Linking debris-flow hazard assessments with geomorphology

Debris-flow hazard assessment schemes are commonly based on empirical, physical, or numerical methods and techniques. Inherent in all methods is generally the assumption of unlimited sediment supply. This study compares model inputs of sediment requirements for debris flows with estimated sediment reproduction from both solifluction and rockwall retreat. The analysis is carried out in Bildudalur, a community in the Westfjords of Iceland. Geomorphic techniques are applied to determine the set of natural processes acting in this landscape to estimate spatial distribution of relevant processes, to approximate level of processes activity, and to provide information for scenario modeling. Debris-flow volumes are determined by coupling rainfall magnitudes and catchment sizes with average sediment contents. Rockwall retreat and solifluction rates are based on literature reviews. For a rainstorm with a 10-year return period, debris-flow volumes are calculated for 12 different creeks. Rates are assumed for solifluction with a velocity of 0.25 m/yr at an average depth of 0.5 in and for rockwall retreat with 2 mm/yr. Comparing sediment requirements with estimated sediment reproduction leads to a factor of deficit ranging between 6.2 and 8.5. Thus, the sediment storage is not refilled as fast as the next potential triggering rainfall occurs. Consequently, if a debris flow has occurred in the past, all sediment is removed, and the following rainstorm event is 'just' causing a flood, which is by far less destructive than a debris-flow event. The challenge of future debris-flow hazard-assessment schemes is to include geomorphic analysis to be able to obtain more sustainable results. (c) 2004 Elsevier B.V. All rights reserved.