A spaceborne SAR-based procedure to support the detection of landslides

Abstract

Abstract. The increasing availability of free-access satellite data represents a relevant opportunity for the analysis and assessment of natural hazards. The systematic acquisition of spaceborne imagery allows for monitoring areas prone to geohydrological disasters, providing relevant information for risk evaluation and management. In cases of major landslide events, for example, spaceborne radar data can provide an effective solution for the detection of slope failures, even in cases with persistent cloud cover. The information about the extension and location of the landslide-affected areas may support decision-making processes during emergency responses. In this paper, we present an automatic procedure based on Sentinel-1 Synthetic Aperture Radar (SAR) images, aimed at facilitating the detection of landslides over wide areas. Specifically, the procedure evaluates changes of radar backscattered signals associated with land cover modifications that may be also caused by mass movements. After a one-time calibration of some parameters, the processing chain is able to automatically execute the download and preprocessing of images, the detection of SAR amplitude changes, and the identification of areas potentially affected by landslides, which are then displayed in a georeferenced map. This map should help decision makers and emergency managers to organize field investigations. The process of automatization is implemented with specific scripts running on a GNU/Linux operating system and exploiting modules of open-source software. We tested the processing chain, in back analysis, on an area of about 3000 km2 in central Papua New Guinea that was struck by a severe seismic sequence in February–March 2018. In the area, we simulated a periodic survey of about 7 months, from 12 November 2017 to 6 June 2018, downloading 36 Sentinel-1 images and performing 17 change detection analyses automatically. The procedure resulted in statistical and graphical evidence of widespread land cover changes that occurred just after the most severe seismic events. Most of the detected changes can be interpreted as mass movements triggered by the seismic shaking.

Modelling the control of groundwater on landslides triggering: the respective role of atmosphere and rainfall during typhoons

“…Our model outcomes also corroborate the preferential location of storm-triggered landslides at the toe of hillslopes (Meunier et al, 2008). As a long-term insight, we believe that a better characterisation of the timing of landslide failure during large storms or typhoons, for instance thanks to the development of SAR imagery (Singhroy and Molch, 2004;Xu et al, 2019;Esposito et al, 2020), could help to separate the respective role of atmospheric pressure drop and rainfall in slope destabilisation. Geol.,197,[172][173][174][175][176][177][178][179][180][181][182][183][184][185][186][187]…”

A new strategy to map landslides with a generalized convolutional neural network

“…This could be a serious problem for some cases as it might take many days for the clouds to clear the sky. Methods working on spaceborne Synthetic Aperture Radar data can provide information even through cloud cover and should be used for such situations 38 , 39 . All the study areas selected for this study had had high coverage of vegetation.…”

Establishing the timings of individual rainfall-triggered landslides using Sentinel-1 satellite radar data

“…Different studies have demonstrated that all three of these backscatter coefficients can be applied to detect vegetation removal due to landslides and other processes such as deforestation and wildfires (e.g. Ban et al, 2020;Belenguer-Plomer et al, 2019;Bouvet et al, 2018;Esposito et al, 2020;Hernandez et al, 2021;Mondini, 2017;Mondini et al, 2019;Motohka et al, 2014). Here we used γ 0 .…”