Rockfall susceptibility and network-ranked susceptibility along the Italian railway

Rockfalls pose a substantial threat to ground transportation, due their rapidity, destructive potential and high probability of occurrence on steep topographies, often found along roads railway routes. Approaches for the assessment rockfall susceptibility range from purely phenomenological methods statistical methods, suitable modeling large areas, deterministic ones, usually easier use in local analyses. A common requirement is need locate detachment points, uphill cliffs, subsequent runout areas rockfalls stemming such points. Here, we apply physically based model calculation trajectories whole Italian network, within corridor total length about 17,000 km varying width. We propose data-driven method location source points expert mapping sample representative locations. Using empirical distributions gridded slope angle values mapped by experts, derived probabilistic maps sources proximity regardless particular trigger. Source act as starting simulated three-dimensional STONE. The program provides pixel-by-pixel trajectory count, covering 24,500 km2 representing largest homogeneous application date. classified raster map into vector map, analyzing track collection segments, which provide segment-wise susceptibility. Eventually, considered an equivalent graph representation helps classifying segments both basis role each segment resulting network-ranked Both are useful hazard assessment, prioritize safety improvements railway, at national scale.