Evaluating catch and mitigating risk in a multispecies, tropical, inshore shark fishery within the Great Barrier Reef World Heritage Area

Additional keywords: Carcharhiniformes, coastal shark fishery, elasmobranch.

Ongoing worldwide fisheries exploitation continues to fuel a growing debate on the future of wild-caught fisheries (Jackson 2008; Worm et al. 2009). Higher trophic-level predators such as elasmobranchs (sharks and rays) have fared particularly poorly, with some collapses, often rapid, of populations where they are targeted or taken as by-catch (Ripley 1946; Olsen 1959; Graham et al. 2001; Devine et al. 2006). Recently, international concern over the ongoing exploitation of sharks has led to the development of the International Plan of Action for Sharks (FAO 2000). The vulnerability of sharks and rays to overfishing stems largely from their life-history characteristics, including late maturation, low fecundity, low natural mortality and long life-spans (Cortes 2000). These characteristics mean there is a close relationship between stock size and recruitment, and consequently long recovery times after overexploitation has occurred (Holden 1974).

Other factors, such as naturally low abundance as well as complex migration patterns and spatial usage (e.g. sex segregation, site fidelity; Heupel and Simpfendorfer 2005; Sims 2005), can further increase the vulnerability of some elasmobranchs to overfishing. This is relevant to carcharhiniform sharks, particularly of the families Carcharhinidae and Sphyrnidae, which occur abundantly throughout inshore continental shelf regions of the tropics and subtropics worldwide (Musick et al. 2004).

Species of these families vary greatly both in their life histories and their utilisation of inshore habitats (Knip et al. 2010). For example, many small to medium-sized carcharhinids (e.g. Rhizopriondon taylori, Carcharhinus sorrah) remain within inshore areas throughout the duration of their lives. These species are often fast growing and relatively short lived (Davenport and Stevens 1988; Simpfendorfer 1993). Other carcharhiniform sharks utilise inshore habitats only during discrete stages of their lives. These species are generally larger in size and have moderate to slow growth rates. Negaprion brevirostris and C. leucas are examples of species that use inshore areas as neonate and juvenile nurseries (Springer 1950; Castro 1993). Conversely, neonates of other species (e.g. Galeocerdo cuvier, Sphyrna mokarran) are absent from close inshore waters whereas adults are present (Hueter and Tyminski 2007). The wide variety of life-history characteristics and space utilisation means inshore shark populations are likely to be affected in a range of different ways and to varying extents by anthropogenic influences such as fishing.

Artisanal and commercial fisheries for carcharhiniform sharks exist in most equatorial and tropical regions and are particularly common throughout Asia, especially in the Indo-Pacific region and the Indian subcontinent (Kasim 1991; Hanfee 1999; Henderson et al. 2007; White 2007), as well as parts of Africa, the Caribbean, and throughout central America (Motta et al. 2005), notably Mexico (Castillo-Géniz et al. 1998). Despite the important contribution these fisheries make to regional economies and food security, management of such fisheries is often neglected (Fowler et al. 2005). Many countries lack the resources to adequately monitor their fisheries (White and Kyne 2010), and even in more affluent states, the inherent low value of inshore shark fisheries often means research and management are given low priority. Where monitoring is conducted, catch composition is rarely established because of the difficulties in identifying many species so, at best, sharks are identified to family or order (Shotton 1999). The paucity of data on most inshore tropical shark fisheries along with wide variation among life histories and complex spatial ecology provides an imposing hurdle to sustainable harvest of carcharhiniform sharks in these fisheries and raises concerns given the vulnerability of elasmobranchs to overfishing.

