European Union legislation, through the Water Framework Directive (2000/60/ EC), as well as national legislation, such as the 'Grenelle Environnement ' (2007) in France, requires restoration of ecological connectivity in streams to improve free circulation of migratory fish. Different methods (e.g. capture by trap or net, telemetry, hydroacoustics) are used to evaluate the efficiency of fish passes to estimate the migratory species abundance and analyse changes in their within-river distributions. Among these methods, hydroacoustics is non-intrusive, allowing long-term observation and description of fish populations based on physical properties of sound in water. However, the main limit to hydroacoustic tools is their difficulty in identifying species. Initially designed for military purposes, dual-frequency identification sonar (DIDSON) has been used in environmental management for a decade. This acoustic camera uses higher frequencies and more sub-beams than common hydroacoustic tools, which improves image resolution and then enables observation of fish morphology and swimming behaviour. The ability to subtract static echoes from echograms and directly measure fish length improve the species-identification process. However, some limits have been identified, such as automatic dataset recording and the low range of the detection beam, which decreases accuracy, but efficient tools are now being developed to improve the accuracy of data recording (morphology, species identification, direction and speed). The new technological properties of acoustic cameras, such as the video-like visualization of the data, have greatly improved monitoring of diadromous fish populations (abundance, distribution and behaviour), helping river and fisheries managers and researchers in making decisions.

No abstract

No abstract

The NERC and CEH trademarks and logos ('the Trademarks') are registered trademarks of NERC in the UK and other countries, and may not be used without the prior written consent of the Trademark owner. e-mail: emmrich@igb-berlin.de 2 Summary 1) Sampling of lake fish assemblages is a challenging task in fish science and the information obtained strongly depends on the choice of sampling gear. The use of more than one sampling technique is generally preferred in order to achieve a comprehensive view on fish assemblage structure. Therefore, knowledge of whether catches between fishing gears are comparable is crucial.2) We compared catches in benthic multi-mesh gillnets with fish biomass estimates obtained by vertical hydroacoustics in 18 European lakes strongly varying in morphometry and trophic status. Separate analyses were conducted for different depth strata and for several fish-length thresholds to account for depth-and sizeselective gillnet catches.3) Gillnet catches and hydroacoustically obtained fish biomass estimates were significantly correlated. The strength of correlations was independent of the fishlength thresholds applied, but varied across different depth strata of the lakes, with the strongest correlations occurring in the shallow strata.4) The results support the applicability of vertical hydroacoustics for the quantification of fish biomass in stratified lakes. Survey designs combining hydroacoustics with limited gillnetting at sampling dates shortly one after the other, the latter for the purpose of inventory sampling only, are a cost-effective strategy for sampling fish assemblages in lakes. However, gillnet sampling does not provide reliable fish density estimates in very deep lakes with separate, pelagic-dwelling fish assemblages.