Resource partitioning in gurnard species using trophic analyses: The importance of temporal resolution

Fisheries Research 186 (2017) 301–310

Contents lists available at ScienceDirect

Fisheries Research journal homepage: www.elsevier.com/locate/fishres

Resource partitioning in gurnard species using trophic analyses: The importance of temporal resolution Joo Myun Park a,∗ , Troy F. Gaston b , Jane E. Williamson a a b

Department of Biological Sciences, Macquarie University, Sydney NSW 2109, Australia School of Environmental and Life Sciences, University of Newcastle, Ourimbah NSW 2258, Australia

a r t i c l e

i n f o

Article history: Received 14 July 2016 Received in revised form 23 September 2016 Accepted 13 October 2016 ˜ Handle by Dr. J Vinas Keywords: Trophic ecology Stomach contents Stable isotope Gurnards Tasmania

a b s t r a c t Dietary habits and intra- and inter-specific trophic ecology of co-occurring Lepidotrigla mulhalli and L. vanessa from south-eastern Australia were analysed using stomach content and stable isotope ratios (␦13 C and ␦15 N). Both species are bottom-feeding carnivores that consumed mainly benthic crustaceans, but teleosts were also abundant in the diet of larger L. vanessa. Non-metric multidimensional scaling (nMDS) ordination and analysis of similarity (ANOSIM) of dietary data revealed significant inter-specific dietary differences; i.e. food resource partitioning. Carbon (␦13 C) and nitrogen (␦15 N) stable isotope values were similar between L. mulhalli and L. vanessa, however, suggesting similar trophic positioning. Ontogenetic changes in diet composition and stable isotope values were evident. As L. vanessa grew, they preyed upon larger individuals, such as teleosts and caridean shrmips, but no such trend was observed in the diets of L. mulhalli. Adults of both species were significantly enriched in 15 N relative to juvenile conspecifics thus supporting these data. Consequently, in this study, both methodologies, i.e. stomach content and stable isotope analyses, provided evidence of inter- and/or intra-specific dietary segregations and trophic niche partitioning between co-occurring L. mulhalli and L. vanessa off Tasmanian waters. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Sympatric species in marine habitats often co-exist by sharing resources such as food and habitat (Platell and Potter, 2001; O’Shea et al., 2013). Among the different resource axes that control community structure, food resource partitioning and niche segregation are fundamental processes in community ecology and play a major role in the co-existence of similar species (Ross, 1986; Platell and Potter, 1999). The study of food utilization and partitioning between co-occurring fish species is often the principal mechanism of niche segregation (Gerking, 1994; Duarte and Garcia, 1999), and also useful for developing management approaches for conservation and sustainability (Micheli and Halpern, 2005; Greenstreet and Rogers, 2006), as ecosystems with greater diversity and niche complexity generally show greater resilience to outside pressures (Elmqvist et al., 2003). Gurnards (Scorpaeniformes: Triglidae) are marine demersal fish commonly found in tropical and temperate waters globally (Richards and Jones, 2002). Triglidae fishes constitute approximately 125 species in 9 genera (Froese and Pauly, 2015) and

∗ Corresponding author. E-mail address: [email protected] (J.M. Park). http://dx.doi.org/10.1016/j.fishres.2016.10.005 0165-7836/© 2016 Elsevier B.V. All rights reserved.

more than 30 species occur in Australian waters (Rowling et al., 2010). The rough-snouted gurnard Lepidotrigla mulhalli (Macleay, 1884) and butterfly gurnard L. vanessa (Richardson, 1839) are widely distributed throughout southern Australian waters from southeastern New South Wales to southwestern Western Australia (Gomon et al., 2008), occurring in depths less than 200 m (May and Maxwell, 1986). Although several large-sized gurnards such as the red gurnard (Chelidonichthys kumu) and latchets (Pterygotrigla andertori and P. polyommata) grow large enough to be marketed, the relatively small gurnards (e.g. Lepidotrigla spp.) are a major component of trawl-bycatch but not economically valuable per se in Australia (Rowling et al., 2010). L. mulhalli have been listed as ecologically important species, however, because of their high abundance and prominence in the diets of other fishes in the southeastern Australian marine ecosystem (Bulman et al., 2001). The diet of Triglidae fishes has been studied worldwide (e.g. Ross, 1978; Moreno-Amich, 1992, 1994; Terrats et al., 2000; Boudaya et al., 2007; Huh et al., 2007; Baeck et al., 2011). Partitioning of resources amongst triglid fishes also has been studied in the North-western Atlantic (Ross, 1977), the Mediterranean Sea (Morte et al., 1997), south-western Australia (Platell and Potter, 1999), and the Cantabrian Sea (Lopez-Lopez et al., 2011). Despite their ecological importance, however, little is known about the dietary habits of Lepidotrigla species in southern Australia. Two papers have

