© 2005 International Council for the Exploration of the Sea
Sources of variation in the feeding ecology of the piked spurdog (Squalus megalops): implications for inferring predator–prey interactions from overall dietary composition
a Southern Seas Ecology Laboratories, Darling Building DP 418, School of Earth and Environmental Sciences, University of Adelaide Adelaide, SA 5005, Australia
b Primary Industries Research Victoria PO Box 114 Queenscliff, Vic 3225, Australia
*Correspondence to J. M. Braccini: tel: +61 8 83036224; fax: +61 8 83034364. e-mail: matias.braccini{at}adelaide.edu.au.
Sources of variation in dietary composition were examined in the piked spurdog (Squalus megalops). The species is an opportunistic predator that consumes a wide range of prey items. When importance of prey was measured by weight or occurrence, S. megalops preyed largely on molluscs and teleosts. However, when number of prey was considered, the main items were crustaceans. A bootstrap analysis showed that considerable variability can be expected in the importance of prey items in the species' overall diet. Regional, seasonal, and ontogenetic differences in dietary composition were found, but there were no differences between mature and immature sharks or between males and females. The spatial and temporal variation in diet exhibited by S. megalops and the intrinsic natural variability of the dietary composition of this opportunistic predator suggest that studies that infer predator–prey interactions from overall diet are likely to miss information on the ecological relationships among species and thus account for only part of these interactions.
Keywords: Australia, diet, predator–prey interactions, shark, variation
Received 12 January 2005; accepted 14 April 2005.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionAppendixReferences |
|---|
The feeding ecology of marine animals has been studied to determine the ecological roles and position of animals within foodwebs and to understand predator–prey interactions (Caddy and Sharp, 1986; Pauly et al., 1998; Cortés, 1999). Interactions among species affect population dynamics and also cause indirect ecological effects (Alonzo et al., 2003). Hence, if interactions among species were determined, ecosystems could be managed with higher certainty (Yodzis, 1994). Traditional single-species fishery management ignores fishery impacts on ecosystems (Agardy, 2000). As an alternative, ecosystem-based fishery management has been proposed to account for such impacts (Gulland, 1978; Caddy and Sharp, 1986; Fulton et al., 2003). Many ecosystem models use dietary information as a proxy for the interactions among species (e.g. Christensen, 1995; Walters et al., 1997; Yodzis, 1998). However, most models use overall diet data, ignoring many sources of variation that can affect the dietary composition of predators.
Natural systems are dynamic and vary in time and space (Paine, 1988). It is, therefore, expected that diet of predators, and hence predator–prey interactions, may also vary in time and space. Trophic interactions are determined by the size of predators and their prey (Floeter and Temming, 2003), but little is known about predator–prey size relationships of large marine predators such as sharks. Also for sharks, the effects of time and space and their interactions with other potential sources of variation in their diet, such as sex or maturity condition, have been little studied. Although some studies have reported regional, seasonal, or ontogenetic differences in diet (see Wetherbee and Cortés, 2004, for a review), many studies on the diet of sharks have been limited to simple lists of prey items (Heithaus, 2004). Moreover, variation in diet has often been reported qualitatively with little statistical support (Ferry and Cailliet, 1996; Cortés, 1997; Wetherbee and Cortés, 2004). Hence, a more rigorous and quantitative approach is required to study the feeding ecology of sharks.
The piked spurdog (Squalus megalops) is a suitable species to test for the effects of potential sources of variation in the dietary composition of predators, as it is a very abundant shark in southern Australia (Jones, 1985; Bulman et al., 2001; Graham et al., 2001). S. megalops inhabits waters of the continental shelf and upper continental slope to 510 m (Last and Stevens, 1994). Off South Africa, females grow larger (782 mm total length, TL) than males (572 mm TL) and attain 50% maturity at 15 years, and 50% of males are mature at 9 years old (Watson and Smale, 1999). Given its high natural abundance, which has remained stable since it was first surveyed (Graham et al., 2001), S. megalops is a dominant and ecologically important species (Bulman et al., 2001) that is likely to make an important contribution to the structure and functioning of an ecosystem. Nevertheless, information on its feeding habits is scarce. Its overall diet has been described for animals caught off South Africa and eastern Australia, where it preys mainly on teleosts and cephalopods, but it also consumes crustaceans and elasmobranchs (Bass et al., 1976; Ebert et al., 1992; Bulman et al., 2001). Although those studies offer a preliminary description of the diet of this shark, more quantitative analyses are needed.
The purpose of this study was to investigate the effects of several sources of variation in the feeding ecology of S. megalops. The specific objectives were to: (i) quantify its overall dietary composition and account for how much variability would be expected when calculating overall prey importance; (ii) examine relationships between prey and predator size; and (iii) test for the effects of region, maturity condition, sex, season, and ontogenetic variation on its dietary composition.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
Sampling
S. megalops were obtained from the bycatch of shark and trawl vessels operating in the Australian southern and eastern scalefish and shark fishery (Figure 1). Samples were collected monthly between October 2002 and April 2004, with the exception of the period July–September (Table 1), when S. megalops seems to move off the fishing grounds and weather conditions restricted sampling. The specimens were sexed, measured (TL ±1 mm) and weighed on an electronic balance (±0.1 g). Maturity of males was determined on the basis of clasper calcification, condition of testes and vas efferens, and presence of semen in seminal vesicles. Maturity of females was determined on the basis of the condition of oviducal glands and ovarian follicles, and the presence of in utero eggs or embryos.
