ABSTRACT
Landslides are a global natural hazard, resulting in severe economic, environmental and social impacts every year. Often, landsliding occurs in areas of frequent and recurrent slope instability. However, residential areas and critical infrastructure are frequently built in the shadow of past landslide deposits and marginally stable slopes, often due to lack of awareness. These hazards, despite their scale and regional tendencies, can be difficult to detect from ground-based surveys, often due to vegetative cover. However, new developments in remote sensing technology, specifically Light Detection and Ranging mapping (LiDAR) are enabling researchers and practitioners with a new lens for viewing our unstable landscape. LiDAR data sets can be post-processed into Digital Terrain Models (DTMs) representative of the terrain beneath the vegetation, revealing the scars of past slope failures. This tool presents a revolutionary technique to mapping landslide deposits and their associated regions of risk; yet, inventorying is often done manually, an approach that is potentially be tedious, time-consuming and subjective. However, derived data sets from LiDAR present the opportunity to use remote sensing technology and typical landslide geometry to begin to automate detection and inventorying of landslide deposits on a landscape scale. An algorithm, called the Contour Connection Method (CCM) implements this function by detecting steep gradients, often associated with headscarps of a failed hillslopes, and initiating a search, marking deposits downslope of the failure. This approach has shown preliminary agreement with manual landslide inventorying in Oregon’s Coast Range, realizing almost 90% agreement with inventorying performed by a trained geologist. Landslide inventories of an Oregon highway corridor are presented demonstrating a consistent and objective mean for hazard mapping and future risk assessment. The global threat of landslides necessitates new and effective tools for inventorying hazardous areas to protect people, infrastructure and the environment from landslide hazards. Use of the CCM algorithm combined with judgment of geologists, geological engineers and geotechnical engineers may help better define these regions of risk in a consistent and objective manner.
