Seabird mortality in the longline fishery for Patagonian toothfish at the Prince Edward Islands: 1996–1997

Longline fishing for Patagonian toothfish Dissostichus eleginoides commenced at the Prince Edward Islands in October 1996, but unsanctioned fishing commenced some months before this date. This paper summarizes the mortality of seabirds associated with this fishery during the year July 1996-June 1997. Data on seabird bycatches were obtained from fishery observers aboard 10 of 12 sanctioned fishing trips. Some 3.2 million hooks were observed during the period November-June, representing 84 % of the estimated 3.8 million hooks set by permit-holders in the South African Exclusive Economic Zone (EEZ). However, the total fishing effort is estimated, based on landings, to be between 20 and 50 million hooks. Average bird bycatch rate was 0.289 birds per 1000 hooks set. However, there was considerable variance between vessels (range 0.004 to 1.468 birds per 1000 hooks) as a function of fishing season, time of setting, distance from the Prince Edward Islands, mitigation measures employed, and type of fishing gear used. Observers reported 923 birds of 10 species killed. The species most affected were White-chinned Petrels Procellaria aequinoctialis (73 % of the total killed), Grey-headed Thalassarche chyrsostoma and Yellow-nosed Albatrosses T. chlororhynchos (together 23 %), and giant petrels Macronectes spp. (4 %). Extrapolating to the total fishery suggests that between 5 000 and 20 000 birds were killed during 1996-97. This excludes birds killed that were not hauled aboard vessels. Ten toothfish hooks were found in birds or in bird pellets on Marion Island during 1996-97. Almost all birds killed were breeding adults, and there was a significant bias towards males in at least three species. More than 1 % of four or five local breeding seabird populations were killed during the 1996-97 season. Given the low reproductive rate of these species, these levels of mortality (exacerbated by mortality in other longline fisheries) are not sustainable, and will result in local population declines. Bird mortality can be largely avoided by setting lines only at night, using a CCAMLR-approved bird-scaring (tori) line during setting, and not fishing during the main summer breeding season (or at least fishing >200 km from the breeding islands during this period). These measures were largely disregarded during most of the 1996-97 season. Unfortunately, the necessity for a presence around the islands to reduce the impact of uncontrolled fishing renders the last option impractical in the short term. It is important that observers continue to be placed aboard all vessels to monitor catches of both fish and birds, and to ensure compliance with regulations to reduce bird bycatch.