PSI - Issue 84

Alice Vitaletti et al. / Procedia Structural Integrity 84 (2026) 191–198

197

bridge and therefore maintain higher coherence, resulting in a more consistent displacement trend. Disregarding these anomalous values, the transverse component reveals a linear trend of progressive uplift, consistent with the downslope movement of the landslide. However, the synthetic PS in cell 6 deviates from this pattern, showing a progressive downward trend suggesting movement opposite to the expected direction (Fig. 5b).

a

b

Fig. 5. Time series of (a) vertical and (b) transverse displacement component.

The InSAR-derived displacement components represent a transient deformation state limited to the 2018-2022 monitoring period. In contrast, the FEM output corresponds to a steady-state condition, providing the deformation pattern associated with the equilibrium reached by the model. Since the two approaches describe different conditions of the system, a quantitative comparison was not feasible. A qualitative assessment was therefore carried out to evaluate the consistency between the results. The synthetic PS located upslope of the mapped landslide (i.e., cells 7 and 11) displayed horizontal displacements aligned with the expected kinematics, confirming the ability of the FEM to identify areas prone to instability, even partially outside the inventoried boundary. A similar agreement was observed in cells 10 and 14, where both methods indicate minor displacements. Some discrepancies were noted in cell 6, where the InSAR-derived transverse component exhibited an opposite trend to that predicted by the numerical simulation. This is likely due to limitations in the resampling procedure, during which PS from the descending geometry outside the landslide boundary were included, combined with the lower coherence associated with slope vegetation cover. Overall, despite the intrinsic differences of the approaches, the correspondence between the deformation patterns supports the robustness of the numerical model and highlights the complementary value of combining InSAR data with FEM simulations to improve the understanding of landslide affecting critical infrastructures, particularly in contexts where ground-based monitoring is limited. 4. Conclusions This study demonstrates how combining satellite-based displacement data with numerical modelling can enhance the analysis of a slow-moving landslide affecting highway infrastructure. The case study was selected following a preliminary assessment of InSAR data coverage along a highway network in a landslide-prone area, which identified a landslide with a suitable distribution of radar targets. The FEM model showed that the largest movements occur in the downslope portion of the landslide-affected area, with notable displacements at the foundations of the piers located within the unstable volume, reflecting the effect of downslope-directed forces on the structure. InSAR data were processed through spatial and temporal interpolation to estimate the transverse and vertical components of the actual displacement. Overall, the deformation patterns aligned with the FEM results, except in a sector where sparse or low coherence radar targets reduced the reliability of the satellite measurements. In conclusion, the study highlights both the benefits and limitations of combining different approaches to enhance the interpretation of landslides affecting highway infrastructure, particularly where in situ measurements are limited. This combined method provides practical support for infrastructure management and can assist decision-makers in identifying areas that may require further monitoring or preventive action.