PSI - Issue 84

Erica Cernuto et al. / Procedia Structural Integrity 84 (2026) 1167–1174

1168

1. Introduction Landslides are complex natural phenomena whose triggering and evolution have been widely investigated at different spatial scales using both empirical and physically based approaches (Ponziani et al., 2013; Salciarini et al., 2016; Salciarini et al., 2019). When such phenomena interact with critical infrastructure, however, their effects become more difficult to interpret and quantify, particularly for bridges located in geomorphologically complex settings, where slope movements may directly affect structural safety and serviceability (Salciarini et al. 2024a, Gabrieli et al. 2024). Understanding how slope movements interact with engineered structures remains challenging, as these processes involve both surface deformation and deeper-seated mechanisms. A promising strategy to improve the assessment of infrastructure exposed to slope instability is the integration of satellite-based deformation measurements with advanced numerical modelling. InSAR offers wide spatial coverage and continuous deformation measurements without requiring ground instrumentation, making it well-suited for remote or inaccessible areas. Its main limitation, however, lies in its measurement geometry: only the Line-of-Sight (LOS) component is detected, offering a one-dimensional projection of a three-dimensional displacement field. This may lead to partial or full loss of information when movements occur largely perpendicular to the LOS (Macchiarulo et al. 2021), a common feature of landslides dominated by horizontal components. Reconstructing the actual displacement direction therefore requires post-processing techniques and combining ascending and descending acquisitions. Several methods have been proposed (Notti et al., 2014, Brouwer, 2021), but no standard solution exists, particularly because spatial and temporal matching between geometries is uncommon and vegetation often reduces PS availability. Resampling strategies, such as grid-based approaches (Casagli et al., 2009) or geometric pairing (Notti et al., 2014), help mitigate these limitations but remain sensitive to local conditions. While InSAR provides valuable surface information, it cannot characterise deep-seated mechanisms or predict the evolution of a landslide. Three-dimensional numerical modelling addresses these gaps by offering a physically based representation of slope deformation and enabling the simulation of soil–structure interaction. Compared to traditional 2D analyses, 3D models can better capture the geometry of the slope, deformation, and kinematics of the moving mass, as well as identify zones of maximum displacement (Tran et al., 2024, Salciarini et al. 2024b, Cernuto et al. 2026, Vitaletti et al. 2025). The Finite Element Method is particularly suited for reproducing the non-linear behaviour of soils, although its reliability depends on the quality of input data, which are often limited (Rainders et al., 2022). In this context, InSAR contributes valuable constraints for model calibration and interpretation, complementing ground based investigations. This study proposes an integrated framework combining InSAR data and 3D numerical modelling to investigate the interaction between a landslide and a bridge. The approach reconstructs two-dimensional displacement components from satellite observations and simulates the landslide–structure interaction through a calibrated numerical model. The combined use of these techniques supports the identification of critical areas and enhances the assessment of infrastructure exposed to landslide-related hazards. It is noted that the data employed derive from previous investigations and do not represent current site conditions; the objective of the study is not to assess present day stability but to demonstrate the applicability and potential of the integrated methodology. 2. Study area The investigated site is located in the hilly sector of the Parma province (Emilia-Romagna, Italy), within the Northern Apennines, where slopes typically range between 10° and 15°. According to the Italian Landslide Inventory (IFFI), the area is affected by several instability phenomena, including a large roto-translational landslide that directly interacts with a roadway bridge (Fig. 1a). The landslide develops within a geological setting characterised by Scabiazza Sandstones overlying the clay-rich units of the Casanova Complex, with interbedded Palombini Shales, a configuration favourable to deep-seated instability. A multi-year monitoring programme conducted between 1975 and 2015 progressively refined the understanding of the phenomenon. Inclinometers and piezometers installed in different phases, together with repeated topographic measurements, identified a sliding surface at approximately 80 m depth and long-term displacement rates on the order of 6–7 cm/year. More recent instruments confirmed the persistence of active movements, recording displacements of 4–6 mm/month at depths of 37–38 m. Laboratory analyses on borehole samples revealed a predominantly silty–clayey matrix, while piezometric data indicated a shallow groundwater table;