Identification of Debris-Flow Channels Using High-Resolution Topographic Data: A Case Study in the Quebrada del Toro, NW Argentina

Resolving Earth's surface at the meter scale is essential for an improved understanding of the dynamics of mass-movement processes. In this study, we explore the applicability and potential of digital elevation models (DEMs) derived from stereophotogrammetry to detect debris-flow channels in the Quebrada del Toro in the northwestern Argentine Andes. Our analysis relies on a high-resolution (3 m) DEM created from SPOT-7 tri-stereo satellite data. We carefully validated DEM quality with similar to 6,000 differential GPS points and identified optimal parameters for DEM generation in high-relief terrain. After multiple processing steps, we achieved an accuracy of 0.051 1.915 m (1 sigma) using n = 3,139 control points with cm precision. Previous studies have used the drainage area and slope framework to identify topographic signatures of debris flows within a catchment. We built upon this and investigated individual river-channel segments using connected-component (CC) analysis on meter-scale topographic data. We define CC as segments of similar slope along the channel profile. Based on seven manually identified debris-flow catchments, we developed a debris-flow similarity index using component length and mean channel-segment slope and identified channel segments that have likely been shaped by debris flows. The presented approach has the potential to resolve intra-catchment variability of transport processes, allows to constrain the extent of debris-flow channels more precisely than slope-area analysis, and highlights the versatility of combined space- and field-based observations for natural-hazard assessments. Plain Language Summary Debris flows are potentially destructive mass movements that involve a mixture of water and sediment. They regularly disrupt infrastructure and threaten local populations in mountainous regions worldwide. For hazard assessment, it is important to locate areas where debris flows occur. Debris flows carve into bedrock, forming continuously steep channels for long distances of several hundred meters. This research locates channels that were shaped by debris flows using gridded elevation models of the Earth's surface. These models are generated using overlapping satellite-image pairs that show the study area from different viewing angles. Similar to what the human brain does by merging two images from our eyes to produce depth perception, we use algorithms to turn pairs of satellite images into height information. From the surface model, we can derive individual river channels and dissect these into segments that have a common channel slope. In a following step, we compare all channel segments to known debris-flow channels in the landscape and evaluate how similar they are. We detect the signature of debris flows primarily in regions near geologic faults where sufficient loose material is available to fuel a debris flow and steep hillsides provide enough gravitational energy to transport this material downstream. Key Points SPOT-7 tri-stereo imagery may be used to retrieve high-quality, 3 m digital elevation models for geomorphic and hazard analyses We derive a new metric for debris-flow mapping using connected components of channel-slope segments We develop a debris-flow similarity index to classify channel pixels and distinguish between debris-flow and fluvial sections