In tropical northern Australia, carcharhiniform sharks make up large components of several small-scale, inshore fisheries targeting a range of teleost and shark species (Stevens 1999; Salini et al. 2007). The low value of tropical shark (AU$2–3 kg^–1 processed weight) means that despite Australia’s status as a developed nation, many of these fisheries are similar to those in developing nations: fishing effort is highly fragmented along those coastline; fishing vessels are usually small in size (<7 m); and nets are frequently hauled by hand. Aside from the period between 1974 and 1986 when Taiwanese gill-net vessels targeted sharks off northern Australia, the total harvest of elasmobranchs in Australia’s tropical fisheries has been between 2000 and 3000 t year^–1 (Bensley et al. 2010). While some components of northern Australian shark fisheries have been reasonably well monitored and formal risk assessments or stock assessments have been used to inform management, other areas, including the east coast of Queensland, have received little attention (Anon 1990; Stobutzki et al. 2002; Salini et al. 2007). This is somewhat surprising given that on the east coast of Queensland, these fisheries occur within the Great Barrier Reef World Heritage Area (GBRWHA), one of the world’s largest networks of marine protected areas (GBRMPA 2009).

Changes to legislative requirements concerning sustainability in Australian fisheries (Environmental Protection and Biodiversity Conservation Act 1999, Cth), combined with a 200% increase in shark landings on Queensland’s east coast between 1993 and 2004 (Bensley et al. 2010) and concern from managers about shark exploitation within the GBRWHA (GBRMPA 2009) recently created a need to describe the shark component of the inshore net fishery. Consequently, between 2006 and 2009 an onboard-vessel observer study recorded the catch composition and harvest practices of the fishery. The aims of this study were to quantify the composition, to species level, of carcharhiniform sharks caught by net fisheries in the GBRWHA and to examine the characteristics of the catch to qualitatively establish patterns of catch susceptibility. To this end, we compared catch rates between three nominal zones (rivers, intertidal and inshore coastal), examined the sex ratio of the catch and compared male and female length-frequency distributions. We discuss emergent patterns in the catch in relation to life-history characteristics and consider the threats to carcharhiniform sharks in the GBRWHA. Given these new data, we also suggest fisheries management strategies aimed at mitigating the risk of overfishing, and we consider the implications for management in data-poor, inshore fisheries for carcharhiniform sharks throughout tropical regions of the world.

Between June 2006 and July 2009, fishery observers monitored vessels operating in the commercial gill-net sector of the Queensland East Coast Inshore Finfish Fishery (ECIFF) within the boundaries of the GBRWHA (between Cape York (10.5°S) and Bundaberg (24.5°S); Fig. 1). Owing to the vast area of the fishery, data were collected simultaneously by two groups, James Cook University Fishing & Fisheries Research Centre and Fisheries Queensland. Data were subsequently combined to provide the most robust dataset. Fisher participation in the observer survey was voluntary. Prior to commencing a trip, the observer interviewed the fisher to determine the length, depth and mesh size of net to be used, so fishing effort could be calculated. Fishing start time for an individual net shot was recorded as the time when the net was completely in the water, and finish time was when hauling of the net began. Location of nets was recorded using a hand-held GPS and depth was measured using an onboard depth-sounder. Catch composition of elasmobranchs was recorded to species level using a species identification key derived from Last and Stevens (1994). Owing to the small-scale nature of the fishery, a single observer assessed each individual trip. When conditions permitted, the stretch-total length, fork length and pre-caudal length (sharks) or disk width (rays) of a subsample of the catch was recorded in mm (Compagno 1984) and weight measured in kg. When possible, sex and maturity stage of sharks processed at sea was also recorded using a standard staging system (Walker 2005a ).

Fig. 1.  Study area showing observed fishing effort (km-net-hours) by one degree squares of latitude and longitude. Within each square, observed effort is shown for the three zones: inshore coastal (upper left), intertidal (centre), and river (bottom right). The dashed black line indicates the outer boundary of the Great Barrier Reef World Heritage Area.

Data were grouped into three broad zones (river, intertidal and inshore coastal) that corresponded to discrete subcomponents of the ECIFF, each with different resident species, targeting and harvest practices and management strategies (Table 1). In river zones, barramundi (Lates calcarifer) was targeted using set nets with stretched mesh sizes of 165–216 mm. Nets were usually set overnight and fishing occurred between February and October. Within intertidal zones (defined as waters <2 m depth), several teleost species (mostly Eleutheronema tetradactylum, Polydactylus macrochir and species of the family Mugilidae) were targeted using set nets with stretched mesh sizes of 114–216 mm. Fishing in intertidal zones occurred throughout all periods of the day and throughout the year. Within inshore coastal zones (defined as coastal waters of between 2 and 25 m depth), Scomberomorus semifasciatus was targeted during winter and spring, while a generalist shark fishery targeting mainly Carcharhinus tilstoni and C. sorrah operated throughout the year. Some fishers were licensed to use up to 1200 m of 165 mm stretched mesh net, although most were licensed to use 600 m.