302

J.M. Park et al. / Fisheries Research 186 (2017) 301–310

described the feeding of guild of these fishes, including L. mulhalli and L. vanessa in the southeastern Australian waters (Coleman and Mobley, 1984; Bulman et al., 2001), and a paper reported carbon and nitrogen stable isotope values for Lepidotrigla species (Devenport and Bax, 2002). However, there is no targeted study on how feeding may change with size and how resources and trophic niche may be partitioned among these species. Historically, food web interactions have been problematic due to the difficulties associated with stomach content analysis. This technique limits dietary analysis to the very short term, and can be strongly influenced by prey choice (soft vs. hard prey) and regurgitation during capture (Cortés, 1997). More recently, stable isotope analysis has become the technique of choice, with fewer limitations than stomach content analysis and more accurate determination of long-term diet choice (Bearhop et al., 2004). When consuming a prey, a predator integrates the carbon (␦13 C, 13 C: 12 C) and nitrogen (␦15 N, 15 N: 14 N) isotopic ratios of its prey into its own tissues. The consumer’s tissue will thus be a time-integrated dietary representation on a scale of weeks to months (e.g., Buchheister and Latour, 2010), unlike stomach contents that only represent a snapshot of diet. Thus, stable isotopes can elucidate the prey groups that are directly responsible for driving tissue growth and production of consumer species (Fry, 2006), and isotopic techniques are useful for identifying broader sources of production and for differentiating between benthic and pelagic trophic pathway (Fry, 2006). However, stable isotope analyses do not attain the same level of taxonomic resolution afforded by stomach content analysis. For example, different prey groups with similar trophic isotopic values are impossible to distinguish from each other by stable isotope analysis. Thus, a combination of stomach contents and stable isotope analysis is increasingly being used to improve interpretation of trophic ecology and aquatic food web (Lin et al., 2007; Cresson et al., 2014; Knickle and Rose, 2014). However, few studies have attempted to describe trophic ecology using both methods in Australian waters. In this study, dietary habits and the intra- and inter-specific trophic niche partitioning between two species of gurnards in the Tasmanian waters, Australia were assessed. Our specific objectives were to 1) investigate the diets of L. mulhalli and L. vanessa; 2) identify any size-related change in dietary composition based on maturity; and 3) assess any inter- and intra-specific dietary overlap and stable isotope signatures between the two species in this area. The results would be highly beneficial both for management purposes and future trophic community analyses locally and globally.

2. Materials and methods 2.1. Study area and sampling Field sampling was conducted in waters off northeastern Tasmania, Australia (40◦ 15 S–42◦ 20 S, 147◦ 05 E–148◦ 35 E; Fig. 1). Samples were collected as by-catch from repeated research trawls on board the Australian Maritime College vessel RV Bluefin in the Australian spring 2014 (September and November). Fish were collected at depths between 30 and 40 m using a 70 mm mesh demersal trawl of 16 m headline length towed at 3 knots. Immediately after capture, individuals were identified to species, snap frozen at −20 ◦ C and transported to the Marine Ecology Laboratory at Macquarie University. Fish were held frozen until processing, which occurred immediately after thawing in the laboratory. This method of preservation did not influence the ability to identify prey or to estimate the gravimetric contributions of the various prey items.

