River capture is thought to trigger abrupt changes in evolving continental drainage systems, but it is almost always inferred rather than observed, and the mechanisms that lead to capture are unclear. We shed light on these mechanisms by documenting an ongoing capture involving major South American rivers. The Rio Casiquiare is a distributary of the Rio Orinoco and a tributary of the Rio Negro; it forms a perennial water connection between the drainage basins of the Amazon and Orinoco, the largest and fourth‐largest rivers on Earth by discharge. This unusual configuration is the result of an incomplete and ongoing river capture, in which the Rio Negro is capturing the upper Rio Orinoco. We describe a positive feedback between diversion of water into the capturing channel and sedimentation within the channel being captured, a mechanism that could drive river capture in the Amazon and elsewhere.

No abstract

Rivers are conduits for aquatic organisms and host an exceptional number of species. Over geologic time, rivers and the aquatic organisms that live in them are subject to changes in topography that can alter where rivers flow. Differences in erosion rates across drainage divides cause some river basins to grow and others to shrink. Occasionally, rivers are abruptly rerouted by river captures that create both new dispersal corridors and barriers for aquatic organisms. These changes in habitat connectivity can lead to the evolution of new species, which has prompted suggestions that river captures may be a mechanism to produce high freshwater biodiversity. We test this hypothesis by building a model, Bio-SLANT (Biodiversity on Simulated LAndscapes using Neutral Theory). Bio-SLANT couples a computational landscape model that simulates river basin reorganization to a macroevolutionary model that simulates the dispersal, speciation, and extinction of organisms. We first show that modeled basin area exerts a primary control on within-basin species richness due to the species-area relationship. We then describe the effects of drainage area exchange between river basins. River capture increases species richness, but only temporarily, whereas elevated rates of speciation and extinction provide a persistent biological record of river network reorganization. When river captures are frequent, speciation rates increase more than extinction rates, resulting in a positive diversification rate under most of the biological parameters tested. We explore the implications of our results for species richness in landscapes with basins of different relative sizes and for diversification in tectonically active and inactive settings.

The high levels of biodiversity supported by mountains suggest a possible link between geologic processes and biological evolution. Freshwater biodiversity is high not only in tectonically active settings but also in tectonically quiescent montane regions such as the Appalachian Mountains. We show that erosion through different rock types drove allopatric divergence between lineages of the Greenfin Darter ( Nothonotus chlorobranchius ), a fish species endemic to rivers draining metamorphic rocks in the Tennessee River basin in the United States. In the past, metamorphic rock preferred by N. chlorobranchius was more widespread, but as erosion exposed other rock types, lineages of this species were progressively isolated in tributaries farther upstream, where metamorphic rock remained. Our results suggest a geologic mechanism for initiating allopatric diversification in mountains long after tectonic activity ceases.