|
|
Diet and data analyses
Diet was studied by prey identification and analysis of stomach contents. The stomach of each fish was removed, and the contents were identified to the lowest taxon practical. When possible, to correlate size of prey and predator, body width (BW) of worms, TL of fish, mantle length (ML) of cephalopods, and shield length (SL) of hermit crabs were measured to the nearest millimetre. Where these lengths could not be measured, TL of fish, ML of cephalopods, and SL of hermit crabs were estimated from hard tissue pieces found in stomach contents by linear and allometric relationships determined by regression, using a personal reference collection and the fish and crustacean reference collections of the South Australian Museum, Australia, and Museum Victoria, Australia. Prey items that digest more speedily than other prey items or soft-bodied prey may be under-represented if the more persistent hard parts are included in the analyses (Bigg and Fawcett, 1985; Bigg and Perez, 1985). Hence, hard parts (e.g. beaks, vertebrae, chelipeds) were only used for estimating prey item size and describing the overall dietary spectrum, but they were excluded from further analyses.
Taxonomic classification of prey items does not account for differences in habitat utilization of a predator. Therefore, data analyses were carried out by main zoological group (Polychaeta, Sipuncula, Crustacea, Mollusca, Chondrichthyes, Teleostei) and ecological group separately. The ecological groups considered were benthic infauna (prey species living in the sediment), benthic epifauna (prey species living on the sediment surface), benthic (prey species living on the bottom), demersal benthic (prey species living near the bottom but not linked to it), demersal pelagic (prey species with extensive diel vertical migration), and pelagic (prey species living in the upper layers of the water column).
Overall diet
Stomach fullness (SF) and number of prey found in each stomach were recorded to determine the feeding pattern of S. megalops. Stomach fullness was recorded using a quarterly scale (0, empty; 1, 0–25% filled; 2, 26–50% filled; 3, 51–75% filled; 4, 76–100% filled). 
2 tests with Yates' continuity correction (Zar, 1999) were used to test for differences in the distribution of SF.
To obtain a precise description of the overall diet of a predator, it is important to determine the minimum number of stomachs required (Ferry and Cailliet, 1996; Cortés, 1997). The number of S. megalops collected was tested to determine whether sufficient sharks were sampled. Items such as sponges, hydroids, and algae were considered incidental, and were excluded from the analysis. The cumulative number of randomly pooled stomachs was plotted against the cumulative diversity of stomach contents. Diversity was calculated using the pooled quadrat method based on the Brillouin Index of diversity (HZ; Pielou, 1966). To ensure that curves reached an asymptotic value, 10 random orders of stomachs (curves) were calculated (Koen Alonso et al., 2002). Diversity curves were considered asymptotic if at least two previous values to the total sample diversity were in the range of asymptotic diversity ±0.05 (Koen Alonso et al., 2002). Diversity curves were calculated for each combination of factors considered in the analyses of variation in dietary composition.
No single method of analysis of stomach contents completely describes the diet of a predator (Hyslop, 1980); hence, the importance of prey items was evaluated using percentage weight (%W), percentage number (%N), percentage frequency of occurrence (%FO), and percentage Index of Relative Importance (%IRI; Pinkas et al., 1971; Cortés, 1997). Bootstrap methods (1000 replicates) were used to estimate confidence intervals (2.5th and 97.5th percentiles) around the dietary parameters (mean %W, %N, %FO, and %IRI; Haddon, 2001). From the original data matrix, random samples of the observations (i.e. each individual stomach) with replacement were generated to obtain the probability distribution of the dietary parameter estimates for each prey item.
Predator–prey size relationship
The relationship between prey size and shark size was determined using the Spearman rank correlation coefficient (rs). The length variables for the different taxonomic groups were considered. Relative and cumulative frequency histograms of prey size:predator size ratios were plotted to examine the patterns of prey size consumed by S. megalops (Bethea et al., 2004). For this latter analysis, only teleost and cephalopod prey were used.
Variation in dietary composition
Regional comparisons of diet were made for large females (471–650 mm TL) collected in autumn (Table 1). A one-way non-parametric multivariate analysis of variance (NP-MANOVA) using Bray–Curtis distances (Anderson, 2001) on weight and number data for sharks collected at the same time (autumn 2004) was used to test for regional effects on the diet of S. megalops. Weight and number data were transformed to fourth root and standardized to z-scores to minimize differences attributable to stomach size. Region was treated as a fixed factor. Equal sample sizes were used (n = 30 for the analysis of zoological groups, n = 28 for the analysis of ecological groups). If significant differences were found, a posteriori pairwise comparisons were made (Anderson, 2001).
Maturity condition was evaluated, and sexual, seasonal, and ontogenetic comparisons were made on sharks collected from Lakes Entrance between October 2002 and March 2004 (Table 1). Non-parametric multidimensional scaling (nMDS) on Bray–Curtis similarity measures using fourth root transformed data (Clarke, 1993) were used to visualize patterns of variation in dietary composition. Mean percentage weight and number of zoological and ecological groups were used.
The relative and interactive effects of maturity condition, sex, season, and size were evaluated in a similar way to the regional analysis using weight and number. S. megalops is sexually dimorphic, females attaining larger size than males; hence, separate analyses were undertaken for each sex to investigate the effects of maturity condition on dietary composition. The effects of maturity condition (mature, immature) and season (summer, autumn, spring; Table 1) were investigated using individuals within the 382–406 and 433–509 mm TL range for males and females, respectively. These ranges covered the sizes of the smallest mature and largest immature specimen of each sex. For the analysis of males, season was not included as a factor because of the low number of replicates for any season except summer. Hence, the analysis was done using data collected only during the latter season. For females, maturity condition was treated as fixed and orthogonal to the random factor season (i.e. every level of the factor "maturity", mature or immature, is present in every level of the factor "season", summer, autumn, or spring; Table 1). Similar sample sizes (n = 7 for males, n = 8 for females) were used for each combination of factors.
