Key-words:silver eel, downstream migration, behaviour, hydroelectric plant, trashracks, radio-tracking Between 2004 and 2007, 116 downstream migrant silver eels (Anguilla anguilla) were monitored at a hydropower plant on the Gave de Pau river in South-West France using radio and PIT telemetry. The objectives of the study were: (i) to determine the environmental conditions when eels arrived and passed the facility; (ii) to determine the rate of eel escapement (passage other than via the turbines); (iii) to describe the behaviour of eels faced with the intake structure and the permeability of the intake trashracks for the different sizes of eel; and (iv) to determine whether surface bypasses originally designed for salmon could be effective for eels. Five types of behaviour of silver eels in the forebay and at the plant intakes were identified. The study showed the key factor influencing both eel behaviour and the route taken through the plant was variation in river discharge. Escapement rate was related to eel length and the spill flow to river flow ratio, which could be described by a logistic regression model. The surface bypasses originally designed for salmon were found to aid downstream eel migration significantly. At velocities < 0.40 m·s −1 , no eels, even the largest, for which the racks are a physical barrier, were found impinged on the trashracks.

-Obstacles in rivers are considered to be one of the main threats to diadromous fish. As a result of the recent collapse of the European eel, the European Commission introduced a Regulation, requiring to reduce all sources of anthropogenic mortality, including those caused by passing through hydropower turbines. Improving knowledge about migration triggers and processes is crucial to assess and mitigate the impact of obstacles. In our study, we tracked 97 tagged silver eels in a fragmented river situated in the Western France (the River Dronne). Using the movement ecology framework, and implementing a Bayesian state-space model, we confirmed the influence of river discharge on migration triggering and the distance travelled by fish. We also demonstrated that, in our studied area, there is a small window of opportunity for migration. Moreover, we found that obstacles have a significant impact on distance travelled. Combined with the small window, this suggests that assessment of obstacles impact on downstream migration should not be limited to quantifying mortality at hydroelectric facilities, but should also consider the delay induced by obstacles, and its effects on escapement. The study also suggests that temporary turbines shutdown may mitigate the impacts of hydropower facilities in rivers with migration process similar to those observed here.

No abstract