<p>Based on geomorphological criteria, large-scale slow gravitational deformation affecting entire mountain flank, often being referred as Deep-Seated Gravitational Slope Deformation (DSGSD), have been shown to affect most of the reliefs worldwide. For instance in the European Alps, these deformation patterns were identified in several areas such as the Aosta Valley (Martinotti et al., 2011) or the Mercantour massif (Jomard, 2006). DSGSD inventories based on visual interpretation of scarps and field mapping were then compiled (e.g. Crosta et al., 2013) revealing the widespread occurrence of DSGSD. However, many aspects of these large-scale gravitational processes remain unclear and in particular their present-day activity and temporal evolution remain largely unknown.</p><p>The present study aims at characterizing the spatial extent of DSGSD, and their velocity, at the scale of Western Alps through InSAR time series analysis using NSBAS processing chain (Doin et al., 2001). We used the whole SAR Sentinel-1 archive, between 2014 and 2018, with an acquisition every 6 days, on an ascending track. The processing was adapted to fit the specific conditions of the Alps (seasonal snow cover, strong local relief, vegetation and strong atmospheric heterogeneities). In particular we implemented a correction using the ERA 5 weather model and we used snow masks in winter allowing to select long temporal baseline interferograms with as little snow as possible. As we specifically aim to study deformation patterns at the scale of valley flanks, an average high-pass filter on moving subwindows has been applied to the interferograms prior to the implementation of time-serie inversions. This step strongly reduced the impact of residual atmospheric delays.</p><p>The resulting velocity map in the line of sight (LOS) of the satellite reveals ubiquitous gravitational deformation patterns over the whole Western Alps, with localized patches of moving slopes showing sharp discontinuities with stable surrounding areas. We used radar geometry and InSAR measurement quality factors as indicators to identify the most trusted areas and to extract an inventory of potential DSGSD with their spatial extent. Doing so, we identified more than two thousands slowly deforming areas characterized by LOS velocities from 4 to 20 mm/year. We then compared the geometries of our &#8220;InSAR-detected-deforming-slopes&#8221; with previously published DSGSD inventories. Good agreements were found for example in the Aosta valley where most of the deforming areas from our velocity map are falling into the DSGSD outlines of Crosta et al. (2013). Currently, we continue to investigate the potential of this large-scale velocity map for DSGSD understanding and we plan to use artificial intelligence to search for possible generic properties between the detected sites.</p>
LiCSBAS: An Open-Source InSAR Time Series Analysis Package Integrated with the LiCSAR Automated Sentinel-1 InSAR Processor
