|
Article Excerpt
Introduction
In some commercial fishing operations, elasmobranchs represent a significant amount of discarded bycatch. Due to the slow growth rate, late maturity, and low fecundity of sharks in general, shark populations are particularly vulnerable to fishing pressure (Pratt and Casey, 1990). The history of directed shark fisheries in North American waters contains many examples of the deleterious effects overfishing can have on shark populations, including the rise and collapse of the porbeagle, Lamna nasus (Casey et al., 1978); soupfin shark, Galeorhinus zyopterus (Ripley, 1946); and spiny dogfish, Squalus acanthias (Rago et al., 1998) fisheries. Even in the case of species not subject to directed fisheries, such as many pelagic sharks, there is concern that bycatch mortality may still be high enough to harm shark populations (Musick et al., 2000). This concern has led to an urgent call for population assessments of elasmobranch species that often appear as bycatch in pelagic commercial fishing operations. (1)
To help increase the amount of management-relevant information available on pelagic sharks, we have examined nine years of fishery observer data to quantify and describe the patterns of shark bycatch in a major U.S. pelagic fishery, the swordfish, Xiphias gladius, and tuna, Thunnus spp., pelagic longline fleet, off the southeastern United States. These results may provide a clearer perspective of the magnitude of shark bycatch, and the distribution, relative abundance, and characteristics of shark populations that utilize the pelagic habitat in this region than has previously been available from fishery-independent scientific cruises alone. The data sources we used for this study, albeit fishery-dependent, offer the advantage of providing a much greater number of observations spread out over various times of the year from which to assess the status of Atlantic pelagic shark populations, and provide information relevant for their management.
Materials and Methods
Description of the Fishery
The major fishery targeting large pelagic species off the southeastern United States is the pelagic longline fishery. Descriptions of this fishery can be found in Berkeley et al. (1981), Berkeley and Campos (1988), and Beerkircher et al. (2002): the pelagic longline gear used in this area consists of a heavy monofilament mainline (7-65 km long), which is suspended at various depths below the surface and from which are suspended numerous lengths of lighter monofilament line with a single large (size 7/0-11/0) hook at the end. Hooks are placed along the line at a ratio of 11-19 hooks/km, resulting in a total of 80-1,200 hooks. The average number of hooks is 400-500 per longline. The gear free-floats on the surface of the ocean, with the hook depths varying from 35 to 60 m (Beerkircher et al., 2002).
Vessels targeting swordfish generally set gear around sunset and haulback around dawn, use chemical light sticks attached near the hooks, and use mackerel or squid for bait. Fishery-dependent data indicate an average of 4,028 longline sets were deployed per year in this area between 1994 and 1999 (Cramer, 1995; Cramer and Adams, 1999; Cramer, 2002). The primary species targeted by these fishermen is swordfish, although tunas, mahi-mahi, Coryphaena spp., and certain shark species are also commercially important portions of the catch. (2)
Bycatch in this fishery includes teleosts, elasmobranchs, and on rare occasions marine mammal and sea turtle species. The greatest percentage of bycatch in this fishery is composed of sharks (Anderson, 1985). Shark species commonly caught in the pelagic longline fishery include the dusky, Carcharhinus obscurus, night, C. signatus; silky, C. falciformis; oceanic whitetip, C. longimanus; tiger, Galeocerdo cuvier; blue, Prionace glauca; shortfin mako, Isurus oxyrinchus; and scalloped hammerhead, Sphyrna lewini (Anderson, 1985; Beerkircher et al., 2002).
Several of these species are neither generally described as "pelagic" in the literature nor defined as pelagic by the National Marine Fisheries Service (NMFS) Shark Fishery Management Plan (FMP). (2) Since several shark species encountered in the pelagic fishery occupy more than one habitat, this paper ignores subjective distribution classifications and describes bycatch of sharks of any species by the pelagic longline fishery off the southeastern United States.
Study Area and Data Set
The primary data we examined were compiled and maintained by the NMFS Southeast Fisheries Science Center (SEFSC) as part of the pelagic observer program and include data collected since the observer program's inception in June 1992 through December 2000. Observer coverage is mandatory for Federal swordfish permit holders, and selection of a vessel for coverage is based on a random draw. The percentage of longline sets observed in any given area and calendar quarter (quarter 1: January-March, quarter 2: April-June, quarter 3: July-September, quarter 4: October-December) was targeted to be 5% of the total reported number of sets for that area and calendar quarter in the previous year.
The northwest Atlantic (including the Gulf of Mexico and Caribbean) is divided into eleven areas thought to represent regions of similar types of fishing effort (Fig. 1). Two areas, the "Florida East Coast" (FEC, NMFS area 3) and the "South Atlantic Bight" (SAB, NMFS area 4) were combined into the study area examined herein. This area is bounded on the north and south by lat. 35[degrees] and 22[degrees]N and on the east and west by long. 71[degrees] and 82[degrees]W, respectively. This area was selected as the spatial limits of the study because the pelagic longline fishery in it has been classified as one of the five distinct U.S. Atlantic pelagic longline fisheries based on the nature of the target species, temporal distribution of effort, and other fishing practices. (2) The rough similarity of fishing effort throughout this area allows some standardization of catch per unit of effort (CPUE) data, which would be more difficult if a larger study area encompassing variable fishing practices were used. One observed shark-directed set that occurred in shallow water during 1996 was not included in the analysis to preserve CPUE standardization and the intent of the study to examine shark bycatch in the tuna-swordfish fishery.
[FIGURE 1 OMITTED]
Quantitative Methods Used for Data Analysis
Bootstrap procedures with 1,000 bootstrap replications (Efron and Tibshirani, 1993) were used to estimate the mean yearly CPUE (expressed as number of sharks caught per 1,000 hooks) for eleven commonly observed elasmobranch species and for unidentified sharks as a group. Upper and lower 95% confidence limits were taken from the 97.5 and 2.5 percentiles of the ranked replicant means, respectively.
For each shark species, mean CPUE was analyzed to test for differences among seasons using analysis of variance (ANOVA);...
|
