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.