Tunnel excavation through mountain ranges can alter groundwater regimes, affecting spring discharge and dependent ecosystems. Anticipating these interferences is essential for environmental risk management, cost–benefit analysis, and effective mitigation planning. Conventional assessment methods are largely parametric, not physically based, and heavily dependent on expert judgment, introducing subjectivity and uncertainty. This study proposes a data-driven approach based on machine learning (ML), specifically a Random Forest (RF) model, to evaluate hydrogeological risk and spring vulnerability during the preliminary design phase of tunnels. The method exploits geological information typically available in early stages and is validated using two well-documented Italian case studies: The Bologna–Florence high-speed railway line, excavated in fractured sedimentary formations of the Northern Apennines, and the Gran Sasso highway tunnels, driven in karstified carbonate rocks of the Central Apennines. Both datasets are published, hydrogeologically validated, and supported by detailed pre- and post-excavation monitoring, ensuring robustness and consistency. The ML framework demonstrates strong predictive performance across standard classification metrics, including accuracy, precision, recall, F1-score, Matthews correlation coefficient (MCC), and area under the curve (AUC). While it employs input parameters comparable to those of parametric methods such as the drawdown hazard index (DHI), it differs by deriving parameter weights directly from data. Moreover, the use of Shapley additive explanations (SHAP) values enhances interpretability, mitigates black-box behaviour, and allows expert knowledge to be effectively integrated. The innovation, therefore, lies in offering a transferable and reproducible tool to support early-stage tunnelling decisions, representing a clear improvement over traditional qualitative or semi-quantitative approaches.

Landslides pose serious risks to infrastructure, particularly railways, due to their rigid construction and essential transport role. Susceptibility mapping is a valuable tool during the feasibility phase of railway projects, helping identify high-risk areas and inform mitigation strategies. However, effective application requires both reliable classification of landslide types and robust reclassification methods for clear communication with stakeholders. This study presents a comprehensive workflow for landslide susceptibility mapping, combining Weight of Evidence (WoE) and a Generalized Additive Model with boosting (GAMB). We generated separate susceptibility maps for five landslide types and evaluated them using AUROC metrics. The maps were merged into an overall susceptibility map using a complementary probability approach, which also allowed assessment of each type's sensitivity to the overall susceptibility. To improve threshold reliability, we implemented an ensemble reclassification method using six approaches and applied the statistical mode to define more objective class boundaries. Visualizations of susceptibility along the railway route and its adjacent sides were developed for practical application. The methodology was applied to a 22 km planned railway section in the Marche region (Italy). Results revealed high spatial variability: rockfall types showed the highest accuracy (AUC = 0.94 WoE, 0.98 GAMB), while slides performed poorest. GAMB consistently outperformed WoE in reliability and smoothness of results. Finally, a comparison with EGMS ground motion data showed no significant correlation (R2 approximate to 0.1), underscoring the temporal disconnect between long-term susceptibility and short-term ground deformation.

Since Active and Capable Faults (ACFs) may generate significant permanent deformation of the topographic surface, a careful evaluation of their spatial and geometric characteristics is essential for seismic hazard assessment when planning new linear infrastructures (e.g., roads, railway lines, pipelines). Although this is generally overlooked, the common structural complexity of fault zones leads to a non-uniform hazard along and across faults' traces, because of deformation partitioning. This study reviews the factors controlling fault rupture and propagation, specifically focusing on fault zone architecture and growth mechanisms. Four scenarios of physical interaction between ACFs and linear infrastructures are analysed. The fault-crossing scenario is likely the most susceptible to ground surface displacement, while the fault-parallel scenario needs evaluation of the width of fault damage zone overlapping with the infrastructure. Near-fault tip and transfer zone-crossing scenarios require specific assessment of the local deformation patterns. Given the importance of a structural geological approach toward the reliable assessment of seismic hazard related to ACFs, we review suitable investigations to derive appropriate deterministic geological constraints on the geometry, kinematics, slip and deformation style of ACF's. Our approach may have significant impact on the legislation regulating the early stages of infrastructural design.

The valley junction of Isarco and Pusteria (Rienza River), located in the Bressanone area, showcases a complex stratigraphic succession that traces back to the Late Pleistocene evolution. Extensive field surveys and numerous drillings conducted between Bressanone and Varna/Sciaves have unveiled the stratigraphic architecture of the valleys. Four distinct glacier advances have been identified, with the thickest deposits attributed to the Last Glacial Maximum (LGM), characterised by a fine-grained subglacial traction till up to 30 metres thick. Additionally, two Lateglacial stadial moraines are linked to the Isarco glacier, indicating that the modern Rienza lower valley was sculpted as the ice retreated at the end of the LGM. A pre-LGM fluvial-lacustrine system, receiving contributions from both valleys, suggests that the junction was located further north than its current position. Below this deposit, an older glacigenic sediment layer consisting of coarse subglacial traction till marks a phase between two major ice advances. At the deepest point, core samples from the Isarco valley reveal fluvial deposits from the Pusteria valley catchment, highlighting the existence of a narrower lower reach across the Rienza River. This evidence indicates the long-standing presence of the river valley at the junction in the Sciaves/Varna area, well before the LGM. The discovery of a large landslide reveals notable slope dynamics due to glacial erosion, with the landslide body covered by LGM glacial deposits, whereas post-LGM slope deposits are related to small-scale slope processes.

The engineering design for tunnel excavation in soils may produces an hydrogeological barrier to the natural groundwater flow. Upstream of the tunnel the groundwater table may rise while downstream may be lowered. This paper examines three case studies, all in a urban context (Turin, Parma and Venice-Italy). For each case a brief geological framework is provided, as well as performed Modflow models for investigating this problematic are presented. The quantified perturbation induced by the hydraulic barriers shows the classical behavior on the natural groundwater hydrodynamic leading to rising/lowering water table.

Landslide processes commonly occur on both sides of the Scascoli Gorge. The valley floor is flanked by vertical sandstone walls 40 to 80 m high, as well as steep slopes of heavily fractured and weathered rocks. The most important rockfalls known up to date occurred from the left valley side in 1992, 2002 and 2005, when a rock wedge called Mammellone 1, collapsed by a large and complex rockslide, with an estimated volume of 30.000 m3. The collapse of the Mammellone 1 shifted the riverbed to the right valley side where, just a few weeks later, a roto-translational earth slide was triggered by the consequent undermining process. In order to rehabilitate the local road, three safety interventions were planned in the following order: (a) reshaping the residual wall of the Mammellone 1; (b) removing the shallow earth slide from the right valley side, and consolidating its detachment zone; (c) carrying out training works on the riverbed, and reconstructing the road embankment.

In October 2002, a rock mass volume of about 20,000 m3 fell into the so-called Scascoli Gorge, south of Bologna (Italy), completely obstructing the Savena riverbed and destroying a 150 m long segment of the local road, named "fondovalle Savena". The intrinsic predisposition to instability, due to the stratigraphic and tectonic features, is made worse by a