Table 1.  Nominal fishery zones in the East Coast Inshore Finfish Fishery

Some of the earlier Fisheries Queensland observer trips were primarily focussed on recording teleost catch, so identification of Carcharhinidae and Sphyrnidae species was limited to family level (e.g. ‘whaler shark’). With the exception of overall catch composition (Table 3), these trips were excluded from further analyses, which focussed only on carcharhiniform sharks. Mean length at capture was calculated and, although not all animals were measured, the recorded lengths were assumed to represent a random subsample of the total catch. Mean weight at capture was calculated using length–weight regressions derived from the present study or, if unavailable, from previous studies in northern Australia (Stevens and Lyle 1989; Stevens and McLoughlin 1991). Catch was standardised to number per unit effort (individuals km-net-hour^–1) and weight per unit effort (kilograms km-net-hour^–1). To further examine characteristics of the overall catch (data pooled between zones), two-sample Kolmogorov–Smirnov (KS) tests were used to test whether length-frequency distributions of males and females of individual species were significantly different. The sex ratio (females/males) of the catch was also calculated and, where there were at least five individuals from each sex, Chi-square tests were used to determine any significant differences in sex ratio. All of the Carcharhiniformes species caught in the present study had a reproductive mode of placental viviparity (except for the tiger shark, Galeocerdo cuvier), so the percentage of neonates in the catch could be inferred from the presence of an open or unhealed umbilical scar, thus indicating recent birth. The catch characteristics above were used to qualitatively establish the susceptibility of different species to the fishery. Capture susceptibility was defined as the culmination of factors that result in an individual of a species being killed by the fishery (e.g. availability, encounterability, selectivity). We considered susceptibility in the general sense of the term and no attempt was made to quantify it (e.g. Stobutzki et al. 2002). Emergent patterns in the catch were further discussed in relation to the life-history characteristics of captured species, such as length at 50% maturity, growth characteristics and habitat preferences. Life-history data were obtained from the published literature or, if available, from unpublished data obtained during the present study.

Table 3.  Catch composition of elasmobranchs caught by the East Coast Inshore Finfish Fishery in the Great Barrier Reef World Heritage Area
Data, grouped by order and sorted by numerical abundance, are from all 149 trips observed across the three nominal zones (river, intertidal, and inshore coastal). The dominance of carcharhiniform sharks in the elasmobranch component of the catch is typical of many tropical, inshore fisheries

Between June 2006 and July 2009, observers were deployed on 149, often multiday, fishing trips within the GBRWHA. Observations were on 1188 separate net shots during 297 days onboard vessels, giving a total of 1452 km-net-hours (Table 2).

Table 2.  Total fishing effort observed from 2006–2009 in the East Coast Inshore Finfish Fishery in the Great Barrier Reef World Heritage Area
The observer coverage was the most comprehensive fisheries observer program ever applied to the East Coast Inshore Finfish Fishery, and included considerable coverage of fishing in river and intertidal zones, the most data-poor sectors of the fishery

Although 60% of trips occurred in intertidal zones, the greatest amount of fishing effort was observed in inshore coastal zones (Table 2). This reflected the generally shorter duration of trips occurring in intertidal and river zones ( = 1.3 days, and  = 2.4 days respectively), compared with those in inshore coastal zones ( = 3.5 days), and also the generally shorter net lengths used in intertidal and river zones. The longest trip observed in all zones was 7 days, while the shortest was <1 day (i.e. a single-day trip). Total effort observed was 202 km-net-hours in river zones, 237 km-net-hours in intertidal zones and 1013 km-net-hours in inshore coastal zones (Table 2). All observed fishing effort was between 13°S and 24°S (Fig. 1).

