Marine Biology (2006) 149: 961–968 DOI 10.1007/s00227-006-0278-4 RESEARCH ARTICLE Aaron J. Wirsing Æ Michael R. Heithaus Lawrence M. Dill Tiger shark (Galeocerdo cuvier) abundance and growth in a subtropical embayment: evidence from 7 years of standardized fishing effort Received: 9 June 2005 / Accepted: 23 January 2006 / Published online: 25 February 2006 Ó Springer-Verlag 2006 Abstract The tiger shark (Galeocerdo cuvier Peron and from other regions, but exceeded those for populations Lesueur 1822) is a widely distributed predator with a elsewhere for sharks >275 cm fork length (FL), perhaps broad diet and the potential to affect marine community because mature sharks in the study area rely heavily on structure, yet information on local patterns of abun- large prey. The data suggest that (1) the threat of dance for this species is lacking. Tiger shark catch data predation faced by animals consumed by tiger sharks were gathered over 7 years of tag and release research fluctuates dramatically within and between years, and fishing (1991–2000, 2002–2004) in Shark Bay, Western (2) efforts to monitor large shark abundance should be Australia (25°45¢S, 113°44¢E). Sharks were caught using extensive enough to detect inter-annual variation and drumlines deployed in six permanent zones (3km2 in sufficiently intensive to account for intra-annual trends. area). Fishing effort was standardized across days and months, and catch rates on hooks were expressed as the number of sharks caught hÀ1. A total of 449 individual tiger sharks was captured; 29 were recaptured. Tiger Introduction shark catch rate showed seasonal periodicity, being higher during the warm season (Sep–May) than during The tiger shark, Galeocerdo cuvier, is a large carcharhinid the cold season (Jun–Aug), and was marked by inter- that often is an apex predator in marine ecosystems annual variability. The most striking feature of the catch (Randall 1992). Commonly found within tropical and data was a consistent pattern of slow, continuous vari- warm-temperate regions of the world’s oceans (Randall ation within each year from a peak during the height of 1992), tiger sharks have a broad diet that shifts ontoge- the warm season (February) to a trough in the cold netically, with small individuals acting primarily as season (July). Annual growth rates of recaptured indi- piscivores and larger ones consuming teleosts and a variety viduals were generally consistent with estimates of large-bodied species (e.g., elasmobranchs, marine reptiles, and marine mammals; Simpfendorfer 1992; Communicated by J.P. Grassle, New Brunswick Lowe et al. 1996; Heithaus 2001; Simpfendorfer et al. 2001). Consequently, the tiger shark has the potential to A. J. Wirsing (&) Æ L. M. Dill influence marine communities via trophically- and Behavioural Ecology Research Group, Department of Biological behaviorally-mediated interactions with a variety of prey Sciences, Simon Fraser University, BC V5A 1S6 Burnaby, Canada species (e.g., Lowe et al. 1996; Heithaus 2001; Simpfen- A. J. Wirsing dorfer et al. 2001; Heithaus et al. 2002; Dill et al. 2003; Honorary Research Associate, School of Animal Biology, Uni- Heithaus 2004). The nature and magnitude of this influ- versity of Western Australia, 35 Stirling Highway, 6009 Crawley, ence may be dynamic if the abundance of tiger sharks WA, Australia fluctuates within a given area (see Holling 1959; Lima M. R. Heithaus 2002); therefore, analyses of temporal variation in the Department of Biological Sciences, Marine Biology Program, abundance of this top predator would improve under- Florida International University, Biscayne Bay Campus AC1, 3000 standing of the marine systems of which they are a part. NE 151 St, North Miami, FL 33181, USA Tiger sharks have been studied in a variety of loca- A. J. Wirsing tions, including the east coast of North America (e.g., Department of Biological Sciences, Northwest Atlantic and Gulf of Mexico, Branstetter Florida International University, et al. 1987; western North Atlantic, Natanson et al. 300 NE 151 St, North Miami, FL 33181, USA 1999; Northwest Atlantic, Baum et al. 2003), the Gulf of E-mail: wirsinga@fiu.edu Tel.: +1-305-9195819 Mexico (Branstetter et al. 1987), the Hawaiian Islands 962 (Polovina and Lau 1993; Holland et al. 1999), the east channels (6.5–12.0 m), and shallow banks (<4.0 m). coast of South Africa (e.g., Wintner and Dudley 2000), Most of the shallow habitat within the bay is dominated and the northeastern (Simpfendorfer 1992) and western by extensive seagrass meadows (Walker 1989). In 1991, coastlines of Australia (Simpfendorfer et al. 2001; the Shark Bay region was listed as a World Heritage Heithaus 2001; Heithaus and Dill 2002). These inquiries Area. The local tiger shark (Galeocerdo cuvier Peron and have revealed that tiger shark populations are comprised Lesueur 1822) population has been free from