Expecting the expected – learning from the past to provide forward scenarios through geomorphic change detection, monitoring and modeling

Abstract

This paper tells the story of a landslide, its origins, its activity and the actions undertaken to mitigate the risks that it poses. The Rotolon landslide is a large Deep-seated Gravitational Slope Deformation (DGSD), located in the Eastern Italian Alps, whose unremitting movements provide a sediment supply for large-scale debris flow events. It has been active since at least 1789, threatening the valley where the thermal baths town of Recoaro Terme is located. In 2010, a major reactivation of the landslide channelled 320,000 m3 of material towards the town causing significant concern, the evacuation of the exposed population and threatening several buildings. A few days after the emergency, the personnel of the Research Institute for Geo-Hydrological Protection of the Italian Research Council (IRPI-CNR) deployed a monitoring system consisting of an automatic total station, several extensometers and web cameras to monitor the evolution of the unstable slope and set up an early warning and alarm system equipped with sirens to warn the local population. Subsequently, after the emergency phase, the 2010 event was studied through multi-temporal LiDAR Digital Terrain Models (DTMs) and modeling. The authors have continued monitoring and actively studying this landslide ever since. In 2020, a new LiDAR survey of the area allowed assessment of the sediment dynamics within the catchment through the comparison with the post-2010 event LiDAR DTM. Based on DEM of difference maps, new insight into the behaviour of the catchment emerged, which appears to be influenced both by natural processes and anthropic activities. The authors were able to assess the amount of sediment that could be stored in the channel without causing overflooding and to calculate the expected damage to the built environment should another event occur. Furthermore, the considerable amount of data collected by the monitoring system throughout the years has allowed identification of two active sectors of the DGSD that may be prone to detachment, and through numerical modeling, future hazard scenarios were produced. Based on these outcomes, a second-tier targeted monitoring campaign was implemented, consisting of a network of four permanent GNSS receivers, additional topographic benchmarks and web cameras, and a new early warning protocol, in an ever-updating cycle of monitoring, modeling and mitigation.