INNOVATIVE EXPERT-BASED TOOLS FOR MULTITEMPORAL GEOENVIRONMENTAL ANALYSES: FROM MAPPING TO MODELLING

Accurate, timely, and scalable geomorphological risk assessment strategies are essential for effective territorial management and mitigation planning. This dissertation addresses the limitations of traditional geomorphological mapping, especially regarding landslide risk assessment by developing and integrating novel, open, and interoperable tools that blend expert knowledge with artificial intelligence. A human-centred approach guided this doctorate project, where automated technologies support but do not replace expert judgment in hazard assessment and response. This research presents a comprehensive framework for spatio-temporal shallow landslide detection and susceptibility analysis. It is structured around three core technological contributions: (i) GOGIRA (Ground Operative-system for GIS Input Remote-data Acquisition), a field-based digital cartographic mapping system that enables the real-time acquisition of landslide spatial coordinates; (ii) ExMAD (Expert-based Multitemporal AI Detector), an artificial intelligence framework using satellite time series and expert supervision to identify the temporal occurrence of shallow landslides; and (iii) GEMMA (Geo-EnvironMental Multivariate Analysis) Toolbox, a user-friendly, open-source software platform for multivariate geo-environmental data analysis, culminating in GEMMA 4 Landslides, a dedicated module for landslide susceptibility mapping. Through systematic field validation and iterative development, this research demonstrates the feasibility and reliability of low-cost, modular systems for Direct Numerical Cartography and hazard analysis. The integration of expert-driven mapping, automated detection, and accessible analytical tools represents a methodological advance over traditional “black-box” models, emphasizing transparency, interpretability, and reproducibility. By prioritizing usability and accessibility for non-coding professionals and decision-makers, this research promotes the democratization of advanced risk assessment techniques. This dissertation contributes to a scalable and adaptable suite of tools for improving the spatial and temporal resolution of landslide risk modelling, supporting evidence-based decision-making in civil protection, infrastructure planning, and environmental management. It lays the foundation for the broader operational use of AI-supported geomorphological tools and sets a precedent for their integration into next-generation geospatial risk management systems.