2.2. Stomach contents analysis In the laboratory, total length (TL, ± 1.0 mm) and wet weight (± 1.0 g) were measured for each specimen. Stomachs were removed and preserved in 70% isopropanol for at least 24 h, and then the contents were analysed using a stereo microscope. Stomach contents were identified to the lowest taxonomic level and the prevalence of each prey item was quantified (numbers and wet weights [±0.001 g] of each prey item were recorded). Cumulative prey curves were constructed for each species to determine if a sufficient number of stomachs were analysed to describe their diets (Ferry and Cailliet, 1996). The order of stomachs was randomized 10 times and the cumulative number of new prey taxa was counted for each randomization. Mean number of prey taxa against the number of stomachs analysed were plotted. Attainment of an asymptote indicated that an adequate number of stomachs were studied. A curve was considered to asymptote if at least ten previous values of the total number of prey taxa were in the range of the asymptotic number of prey ± 0.5 (Huveneers et al., 2007). Diet was quantified by frequency of occurrence (%F = 100 × Ai × N−1 ), as a numerical percentage (%N = 100 × Ni × NT −1 ), and as a mass percentage (%M = 100 × Mi × MT −1 ), where Ai was the number of fish preying on species i, N was the total number of fish examined (excluding those with empty stomachs), Ni (Mi ) was the number (mass) of prey individuals i, and NT (MT ) was the total number (mass) of prey individuals. Next, the index of relative importance (IRI) (Pinkas et al., 1971) was calculated for each prey item as follows: IRI = (%N +%M) × %F, and expressed as a percentage (%IRI). To infer ontogenetic trends, individuals were categorised into two size classes (juvenile and adult) based on their gonadosomatic index (GSI = gonad weight/body weight × 100). GSI values of females were plotted against their TL (Fig. S1). The TL with a dramatic increase in GSI was considered to be the minimum maturity size (Tuuli et al., 2011), i.e. criteria to divide between juvenile (immature) and adult (mature). Juveniles were defined as 105–154 mm (n = 13) for L. mulhalli and 119–201 mm (n = 89) for L. vanessa, and adults 155–218 mm (n = 155) for L. mulhalli and 202–248 mm (n = 73) for L. vanessa. Diet diversity and niche breadth were calculated from number of the lowest possible taxonomical level using the Shannon–Wiener diversity index (H ) and Levin’s standardized niche breadth (BN ) (Krebs, 1989). 2.3. Multivariate analysis-Stomach contents Stomach content data were analysed using non-metric multidimensional scaling (nMDS), permutational multivariate analysis of variance (PERMANOVA), analysis of similarity (ANOSIM) and similarity percentages (SIMPER) to make intra- and inter- specific comparisons between these two species based on percentage mass (%M) data between two species (Clarke and Gorley, 2006; Anderson et al., 2008). To examine the relative extent to which the dietary composition of fish differed between species, and between size classes, stomachs were randomly sorted into groups of five or six (depending on the sample size). The proportion of mass for each of the major prey categories were determined for each group for dietary composition analyses. Randomization and subsequent grouping of mass data were designed to reduce the number of prey categories in the samples with zero values, thus increasing the effectiveness of multivariate analysis (White et al., 2004; Marshall et al., 2008). Mass data were square root transformed to avoid any tendency for the main dietary components to be excessively dominant and Bray–Curtis similarity matrices were constructed for each of the three gurnards (Platell and Potter, 2001). The Bray–Curtis

J.M. Park et al. / Fisheries Research 186 (2017) 301–310

303

Fig. 1. Location of the study area in the northeastern Tasmanian waters, Australia. Samples were collected within trawl grounds (shaded area).

similarity matrices were visualized via nMDS ordination plots. To test whether significant differences in dietary composition exist between the two species, dietary data were subjected to a two-way PERMANOVA, with species (two levels) × size class (two levels). PERMANOVA is a non-parametric distance-based analysis of variance that uses permutation procedures to test hypotheses. PERMANOVA assigns components of variation (COV) of differing magnitudes to the main factors and interaction between combinations of main factors. The larger the component of variation, the greater the influence of a particular factor or interaction term on the structure of the data (Anderson et al., 2008; Linke 2011). Two-way crossed ANOSIMs were then used to determine the relative importance of species and size class based on the same factors as used in the PERMANOVA (see above). A one-way ANOSIM test was used to determine whether the diets of size class of each species was significantly different (Clarke and Gorley, 2006). Global R-statistic values from the ANOSIM verified similarities (distance) within defined groups vary between −1 and +1. An R value of zero represented no differences between factor groups or subset samples (stomachs); and R values of −1 and +1 indicated significant separation of subset samples (stomachs) within species and that size classes (factors) were similar (␣ = 0.05). Similarity of percentages (SIMPER) was used to estimate the contribution of each prey category to species differences in diets. Analyses were performed using the routines in the PRIMER v6 multivariate statistics package (www. primer-e.com) and the PERMANOVA+ add-on module (Anderson et al., 2008).

2.4. Stable isotope analysis Stable isotope analysis was performed on a randomly selected subset of gurnards. White dorsal muscle tissue from each species (n = 21 L. mulhalli and n = 22 L. vanessa) was excised, because it has low lipid and inorganic carbonate content which yield less variability of ␦13 C and ␦15 N values than other tissues (Pinnegar and Polunin, 1999). Each muscle tissue rinsed with RO (reverse osmosis) water, dried at 60 ◦ C for 24 h, and then ground to a powder using a mortar and pestle. Ground samples were weighed (1–2 mg) into tin capsules and analysed using a Sercon Hydra 20–22 automated Isoprime Isotope Ratio Mass Spectrometer at Griffith University. Results are expressed in ␦ notation relative to the standards Pee Dee Belemnite and atmospheric N2 for ␦13 C and ␦15 N, respectively, according to the equation; ␦X(‰) = [(Rsample /Rstandard ) − 1] × 1000, where X is 13 C or 15 N and R is the mass ration of the heavy to light stable isotopes 13 C/12 C or 15 N/14 N for either the sample or standard, respectively. Differences in mean stable isotope values (␦13 C and ␦15 N) between species, and between size classes within species were assessed using two-way analysis of variance (ANOVA) with species and size class as fixed effects. Relationships between standard length and both stable isotope values of each species were analysed using linear regressions. These were performed using SYSTAT software (Systat version 18. SPSS Inc., USA); significant differences were based on the 0.05 significance level.

