Comparison of sonar-, camera- and net-based methods in detecting riverine fish-movement patterns

Abstract

Monitoring of fish movement is important in understanding and optimising the functionality of fishways and in restoring riverine connectivity. This study compared fish monitoring data (ARIS sonar-based and GoPro camera-based), with catches in a multi-mesh stow net following downstream passage in a small river in Bavaria, Germany. In terms of the number of individuals, the sonar-based system (detection rate=62.6% of net-based catches) outnumbered the counts of the camera-based system (45.4%). Smaller specimens of <100 and <150mm were under-represented with the sonar and the camera-based systems respectively. Species identification based on the camera system was similar to that for net-based catch, whereas no proper species identification could be performed with sonar data. In conclusion, the sonar-based system can be recommended for the counting of fish >100mm during night and turbid conditions, unless species identification is necessary. During daylight and with clear water, cameras can be a cheaper and promising option to monitor species compositions of fish >150mm.