In total, 18 625 fish were recorded by observers including 6828 elasmobranchs that constituted 37% of the catch by number. Overall, 38 species of elasmobranchs from 11 families and 4 orders were identified (Table 3), of which Carcharhiniformes was both the most diverse order (25 species) and the largest component of the catch by number (94.5%). Rajiformes was the next most diverse order (>10 species) but only contributed 3.9% of the elasmobranch catch by number. The remaining 1.6% came from two species of Pristiformes, three species of Orectolobiformes and a small number of unidentified sharks. After the removal of trips that contained fish identified only to family level, species-level catch composition was determined for 126 trips with a total effort of 905 km-net-hours. Among the Carcharhiniformes (Table 4), the morphologically identical blacktip sharks C. tilstoni and C. limbatus were the most numerous (28%, Table 4). These species could not be separately identified in the field and were therefore grouped together. The spot-tail shark (C. sorrah) and scalloped hammerhead (S. lewini) were also relatively large contributors to the catch number (17 and 11%, Table 4). By weight, the target species of the fishery, C. tilstoni/C. limbatus and C. sorrah, also dominated the catch, contributing ~51% of the catch (Table 4). Despite being only 2.4% of the catch by number, the great hammerhead shark (Sphyrna mokarran) was the third largest component of the total weight (9.64%), owing to its large mean size at capture (Table 4). Conversely, catch by weight of some smaller species (e.g. R. acutus, R. taylori) as a proportion of total catch was lower than their respective proportion of catch by number.

Table 4.  Catch per unit effort and catch composition of carcharhiniform sharks caught by the East Coast Inshore Finfish Fishery within the boundaries of the Great Barrier Reef World Heritage Area
Species are sorted by the proportion of the total observed catch across all habitats by weight. Data are from a subsample of 126 observer trips, where all individuals were identified to species level. Blank records indicate no recorded occurrence in catch

Diversity of carcharhiniform sharks captured increased with distance from the coast, with 7 species recorded in rivers, 17 species in intertidal zones, and 25 species recorded in inshore zones (Table 3). Number and weight per unit effort of Carcharhiniformes also increased with distance from the coast (Table 4). Compared with river zones, the catch of Carcharhiniformes was ~5 times greater in intertidal zones, and 9 times greater in inshore coastal zones. C. tilstoni/C. limbatus were by far the most captured species in both the intertidal and inshore coastal zones, and also accounted for the greatest weight. C. sorrah accounted for a large component of the catch in inshore coastal zones, but was rarely caught in intertidal zones. Although few were caught, the large size of S. mokarran meant that it accounted for a relatively large component of weight in both intertidal and inshore coastal zones. The bull shark (C. leucas) was the only species regularly captured in river zones.

Mean lengths of species within the overall catch (all zones pooled) ranged from 637 mm for R. taylori males to 1544 mm for S. mokarran females (Table 5, Fig. 2). Sex-specific differences in the length-frequency distributions were found for 6 of the 14 species where there were sufficient data to carry out the KS test (Table 5). A significant difference in the sex ratio of the catch was also found for 6 of the 14 species tested (Table 5). No clear trends in sex ratio were evident, with females greatly outnumbering males in some species such as S. mokarran and R. taylori, and males greatly outnumbering females in other species such as S. lewini and R. acutus. There was also considerable interspecific variation in the different life stages present within the catch (Table 5). Percentages of neonates recorded in the catch ranged from 0% for many species up to 62.1% for C. leucas. The percentages of mature animals in the catch was inversely related to maximum size for many species, as small species (<1000 mm) were typically caught as adults, and moderate to large species (>1000 mm) were caught as juveniles. Exceptions to this trend included the snaggletooth shark (Hemipristis elongata) and the winghead shark (Eusphyra blochii), which were both moderate sized species (up to 2000 mm) mainly caught as adults. Large sex-specific differences were also found in the percentage of the catch that was mature for some species, including R. acutus and S. lewini, where adult males were present in the catch, but adult females were either rare or absent. For the blacktip complex of C. tilstoni/C. limbatus, the percentage of mature animals should be considered an approximation, as it was based on length at maturity of C. tilstoni and likely included some C. limbatus specimens. Off eastern Australia, C. limbatus is known to mature at sizes >2000 mm (Macbeth et al. 2009), which is larger than any individuals measured during the observer survey. Therefore, although an accurate estimate of the percentage of mature C. tilstoni specimens was not possible, we have a high level of confidence that no C. limbatus adults were caught.

