Mitigation of ecological impacts on fish of large reservoir sediment management through controlled flushing – The case of the Verbois dam (Rhône River, Switzerland)

Highlights: • Controlled flushing (CSFO) are new practices to tackle sediment trapping in reservoirs. • Impacts on fish of the first CSFO of the Verbois Dam (Switzerland) were assessed. • From hydroacoustics, fish density in the reservoir did not change after the CSFO. • From radiotelemetry, impacts below the dam were mainly behavioural rather than lethal. • CSFO should be promoted to recover storage volume while lowering ecological impacts. Sediment trapping within reservoirs is a worldwide phenomenon which impairs the ecological functioning of upstream and downstream ecosystems. It also reduces reservoir water storage volume, which lessens the services dams provide such as hydropower production or flood control and questions their sustainability. Hydraulic flushing is a widely used operation to recover the reservoir volume, but ecological impacts are massive. Recently, environmental awareness led dam operators to modify their management practices: 'Controlled Sediment Flushing Operations' (CSFOs) include environmental objectives in their implementation and are designed to be less harmful for aquatic ecosystems by controlling the flow and Suspended Sediment Concentration (SSC) downstream. However, CSFOs are not yet widespread, their ecological impacts are poorly documented, and comparisons with 'classical' flushing operations are unreported. Here, we analysed impacts on fish of the first CSFO of the Verbois reservoir in 2016, both upstream and downstream of the dam, and compared these with those from the empty flushing of 2012 using the same methodology (Grimardias et al., 2017). Time-series of hydroacoustics surveys enabled us to estimate the fish abundance in the reservoir, while radiotelemetry measured movements and apparent survival below the dam for four representative species. The 2016 CSFO lasted 10 days, and released a mean Suspended Sediment Concentration (SSC) of 3.47 g·L⁻¹ (peak = 11.98 g·L⁻¹). The fish density as assessed by the mean acoustic scattering strength S_A in the reservoir did not change significantly pre- and post-CSFO, and S_A seasonal estimates of year 2016 did not differ from those of 2015 and 2017. The apparent survival estimated from capture-recapture survey of marked fish (N = 118) decreased significantly during the CSFO for all species and differed across species, while the distances moved downstream increased. By comparison with the 2012 empty flushing, the 2016 CSFO allowed fish to remain in the reservoir, while impacts below the dam were mostly behavioural rather than lethal. Overall, despite significant impacts, the CSFO advantageously replaced 'classical' flushing from an ecological viewpoint. Provided that an acceptable balance between economical, ecological and technical aspects is found, CSFO can be considered for many reservoirs while accounting for their biological and physical site-specificity.