Variation in codend selectivity parameters between hauls and years in trawl nets for swordtip squid (Loligo edulis)

Abstract

Background: Trawl net fisheries are effective fishing activities, but often cause bycacth and discards issues for commercial species of unmarketable small size and endangered species. It is well accepted that codend mesh has size sorting properties for fish and other animals, and the codend selectivity is usually used as one of fisheries management measures to protect small animals. In the ocean, squids are abundant and still prospective fisheries resources, and especially swordtip squid is commercially important because of its good taste and high price in Japan and China. Objectives: Difference in selectivity parameters a, b and p could occur between years. This study objective is to analyze the variation between years and between hauls in selectivity parameters for swordtip squid (Loligo edulis) having a peculiar soft body type. Methods: A total of 23 experimental hauls were carried using trouser trawl onboard the research and training vessel Umitaka-Maru of the Tokyo University of Marine Science and Technology from 2006 to 2011, with 60mm mesh size test codend and 8.3mm mesh size control codend. Mantle length and girth of the swordtip squid caught were measured in cm. Catches were modeled by fitting codend selectivity parameters (a, b and p) in the Share Each LEngth Catch to the Total (SELECT) process and its variation between hauls and between years tested by the Akaike's Information Criterion (AIC) model selection. Results: The model with common logistic parameters a and b in each year and different split parameter p for each haul had a better estimation, confirmed by AIC. There was a variation in the 50% retention length (l50) from 6.04 to 7.11 cm and in the selection range, (S.R.) from 0.62 to 5.05 cm, which resulted in different selection curves annually. Conclusion: The difference in selection curves occurred although the codend used was the same. The fluctuation in S.R. was large compared with the l50, which could result from the small sample size. Others possible reasons are the yearly change in mesh size, catch size, and so on. This result is useful for fisheries managers to consider mesh size regulation for squid resource conservation.