New coupled quantitative scenario based approaches in a geohazard mitigation measures: The case Study of the Village of San Vito Romano (RM)

The development of tools for quantitative scenario is a robust methodology for risk mitigation, especially in the landslide risk framework. In the case proposed below, this approach is crucial for urban planning and protection of historical heritage. Its application requires a thorough analysis of preparatory and triggering factors. It is also an effective strategy for detecting and monitoring seasonal ground instability effects, while providing a useful instrument to calibrate numerical models aimed at predicting multi-hazard scenarios. The case study selected concerns the village of San Vito Romano (RM) which is a historical place representative for the countryside contexts near to Rome in the Latium region (Italy). The village hosts an historical center with a convent and a historical building for the city hall. The village is for the main part built on a active landslide of almost 1km2 with a rototranslational mechanism and a slow kinematics. The landslide involves silicoclastic deposits from the Frosinone Formation (Upper Tortonian), characterised by alternated layers of clayey and arenaceous marls. This geological setting represents a predisposing factor for slope instability; on the other side, preparatory factors include cumulative precipitation, affecting the soil moisture, and triggering factors include seismic events and intense rainfalls. Landslide activity is documented by the presence of fractures in buildings that have been declared unusable. This phenomenon is causing not only a progressive demographic decline in the town, but also a decrease in tourism, which is one of the fundamental pillars of the local economy. To obtain a detailed geomorphological layout, a LIDAR flight will be conducted, allowing the creation of a Digital Terrain Model (DTM). In addition, regarding satellite techniques, the DInSAR approach was applied, specifically the SBAS (Small Baseline Subset) methodology, an interferometric technique that generates maps of average deformation velocity and provides time series on the displacements of individual points within the image. The processed data, covering the period from 2022 to the present, has shown an increase in displacement velocity, particularly in some areas of the landslide. For the analysis of preparatory and triggering factors, the present research proposes the integration of passive seismic geophysical techniques and satellite interferometric methods. Regarding the first technique, four three-component velocimeters with a sampling frequency of 250 Hz were installed, two of which were positioned inside the landslide and two outside. Continuous monitoring of ambient seismic noise allows the calculation of landslide mobility indices (LSMI), such as natural period variation (dT/T), peak polarization variation (dP/P), variation in the velocity of Rayleigh waves (dV/V) over time. The measurement of surface wave velocity is obtained through the innovative technique of seismic interferometry of ambient noise, which, through the cross-correlation of the recorded signal, makes it possible to determine the variation of surface wave velocity between different monitoring stations. From the variation of these parameters over time, we can continuously monitor changes in rigidity, amplification, and non-linear elastic properties of the soil to understand how the slope is prepared for possible instability. The project will lead to the creation of a quantitative tool capable of correlating interferometric displacements of the landslide and proxies of displacements deduced by passive seismic techniques (LSMI) to point out the role the preparatory and triggering factors in the landslide activity. The influence of seismic events on landslide movement will also be analyzed in relation to magnitude and distance, and the amount of precipitation that shows the most significant correlation with displacements will be identified. Once this information has been assimilated it will be possible to train the model in a forward mode for depicting future scenarios, in a multi-hazard perspective .