Enhancing Landslide Prediction Through Advanced Transformer-Based Models: Integrating SAR Imagery and Environmental Data

Landslides, a significant natural hazard driven predominantly by rainfall infiltration, pose a continuous threat to the terrain, buildings, and ecosystem of Caiazzo, located in southern Italy (Campania Region). This research under-takes a novel approach to predict and understand the complex patterns of landslide deformations in this hamlet. Our study leverages the advanced capabilities of COSMO-SkyMed (CSK) satellite imagery, integrating a comprehensive dataset that includes landslide predisposing factors. We employ Transformer-based models, renowned for their effectiveness in capturing long-range dependencies within sequential data. The Transformer models in our study are adept at analysing the temporal sequences of environmental factors and their intricate interactions, thereby offering a more nuanced understanding of the temporal patterns leading to landslides. In order to validate our results, different metrics, including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R2 score, have been applied, demonstrating the superior performance of the Transformer-based approach, with significant improvements over the conventional Deep Learning model. The visual correlation of our predictions with actual landslide occurrences further corroborates the effectiveness of this method. This transformative approach not only enhances our understanding and predictive capability for landslides in Caiazzo but also sets a benchmark for landslide prediction in geologically vulnerable regions worldwide.