Spatio-temporal-dependent characteristic evaluation of yellowtail (Seriola quinqueradiata) in aquaculture cages based on stereo camera measurements

Yellowtail ( Seriola quinqueradiata) is one of the most important aquaculture species in Japan. Stereo cameras are commonly used for monitoring fish growth, and their accuracy is crucial. However, growth monitoring with stereo cameras faces challenges, including depth-dependent and multiple measurement biases, which may be influenced by swimming speed. Swimming speed itself is also an important indicator of fish welfare. This study aimed to elucidate the spatio-temporal characteristics of yellowtail in aquaculture, focusing on average fork length accuracy and swimming speed variation. Experiments were conducted in the same cage across four different seasons. Fork length and swimming speed were measured using four stereo cameras placed at different depths. The results indicated a depth-dependent bias in fork length measurements of less than 1 % throughout the year, except under specific conditions such as feeding times. In contrast, swimming speed varied with depth, season, and special conditions. These findings suggest that fork length measurement accuracy remains relatively stable over time and across depths. To accurately estimate the average weight of fish in a cage, measurements at the central depth of the fish school under optimal lighting conditions are sufficient. However, swimming speed showed greater variability when measured with stereo cameras. A novel finding of this study is that swimming speed is influenced by fish size, depth, and season. These results indicate that monitoring swimming speed and implementing appropriate measures may help improve the welfare of fish in aquaculture cages. • Stereo cameras assessed fork length and swimming speed in yellowtail cages • Fork length bias was <1 % across depths and seasons, except at feeding times • Swimming speed varied with fish size, depth, season, and temporal conditions • Measurements at central depth under good lighting gave accurate weight estimates • Monitoring swimming speed can help improve fish welfare in aquaculture cages