Rivers in reverse: Upstream-migrating dechannelization and flooding cause avulsions on fluvial fans

Abstract

The process of river avulsion builds floodplains and fills alluvial basins. We report on a new style of river avulsion identified in the Landsat satellite record. We found 69 examples of retrogradational avulsions on rivers of densely forested fluvial fans in the Andean and New Guinean alluvial basins. Retrogradational avulsions are initiated by a channel blockage, e.g., a logjam, that fills the channel with sediment and forces water overbank (dechannelization), which creates a chevron-shaped flooding pattern. Dechannelization waves travel upstream at a median rate of 387 m/yr and last on average for 13 yr; many rivers show multiple dechannelizing events on the same reach. Dechannelization ends and the avulsion is complete when the river finds a new flow path. We simulate upstreammigrating dechannelization with a one-dimensional morphodynamic model for open channel flow. Observations are consistent with model results and show that channel blockages can cause dechannelization on steep (10–2 to 10–3), low-discharge (~101 m3 s–1) rivers. This illustrates a new style of floodplain sedimentation that is unaccounted for in ecologic and stratigraphic models.

Coseismic river avulsion on surface rupturing faults: Assessing earthquake-induced flood hazard

“…The potential for overbank flow and dechannelization upstream of an FSB, and thus relative FIRA hazard, is dependent on the characteristics of the backwater that forms after faulting. Higher backwater surface elevations over a longer reach maximize the potential for overbank flow and eventual avulsion ( 5 ). A prior study on retrogradational avulsions using sediment transport models and observations of natural avulsions demonstrated that dechannelization is dependent on backwater evolution upstream of a channel blockage.…”

“…Topography surrounding a faulted river channel (i.e., levees, terraces, or antecedent structures), footwall subsidence, and downstream vertical throw may obstruct rivers and promote the formation of a local backwater upstream of the fault. Resultant increases in water surface elevation upstream of the fault may consequently promote overbank flow, wherein water is directed across the floodplain ( 5 ). The nature of any local topographic constriction (or lack thereof), surrounding vegetative cover, and the erosional resistance of the floodplain will contribute to whether fault-impounded river flow is able to effectively mobilize across the floodplain ( 24 , 25 ).…”

Quantitative constraints on flood variability in the rock record.

“…These occur in extreme discharge events, with overbank flow high enough to entrain large stems and braches from the floodplain, with the potential to uproot and incorporate additional living plant material. Previous work also suggests that debris beds like these may be linked with channel avulsion events, and associated with avulsion-dominated systems 44 . Large woody debris and log-jam deposits have been documented in very similar tectono-climatic settings: Trümper et al 5 300 Ma.…”

“…Debris bed units are between 1 and 4 m in thickness, and exist at the bases of barforms, accompanied by abundant upper plane bed sub-parallel lamination and some lower flow regime cross-bedding, indicating high flow transport stages. The chaotic orientation and sorting of these debris beds is typical of storm event beds that occur when extreme precipitation causes high-energy flood discharges, which were clearly capable of recruiting large clasts from the river levees and floodplain 5,44,45 .…”

A method to detect abrupt shifts in river channel position using a Landsat‐derived water occurrence record

“…Developing methods to detect avulsions and cutoffs across basin‐ to continental‐scale areas would add to our understanding of fluvial processes. Regional‐scale datasets of manually compiled, remotely sensed avulsion observations have already led to insight on path selection (Edmonds et al, 2016, 2022) and crevasse splay production on avulsing rivers (Lombardo, 2017). Additionally, these smaller datasets suggest that avulsive behaviour is related to stream morphology and systematically changes downstream (Valenza et al, 2020), possibly in response to downstream fining (Dingle et al, 2020).…”