Over the past decade, the farming of bluefin tuna has played an increasingly important role in the aquaculture of Japan. According to a recent report with respect to tuna farming, there are 160 tuna farms and 1432 cages of bluefin tuna in Japan. For the bluefin tuna farmer in Japan, to determine the number of farmed fish is one of the key issues to monitor the number of fish in addition to feed waste, escapement, behavior, and dead fish. However, there presently exists no reliable method to count the number of bluefin tuna in a cage. The most popular type of counter is currently the underwater stereoscopic camera system which has mainly been used by a diver for counting. However, this kind of counting method is not only labor-intensive, but its accuracy is low in dark or turbid water. The purpose of this study is to develop an accurate counting method for the farmed bluefin tuna using the multitransducer sonar and pinger. This newly developed a multi-transducer sonar system is based on counting the individual fish that has passed through “the sound curtain” consisting of 15 transducers（ 460 kHz）. In addition, a pinger was used to clarify the behavior and swimming speed of the caged fish. As a result, it was found that all of the fish regularly swim in a concentric circle in the cage space, and the lap time in each lane of 1m from cage center for one round was estimated by a linear regression equation. The number of fish in each lane for one round could be calculated by multiplying this lap time and the number of fish that passed through “the sound curtain” per unit time. The total number of fish could then be calculated by adding up the number of fish in each lane. The accuracy of this method verified by multiple actual tests was 1% or less and 1 fish. It was considered that this approach is effective for counting caged bluefin tuna with the objective of practical use.
