Species-separation efficiency of small beam trawl for mantis shrimp in Tokyo Bay

Abstract

A separator-trawl used for a mantis shrimp catch in Tokyo Bay has a separator panel and two codends; separator net-panel (100 mm stretched mesh length) on part of the top of the first codend (62 mm mesh length) which is covered with the second codend (33.6 mm mesh length). The fishermen hoped mantis shrimp as a main target species would be caught in the second codend and be separated from starfishes retained in the first codend, because large amounts of starfishes cause difficulty in fishermen's hand-sorting on deck. Separation efficiencies of the trawl were obtained from the catch numbers in the two codends in a fishing experiment. Separation efficiency between species i and j is defined as γi+γj-1, where γi is the recovery ratio in the codend for i-species, that is, a proportion of the i-species catch number in the i-species codend. While the recovery ratio for starfish in the first codend was 0.96, that for mantis shrimp in the second codend was 0.66 which was not so high as expected. Thus, separation efficiency between mantis shrimp and starfish was 0.62. Although the separator net-panel has enough large mesh size for small animals, e.g. small crab Carcinoplax vestina, some species of low swimming ability failed to pass through the separator panel and were retained in the first codend. This implies that separation performance of the separator-trawl depends on the swimming ability of each species in the net as well as mesh selectivity of the separator net panel.