Reconceptualising coarse sediment delivery problems in rivers as catchment-scale and diffuse

This paper assesses river channel management activities in the context of the interaction between coarse sediment delivery, climate change, river channel response and flood risk. It uses two main sources of evidence: (1) an intensive instrumentation of an upland river catchment using both traditional hydrometric and novel sediment sensing methods; and (2) a sediment delivery model that combines a treatment of sediment generation from mass failure with a treatment of the connectivity of this failed material to the drainage network. The field instrumentation suggests that the precipitation events that deliver sediment from hillslopes to the drainage network are different to those that transfer sediment within the network itself Extreme events, that could occur at any time in the year (i.e. they are not dependent on wet antecedent conditions), were crucial for sediment delivery. However, sustained high river flows were responsible for the majority of transfer within the river itself. Application of three downscaling methods to climate model predictions for the 2050s and 2080s suggested a significant increase in the number and potential volume of delivery events by the 2050s, regardless of the climate downscaling scenario used. First approximations suggested that this would translate into annual bed level aggradation rates of between 0.10 and 0.20 in per year in the downstream main channel reaches. Second, the importance of this delivery for flood risk studies was confirmed by simulating the effects of 16 months of measured in-channel simulation with river flows scaled for climate change to the 2050s and 2080s. Short-term sedimentation could result in similar magnitude increases in inundated area for I in 0.5 and I in 2 year floods to those predicted for the 2050s in relation to increases in flow magnitude. Finally, we were able to develop an alternative approach to river management in relation to coarse sediment delivery, based upon reducing the rates of coarse sediment delivery through highly localised woodland planting, under the assumption that reducing delivery rates should reduce the rate of channel migration and hence the magnitude of the bank erosion problem. Thus, the paper demonstrates the need to conceptualise local river management problems in upland river environments as point scale manifestations of a diffuse sediment delivery process, with a much more explicit focus on the catchment scale, if our river systems are to become more insulated from the impacts of future climate changes. (C) 2007 Elsevier B.V. All rights reserved.