A method for evaluating the impacts of fishing mortality and stochastic influences on the demography of two long-lived shark stocks

Rory B. McAuley 0 Colin A. Simpfendorfer 0 Norm G. Hall 0 0 R. B. McAuley: Department of Fisheries, Government of Western Australia, Western Australian Fisheries and Marine Research Laboratories , PO Box 20, North Beach, WA 6920 , Australia , and Centre for Ecosystem Management, School of Natural Sciences, Edith Cowan University , 100 Joondalup Drive, Joondalup, WA 6027 , Australia. C. A. Simpfendorfer: School of Earth and Environmental Sciences, James Cook University , Townsville, Qld 4811 , Australia. N. G. Hall: School of Biological Sciences and Biotechnology, Murdoch University , Murdoch, Western Australia 6150 Stochastic demographic models were developed for Carcharhinus obscurus and C. plumbeus populations off the west coast of Australia by resampling the input parameters for life tables from empirical biological data collected from commercial target fisheries and fishery-independent surveys. The models were used to examine the effects of multiple scenarios of age-specific survival, derived from the fishing mortality rates estimated from a tagging study on sharks and indirect estimates of natural mortality. In the absence of fishing , median estimates of the rates of intrinsic population increase (r) were 0.025 for both species. Inclusion of the age-specific fishing mortality rates estimated for C. obscurus recruits born in 1994 and 1995 resulted in the median estimates of r declining to 0.007 and 0.012, respectively, suggesting that recent harvest levels of mainly neonates by the target fishery were probably sustainable. However, the model also suggested that the population was more susceptible to exploitation of older sharks than was previously believed. The C. plumbeus model indicated that fishing mortality between 2001 and 2004 was probably unsustainable. The increasingly negative trend in median r estimates (from - 0.032 to - 0.049), and the population's apparently limited capacity for density-dependent compensation through changes in fecundity, somatic growth and longevity, suggests that management intervention is necessary to prevent continued stock depletion. Introduction The dusky shark (Carcharhinus obscurus) and the sandbar shark (C. plumbeus) are relatively abundant medium to large carcharhinids that share similar circumglobal distributions in tropical and temperate coastal and adjacent oceanic waters (Compagno, 1984; Last and Stevens, 1994). Both species are distributed along the east and west coasts of Australia (Last and Stevens, 1994), but neither is seemingly common north of 168S or in southeastern Australian waters (Last and Stevens, 1994; McAuley et al., 2005, 2007a). Therefore, each species is considered to be represented by two distinct regional populations. Western Australian populations of C. obscurus and C. plumbeus co-occur throughout most of their ranges and both exhibit similar size-segregated structure, juveniles being most common in temperate latitudes and adults in warmer northern latitudes (McAuley and Simpfendorfer, 2003; McAuley et al., 2007a). Because of their occurrence nearshore and their high-quality flesh and fins, both species support significant commercial and artisanal shark fisheries around the world (Bonfil, 1994; Joung and Chen, 1995; Castro et al., 1999; Chan A Shing, 1999; Fowler et al., 2005; McVean et al., 2006). Neonate and young juvenile C. obscurus have been the primary target of a demersal gillnet fishery in southwestern Australian waters since at least the 1970s (Heald, 1987; Simpfendorfer and Donohue, 1998; Simpfendorfer, 1999a, b). Their catches by that fishery escalated rapidly from under 100 t (estimated live weight) per year in the late 1970s to a peak of just under 600 t in 1988/1989 before management restrictions reduced and stabilized catches at 300 t year21 (McAuley, 2006a). Juvenile C. plumbeus are also caught by that fishery, particularly off the lower west coast, but they were not heavily targeted until the midto late 1990s (McAuley and Simpfendorfer, 2003). Demersal gillnet catches of C. plumbeus more than doubled between 1994/1995 and 2003/2004, to .200 t year21 (McAuley, 2006a). In addition to the growth in the temperate gillnet fisherys catch, the concurrent development of a demersal longline shark fishery off the northwest coast of Australia during the late 1990s resulted in a similar escalation of C. plumbeus catches from the northern part of the stocks range. By 2003/2004, reported longline catches by this northern fishery had also exceeded 200 t year21, resulting in a combined C. plumbeus catch by these two target fisheries of 415 t in that year (McAuley 2006b). Although the northern shark longline fishery has a known bycatch of C. obscurus (McAuley et al., 2005), it is poorly identified in reported catches, though the catch is thought to be reasonably small (McAuley, 2006b). Owing to their slow rates of growth, high ages at maturity (Natanson et al., 1995; Sminkey and Musick, 1995; Natanson and Kohler, 1996; Simpfendorfer et al., 2002; McAuley et al., 2006) and low fecundity (Bass et al., 1973; Castro, 1983; McAuley et al., 2007a), both C. obscurus and C. plumbeus are highly susceptible to overfishing and are likely to require many decades to recover from periods of overexploitation (Hoff, 1990; Sminkey and Musick, 1996; Cortes, 1998, 1999, 2002; Smith et al., 1998; Simpfendorfer, 1999a; Brewster-Geisz and Miller, 2000). The difficulties associated with sustainably harvesting K-selected fishery resources such as these are indicated by the long history of rapid overexploitation of targeted shark fisheries around the world (Ripley, 1946; Parker and Stott, 1965; Holden, 1968, 1974, 1977; Casey et al., 1978). Severe declines in C. obscurus and C. plumbeus abundance indices in the western North Atlantic and Gulf of Mexico (Musick et al., 1993; Ulrich, 1996; Fowler et al., 2005) and highly pessimistic predictions of continuing stock depletion (Sminkey and Musick, 1996; Cortes, 1999; Brewster-Geisz and Miller, 2000; Cortes et al., 2006; NOAA/ NMFS, 2006), emphasize the importance of carefully managing these low-productivity shark resources to avoid the need for drastic conservation responses and associated economic loss to fisheries. An often cited obstacle to the successful assessment and management of fisheries exploiting long-lived shark stocks is that available time-series of fishery data (i.e. catches, fishing effort, and catch rates) are typically noisy and insufficient in duration to reflect trends in stock abundance (Anderson, 1990; Punt and Smith, 1999; Simpfendorfer, 1999a, b, 2005; Stevens et al., 2000; Bonfil, 2005). These data limitations, as well as the generally low economic value of shark catches, usually preclude the use of dynamic modelling techniques in assessing the effects of fishing on shark stocks. Although there are some notable examples of the application of dynamic assessment techniques to shark stocks (Walker, 1992; Punt and Walker, 1998; Punt et al., 2000; Simpf (...truncated)