The relationship between channel avulsion, flow occupancy and aggradation in braided rivers: insights from an experimental model

Experimental modelling of an aggrading braided river has allowed investigation of the relationship between the frequency of channel avulsion ( A _f), the duration of time that the braidplain is occupied by flow, the spatial pattern of braidplain sedimentation and how these respond to a change in sediment supply ( S _s). Model results demonstrate a strong, positive relationship between S _s and A _f and that there is no downstream change in A _f over the short braidplain distances ( ca 100 m) modelled herein. Although A _f is strongly dependent on S _s, the degree of channel switching does not influence the rate, or spatial pattern, of braidplain sedimentation. All experiments used a single, central input for water and sediment, and the channels occupied the centre of the alluvial basin for a longer period of time than the margins for all sediment supply rates and distances downstream. Despite this spatio-temporal pattern in flow occupancy, braidplain sedimentation rates were nearly uniform both downstream and across the basin, and increased at approximately the same rate as increases in S _s. As a consequence, less frequent, and possibly short-lived, flows at the margins of the braidplain deposited and preserved more sediment per unit time in comparison with the centre of the basin where flow occupancy was higher. An approximate order of magnitude change in sediment supply resulted in only a factor of two change in bed slope, probably due to both an increase in channelization and adjustment of the channel form that maintained sediment transport through the basin. This result suggests that linear diffusion models are unlikely to be applicable in landscape evolution models that possess aggrading multi-thread rivers, which are capable of self-adjustment in channel number and form.