Assessing Disturbance in Forest Opening-Up Infrastructure through Geoinformatics-Based UAV-RTK Terrain Modelling and Hydrological Analysis

Forest opening-up infrastructure is essential for forest accessibility, timber extraction, fire suppression, and post-disturbance intervention. However, forest roads may become highly sensitive to disturbance when topographic relief, surface runoff, slope gradients, and road-surface deformation interact. In Mediterranean forest landscapes, road-surface degradation may be amplified by relief, slope, concentrated runoff, intense rainfall, drought, wildfire-related soil exposure, and increasing anthropogenic pressure. This study develops a geoinformatics-based framework for assessing disturbance in forest road infrastructure through UAV-RTK terrain modeling and DEM-based hydrological susceptibility analysis. High-resolution imagery was acquired over a representative forested pilot road network using a fixed-wing UAV with real-time kinematic (RTK) positioning. The orthomosaic, Digital Surface Model, Digital Elevation Model, slope, and aspect layers were integrated in GIS to delineate deformation, runoff-concentration zones, and road segments requiring intervention. Disturbed road sections were operationally defined as pavement areas where focal RANGE statistics and raster reclassification identified microtopographic anomalies, including rutting, depressions, surface roughness, and material displacement. The deformation model used 3 × 3, 4 × 4, and 5 × 5 pixel windows, producing 30 DEM-based classes. Hydrological susceptibility was derived using DEM filling, flow direction, flow accumulation, and Strahler stream ordering; therefore, the resulting flood/erosion map is interpreted as a susceptibility layer rather than a deterministic flood-risk model. The results demonstrate that at a centimeter resolution, UAV-RTK-derived terrain products detected moderate but operationally significant disturbance traces in forest road networks. The survey calibrated 99.78% of the images and achieved 0.0225, 0.0196, and 0.0203 m RMS errors for X, Y, and Z, respectively. A rule-based overlay prioritized segments where deformation classes coincided with runoff concentration, slope exposure, and cut-and-fill needs. The workflow supports targeted drainage redesign, surface reshaping, erosion-control works, and transferable disturbance-resilient forest infrastructure planning. Overall, the approach strengthens preventive maintenance and adaptive forest infrastructure management under changing environmental and climatic conditions.