Improving Geometric Accuracy and Classification of Forest Roads in the Alpine context through ALS-Based Analysis and the ALSroads R Package

Reliable data on forestry roads is essential for effective forest management, both economically (e.g., accessibility, cost-efficiency, and route optimization) and ecologically (e.g., wildfire prevention and wildlife conservation). However, official maps are often outdated, may lack critical road attributes, and can be inaccurate. Airborne Laser Scanning (ALS) has emerged as a proven technology for detailed mapping of large areas by generating dense 3D point clouds.This study presents a solution for the automatic updating, correction, and enhancement of vector-based forestry road maps over extensive regions. We evaluated a ready-to-use, open-source, and well-documented software tool developed for use in Alpine regions. The method analyzes ALS-derived metrics to generate a raster map indicating the probability of road presence. It then uses an existing, imprecise road map to estimate the approximate start and end points of each road segment. A routing algorithm computes the most probable road alignment by determining the least-cost path on the probability surface. Based on the refined road geometry, the software also estimates road width and condition using characteristics of the point cloud. Our results demonstrate that this fully automated approach accurately reconstructs road centrelines with high geometric precision and reliable classification, enabling precise assessment of road conditions.