Table 5.  Tabulation of sex-specific length at capture details, sex ratio and the percentage of catch mature or neonate for carcharhinid sharks caught in the the Great Barrier Reef World Heritage Area
The results of Kolmogorov–Smirnov (KS) tests comparing length-frequency distributions of males and females and chi-square tests on the sex ratio of the catch are also given. Where only a single length measurement was available it was given as maximum size at capture and other fields were left blank. Bold text indicates statistical significance (P < 0.05)

Fig. 2.  Length-frequency distributions of the top 10 carcharhiniform sharks by weight (Table 4). Bar colour denotes the capture zone: solid black, river; dark grey, inshore coastal; and light grey, intertidal. Length at 50% maturity is denoted by the dashed black line.

The present survey of the mesh-net commercial fishery operating within the GBRWHA revealed the complex nature of tropical shark fisheries. At least 38 species of elasmobranchs were recorded in the catch, with catch rates varying between habitats, life stages (neonate, juvenile, adult) and by sex. Despite this complexity, some broad trends in capture susceptibility were seen among the carcharhiniforms (Table 6). For example, small species <1000 mm in total length (e.g. R. acutus, R. taylori) were, by virtue of their small size relative to the net mesh size, almost exclusively susceptible to capture as adults in the fishery. Moderate sized species 1000–2000 mm total length (e.g. C. tilstoni, C. sorrah) were susceptible to some extent at all sizes, with neonates large enough and young adults small enough to be caught by the nets. In contrast, large species >2000 mm total length (e.g. C. amboinensis, C. brevipinna) were subject to a gauntlet effect by the fishery (Simpfendorfer 1999; Prince 2005). Large species were frequently captured as neonates and juveniles and rarely caught as adults, possibly due to adults migrating away from fished habitats or growing too large to be meshed or entangled by the nets. Hammerhead sharks typified another group of species susceptible in similar ways. Despite growing to a large size, they were susceptible to capture at all sizes due to their head morphology.

Table 6.  Groupings of similarly susceptible shark species caught in the East Coast Inshore Finfish Fishery

The results of this study confirm that Carcharhiniformes dominate the catch of the ECIFF and it is elasmobranchs of this order that are probably most at risk from the fishery. Many of the species caught by the ECIFF were also identified by risk assessments as among the least likely to be sustainable across other northern Australia fisheries (Stobutzki et al. 2002; Salini et al. 2007) and are also probably affected to some extent by recreational fishing within the GBRWHA (Lynch et al. 2010). Stocks of some species are known to be shared with other nearby jurisdictions, so unsustainable fishing practices in these areas would also potentially affect GBRWHA populations (Ovenden et al. 2009), as would illegal fishing encroaching on northern Australian waters (Field et al. 2009). In contrast to the threats posed by fishing, an integrated risk assessment for climate change of the GBRWHA suggested that most of the carcharhiniforms caught in the ECIFF were unlikely to have a high vulnerability to climate change owing to their high adaptive capacities (Chin et al. 2010).

Of the non-carcharhiniform elasmobranchs, most were caught in relatively low numbers, with the exception of cownose rays (Rhinoptera spp.), narrow sawfish (Anoxypristis cuspidata) and wedgefish (Rhynchobatus spp.), all of which were at least 1% of the overall catch by number. The record of seven green sawfish (Pristis zijsron) in the catch indicates this species is still present on the east coast of Queensland at least as far south as the Whitsundays (20°S), even though it is now considered to be extinct in New South Wales waters (Fisheries Management Act 1994, NSW, No. 38). Recent protection of sawfish in Australian waters, as well as catch restrictions imposed on wedgefish in the ECIFF, are likely to mitigate the threats posed to at least two of the families listed above.

