The Mechanics of River Capture: How the Barrier Between Rivers Fails
River capture is commonly explained by headward erosion, yet channel heads end below drainage divides. How does a new connection open across the unchanneled surface that remains? I use analytical solutions and numerical experiments to examine the surface between a lower ‘captor’ and a higher ‘threatened’ river as a ‘barrier’ whose local crest must be lost before capture. Its ‘holding relief’ is the greatest difference between the riverbed elevations that a steady crest can sustain. Under uniform linear diffusion it grows with the square of the width and with the incision of both rivers, the threatened river’s counting twice. Limiting slopes, cohesion and channels to the crest set other limits, and a large difference between the beds can leave a crest under a meter high. The crest outlives the limit. Under slow approach and equal incision it lasts about a further sixth of the barrier’s diffusive time, whatever the rate of approach, and less when the captor incises faster or slopes steepen toward their limit. Under fast approach a gentle barrier lasts almost until the captor’s head arrives. A threatened river whose incision slows is weakened twice, as the elevation difference grows and its holding relief falls. A capture can therefore lag its cause, a delay omitted by instantaneous rules for an unresolved barrier. In flow-routing experiments, a breach captured drainage only where it reached an established river. Erosion of the whole surface between two riverbeds thus permits capture without a channel cutting through the divide.
Authors
- Daniel Peifer (ORCID: https://orcid.org/0000-0001-9238-5072)
Institutions
- Universidade Federal de Minas Gerais (BR)
Publication Details
- Journal
- EarthArXiv (California Digital Library)
- Published
- 2026-11-19
- DOI
- https://doi.org/10.31223/x5t23p
- Primary Topic
- Hydrology and Sediment Transport Processes
- Type
- preprint