Linking debris flow erosion to channel-bed parameters: Geotechnical and UAS-based investigation of ten channels in Switzerland

Debris flow dynamics are governed by complex interactions between channel topography and the hydro-geomechanical characteristics of both the flow and its erodible bed sediments. However, systematic field data linking sediment characteristics to debris flow erosion and deposition remain scarce, largely because steep channels are difficult to access and their sediments are heterogeneous and poorly sorted. In this study, we investigated ten debris flow channels in Switzerland using repeated Unmanned Aerial Systems (UASs) surveys combined with geotechnical field and laboratory measurements. Multi-epoch digital surface models showed that erosion generally prevails over deposition where thalweg slopes exceed ≈ 2 0 ∘ , although both the magnitude and spatial patterns of erosion vary substantially among channels and between events. To explore the causes of these differences, we measured grain size distributions, saturated hydraulic conductivity ( K sat ≈ 1 0 − 7 to 1 0 − 3 m/s ), and in-situ mechanical resistance. Channels composed of fine-rich, less permeable sediments exhibited greater average erosion, possibly because reduced permeability hinders pore pressure dissipation during undrained loading. Penetrometer and Vane Shear Tests further indicated that freshly deposited wet sediments were mechanically weaker than coarser materials left after erosion. The strength of the deposits can then increase rapidly upon drying. Erosion was also greater in larger channels and catchments. By integrating geomorphological and geotechnical parameters, we classified the studied channels into four types, ranging from small, coarse gullies to large, fine-rich channels. This multi-site dataset offers rare field constraints on the relationship between sediment properties and debris flow erosion, while also demonstrating the potential and limitations of applying geotechnical field methods in debris flow channels. The results provide essential parameters for the calibration and validation of next-generation hydro-mechanical debris flow models.