To test for sexual, ontogenetic (size), and seasonal differences, sharks of similar size (<471 mm TL) were used in a three-way NP-MANOVA (factors: sex, size, and season). Sex (males, females) and size (small and large males, small and medium-sized females) were treated as fixed and orthogonal to the random factor season (summer, autumn, spring; Table 1). Equal sample sizes (n = 6) were used for each combination of factors. As small and large males and small and medium-sized females had similar diets (see below), data were pooled to test for ontogenetic and seasonal differences between small (<471 mm TL) and large (
471 mm TL) animals. A two-way NP-MANOVA (factors: size and season) with equal sample sizes (n = 26) was used for each combination of factors. Finally, winter samples could only be collected for small specimens, so to include winter in the seasonal study, a one-way NP-MANOVA was undertaken for small S. megalops using a balanced design (n = 24).
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
The stomach contents and fullness of 937 S. megalops were examined. In all, there were 77 small males (274–400 mm TL), 105 small females (270–400 mm TL), 129 large males (400–470 mm TL), 193 medium-sized females (401–470 mm TL), and 433 large females (471–650 mm TL).
Overall diet
Of the 937 stomachs examined, 603 (65.3%) contained food, from which >60% contained a single prey item. For stomachs with >1 item, the number of prey items ranged from two to ten. For stomachs with prey, the distribution of stomach fullness was relatively even (
25%) and there were no significant differences among the frequency of individuals in each SF category (2 = 2.150, n = 603, p = 0.542).
Of the 603 stomachs with food, 111 were excluded because they contained only hard parts, sponges, hydroids, algae, or unidentified material. The prey diversity curve for the overall diet reached a stable level at about 350 stomachs (Figure 2a), so the sample size of 492 was large enough to describe the overall diet of S. megalops.
|
The stomachs contained 107 taxonomic levels of prey item: six polychaetes, two sipunculids, 29 crustaceans, 17 molluscs, 47 fish, remains of sea lion, and other items such as echiurids, algae, sponges, hydroids, and brittle stars (a Appendix). Arrow squid (family Ommastrephidae) was the dominant prey item, contributing the highest values of %W (20.03%), %N (7.54%), %FO (8.76%), and %IRI (32.05%). Octopus (Octopus sp.) was the second most important prey item by weight (12.55%), frequency of occurrence (7.66%), and relative importance (19.37%). The third major prey was fish of the family Triglidae (gurnards) in terms of weight (9.77%), number (5.33%), frequency of occurrence (5.97%), and relative importance (12.00%). Shrimps (Caridea) and hermit crabs (Diogenidae) were important by number (6.88% and 5.90%, respectively), but not in terms of weight or frequency of occurrence.
A similar pattern was observed when data were analysed by main zoological group (Appendix). Molluscs were the most important item by weight (56.43%), frequency of occurrence (35.89%), and relative importance (50.31%). However, the most numerous items were crustaceans (31.61%). Teleosts were the second most important item in terms of weight (38.32%), frequency of occurrence (34.03%), and relative importance (37.27%).
When data were analysed by ecological group, the most important group by weight was demersal pelagic prey (40.25%), followed by benthic (36.95%), and demersal benthic (11.04%) prey (Appendix). In contrast, benthic epifauna dominated by number (41.15%) and frequency of occurrence (29.41%), followed by benthic prey (21.10% by number and 25.35% by frequency of occurrence). Finally, for %IRI, the main ecological group was benthic prey (33.96%), followed by benthic epifauna (30.70%), demersal pelagic (26.52%), and demersal benthic (6.27%) prey. Pelagic and benthic infauna were less important.
Irrespective of analysing prey items by zoological or ecological group, considerable variability was found around the estimation of overall mean prey importance (Appendix). For important prey such as molluscs or teleosts, there was 20% of variability within the upper and lower 95% confidence intervals. However, for less important prey such as crustaceans, variability was 50%. When the mean values obtained from bootstrapping were compared with those obtained from point estimates of overall diet, variability ranged from 1 to 14% (not shown). A similar pattern was observed for ecological groups.
Predator–prey size relationship
S. megalops consumed prey of a wide range of sizes (Figure 3). More than 60% of teleosts and cephalopods consumed were less than 30% and 24% of S. megalops total length (TL), respectively, but S. megalops also consumed fish and cephalopods up to 60% of its TL.
|
No correlation was found between predator TL and shield length of hermit crabs (rs = 0.119, n = 65, p > 0.05), TL of teleosts (rs = 0.157, n = 39, p > 0.05), or body width of worms (rs = 0.273, n = 14, p > 0.05). However, there was a positive correlation between predator TL and mantle length of cephalopods (rs = 0.455, n = 43, p < 0.05; Figure 3).
Variation in dietary composition
Prey diversity for sharks collected from WWP (3.10; Figure 2b) and NSW (2.64; Figure 2d) was lower than that for sharks from EWP (3.85; Figure 2c), suggesting a more diverse diet at EWP. The prey diversity curves reached a stable level for each of the three regions analysed, indicating that the sample was large enough to describe the diet of sharks from each region.
