HR: 14:25h
AN: H42J-04    [PDF]
TI: Runoff Variability and Erosional Landscape Evolution: Insights from Numerical Modeling
AU: * Lague, D
EM: dlag02@esc.cam.ac.uk
AF: Dpt of Earth Sciences, University of Cambridge, Downing Street, Cambridge, CB2 3EQ United Kingdom
AU: Davy, P
EM: philippe.davy@univ-rennes1.fr
AF: Geosciences Rennes, UMR 6118 CNRS, Campus de Beaulieu, Rennes, 35042 France
AU: Hovius, N
EM: nhovius@esc.cam.ac.uk
AF: Dpt of Earth Sciences, University of Cambridge, Downing Street, Cambridge, CB2 3EQ United Kingdom
AB: Fluvial incision and sediment transport are driven by variable precipitation, with a common emphasis on rare, large events. The frequency of such events has changed with climate. Records of past runoff and its variability, and associated landscape evolution are scarce, and do rarely permit evaluation of the geomorphic importance of extreme events. This problem can be addressed from a theoretical point of view. Recent studies have shown that in the case of a mechanical threshold for particle detachment and transport, increasing runoff variability gives rise to elevated incision and sediment transport rates, even when the intensity of geomorphic processes depends less than linearly on runoff. Thresholds are a fundamental characteristic of many geomorphic processes. We have studied the effects of runoff variability and changes in runoff variability on incision and sediment transport, and the geomorphic response time, using a numerical surface process model (EROS). This model pairs a stochastic magnitude-frequency distribution of runoff events with deterministic erosion and transport formulations. Computationally inexpensive, the stochastic runoff distribution provides a good description of daily discharge datasets. It can be tuned for various watersheds. Incision and transport laws were calibrated to predict realistic steady-state slope-area relationships. Given the uncertainties on the descriptions of erosion and sediment transport in natural systems, we have explored different cases thought to be characteristic of bedrock or alluvial rivers. Our results imply that the response time of alluvial rivers is more sensitive to change in runoff distribution (mean and variability) than bedrock rivers. In particular, bedrock river response time diminishes as the inverse square root of the mean runoff rate, and decreases with runoff variability, when a mechanical threshold exists. In the case of very variable runoff, alluvial rivers can enter a regime where the mean long-term sediment flux is completely governed by the largest runoff event. This regime is not found for bedrock rivers where each erosive event contributes to the long-term incision rate. This is consistent with field observations of bedrock incision in Taiwan.
DE: 1815 Erosion and sedimentation
DE: 1821 Floods
DE: 1824 Geomorphology (1625)
DE: 1869 Stochastic processes
DE: 3210 Modeling
SC: Hydrology [H]
MN: 2003 Fall Meeting