The diversity of elasmobranchs within the tropical coastal regions of the world, combined with the complex spatial ecology and behaviour patterns they exhibit, clearly provides a major challenge for sustainable management of extractive fishing. It has long been recognised that the idiosyncrasies of shark populations and fisheries require alternative approaches to management compared with teleost resources (Holden 1974). More recently it has also been shown that the features of elasmobranchs that make them vulnerable (e.g. close stock-recruitment relationships) can also be advantageous when properly managed (Walker 1998). Indeed, the idiosyncrasies of shark populations may provide many under-utilised opportunities for designing management strategies and, if properly understood, may help reconcile some of the impediments to sustainable management.

Perhaps one of the simplest observations that can be taken from the present study is that despite the large number of species caught within the ECIFF, there were only a few patterns in the way they were susceptible (Table 6). In many instances, similarly susceptible species also shared similar life-history traits. For example, many small, coastal tropical carcharhiniform sharks (<1000 mm, e.g. R. acutus, R. taylori) are amongst the fastest growing and most productive elasmobranchs studied (Simpfendorfer 1993; Harry et al. 2010). Medium-sized (1000–2000 mm, e.g. C. cautus, C. sorrah, C. tilstoni) coastal tropical species are somewhat less productive, typically living 10–20 years and maturing relatively young (Davenport and Stevens 1988; White et al. 2002). In contrast, large tropical Carcharhiniformes (>2000 mm e.g. C. leucas, N. brevirostris) typically conform to the slow-growing, long-lived paradigm more frequently associated with elasmobranchs (Brown and Gruber 1988; Neer et al. 2005). These similarities also extend to habitat and spatial usage. Most species within the small and medium-sized groups are restricted to coastal waters, while most large species are semi-pelagic, migrating offshore at larger sizes.

These life-history patterns have been recognised and described by a variety of authors (Hoenig and Gruber 1990; Cortes 2000; Frisk et al. 2001). While actual groupings are arbitrary (e.g. small, medium, large), the underlying concept of a life-history continuum, ranging from ‘slow’ to ‘fast’ species (Cortes 2002), has important implications in terms of simplifying management of multispecies fisheries. Although species-specific management may be unfeasible, it may be possible to direct management strategies at species that are not only susceptible in the same way, but also have similar life-history traits (e.g. the management of ‘Small Coastal Shark’ and ‘Large Coastal Shark’ complexes in the United States). In the case of the ECIFF, management of the fishery could potentially be simplified by directing management strategies at the four nominal groups identified in Table 6.

Examples of specific management strategies that could be used to mitigate the risk of overfishing to tropical carcharhiniform sharks may involve the use of gear restrictions and spatial and temporal closures. Modifying the gear selectivity in a fishery to take advantage of particular life-history traits may be one of the most effective measures for mitigating risk. This is especially relevant in gill-net fisheries for sharks where size-selectivity dynamics are well understood compared with other gear types, such as trawl and line (Kirkwood and Walker 1986). In the present study, the exclusive use of small-mesh gill-nets (typically <165 mm) by the ECIFF meant that sharks >1500 mm were rarely captured (with the exception of hammerheads). This in itself may be a good strategy for multispecies tropical shark fisheries, as only neonates and juveniles of the largest (and often least productive) species are captured by the fishery, while adults are subject to a ‘gauntlet’ effect and effectively excluded. The concept of the gauntlet fishery has been proposed as an effective method of harvesting long-lived species, providing that fishing mortality on adults remains extremely low (Simpfendorfer 1999; Prince 2005). Although such a harvest strategy is unlikely to provide the maximum sustainable yield (Gallucci et al. 2006), it may be preferable, depending on the goals of the fishery. In an artisanal fishery, for example, the harvest of large sharks is unlikely to contribute to food security given that the flesh from these animals often contains high levels of mercury and may not be suitable for human consumption (Lyle 1984; Clarkson 1997). In the case of the ECIFF, the use of a gauntlet-style harvest strategy potentially provides a lower-risk method of harvesting the large coastal/semi-pelagic species, but at the same time allows for concurrent harvest of the more productive small and medium coastal sharks and teleosts.

