Separating Geometry, Measurement Uncertainty, and Return Observability in LiDAR-Based Black-Ice Road Monitoring: A Simulation Study

Winter-road monitoring requires evidence that separates surface change from uncertain or unavailable observations. A single-beam range shift alone leaves the cause of that change unresolved. This simulation sensitivity study evaluates how geometry, measurement and dry-reference uncertainty, and return availability affect the interpretation of published black-ice range evidence. Published distances are reassessed using planar geometry, Monte Carlo comparisons, paired return controls, and prescribed local-plane perturbations. Nine published rows yield seven 30-mm and two 31-mm differences; one reported 51-mm difference recomputes to 31 mm. For the selected uncertainty comparison, conditional detection rises from 0.2684 to 0.8994 between normal and 70-degree incidence under constant uncertainty, but falls from 0.2478 to 0.1100 under moderate combined angle/range uncertainty. In a same-geometry return control at 70 degrees, coverage falls from 0.1950 to 0.0044 and all-attempt opportunity from 0.0142 to 0.0002. These results show why geometric displacement, conditional detection, and usable observation frequency require separate interpretation. The contribution is a reproducible mechanism-separation framework and calibration pathway, not a validated detector: neither range shift nor missing return establishes black-ice material identity.