Modeling Debris Flow Transitions: Experimental Validation and Field-Scale Application

Debris flows, prevalent in mountainous regions, exhibit distinct dynamics depending on whether they occur over bedrock (rigid bed) or accumulated deposition (erodible bed). Understanding the transition between these bed types is essential for hazard prediction and mitigation. This study improves an existing unsteady, non-uniform debris flow model to more accurately simulate the evolution of flow depth and velocity under varying boundary conditions. The improved model is grounded in mass, momentum, and kinetic energy conservation principles, incorporating a linearized μ(I) rheology to describe granular flow behavior and Coulomb friction along sidewalls, ensuring a realistic representation of debris flow mechanics.

To validate the improved model, granular dam break experiments were conducted in a narrow glass channel (3.5 m long, 0.04 m wide) with varying downstream deposit depths to establish different basal boundary conditions. High-speed camera footage and Particle Tracking Velocimetry (PTV) were employed to capture granular motion and generate velocity fields. The model exhibited good agreement with experimental results, accurately predicting the flow depth and velocity evolution during the transition between rigid and erodible beds.

Furthermore, the model was applied to field-scale debris flows at the PuTunPunas River in southern Taiwan, a site that has experienced several debris flow events over the past decades. Channel width variations at this site were incorporated into the model to assess erosion potential and flow behavior under real-world conditions. Comparisons with field observations confirmed the model’s capability to simulate debris flow transitions and erosion processes in natural channels, offering valuable insights for hazard assessment and mitigation in mountainous regions.