304

J.M. Park et al. / Fisheries Research 186 (2017) 301–310

2.5. Trophic position Data on dietary composition were used to estimate the trophic position with the aid of TrophLab software (June 2000 version; Pauly et al., 2000). This software yields a trophic level (or TROPH value) that reflects the position of the organism within the food web that largely defines the aquatic ecosystem. To estimate fish TROPH values, it is necessary to consider both the dietary composition and the TROPH values of their food item(s). The TROPH of a fish species “i” is next estimated to be: TROPHi = 1 + DCij × TROPHj , where DCij is the proportion of prey “j” in the diet of predator “i” and TROPHj is the trophic level of prey “j”. The standard error (SE) of each TROPH value is estimated using the weight contribution and trophic level of each prey item in the diet. The trophic levels of various prey organisms have been determined by calculating TROPH values using the TrophLab database of FishBase (Froese and Pauly, 2015). In terms of stable nitrogen isotope (␦15 N), ␦15 N values provide an indication of the trophic position of consumers (Post, 2002). Thus, the trophic position can be calculated by: Trophic position = 2 + [(␦15 Nfish − ␦15 Nbaseline )/3.4)], where 2 represents trophic position of benthic primary consumers, ␦15 Nbaseline is the average value of seven benthic primary consumers, sea urchin (Clypeaster), sea cucumber (Holothuroidea), sea mouse (Aphrodita), ophiuroids, bivalve (Glycymeris striatularis), polychaete (Sabellidae) and gastropod (Maoricolpus roseus) (i.e. 7.43 ± 1.37) in the southeastern marine ecosystem (Davenport and Bax, 2002), and 3.4 indicates increment in ␦15 N per 1.0 trophic level.

3. Results 3.1. General stomach contents In total, 170 Lepidotrigla mulhalli (TL = 105–218 mm, mean ± SD = 169 ± 13 mm) and 162 L. vanessa (TL = 119–248 mm, mean ± SD = 188 ± 34 mm) were studied. Percentages of empty stomach were 10.6% for L. mulhalli and 17.9% for L. vanessa. Cumulative prey curves in terms of overall diet (number of prey species found) attained asymptotes in 83 stomachs for L. mulhalli and 98 stomachs for L. vanessa. Thus, sample sizes were sufficiently large to allow us to adequately describe the diet of both fishes in our study area (Fig. 2). A total of 152 L. mulhalli stomachs that were assessed contained at least 15 identifiable prey taxa (Table 1). Crabs accounted for almost the entire diet by%IRI (69.78%). Juvenile crabs (megalopa stage) were the most common crab preys, consisting of 56.82% by mass and 29.76% by number, and occurring in 66.45% of all stomachs examined. Cumaceans and gammarid amphipods were second in importance, comprising 15.29% and 10.22% by IRI, respectively. For L. vanessa, the stomachs (n = 133) contained a total of 16 prey taxa. Most of these were crabs and teleosts, which comprised 49.67% and 31.55% by IRI of all stomachs examined, respectively. Ebalia crabs (16.41% by mass) and juvenile crabs (megalopa stage sp.A, 16.64% by mass) were the most common crustaceans consumed, while morid fishes (20.67% by mass) were the most abundant fish prey consumed. Caridean shrmips were the next most abundant prey item, and accounted for 9.94% by IRI. Leptochela sydniensis was the principal shrimp prey species. Remaining prey items accounted for <5.0% by IRI in the diets of both species. Gravel was also found frequently in the diets of both species, but its occurrence was twice as high in L. mulhalli diets (59.21%) than L. vanessa (28.57%). Overall diet diversity and niche breadth differed between the two species. Values of diversity and dietary breadth of L. mulhalli (H = 2.37, BN = 0.117) were lower than those of L. vanessa (H = 2.89,

Fig. 2. Cumulative prey curves (prey taxa per stomach) for a) L. mulhalli and b) L. vanessa collected off northeastern Tasmania, Australia. Vertical bars represent standard deviations.