commer- at least in part of individuals that maintain defined, cial fishing since 1994, and has never been subject to though likely very large, home ranges and return to heavy harvest (Heithaus 2001). Thus, the study site has specific areas on a regular basis (Holland et al. 1999; an intact and protected assemblage of tiger sharks and Heithaus 2001), that relative use of particular areas by their major prey species (dugongs, Dugong dugon; pied tiger sharks may be size- and sex-biased (Lowe et al. cormorants, Phalacrocorax varius; sea snakes, primarily 1996; Heithaus 2001), that local tiger shark abundance Hydrophis elegans; and sea turtles, Caretta caretta, may be characterized by substantial variability as well as Chelonia mydas; Heithaus 2001). seasonal periodicity (e.g., between 1997 and 1999 in Shark Bay, Western Australia, tiger shark numbers consistently peaked during the warm season, Septem- Field methods ber–May; Heithaus 2001; Heithaus and Dill 2002; see also Simpfendorfer 1992), and that many tiger shark Over the course of the investigation (1997–2000, 2002– populations in areas exploited by commercial fishing 2004), we sampled tiger sharks in six deep-water fishing apparently are in sharp decline (Baum et al. 2003; Ward zones (Fig. 1b). Individual sharks were captured using and Myers 2005). Several studies have addressed local- drumlines equipped with a single hook (Mustad Shark ized tiger shark population dynamics over an extended Hook; size 13/0 used predominantly, with occasional use duration (>5 year) (e.g., Simpfendorfer 1992; Baum of sizes 12/0 and 14/0) fished at a depth of 0.7–2.0 m. et al. 2003), but no protracted investigation to date has Hooks were baited predominantly with Australian sal- carefully controlled for fishing effort in an area where mon (Arripis truttaceus, AS; 9,771 h; 70% of total fish- tiger sharks are neither commercially harvested nor ing time). Between 2002 and 2004, four other bait species subjected to control measures. Consequently, baseline were used when AS was not available: baldchin groper data allowing for rigorous, inter-annual assessments of (Choerodon rubescens, G; 189 h), sea mullet (Mugil regional trends in cohort-specific and overall abundance cephalus, M; 256 h), pink snapper (Pagrus auratus, PS; in this species are lacking. 2,451 h), and tailor (Pomatomus saltatrix, T; 1,273 h). Here, we present results from a 7 year (1997–2000, Ten drumlines were deployed concurrently in one or two 2002–2004) study of tiger sharks in Shark Bay, Western zones and spaced 300 or 700 m apart (Heithaus 2001). Australia. Over the course of the investigation, which is Shark catch rates may in part be a function of sampling an extension of the work of Heithaus (2001), sampling design (Heithaus 2001). However, catch rates (sharks location, timing, and effort were held relatively constant. hÀ1, see below) for the two deployment protocols (one Thus, we were able to (1) determine whether the seasonal zone with 300 m spacing versus two zones with 700 m pattern of shark abundance documented by Heithaus spacing) did not differ (paired t test based on catch (2001) remained consistent over a longer time interval, values for days falling within the same week during (2) test for variation in shark abundance within both which one protocol was used exclusively, t6=À0.542, warm and cold seasons, (3) address the magnitude of P=0.607), so all catch rate data have been pooled for inter-annual variation in tiger shark abundance within a the purposes of this analysis. local area, and (4) compare growth rates of recaptured Following deployment at dawn, drumlines were individuals to those of tiger sharks elsewhere. Because checked every 2–4 h. During each check, we noted bait used in the study varied over time, we also (5) asked whether bait was present or absent on all hooks that whether tiger shark catch rates were affected by different failed to catch sharks. We defined soak time as the baits. number of hours elapsing between deployment and re- moval for hooks that retained bait for an entire fishing day. For hooks from which bait was lost or on which a Materials and methods shark had been caught, soak time was considered to have ended half way between the previous point at Study site which bait presence was verified and the time when bait loss or a shark was detected. This study was undertaken in the Eastern Gulf of Shark We used a shark handling procedure consistent with Bay, offshore of the Monkey Mia Dolphin Resort that of Heithaus (2001). Briefly, hooked sharks were (25°45¢S, 113°44¢E). Shark Bay is a large (13,000 km2), brought alongside a 4.5 m vessel and allowed to swim semi-enclosed basin located roughly 900 km north slowly while the boat idled forward. Each shark was of Perth, Western Australia (Fig. 1a, b). The study then measured (pre-caudal length [PCL], fork length site (160 km2; Fig.