Hydroacoustic estimates of fish densities in comparison with stratified pelagic trawl sampling in two deep, coregonid-dominated lakes

Reliable information on fish stocks is given crucial importance in fisheries science and ecosystem management. Among others, the European Water Framework Directive calls for consistent sampling methods that provide comparable and reproducible results. However, the standardized sampling of fish in lakes by gillnets is conducted by low effort only in the pelagic areas, thus potentially not reflecting between-lake variability in fish density. Here, we compared two additional fishing methods, hydroacoustics and midwater trawl sampling, to estimate pelagic coregonid densities in two deep lakes. Night-time trawling was conducted in four different depth layers and compared with the simultaneously obtained abundance and biomass estimates from echo recordings. Overall areal fish biomass from both methods corresponded to each other (R2 = 0.40), with slope and intercept of major axis regression not differing from unity. However, we found differences in estimates for the deepest layers fished, and for very small and the largest coregonids, most likely to be explained by the specific technological details of either method. Our study supports that either hydroacoustics or trawl sampling should accompany gillnet fishing to monitor pelagic fish assemblages. Because both methods result in estimates of absolute fish densities per volume or area, their application may uncover subtle responses of fish assemblages to anthropogenic disturbance in deep lakes.