Landslide-Induced River Disruption : Geomorphic Imprints and Scaling Effects in Alpine Catchments of South Westland and Fiordland, New Zealand

Abstract

This study examines the geomorphic long-term and long-range implications of large landslides within alpine valley trains of South Westland and Fiordland, New Zealand. A regional-scale reconnaissance suggests that slope instability has forced reach-scale adjustments to landslide-induced river disruption such as obliteration and burial of alpine valley floors, stream blockage, occlusion, and excessive lateral sediment input. The preservation of numerous large and prehistoric deposits from deep-seated failures in bedrock is seemingly at odds with extremely high rates of erosion and local rock mass strength in alpine landscapes dominated by excessive precipitation and neotectonic activity. The geomorphic implication of these large-scale mass movements is significant in that it attests their role not only in representing important sources of sediment, but also scale-dependent controls on alpine sediment flux. It is argued that, apart from being dominant controls on alpine landscape denudation by total affected area, the temporal up- and downstream scaling effects of large landslides may modify or control catchment response on time-scales ranging from 10 1 to at least 10 4 years. Investigation of geomorphic landslide impact signals on these scales thus bridges a significant gap between process-dominated local case studies and coupled uplift/erosion models in long-term mountain belt evolution. Geomorphometric analysis of landslide dams and related phenomena underlines the multivariate character of processes and landforms related to the coupling of landslides and mountain river channels. Using a simplistic approach, several quantitative indices as well as modelling based on digital elevation data are presented as a preliminary means to objectively rank and potentially predict the magnitude and stability of landslide dams. In terms of assessing the geomorphic hazard and risk from landslide-induced river disruption, the varying degrees of data constraints strongly necessitate the integration of quantitative analysis with site-specific field investigations and qualitative observations.