Morphodynamic evolution following sediment release from the world’s largest dam removal

Abstract

Sediment pulses can cause widespread, complex changes to rivers and coastal regions. Quantifying landscape response to sediment-supply changes is a long-standing problem in geomorphology, but the unanticipated nature of most sediment pulses rarely allows for detailed measurement of associated landscape processes and evolution. The intentional removal of two large dams on the Elwha River (Washington, USA) exposed ~30 Mt of impounded sediment to fluvial erosion, presenting a unique opportunity to quantify source-to-sink river and coastal responses to a massive sediment-source perturbation. Here we evaluate geomorphic evolution during and after the sediment pulse, presenting a 5-year sediment budget and morphodynamic analysis of the Elwha River and its delta. Approximately 65% of the sediment was eroded, of which only ~10% was deposited in the fluvial system. This restored fluvial supply of sand, gravel, and wood substantially changed the channel morphology. The remaining ~90% of the released sediment was transported to the coast, causing ~60 ha of delta growth. Although metrics of geomorphic change did not follow simple time-coherent paths, many signals peaked 1–2 years after the start of dam removal, indicating combined impulse and step-change disturbance responses.

Geomorphic Evolution of a Gravel‐Bed River Under Sediment‐Starved Versus Sediment‐Rich Conditions: River Response to the World's Largest Dam Removal

“…Our observations indicated little river‐corridor storage on the Elwha River, at least relative to the sediment proportion that reached the river mouth; sediment‐budget calculations indicate that only 10% of the evacuated reservoir sediment remained stored in the river corridor as of 2016 (Ritchie, Warrick, et al, ). However, the two Elwha reservoirs occurring in sequence meant that some sediment from Lake Mills was depositing within the Aldwell reservoir reach and in the lowest 2 km of the middle Elwha River, while the Aldwell sediment deposit was evolving, slowing net sediment export (longitudinal profiles in supporting information; East et al, ; Ritchie, Warrick, et al, ). The role of fluvial network structure in sediment‐pulse evolution and the position of dam removals relative to other, extant reservoirs are topics requiring additional research (Foley, Magilligan, et al, ).…”

Formation and Removal of a Coastal Flood Deposit

“…In some cases, flood‐derived sands may be transported distally by gravity flows to form sand bodies on the middle shelf (e.g., Steel et al, ), and thus bypass beaches (Warrick & Milliman, ). The fate of sands eroded from the Elwha flood deposit is not well constrained, but based on the tapering of the 2014 flood deposit between T1 and T2 (Figure ) as well as multibeam evidence of a sand body migrating east from the river mouth within ~2 km of shore (Ritchie, Warrick, et al, ), it seems likely that some of this sediment has become incorporated into the longshore transport system.…”

“…Not all rivers or floods are created equal, however, and some events may leave little stratigraphic record. Beginning in 2011, two unprecedented dam removals on the Elwha River in Washington State released ~10.5 Mt of sediment in 2 years (Ritchie, Warrick, et al, ; Warrick et al, ), and fluvial SSCs were frequently several grams per liter during rainstorms (Curran et al, ). In order to determine the mechanisms of dispersal and deposition related to intense sediment discharge, and to measure any gravity‐flow processes, we deployed 1–2 instrumented tripods near the river mouth nearly continuously from November 2011 to April 2014.…”

A decadal‐scale numerical model for wandering, cobble‐bedded rivers subject to disturbance

“…When two dams on the Elwha River were removed, the channel responded with increased bank erosion, the reactivation of floodplain channels, and local avulsion events (East et al ., ; Ritchie et al ., ). While much of the geomorphic change associated with the sediment pulse occurred within the first two years following removal, the channel continued to adjust after five years (Ritchie et al ., ). It is unclear how long it will take for the river to reach a steady state with its new sediment regime and whether it will trend to pre‐dam conditions or adopt a new steady state.…”