GSA Connects 2021 in Portland, Oregon

Paper No. 207-14
Presentation Time: 11:45 AM

MASS WASTING RUNOUT: A SIMPLE MODEL FOR PREDICTING LANDSLIDE RUNOUT AND THE TOPOGRAPHIC EVOLUTION OF HEADWATER CHANNELS


KECK, Jeffrey, Forest Resources Division, Washington Department of Natural Resources, Olympia, WA 98502; Civil & Environmental Engineering, University of Washington, Seattle, WA 98115 and ISTANBULLUOGLU, Erkan, Civil & Environmental Engineering, University of Washington, Seattle, WA 98195

A new model for routing a debris flow or other mass wasting volume through a watershed and tracking the topographic evolution of the channel network is proposed. The model was developed for headwater sediment production and transport studies but may be useful for landslide (and subsequent debris flow) hazard mapping as well. It uses a cellular-automata modeling framework and a precipiton conceptualization of the mass wasting volume to predict downslope mass wasting movement and interaction with basin topography. Like many previous cellular-automata models, two dimensional flow rates are approximated using a multi-direction slope algorithm and volumetric flux at an individual grid cell is controlled by established geomorphic transport laws for deposition and scour. The model is coded to run in the Python module Landlab but could be transferred to another coding platform such as ArcGIS. Model behavior is controlled with three parameters: mass wasting material angle of repose, a threshold flux value below which the mass wasting material is forced to deposit and a coefficient that controls erosion rates as a linear function of the depth-slope product of the mass wasting flow. Initial values for all three parameters can be obtained from a geologic field review of existing mass wasting deposits near the study location. Early comparisons with actual landslides and debris flows in the Olympic Mountains and Capital State Forest of Washington State show that with a little trial and error, the Mass Wasting Runout model can be parameterized to reproduce observed mass wasting runout behavior. Since the model is coded using a large-scale, mass balance conceptualization (precipiton) and cellular-automata framework, it can be run quickly and may be suitable for Monte Carlo type simulations.