Estimates of marine mammal, sea turtle, and seabird mortality in the California drift gillnet fishery for swordfish and thresher shark, 1996-2002

Marine Fisheries Review, Spring, 2004 by James V. Carretta, Tim Price, Don Petersen, Robert Read

Introduction

The California drift gillnet fishery for broadbill swordfish, Xiphias gladius, and common thresher shark, Alopias vulpinus, developed in the late 1970's when incidental catches of pelagic sharks in small-mesh coastal drift gillnets targeting barracuda, Sphyraena argentea, and white seabass, Atractoscion nobilis, motivated fishermen to experiment with large-mesh nets targeting thresher shark. (1) In 1979, 40 vessels participated in the fishery, and by the 1981-82 fishing season, over 6,000 sets were made by about 200 active permit holders (Herrick and Hanan, 1988; Hanan et al., 1993; Holts et al., 1998). Fishing effort peaked at over 10,000 sets in the 1982-83 and 1986-87 fishing seasons, after which effort declined, largely due to a combination of a limited entry system, net length restrictions, and a series of time/area closures (Hanan et al., 1993; Holts et al., 1998). NOAA's National Marine Fisheries Service (NMFS) began placing observers on drift gillnet vessels in 1990 to monitor marine mammal bycatch. Fishing effort at that time was about 4000-5000 sets. Effort had declined to about 1,700 sets by 60 vessels by 2002.

In the early to mid 1990's, the fishery was responsible for taking a number of marine mammal species at levels where estimated mortality exceeded potential biological removal (PBR) limits set under the Marine Mammal Protection Act (MMPA) (Barlow et al., 1997; Julian and Beeson, 1998). At that time, the fishery was considered to be a MMPA Category I fishery, where incidental mortality and serious injury of marine mammals is frequent and annual mortality and serious injury for a given stock or stocks exceeds 50% of PBR. A Take Reduction Team (TRT) was convened in 1996 with the goal of reducing marine mammal interactions with the fishery. The TRT recommended the experimental use of acoustic warning devices or "pingers," to test their effectiveness in reducing marine mammal bycatch. Experimental results demonstrated that entanglement rates of short-beaked common dolphins, Delphinus delphis, and California sea lions, Zalophus californianus, were significantly reduced in pingered nets (Barlow and Cameron, 2003). The effectiveness of acoustic pingers in this fishery led to their mandatory use in late 1997 (U.S. Dep. Commer., 1997).

Typical gear used in this fishery is an 1,800 m (1,000 fm) gillnet with a stretched mesh size ranging from 46 to 56 cm (18 to 22 in), with a required 36 cm (14 in) minimum. The net is attached to one end of the vessel, set at dusk, and allowed to drift during the night for 12 to 14 h. In the 1997-98 fishing season, a net extender length of 11 m (36 ft, the minimum depth at which the top of the net may be fished) became mandatory. Vessel trips usually range from 5 to 18 days, depending on the area to be fished, weather, and fish availability. Effort in this fishery is highly seasonal, with > 70% of sets occurring between October and December. Season-area closures for this fishery require that effort must be further than 200 nautical miles (n.mi.) from shore from 1 February to 30 April; inclusive, and that effort must be further than 75 n.mi. from shore from 1 May to 14August. Since August 2001, a season/area closure to protect leatherback turtles prohibits drift gillnet fishing from 15 August through 15 November in the area bounded by straight lines from Point Sur, Calif., (lat. 36[degrees]17'N) to lat. 34[degrees]27'N, long. 123[degrees]35'W, west to long. 129[degrees]W, north to lat. 45[degrees]N, then east to the Oregon coast (Fig. 1). Due to current season/area closures, fishing effort is concentrated in southern California waters (Fig. 2). An additional season area closure south of Point Conception, Calif., and east of long. 120[degrees]W is effective during the months of June, July, and August during El Nino years to protect loggerhead turtles, Caretta caretta.

[FIGURES 1-2 OMITTED]

Marine mammal, seabird, and sea turtle mortality in this fishery has been described for the period 1990-95 by Julian and Beeson (1998). Those authors reported estimates of 400-650 cetaceans and 100-200 pinnipeds killed annually, based on annual observer coverage that ranged from 4% to 18%. This paper summarizes the number of marine mammals, seabirds, and sea turtles observed killed annually in the California drift gillnet fishery for swordfish and thresher shark from 1996 to 2002, annual estimates of mortality, and total estimated mortality for the whole period.

Materials and Methods

Data Collection

Data on incidental entanglement and mortality of protected species is collected by observers that are placed on board drift gillnet fishing vessels. An attempt is made to sample at least every fifth vessel trip, with an overall goal of 20% observer coverage for the fishery. Not all vessels are observed, because smaller vessels have no berthing spaces to accommodate observers. During the 2002 fishing season, at least 13 vessels were unobserved, which were responsible for a minimum of 550 days of fishing effort (NMFS Southwest Regional Office, unpubl, data). Observers record information for each set, including location, presence, and functionality of pingers, and bycatch of protected species. During the 1996-97 pinger experiment, observers carried acoustic pingers with them, and, therefore, all pingered sets were observed during this period. For each marine mammal incidentally killed, observers record the species and gender of the animal. Additional biological data and samples (e.g. total body length, gonads, teeth, skin sample) are collected whenever possible. When practical, the entire carcass of the marine mammal is retained for life history studies. Species identifications made in the field are validated, if necessary, and corrected using molecular genetic methods when the species identification is in doubt (Chivers et al., 1997; Henshaw et al., 1997). A summary of the number of species identifications corrected using genetic methods is presented in the Results section. A description of the data collected and life history information available from incidental kills in this fishery has been summarized in Chivers et al. (1997) and Henshaw et al. (1997). Occasionally, entangled animals were released with injuries that made future survival doubtful. These cases of "serious injuries" were defined by reviewing observer notes and comparing the extent of the injuries with the serious injury guidelines used by NMFS (Angliss and DeMaster, 1998). A serious injury is defined as "any injury that will likely to lead to mortality". (2) Serious injuries may include, but are not limited to, the following: animals released with trailing gear that would impair the animal's mobility or ability to feed, ingested hooks, visible blood flow, loss or damage to an appendage, listless appearance or inability to defend itself, inability to swim or dive upon release from fishing gear, signs of equilibrium imbalance, perforation of any part of the body by fishing gear, and animals that swim abnormally after release. Serious injuries were treated as observed mortalities and are included as such when estimating overall mortality.