There was a regional pattern in the diet of S. megalops (Table 2). Significant differences in dietary composition were found between sharks collected from WWP and EWP, irrespective of the use of weight or number of zoological or ecological groups (Figure 4; pairwise comparisons). Significant differences were also found between sharks collected from WWP and NSW when weight of zoological group and weight or number of ecological group were used (Figure 4; pairwise comparisons). No differences were found between the diets of sharks collected from EWP and NSW (Figure 4; pairwise comparisons). For EWP and NSW, S. megalops consumed mainly teleosts, molluscs, and crustaceans, and also small amounts of worms and chondrichthyans for EWP. However, for WWP, S. megalops preyed largely on molluscs and, to a lesser extent, teleosts. For ecological groups, S. megalops collected from WWP preyed mostly on demersal pelagic prey, whereas those collected from EWP and NSW preyed mostly on benthic organisms.
|
|
Most prey diversity curves (not shown) showed asymptotes or trends towards an asymptote for each combination of maturity condition and season. Irrespective of the use of weight or number of a zoological or an ecological group, there were no significant differences in dietary composition between immature and mature S. megalops (Table 3). Therefore, immature and mature sharks were pooled for subsequent analyses.
|
Most prey diversity curves (not shown) showed asymptotes or trends towards an asymptote for each combination of sex, size, and season. A significant seasonal pattern in the dietary composition of S. megalops was found for the three-way analysis, but there were no sexual or ontogenetic differences (Table 4). Therefore, both sexes and sizes (small and large males, and small and medium-size females) were pooled for subsequent analyses.
|
Prey diversity curves for each size–season combination reached a stable level and had similar values of diversity, except for small sharks collected in winter that showed lower values (Figure 5). After including all sizes in the analysis, significant ontogenetic and seasonal effects were detected. Also, a significant interaction between size and season was found for weight and number of prey items for both zoological and ecological groups (Figures 6, 7; Table 4). The ordination showed two separate groups when zoological data and ecological number data were used (Figure 6). Large sharks tended to be separated from small ones, and samples collected in summer and autumn were separated from those collected in spring. However, no clear visual pattern was observed when the analysis was done for ecological groups using weight data. Large and small S. megalops had different diets in summer and autumn but similar diets in spring (Figure 7; pairwise comparisons). In summer and autumn, large sharks consumed mainly molluscs, whereas small sharks consumed mainly crustaceans. For ecological groups in summer, large S. megalops preyed mainly on demersal pelagic prey whereas small sharks preyed on benthic organisms. In spring, both size classes had a similar feeding pattern, consuming mainly teleosts, followed by molluscs and crustaceans. By ecological group, large and small sharks collected in spring preyed mainly on benthic organisms. When winter was included in the seasonal analyses, the seasonal pattern was similar (Table 5). Dietary composition in summer, autumn, and winter was similar, but significant differences were observed among these three seasons and spring (pairwise comparisons).
|
|
|
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
Dietary studies of sharks commonly report a high proportion of empty stomachs and few prey items per stomach, most of them in advanced stages of digestion (Wetherbee et al., 1990; Ebert et al., 1992; Simpfendorfer et al., 2001a). Therefore, many shark species are considered intermittent feeders. For such species, short periods of active feeding are followed by longer periods of reduced predatory activity (Wetherbee et al., 1990; Wetherbee and Cortés, 2004). The present study supports this hypothesis. Almost 35% of stomachs examined were empty, and for stomachs with prey, >60% contained a single prey item, suggesting that feeding is intermittent. However, further research on the feeding duration, total digestion time, and gastric evacuation rates using captive S. megalops would allow estimates of feeding frequency and feeding periodicity.
There was a wide range of food items in the stomachs of S. megalops, which meant that many stomachs were needed to describe overall diet. When diversity curves have been used to determine the sample size required for a precise description of the diet of sharks, most studies have found stable levels of diversity at <200 stomachs sampled (Carrassón et al., 1992; Gelsleichter et al., 1999; Koen Alonso et al., 2002; Morato et al., 2003; Bethea et al., 2004). However, prey diversity was high for S. megalops, and at least 350 stomachs had to be sampled to describe its overall diet. S. megalops can be considered a generalist and opportunistic feeder given that portions of large teleosts, cephalopods, and sharks were found in many stomachs, and that they consumed abundant prey such as arrow squid (Triantafillos et al., 2004) and gurnards (Triglidae; M. Gomon, pers. comm.). Other studies also suggest that sharks are generalist and opportunistic feeders that consume the most abundant prey (Wetherbee et al., 1990; Hanchet, 1991; Ellis et al., 1996; Koen Alonso et al., 2002).
Overall, results differed when average prey importance was analysed using weight, number, or frequency of occurrence of prey groups. If importance of prey is to be deduced on the basis of weight or frequency of occurrence, S. megalops preyed largely on molluscs and teleosts. However, if number of prey is to be used, the main items were crustaceans. Analyses done by ecological group showed that S. megalops was a versatile predator that used a wide range of habitats. The most important items by weight were demersal pelagic and benthic prey, whereas benthic epifauna and benthic prey were the most consumed items by number and occurrence. Therefore, number, weight, and frequency of occurrence measures provided different information on feeding habit (MacDonald and Green, 1983; Bigg and Perez, 1985; Cortés, 1998). Ferry and Cailliet (1996) suggest using multiple measures when prey items differ in size. For generalist and opportunistic feeders that consume a wide range of prey, like S. megalops, the use of multiple measures allows better representation of overall diet.
