The Impact of the Toothfish Longline Fishery on Skate Populations around South Georgia

Abstract

Skate have been taken as by-catch in the South Georgia deepwater bottom longline fishery for toothfish since the fishery began in the late 1980s. However, the species impacted, their biology, the nature of the interaction with longlines, the impact of the fishery on them, and the likely methods of reducing these impacts have been little studied. The objectives of this study are to explore the impact of the longline fishery on skate populations around South Georgia and to propose methods for reducing the number of skate deaths. The study was carried out over the period 2001–2004, although data from before and after this time have been used in the analysis. The methodology involved observing the capture of skate on fishing vessels around South Georgia, a sub-Antarctic island in the south Atlantic; recording biological and fishery characteristics; designing field sampling protocols for a large group of scientific observers operating on other fishing vessels, undertaking taxonomic analysis and further biological analysis in shore-based laboratories, designing and executing a field experiment on skate discard survivorship; and statistical analysis of a large volume of scientific observer data. Three species were found to be caught on longlines. Amblyraja georgiana had a shallow distribution, being found predominantly between 200 and 500m, mostly on the continental shelf. A second species, referred to here as A. species anon, was clearly closely related to but significantly different from A. georgiana, having quite different colouration, a different clasper and pelvic structure, different length and age at maturity, and different bathymetric range (predominantly between 1200 and 1600 m, on the shelf slope but with significant numbers occurring in even deeper waters). The third species was Bathyraja meridionalis, which occurred between 300 and 2000 m. Of the three species caught within the fishery around South Georgia, B. meridionalis was found to grow to the largest size and weight of the species caught, reaching estimated, approximate total lengths at 50% maturity of 121 cm for males and 126 cm for females. Of the two Amblyraja species, A. sp. anon was the larger of the two, reaching estimated total lengths at 50% maturity of 96 cm for males and 98 cm for females. Insufficient data were collected for either sex to allow for estimates of length at 50% maturity for A. georgiana, but a mixed-sex estimate of total length at 50% maturity was 82 cm. For the survival experiment 102 skate from the three species were opportunistically selected from hauled longlines and placed into specially constructed circulating seawater tanks on a fishing vessel for 12 hours. Skate caught at depths between 1200 and 1300 m had an estimated 75% chance of survival, whereas the predicted survival possibility for skate caught at depths between 1300 and 1500 m dropped to 46%. Skate taken from depths greater than 1500 m were estimated to only have a 24% chance of recovery from capture. There was some suggestion that male skate survive better than females. Generalised linear models were constructed to examine the influence of a number of factors on the probability of capturing skate on a longline, and the catch rate of skate. Skate presence on the line and numbers caught were found to increase almost exponentially with depth, with no sign of either the probability of capture or the numbers caught falling off at depths to 1750 m. Month had a significant effect of the number of skate caught, with July having the highest skate catch per unit effort (numbers per hook) of the four month (May to August) fishing season. Soak time had a significant effect on the probability of skate capture which increased the longer a line was left on the bottom. Of the four subdivisions of the South Georgia fishery examined in this study, two areas, Shag Rocks and the north-west of South Georgia were found to have significantly higher skate presence and skate CPUE than the other two areas. Catch rates were significantly elevated when squid bait were used as opposed to fish bait. A number of mitigation measures could be employed to reduce the by-catch and mortality of skates taken as by-catch in the longline fishery. The easiest and most effective method would be to set a maximum depth limit for fishing, which would have the effect of catching fewer skate and ensuring that more of them survived being returned to the sea. A further easy mechanism would be to add a second swivel at the hook end of the hook line, which would allow the skate to rotate freely as they are raised to the surface from the fishing depth of over 1000 m, so reducing injury to the jaw. Other potential options, such as restricting the fishery in July and avoiding the use of squid bait are likely to be met with less enthusiasm by the fishing industry.