Showing papers by "Richard M. Iverson published in 2003"

The debris-flow rheology myth

Abstract: Models that employ a fixed rheology cannot yield accurate interpretations or predictions of debris-flow motion, because the evolving behavior of debris flows is too complex to be represented by any rheological equation that uniquely relates stress and strain rate. Field observations and experimental data indicate that debris behavior can vary from nearly rigid to highly fluid as a consequence of temporal and spatial variations in pore-fluid pressure and mixture agitation. Moreover, behavior can vary if debris composition changes as a result of grain-size segregation and gain or loss of solid and fluid constituents in transit. An alternative to fixed-rheology models is provided by a Coulomb mixture theory model, which can represent variable interactions of solid and fluid constituents in heterogeneous debris-flow surges with high-friction, coarse-grained heads and lowfriction, liquefied tails.

Prediction in geomorphology

Abstract: Widespread availability of computational power and the ability to acquire voluminous data have the potential to improve the accuracy and relevance of geomorphic prediction, although such improvement is by no means guaranteed. A mix of fast and slow geomorphic processes, unknown initial and forcing conditions, unresolved heterogeneities, and unanticipated, typically nonlinear mechanisms make prediction difficult and problems of scale paramount. A variety of approaches to formulating geomorphic models exist, reflecting the diversity of modeler's backgrounds and goals and our incomplete understanding of many earth surface processes. Progress is facilitated by a continuing dialog among those pursuing different approaches to making and testing predictions. This paper provides a synopsis of contributions to a volume that focuses on the broader objectives, methods, and testing of geomorphic models. Why do we model and to what purpose? How do we formulate and test models? How do we evaluate whether a model has predictive capability or any utility at all? Is prediction even necessary? Despite the diverse approaches to making predictions in geomorphology, common themes in the volume highlight the need for models that unambiguously link cause and effect, the importance of rigorous and flexible model tests, and the utility of models in focusing and organizing understanding and expressing concepts in unambiguous and testable ways.

Surge dynamics coupled to pore-pressure evolution in debris flows

Abstract: Temporally and spatially varying pore-fluid pressures exert strong controls on debris-flow motion by mediating internal and basal friction at grain contacts. We analyze these effects by deriving a one-dimensional model of pore-pressure diffusion explicitly coupled to changes in debris-flow thickness. The new pore-pressure equation is combined with Iverson’s (1997) extension of the depth-averaged Savage-Hutter (1989, 1991) granular avalanche equations to predict motion of unsteady debris-flow surges with evolving pore-pressure distributions. Computational results illustrate the profound effects of pore-pressure diffusivities on debris-flow surge depths and velocities.

When Models Meet Managers: Examples from Geomorphology

Abstract: Geomorphic models are increasingly used to support public policy and natural resources management. We present five examples of the interaction between models and managers and consider factors that influence their success or failure. Essential elements include common objectives for management and models and clear communication of the assumptions, limitations, and uncertainty of models and their predictions. Where management and modeling objectives cannot be matched, it may be possible to define management actions that do not depend on exact predictions or to pursue alternatives to modeling such as monitoring or environmental history. In some cases, model predictions may be less important than the educational value of model construction and operation. An adaptive modeling process, in which the objectives, mechanisms, and tolerances of a model are adjusted interactively in an ongoing model-manager dialogue may be useful, particularly when the policy context is contested or incompletely defined or when the social mandate is ahead of the science.