Irrespective of which diet descriptor was used, the bootstrap analysis showed a wide range of variability around the estimate of overall importance of prey. In general, studies on the diet of sharks obtain samples opportunistically, and in many cases small sample sizes are collected. However, as sharks are considered opportunistic predators (Wetherbee et al., 1990), large sample sizes would be needed for a comprehensive description of diet. Also, many studies have reported a high proportion of empty stomachs (Wetherbee et al., 1990), and some studies only described diet in terms of number or occurrence of prey, whereas other studies used only weight. However, for S. megalops, number, occurrence, and weight of prey showed different patterns of importance of prey. Therefore, a combination of small sample size, high proportion of empty stomachs, the use of different descriptors of importance of prey, and the opportunistic predatory nature of many shark species, is likely to result in high variability in the dietary composition and hence in evaluation of predator–prey interactions. Accurate characterization of predator–prey interactions inferred from diet data is crucial for ecosystem-based models and in their increasing use as tools for fisheries management. However, if overall diet data do not incorporate a measure of the natural variability in dietary composition exhibited by many shark species, predatory interactions and hence model predictions may be misleading. For example, if overall diet data are used to describe the predatory relationships of S. megalops in southern Australia, the main interactions will be with molluscs, in terms of %W, or with crustaceans, in terms of %N. However, the main interactions will be with teleosts, if sampling is done only in spring, or with molluscs, if only large sharks are collected in summer and autumn, or with crustaceans, if only small sharks are collected in summer and autumn. The same pattern of variability is reported for other shark species. Simpfendorfer et al. (2001b) compared the diet of tiger sharks from four sites off Western Australia. Overall, the main predatory interactions by %FO were with turtles, teleosts, and sea snakes. However, for one site, North West Shelf, the interactions with teleosts and sea snakes were not as important as with dugongs, and for another site, Ningaloo, tiger sharks interact almost exclusively with turtles. The observed variability in the diet of sharks is particularly relevant when using overall diet data as a descriptor of predator–prey interactions, because the use of overall data may obscure site-, size-, or sex-specific interactions. Also, given that ecosystem-based models tend to use %W data from overall diet as inputs, the occurrence of a few heavy prey items, for example, may overestimate the importance of the interaction between the predator and those particular prey, and underestimate the importance of interactions with other prey.
Size-dependent predation can regulate population and community level dynamics (Brooks and Dodson, 1965), but size-selective feeding has been little studied in sharks. In the present study, S. megalops preyed on a wide range of prey size (4–60% of its TL) and, except for cephalopod items, the total length of S. megalops was not correlated with size of prey. Other studies found that shark diets consisted of relatively small prey (in most cases, <36% of the sharks TL), and that prey size was correlated to predator size (Cortés et al., 1996; Scharf et al., 2000; Bethea et al., 2004). However, the present study showed that S. megalops had little size preference for prey, supporting the belief that this shark is a generalist and opportunistic predator.
Predation can be highly variable in space and time (Bax, 1998). There was regional, seasonal, and ontogenetic variation in the diet of S. megalops, and this pattern was consistent despite analyses being conducted on weight or number of zoological or ecological prey groups. Variation was not explained by the effects of sex or maturity condition, but this could be due to the low number of replicates for each combination of factors (e.g. n = 6 for the sex x size x season analysis), and hence low statistical power (Ferry and Cailliet, 1996). Some authors have found differences in the diet of sharks between sexes (Hanchet, 1991; Stillwell and Kohler, 1993; Simpfendorfer et al., 2001b; Koen Alonso et al., 2002) and maturity condition (Koen Alonso et al., 2002). However, some of these studies may have confounded the effects of sex or maturity condition with other factors such as space and time because, although samples were obtained opportunistically across a wide spatial and temporal scale, space and time were not considered in the analyses.
Feeding plasticity of sharks results in regional, seasonal, and ontogenetic variation in diet that complicates an accurate description of feeding ecology (Wetherbee and Cortés, 2004). However, most studies on the feeding ecology of sharks have described only overall dietary composition. Some studies have reported regional, seasonal, or ontogenetic variation (Jones and Geen, 1977; Lyle, 1983; Laptikhovsky et al., 2001; Simpfendorfer et al., 2001a; Ebert, 2002), but most have done so qualitatively (Wetherbee and Cortés, 2004). When a quantitative approach was taken (Cortés et al., 1996; Simpfendorfer et al., 2001b; Vögler et al., 2003; White et al., 2004), region, season, or ontogeny were evaluated independently of each other even though samples were collected across wide spatial and temporal scales. When sampling is opportunistic across wide spatial and temporal scales, if the interactive effects of space and/or time are not considered, it is likely that differences in diet attributed to a certain factor (e.g. size) are unknowingly confounded by the effects of other factors (e.g. region) not included in the analysis. Furthermore, if a factor is analysed independently but many factors are involved, the analysis should, at least, be undertaken on standardized data to remove the effects of the other factors not considered.
Standardized data for the effects of season, sex, and size showed regional variation in the diet of large females collected in autumn. Sharks from WWP fed largely on demersal pelagic prey (mainly ommastrephid squid), but those from EWP and NSW had a more varied diet, also consuming benthic prey (teleosts and crustaceans). A demersal pelagic diet implies that a demersal shark such as S. megalops undergoes vertical feeding migrations to exploit pelagic prey such as squid or preys on squid while aggregated near the seabed (Roper and Young, 1975). These findings suggest that S. megalops would have different patterns of habitat utilization in different areas, interacting in different ecological communities and acting as an energy linkage between them. Although squid occur across the three regions (Norman and Reid, 2000), information on their abundance at a lower scale (regional level) is scarce. Several other shark species show regional variation in dietary composition, switching between prey types with changes in prey availability (Medved et al., 1985; Cortés and Gruber, 1990; Stillwell and Kohler, 1993; Simpfendorfer et al., 2001b). Therefore, it is unclear whether the regional differences found in the diet of S. megalops reflect different patterns in feeding and habitat utilization or rather the natural pattern of prey availability. In any case, the present findings reinforce the importance of considering spatial variation as a common phenomenon affecting the feeding ecology of sharks.
