How and where to pass? Atlantic salmon smolt’s behaviour at a hydropower station offering multiple migration routes

We studied downstream migration behaviour of Atlantic salmon smolt at a hydropower station that offers four safe (weir, new incision gate on weir, downstream bypass, Archimedes screw) and one potentially unsafe (Kaplan turbines) migration routes. We followed hatchery smolts using radio (n = 17) and RFID (n = 200) telemetry. They were released 1.2 km upstream of the hydropower station in spring 2018, in four groups, at different water discharge (18 to 37.2 m3s-1) and temperature (12.2–17.6 °C) conditions. For radio-tagged smolts, the repartition of the migration routes was 41.2% for the bypass, 17.6% for both the Kaplan turbine and the weir, 11.8% for the Archimedes screw, 0% for the new incision and 11.8% of unknown route. For the RFID-tagged smolts, the repartition was 38.0% for the bypass, 56.5% for the weir and/or the Kaplan, 4.5% for the new incision and 1.0% for the Archimedes Screw. The median time to cross the hydropower station was 58 min, and 88.2% of the smolts succeed to continue their downstream migration. The results demonstrate a diversity of choices of migration routes and underline the importance to optimise the position of fish-friendly turbines and fish passes at hydropower stations to increase their attractiveness.