Landslide susceptibility mapping in the Northern part of Los Glaciares National Park, Southern Patagonia, Argentina using remote sensing, GIS and frequency ratio model

Abstract

The valleys in the northern part of Los Glaciares National Park, Southern Patagonia, Argentina, are highly dynamic due to glacier shrinkage. To improve our understanding of slope instability in deglaciated environments, we generated and analyzed a Landslide Susceptibility Map (LSM). The proposed methodology includes remote sensing, interpretation of geological maps, both performed in a Geographic Information System (GIS), and finally a Frequency Ratio Model (FRM). Factors included in the slope instability analysis were lithology, geomorphology, land cover/land use, slope, aspect, elevation, curvature, distance to geologic fault, and distance to roads and trails. Slope instability events identified in the inventory include moraine landslides, debris flows, and rockfalls. The area is modeled by glacial action and has numerous geographic features related to erosional and depositional processes of glaciofluvial, cryogenic, and paraglacial origin. According to their high frequency ratio values, geomorphology, land cover and land use, lithology, and elevation are the most influential predisposing factors. The three best performing classes are Quaternary deposits, ice-contact topography, and the innermost lateral moraines on the western side of the study area, which surround glaciers and proglacial lakes. The performance and accuracy of the LSM was evaluated and verified using the Area Under Curve (AUC) with an accuracy of 0.804. The study provides knowledge of the deglaciated environment and landslide-prone areas in the study area. In this way it will help to provide tools for action to prevent and manage potential hazards to infrastructure, residents and/or tourists.

Keywords

• Predisposing factors
• Remote sensing
• Frequency ratio
• Landslide susceptibility map
• Los glaciares national park