Large and small S. megalops exploited different resources during part of the year. In summer and autumn, large sharks preyed mostly on demersal pelagic prey (mainly ommastrephid squid), whereas small sharks consumed mainly benthic crustaceans. These ontogenetic differences may be attributed to morphological limitations of small sharks (e.g. gape-limited), better foraging ability of large fish, or differences in the habitat occupied by the two size classes. In spring, however, both size classes had a more varied diet, consuming mainly benthic organisms. Demersal pelagic prey such as squid occur throughout the year, but they show large, unpredictable fluctuations in abundance (Anderson and Rodhouse, 2001). Therefore, the decline in squid consumption shown by large S. megalops during spring may be due to a decline in the availability of squid. Collection of data on the seasonal variation in the abundance of squid in the studied area is needed for a better understanding of the seasonal pattern exhibited by large S. megalops. Seasonal and ontogenetic variation in diet is common, and it has been reported for a related species, the spiny dogfish (Squalus acanthias; Jones and Geen, 1977; Hanchet, 1991; Koen Alonso et al., 2002), and for many other shark species (e.g. Cortés and Gruber, 1990; Simpfendorfer et al., 2001b; White et al., 2004). Cortés et al. (1996) found an interaction between season and size of shark in the diet of the bonnethead shark. However, no other study on the diet of sharks has analysed the interaction of these factors when samples from different seasons and size classes were compared. In the present study, an interaction between size and season was found; large and small S. megalops had different diets in summer and autumn, but consumed similar prey items in spring. Therefore, the differences found in the dietary composition of large and small S. megalops suggest that large and small dogfish would exhibit, at least during part of the year, different predator–prey interactions and ecological roles within the marine ecosystem. Hence, if only the overall diet data are used in an ecosystem model as a proxy for the predator–prey interactions of S. megalops, some of the interactions exhibited by this species throughout its lifespan would be ignored.
In conclusion, high variability was found when the overall importance of prey items was estimated. Furthermore, the dietary composition of S. megalops varied in space and time, exhibiting differences among regions, seasons, and size classes. Therefore, the intrinsic natural variability in the dietary composition of S. megalops, and the spatial and temporal variation in diet exhibited by this opportunistic predator, suggest that studies that infer predator–prey interactions from overall diet are likely to miss information on the ecological relationships among species and therefore account for only part of these interactions. Understanding predator–prey interactions is required for long-term strategic ecosystem management (Bax, 1998). Hence, given that natural variability is intrinsic to ecological systems, the natural variability of predation should be considered when predatory interactions are used to model ecosystem dynamics.
|
Appendix |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
Overall dietary compositions. Prey item sorted by (upper panel) taxonomic, and (lower panel) ecological group. Mean percentage weight (%W), mean percentage number (%N), mean percentage frequency of occurrence (%FO), and mean percentage Index of Relative Importance (%IRI), and 95% confidence intervals. Unid.: unidentifiable; n = 492.
|
|
Acknowledgements |
|---|
We are grateful to M. Gomon, G. Poore, M. Norman, R. Wilson, W. Ponder, and G. Rouse for help in identifying prey items, to M. Koen Alonso for help with data analyses, and to P. Risley, G. Richardson, and the crew of the FV "Nungurner" for help in sample collection. E. Cortés and an anonymous referee provided insightful comments that improved this manuscript. This research was supported by an IPRS and UAPRS to JMB and a FRDC grant (FRDC 2002/033) to TIW. BMG was supported by an ARC QEII Research Fellowship. Funding for the field and laboratory components was provided by Sea World Research and Rescue Foundation, Royal Zoological Society of New South Wales, Nature Foundation SA, and Royal Zoological Society of South Australia.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
-
Agardy T. (2000) Effects of fisheries on marine ecosystems: a conservationist's perspective. ICES Journal of Marine Science 57:761–765.
Alonzo S.H., Switzer P.V., Mangel M. (2003) An ecosystem-based approach to management: using individual behaviour to predict the indirect effects of Antarctic krill fisheries on penguin foraging. Journal of Applied Ecology 40:692–702.[CrossRef][Web of Science]
Anderson C.I.H. and Rodhouse P.G. (2001) Life cycles, oceanography and variability: ommastrephid squid in variable oceanographic environments. Fisheries Research 54:133–143.[CrossRef][Web of Science]
Anderson M.J. (2001) A new method for non-parametric multivariate analysis of variance in ecology. Australian Ecology 26:32–46.[CrossRef]
Bass A.J., D'Aubrey J.D., Kistnasamy N. (1976) Sharks of the east coast of southern Africa. 6. The families Oxynotidae, Squalidae, Dalatiidae and Echinorhinidae. Investigational Report of the Oceanographic Research Institute of South Africa 45: 103 pp.
Bax N.J. (1998) The significance and prediction of predation in marine fisheries. ICES Journal of Marine Science 55:997–1030.
Bethea D.M., Buckel J.A., Carlson J.K. (2004) Foraging ecology of the early life stages of four sympatric shark species. Marine Ecology Progress Series 268:245–264.[Web of Science]
Bigg M.A. and Fawcett I. (1985) Two biases in diet determination of northern fur seals (Callorhinus ursinus). In Beddington J.R., Beverton R.J.H., Lavigne D.M. (Eds.). Marine Mammals and Fisheries(George Allen & Unwin, London) pp. 284–291 354 pp.
Bigg M.A. and Perez M.A. (1985) Modify volume: a frequency–volume method to assess marine mammal food habits. In Beddington J.R., Beverton R.J.H., Lavigne D.M. (Eds.). Marine Mammals and Fisheries(George Allen & Unwin, London) pp. 277–283 354 pp.
