A MODEL FOR THE MECHANISM OF ALLUVIAL RIVER CAPTURE

River networks are dynamic features of the Earth’s surface that evolve through time via divide migration and drainage captures. River capture occurs when a drainage divide is breached, and the upstream segment of a drainage basin is diverted into a neighboring basin. This geomorphological process abruptly changes the routing of water and sediment to depositional basins and directly affects the dispersal pathways of aquatic organisms. Recent efforts to understand the mechanisms of river capture have focused on high-relief bedrock rivers, while river capture in low-relief alluvial settings has received less attention. Geomorphic observations and biological data suggest that alluvial river captures have occurred in the past, yet it is unclear how an alluvial river capture starts and what conditions lead to a successful capture. Alluvial rivers, in contrast to bedrock rivers, tend to have low-relief drainage divides that can sometimes be overtopped by high flows. We propose that an alluvial river capture begins as a partial diversion of flow into an adjacent basin, creating a bifurcated channel. We adapted a morphodynamic model of a river bifurcation to study the evolution of an ongoing alluvial river capture. We performed numerical experiments to test the influence of initial conditions on the fate of the bifurcation and capture. The results indicate that initial conditions such as slope differences and discharge partitioning have important consequences for the fate of the river capture. Similar slopes and a small fraction of discharge diverted into the intruding basin will not result in capture, whereas a steeper slope and larger initial discharge diversion into the intruding basin will make a successful capture more likely. We apply our numerical model to the ongoing capture of the Orinoco River by the Casiquiare River, a tributary of the Amazon River in South America.