KIRK BRYAN AWARD: HOW DEBRIS FLOWS

Field observations, large-scale experiments, and physical theory provide compelling evidence that traditional models of debris flows and related phenomena are inappropriate. Unlike small-scale laboratory mudflows, debris flows with abundant coarse sediment do not behave as simple rheological materials that have fixed viscosities and yield strengths. Moreover, no rheological model provides an adequate framework for understanding debris-flow dynamics and deposits, because the apparent rheology of debris flows varies strongly as a function of time and position.

Key characteristics of debris flows, which are observed in the field but are more readily measured in controlled experiments, include (1) mobilization from wet but nearly rigid soil or sediment that liquefies as a result of pore collapse and/or granular agitation during frictional shear failure, which may occur in a single mass or in numerous masses that subsequently coalesce; (2) efficient grain-size segregation that produces concentrations of coarse clasts at flow margins; (3) unsteady flow surges with steep, highly concentrated heads and gradually tapered, dilute tails; (4) strong flow regulation by three-dimensional path topography; (5) deposition that occurs preferentially where flow escapes lateral confinement; and (6) formation of depositional levees and lobes that are constrained by friction concentrated at flow margins.

All of these characteristics are consistent with a new, two-phase debris-flow model that de-emphasizes rheology and establishes clear mechanical connections between debris flows and related phenomena such as rock avalanches and flash floods. The model idealizes debris flows as variably liquefied mixtures of Coulomb frictional solids and Newtonian viscous fluid, which obey Terzaghi's effective-stress principle and Darcian pore-pressure diffusion. The model contains no unconstrained parameters, conserves mass and momentum in four (space + time) dimensions, and produces good matches with experimental data. A great remaining challenge involves evaluation of mass-change terms that account for progressive erosion and deposition by debris flows.