Brooks J.L. and Dodson S.I. (1965) Predation, body size, and composition of plankton. Science 150:28–35.
Bulman C., Althaus F., He X., Bax N.J., Williams A. (2001) Diets and trophic guilds of demersal fishes of the south-eastern Australian shelf. Marine and Freshwater Research 52:537–548.[CrossRef][Web of Science]
Caddy J.F. and Sharp G.D. (1986) An ecological framework for marine fishery investigations. FAO Fisheries Technical Paper 283: 152 pp.
Carrassón M., Stefanescu C., Cartes J.E. (1992) Diets and bathymetric distributions of two bathyal sharks of the Catalan deep sea (western Mediterranean). Marine Ecology Progress Series 82:21–30.[Web of Science]
Christensen V. (1995) A model of trophic interactions in the North Sea in 1981, the year of the stomach. Dana 11:1–28.
Clarke K.R. (1993) Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecology 18:117–143.[CrossRef][Web of Science]
Cortés E. (1997) A critical review of methods of studying fish feeding based on analysis of stomach contents: application to elasmobranch fishes. Canadian Journal of Fisheries and Aquatic Sciences 54:726–738.
Cortés E. (1998) Methods of studying fish feeding: reply. Canadian Journal of Fisheries and Aquatic Sciences 55:2708.
Cortés E. (1999) Standardized diet compositions and trophic levels of sharks. ICES Journal of Marine Science 56:707–717.
Cortés E. and Gruber S.H. (1990) Diet, feeding habits and estimates of daily ration of young lemon sharks, Negaprion brevirostris (Poey). Copeia 1990:204–218.[CrossRef]
Cortés E., Manire C.A., Hueter R.E. (1996) Diet, feeding habits, and diel feeding chronology of the bonnethead shark, Sphyrna tiburo, in southwest Florida. Bulletin of Marine Science 58:353–367.[Web of Science]
Ebert D.A. (2002) Ontogenetic changes in the diet of the sevengill shark (Notorynchus cepedianus). Marine and Freshwater Research 53:517–523.[CrossRef][Web of Science]
Ebert D.A., Compagno L.J.V., Cowley P.D. (1992) A preliminary investigation of the feeding ecology of squaloid sharks off the west coast of southern Africa. South African Journal of Marine Science 12:601–609.
Ellis J.R., Pawson M.G., Shackley S.E. (1996) The comparative feeding ecology of six species of shark and four species of ray (Elasmobranchii) in the north-east Atlantic. Journal of the Marine Biological Association of the United Kingdom 76:89–106.[Web of Science]
Ferry L.A. and Cailliet G.M. (1996) Sample size and data analysis: are we characterizing and comparing diet properly? In MacKinlay D. and Shearer K. (Eds.). Feeding Ecology and Nutrition in Fish: Proceedings of the Symposium on the Feeding Ecology and Nutrition in Fish, International Congress on the Biology of Fishes, San Francisco, California, 14–18 July 1996American Fisheries Society pp. 71–80.
Floeter J. and Temming A. (2003) Explaining diet composition of North Sea cod (Gadus morhua): prey size preference vs. prey availability. Canadian Journal of Fisheries and Aquatic Sciences 60:140–150.
Fulton E.A., Smith A.D.M., Johnson C.R. (2003) Effect of complexity on marine ecosystem models. Marine Ecology Progress Series 253:1–16.[Web of Science]
Gelsleichter J., Musick J.A., Nichols S. (1999) Food habits of the smooth dogfish, Mustelus canis, dusky shark, Carcharhinus obscurus, Atlantic sharpnose shark, Rhizoprionodon terraenovae, and the sand tiger, Carcharias taurus, from the northwest Atlantic Ocean. Environmental Biology of Fishes 54:205–217.[CrossRef][Web of Science]
Graham K.J., Andrew N.L., Hodgson K.E. (2001) Changes in relative abundance of sharks and rays on Australian south east fishery trawl grounds after twenty years of fishing. Marine and Freshwater Research 52:549–561.[CrossRef][Web of Science]
Gulland J.A. (1978) Fishery management: new strategies for new conditions. Transactions of the American Fisheries Society 107:1–11.[CrossRef]
Haddon M. (2001) Modelling and Quantitative Methods in Fisheries(Chapman & Hall/CRC, Boca Raton) 406 pp.
Hanchet S. (1991) Diet of spiny dogfish, Squalus acanthias Linnaeus, on the east coast, South Island, New Zealand. Journal of Fish Biology 39:313–323.[CrossRef][Web of Science]
Heithaus M.R. (2004) Predator–prey interactions. In Carrier J.F., Musick J.A., Heithaus M.R. (Eds.). Biology of Sharks and their Relatives(CRC Press, Boca Raton) pp. 487–521 596 pp.
Hyslop E.J. (1980) Stomach contents analysis – a review of methods and their application. Journal of Fish Biology 17:411–429.[CrossRef][Web of Science]
Jones G. K. (1985) An exploratory dropline survey for deepsea trevalla (Hyperoglyphe antarctica) in continental slope waters off South Australia. Fisheries Research Paper, Department of Fisheries, 15. 20 pp.
Jones B.C. and Geen G.H. (1977) Food and feeding of spiny dogfish (Squalus acanthias) in British Columbia waters. Journal of the Fisheries Research Board of Canada 34:2067–2078.[Web of Science]
Koen Alonso M., Crespo E.A., Garcia N.A., Pedraza S.N., Mariotti P.A., Mora N.J. (2002) Fishery and ontogenetic driven changes in the diet of the spiny dogfish, Squalus acanthias, in Patagonian waters, Argentina. Environmental Biology of Fishes 63:193–202.[CrossRef][Web of Science]
Laptikhovsky V.V., Arkhipkin A.I., Henderson A.C. (2001) Feeding habits and dietary overlap in spiny dogfish Squalus acanthias (Squalidae) and narrowmouth catshark Schroederichthys bivius (Scyliorhinidae). Journal of the Marine Biological Association of the United Kingdom 81:1015–1018.[Web of Science]
Last P.R. and Stevens J.D. (1994) Sharks and Rays of Australia(CSIRO Publishing, Melbourne) 513 pp.