Spatial and temporal closures may also be used to mitigate the risk to sharks in multispecies fisheries and may be the only way to protect some species that are particularly susceptible to certain gear types (e.g. hammerheads in the present study). Closures of inshore nursery areas have historically been used as a way of protecting sharks and have been considered a critical tool in managing shark populations (Olsen 1959; McCandless et al. 2007), although their usefulness has also been contested (Kinney and Simpfendorfer 2009). Demonstrating the effectiveness of spatial closures for protecting wide-ranging, migratory species (e.g. the large coastal/semi-pelagics and hammerheads) is challenging, although evidence suggests that these species may receive some benefits from spatial closures (Claudet et al. 2010). Many of the patterns observed in this study (e.g. segregation by size, sex and habitat) may also present further opportunities for spatial or temporal closures. Capitalising on the seasonal nature of reproduction displayed by many elasmobranchs could be one way to achieve this. Most carcharhiniform sharks across northern Australia give birth during a relatively restricted time period over summer (Stevens and Wiley 1986; Stevens and McLoughlin 1991). This is also true within the GBRWHA, where neonates of several species such as S. lewini, C. tilstoni/C. limbatus and C. fitzroyensis were captured in intertidal zones at the beginning of summer, but apparently moved away from this zone soon afterwards. Temporal closures of nursery habitats during this brief period may therefore be effective in protecting both neonates and adult females of the medium coastal and large coastal/semi-pelagic groups, should they be vulnerable at this time.

Perhaps one of the most promising and as yet under-utilised risk mitigation strategies for sharks is sex-differential harvest. Strong segregation by size, sex and reproductive stage are well documented characteristics of most shark populations (Springer 1967; Sims 2005). Mucientes et al. (2009), for example, reported strong sex segregation at large scales in the Pacific Ocean for shortfin mako (Isurus oxyrinchus) and suggested that differential exploitation of males was possible. Camhi et al. (1998) also suggested selective take of males only as a potential management measure for sharks. In the present study, sex-differential harvest was seen to be already occurring for some species. Where this was occurring for males (e.g. R. acutus and S. lewini) it may allow higher catches of these species with minimal effect on population growth rates. Conversely for species such as S. mokarran, the high bias towards catching females must also be recognised by managers as it is likely to have a disproportionately negative effect on the population growth rate compared with equal harvest of both sexes. A sex differential harvest strategy would probably be most suited to the wide-ranging large coastal/semi-pelagic species and hammerheads, where sex segregation is likely to be occurring over large spatial scales. Such a management strategy could be formalised by restricting fishing to depths or regions where high numbers of males occur. Sex differential harvest is also appealing because, in fisheries where sharks are landed live (and assuming low post-release mortality), it can be incorporated without the need for any spatial closures, as sex can easily be established in sharks via examination of the pelvic fins. Enforcement of this management technique can also be achieved by requiring fishers to land sharks with pelvic fins intact (Walker 2005b ).

The present study was the most comprehensive observer survey ever applied to the Queensland ECIFF. The high elasmobranch species diversity, dominated by Carcharhiniformes, was characteristic of many inshore, tropical fisheries. The data-poor and highly complex nature of the ECIFF and similar fisheries means that quantitative, species-specific management is unlikely to be possible. However, close scrutiny of the catch characteristics show that there are many aspects of elasmobranch biology that are likely to be useful in designing management strategies that can mitigate the risk of overexploitation posed by such fisheries. These include the tendencies of elasmobranchs to show strong segregation by size and sex, along with the use of discrete areas during different life stages (e.g. nurseries) and the existence of many interspecific patterns in life-history traits. Uptake and implementation of practical management strategies using this information is currently limited by a poor knowledge of life history and spatial ecology of sharks. Even across northern Australia, where the tropical carcharhiniform shark assemblage and fisheries are arguably among the best-studied worldwide, age and growth information is currently available for only a limited number of species. The location and movement of adult stocks of many of the large, semi-pelagic species are also poorly documented. This highlights the ongoing requirement for the study of life history and spatial ecology in elasmobranchs.