has been systematically implemented and evaluated in Peru, nor has its behavior been assessed under spatial resolution changes relevant to mining-scale applications. This study addresses these gaps by developing a national-scale Landslide Susceptibility Map (LSM) for Peru using an Artificial Neural Network (ANN) framework trained on 105,181 samples derived from the INGEMMET landslide inventory. The model integrates DEM-derived morphometric variables (slope, aspect, curvature), hydrological indices (SPI, TWI), and land cover, and is implemented using resilient backpropagation. Beyond national hazard assessment, the proposed framework is designed to support operational decision-making in mining contexts by enabling quantitative identification of highsusceptibility zones within and around mining units. Such capability allows prioritization of stabilization measures, optimization of infrastructure layout, reduction of geotechnical risk, and minimization of production interruptions associated with mass movement events. The methodology further evaluates model behavior under spatial downscaling, supporting its applicability across multiple scales, from national planning to basin and mine-site analysis. 2. MATERIALS & METHODS 2.1 Study Area The study area covers the Peruvian territory (≈1.285 million km²), extending between 0°02′–18°21′34″ S and 68°39′07″–81°20′13″ W, and is dominated by the Central Andean Cordillera. Its relief results from long-term tectonic processes associated with Pacific–South American plate convergence, generating uplift, active faulting, and structural complexity. This geological framework, combined with intense precipitation and seismic activity, favors widespread landslide occurrence across the Andes (Figure 1). Figure 1 – Study area location and data classification map. Inset shows the distribution of testing, training, and new data (1995–2023) overlaid on a digital elevation model (DEM) of the Cusco province. Within this area, a total of 29,164 landslide events have been recorded by INGEMMET
RkJQdWJsaXNoZXIy MTM0Mzk2