A Topography-Informed Morphology Approach for Automatic Identification of Forest Gaps Critical to the Release of Avalanches

Human assets in Alpine regions are prone to gravitational natural hazards such as rock fall, shallow landslides and avalanches. Forests make up a substantial share in that landscape and can mitigate those hazards. Management of avalanche protection forests must cope with avalanches potentially released in forest gaps, which can damage downslope forests. The Swiss guidelines Sustainability and success monitoring in protection forests prescribe forest-gap extents in slope-line direction critical to the release of avalanches in forested areas. This article proposes a topography-informed morphology approach (TIMA) to automate the detection of critical gaps based on a digital terrain model and a canopy height model (CHM) derived from airborne LiDAR-data. TIMA uses complementary information about topography to probe forest gaps computed from the CHM with templates meeting critical-gap extents adjusted to local topography. The method was applied to a test site in Klosters-Serneus (Switzerland). The comparison of a critical-gap map with the results of a field assessment at 19 sample locations resulted in 84% overall accuracy. Moreover, plausibility of gap detection could be improved by including linear features forest roads and torrent channels in TIMA to account for decoupled snow layer resulting from abrupt breaks on the hillslope. If the TIMA concept can be successfully applied to the case of avalanches, this would encourage its use in assessing other gravitational natural hazard processes.