Heifangtai during 2007-2010 and 2014-2017 was catalogued.Today,the landslides mainly occurred in the dangchuan landslide group,chenjia landslide group,and moshigou landslide group.(4)Using the SBAS-InSAR technology,three landslides occurred in the Dangchuan in 2015,2016 and 2017 were monitored and the characteristics of deformation before landslides were analyzed.For the dangchuan 2# landslide,high-resolution DEM and SAR data are mainly used to estimate the change of the terrain after sliding,and the results are in good agreement with the UAV measurement results.For Dangchuan 3# landslide,three kinds of SAR data were used,including TerraSAR data with both ascending and descending tracks and Sentinel descending data to calculate the three-dimensional deformation of the landslide.For the Dangchuan 4# landslide,the Sentinel data was used to calculate the pre-slide deformation rate,and the starting and accelerating time point of the landslide was found.This study provides a relatively complete set of technical solutions for the investigation,monitoring and early warning of loess landslides which are widely distributed in China,and make important contributions to disaster prevention and reduction.