Landslide Mapping and Monitoring Using Persistent Scatterer Interferometry (PSI) Technique in the French Alps

Gokhan Aslan,Ziyadin Cakir,Marcello De Michele,Michael Foumelis,Daniel Raucoules,Severine Bernardie

doi:10.3390/rs12081305

Abstract

Continuous geodetic measurements in landslide prone regions are necessary to avoid disasters and better understand the spatiotemporal and kinematic evolution of landslides. The detection and characterization of landslides in high alpine environments remains a challenge associated with difficult accessibility, extensive coverage, limitations of available techniques, and the complex nature of landslide process. Recent studies using space-based observations and especially Persistent Scatterer Interferometry (PSI) techniques with the integration of in-situ monitoring instrumentation are providing vital information for an actual landslide monitoring. In the present study, the Stanford Method for Persistent Scatterers InSAR package (StaMPS) is employed to process the series of Sentinel 1-A and 1-B Synthetic Aperture Radar (SAR) images acquired between 2015 and 2019 along ascending and descending orbits for the selected area in the French Alps. We applied the proposed approach, based on extraction of Active Deformation Areas (ADA), to automatically detect and assess the state of activity and the intensity of the suspected slow-moving landslides in the study area. We illustrated the potential of Sentinel-1 data with the aim of detecting regions of relatively low motion rates that be can attributed to activate landslide and updated pre-existing national landslide inventory maps on a regional scale in terms of slow moving landslides. Our results are compared to pre-existing landslide inventories. More than 100 unknown slow-moving landslides, their spatial pattern, deformation rate, state of activity, as well as orientation are successfully identified over an area of 4000 km2 located in the French Alps. We also address the current limitations due the nature of PSI and geometric characteristic of InSAR data for measuring slope movements in mountainous environments like Alps.

Highlights

Landslides are among the most widespread geological hazards that pose a major threat to human life, infrastructure and natural environment in most mountainous regions with steep slopes
In order to better illustrate the conversion from VLOS into VSLOPE and the various steps involved for the extraction of the active deformation areas, we focused on a small area characterized by relatively dense slow moving landslides
In this work, mapping and monitoring of slow-moving landslides over an area of 4000 km2 located in the French Alps has been carried out using the Persistent Scatterer Interferometry (PSI) approach applied to Sentinel 1 TOPSAR data in ascending and descending orbits

Summary

Introduction

Landslides are among the most widespread geological hazards that pose a major threat to human life, infrastructure and natural environment in most mountainous regions with steep slopes. Landslide inventory databases and maps are documenting the identification of landslide processes including the spatial distribution and characteristics such as geometry, volume, total length of area, causal factors, temporal frequencies, and sliding rates of different types of past landslide activities [4,5]. This information might provide a clue to the locations of future mass movements that are very helpful in predicting the landslide susceptibility and vulnerability in order to support local authorities for hazard mitigation [6,7,8]. In France, for example, national landslide databases provided by French Geological Survey (BRGM) are mostly based on historical records, field inventory and detailed geomorphological formation surveys and aerial orthophotos

Objectives

Methods

Results

Discussion

Conclusion