MT-InSAR Technology Combined With Inverse-velocity Method For Highlevel Landslide Monitoring And Early Warning

Landslides have always been a frequent and harmful geological disaster that plagues our country and the world.How to prevent and control landslides is of great significance to the protection of people’s life and production safety.Traditional landslide monitoring,such as the use of GNSS,level and total station,to collect data is mostly done by manpower.It is particularly difficult to monitor large areas of high mountain deformation in remote mountainous areas.With the development of radar interferometry technology,in addition to achieving many reliable results in urban settlement monitoring,the In SAR method has gradually become a reliable and efficient method for landslide monitoring.The classic SBASIn SAR method can be used in mountain slopes.A certain number of high-coherence point targets are detected on the body for landslide deformation analysis.This deformation information can not only reflect the geographic location and scope of the landslide deformation,but also can make a certain judgment on the subsequent deformation trend of the slope through the analysis of its time series deformation.It can play a positive role in landslide prevention.We expect a technology not only to have the ability to monitor landslides,but also to use the monitored deformation information for early warning.However,the current research on this technology in the field of landslide early warning is still in its infancy and existing methods have some shortcomings.Many scholars have stated that the inverse-velocity method is a simple to use method that can accurately predict the time of a landslide based on time series deformation data.Displacement deformation is an obvious feature in the evolution of the slope.The measurement of this parameter is simple and easy to obtain.Displacement-based landslide predictions occupies most of the cases of successful landslide prediction,which shows that displacement is extremely important in landslide prediction.In this paper,in order to fully excavate the application of the LOS direction displacement deformation data information of the slope body obtained by the SBAS method in the landslide warning,it is combined with the inverse-velocity method to detect the landslide deformation first to obtain the deformation data and then use the method to conduct the warning research.The main work of the thesis is as follows:(1)A summary analysis of the 26 landslide prediction data obtained by using the inversevelocity method in the existing literature shows that the method is reliable and practical in landslide time prediction,and pointed out that the time span will affect the prediction results to a certain extent.Generally,the closer the time series data collection time is to the landslide,the more accurate the prediction results.Therefore,the deformation data of the pre-sliding stage should be selected for landslide prediction as much as possible.(2)Using 24 sentinel1A/B images covering about 14 months in Xinmo Village,Maoxian County,as the data source,the experiment was carried out.Based on the SBAS-In SAR method,a total of 2 obvious deformation areas were detected.One of them is the sliding source area of Xinmo Village,Maoxian County,and the other deformation is located in the mountains near Songpinggou Scenic Area.Extracting two time-series deformations,it is found that the Maoxian landslide has the phenomenon of continuous acceleration in the imminent sliding stage before occurrence.There are displacement changes in the area but no continuous acceleration;the analysis of Maoxian landslide combined with rainfall data and historical earthquakes shows that earthquakes are the internal cause of landslides and rainfall is the inducing factor of landslides.(3)The three methods to solve the slope deformation rate given in the original literature are not accurate enough due to the large time interval when applied to the time series deformation data obtained by the MT-In SAR method.This will lead to large errors in the prediction results of the inverse-velocity method.Therefore,this paper proposes the cubic spline interpolation method to solve the slope deformation rate.This rate is used in the inversevelocity method and combined with the Maoxian landslide time series data to retrospectively predict the landslide time,and compared with the inverse-velocity method landslide prediction results based on other methods to obtain different rates.In addition,a foreign landslide data obtained by the MT-In SAR method in other documents is used to compare the landslide time prediction based on the different rates solution method.The two comparisons show that the velocity calculated by the cubic spline method proposed in this paper can obtain better landslide time prediction results than other methods when used in the inverse-velocity method,and the original time series displacement data used in the landslide prediction is better than the smoothed data.Finally,the inverse-velocity landslide time prediction method based on the improved rate calculation method and the classic gray-Verhulst landslide time prediction method were compared for landslide prediction.The result was once again The feasibility and reliability of the inverse-vetocity method in the application of landslide time prediction are verified.