Differential Bank Migration Limits the Lifespan and Width of Braided Channel Threads

Successful management of flooding and erosion hazards on floodplains depends on our ability to predict a river channel's shape and the lifespan during which it will continue to flow. Recent progress has improved our understanding of what sets the lifespan and width of single-thread channels; the next challenge is to extend this knowledge to braided channels and their interwoven sub-channels (threads). In this study, we investigate the lifespan and width of braided channel threads in a large experimental data set, coupled with particle-image velocimetry-derived measurements of riverbank erosion and accretion. We find that, unlike single-thread channels, braided channels in the experiment do not exhibit an equilibrium between bank erosion and accretion. Instead, bank erosion outpaces lateral accretion, causing individual threads to widen and infill until they are abandoned. Thread lifespan is limited to the time it takes for threads to triple their width: tripling of the width yields enough bank material to aggrade more than half the channel depth, at which point flow is rerouted to a narrower thread. In consequence the width of active threads is limited to three times their initial width. Threshold channel theory accurately predicts the median thread width, which is roughly double the initial width and two-thirds the limiting width. The results are consistent with existing field data and suggest that differential bank migration is sufficient to explain why braided channels show greater width variability and higher width-to-depth ratios than their single-thread counterparts.