Lyle J.M. (1983) Food and feeding habits of the lesser spotted dogfish, Scyliorhinus canicula (L.), in Isle of Man waters. Journal of Fish Biology 23:725–737.[CrossRef][Web of Science]
MacDonald J.S. and Green R.H. (1983) Redundancy of variables used to describe importance of prey species in fish diets. Canadian Journal of Fisheries and Aquatic Sciences 40:635–637.
Medved R.J., Stillwell C.E., Casey J.G. (1985) Stomach contents of young sandbar sharks, Carcharhinus plumbeus, in Chincoteague Bay, Virginia. Fishery Bulletin US 83:395–402.
Morato T., Solá E., Grós M.P., Menezes G. (2003) Diets of thornback ray (Raja clavata) and tope shark (Galeorhinus galeus) in the bottom longline fishery of the Azores, northeastern Atlantic. Fishery Bulletin US 101:590–602.
Norman M. and Reid A. (2000) A Guide to Squid, Cuttlefish and Octopuses of Australasia(CSIRO Publishing and the Gould League of Victoria, Melbourne) 96 pp.
Paine R.T. (1988) Food webs: road maps of interactions or grist for theoretical development? Ecology 69:1648–1654.[CrossRef][Web of Science]
Pauly D., Trites A.W., Capuli E., Christensen V. (1998) Diet composition and trophic levels of marine mammals. ICES Journal of Marine Science 55:467–481.
Pielou E.C. (1966) The measurement of diversity in different types of biological collections. Journal of Theoretical Biology 63:131–144.
Pinkas L.M., Oliphant S., Iverson I.L.K. (1971) Food habits of albacore, bluefin tuna and bonito in Californian waters. California Fish and Game 152:1–105.
Roper C.F.E. and Young R.E. (1975) Vertical distributions of pelagic cephalopods. Smithsonian Contributions to Zoology 209:1–51.
Scharf F.S., Juanes F., Roundtree R.A. (2000) Predator–prey relationships of marine fish predators: interspecific variation and effects of ontogeny and body size on trophic–niche breadth. Marine Ecology Progress Series 208:229–248.[CrossRef][Web of Science]
Simpfendorfer C.A., Goodreid A., McAuley R.B. (2001) Diet of three commercially important shark species from Western Australian waters. Marine and Freshwater Research 52:975–985.[CrossRef][Web of Science]
Simpfendorfer C.A., Goodreid A.B., McAuley R.B. (2001) Size, sex and geographic variation in the diet of the tiger shark, Galeocerdo cuvier, from Western Australian waters. Environmental Biology of Fishes 61:37–46.[CrossRef][Web of Science]
Stillwell C.E. and Kohler N.E. (1993) Food habits of the sandbar shark Carcharhinus plumbeus off the U.S. northeast coast, with estimates of daily ration. Fishery Bulletin US 91:138–150.
Triantafillos L., Jackson G.D., Adams M., McGrath Steer B.L. (2004) An allozyme investigation of the stock structure of arrow squid Nototodarus gouldi (Cephalopoda: Ommastrephidae) from Australia. ICES Journal of Marine Science 61:829–835.
Vögler R., Milessi A.C., Quiñones R.A. (2003) Trophic ecology of Squatina guggenheim on the continental shelf off Uruguay and northern Argentina. Journal of Fish Biology 62:1254–1267.[CrossRef][Web of Science]
Walters C., Christensen V., Pauly D. (1997) Structuring dynamic model of exploited ecosystems from trophic mass–balance assessments. Reviews in Fish Biology and Fisheries 7:139–172.[CrossRef][Web of Science]
Watson G. and Smale M.J. (1999) Age and growth of the shortnose spiny dogfish Squalus megalops from the Agulhas Bank, South Africa. South African Journal of Marine Science 21:9–18.
Wetherbee B.M. and Cortés E. (2004) Food consumption and feeding habits. In Carrier J.F., Musick J.A., Heithaus M.R. (Eds.). Biology of Sharks and their Relatives(CRC Press, Boca Raton) pp. 225–246 596 pp.
Wetherbee B.M., Gruber S.H., Cortés E. (1990) Diet, feeding habits, digestion, and consumption in sharks, with special reference to the lemon shark, Negaprion brevirostris. In Pratt H.L., Gruber S.H., Taniuchi T. (Eds.). Elasmobranchs as Living Resources: Advances in Biology, Ecology, Systematics, and the Status of the Fisheries pp. 29–47 NOAA Technical Report, NMFS 90. 518 pp.
White W.T., Platell M.E., Potter I.C. (2004) Comparisons between the diets of four abundant species of elasmobranchs in a subtropical embayment: implications for resource partitioning. Marine Biology 144:439–448.[CrossRef]
Yodzis P. (1994) Predator–prey theory and management of multispecies fisheries. Ecological Applications 41:51–58.[Medline]
Yodzis P. (1998) Local trophodynamics and the interaction of marine mammals and fisheries in the Benguela ecosystem. Journal of Animal Ecology 67:635–658.[CrossRef][Web of Science]
Zar J.H. (1999) Biostatistical Analysis 4th edn (Prentice Hall, London) 663 pp.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||