BN = 0.177) indicating that L. vanessa showed a more diversified diet and consumed a wider range of prey species than L. mulhalli. 3.2. Size-related changes in dietary compositions The two species of gurnards showed correlative trends in diet as they grew (Fig. 3). L. mulhalli showed a similar dietary composition between juvenile and adult groups, with crabs representing more than 67.0% by mass. While euphausia were relatively abundant in the diets of small size class of L. mulhalli, and cumaceans and amphipods were consumed more by larger L. mulhalli. For L. vanessa, although crabs and teleosts together constituted almost all of the diets of both size classes (more than 78.0% by mass), the relative abundance of crabs, shrimps and teleosts increased as L. vanessa grew, while consumption of anomurans greatly decreased. Size-related diet diversity and niche breadth differed between juveniles and adults of two species, but not in a consistent manner. Diet diversity of juvenile L. mulhalli (H = 2.12) was lower than that in the adults (H = 2.35), whereas niche breadth was higher in the diets of juvenile L. mulhalli (BN = 0.248) than those of large specimens (BN = 0.121). For L. vanessa, both indices were higher in the diets of larger fishes (H = 2.63, BN = 0.163) than in those of small fishes (H = 2.51, BN = 0.125). 3.3. Inter- and intra-specific diet overlap Two-way PERMANOVA revealed that the dietary compositions differed significantly between species, with no relationship to size class or interaction between these two factors (Table 2). The components of variation (COV) for species were four or five times

J.M. Park et al. / Fisheries Research 186 (2017) 301–310

305

Table 1 Percentage frequency of occurrence (%F), number (%N), mass (%M), and index of relative importance (%IRI) of prey in the diets of Lepidotrigla mulhalli and L. vanessa inhabiting northeastern Tasmanian waters, Australia. Total dietary values of prey taxa are given in boldface. Taxa

ANNELIDA Polychaeta

MOLLUSCA Gastropoda Bivalve CRUSTACEA Copepoda Amphipoda

Ostracoda Cumacea Tanaidacea Caridea

Anomura

Brachyura

Stomatopoda Euphausiacea Isopoda

Mysidacea Unidentified PISCES Teleostei

Other material

Prey organisms

Lepidotrigla mulhalli

L. vanessa

%F

%N

%M

%IRI

%F

%N

%M

%IRI

Total Ampharetidae Nephtyidae Unidentified

2.00 0.67 0.67 0.67

0.48 0.05 0.37 0.05

0.32 0.07 0.25 <0.01

0.01

0.75 0.75

0.08 0.08

0.06 0.06

<0.01

Unidentified Unidentified

3.95 7.24

0.37 0.74

0.11 0.47

0.02 0.07

1.50 2.26

0.17 0.25

0.02 0.18

<0.01 0.01

54.00 17.11 5.26 1.32

17.50 6.17 0.79 0.11

5.76 3.87 0.24 0.04

10.22

4.17 2.11 0.11 0.58 3.27 6.24 27.23 0.58 3.74

0.33 0.30 0.02 0.32 0.53 2.48 7.65 0.10 4.38

0.17 11.18 1.50 0.75 0.58 0.08 5.75 0.42

0.01 1.14 0.28 0.09 0.18 0.04 0.11 0.02

<0.01 4.79

22.37 14.47 1.32 1.97 19.74 16.00 53.95 4.61 12.67

0.75 36.09 9.77 4.51 3.01 0.75 10.53 3.01

0.67

0.05

0.13

7.33 7.33 7.24 0.66

2.93 0.75 0.74 0.05

3.36 0.89 3.35 0.08

2.63 3.95 84.21 6.58 1.32

0.26 0.42 34.04 1.74 0.16

0.58 2.69 67.96 5.56 0.15

66.45 1.32 2.63 3.95 16.45

29.66 0.11 0.32 0.37 1.69

56.29 0.53 0.19 0.34 4.90

13.53 1.50 6.77 0.75 39.85 0.75 3.76 12.78 5.26 1.50 18.05 9.02 21.80 5.26 7.52 1.50 3.76 6.02 1.50 60.90 21.05 3.76 3.76 1.50 35.34

2.09 0.25 16.10 0.08 14.85 0.08 0.75 3.59 0.92 0.25 7.92 1.33 3.84 0.58 1.33 0.25 0.67 0.83 0.17 36.86 9.84 1.42 0.58 0.92 22.02

0.41 0.03 1.37 0.02 8.27 0.52 0.68 1.26 0.34 0.24 4.52 0.70 5.34 0.80 2.70 0.22 0.50 0.86 0.26 38.72 16.41 1.40 1.68 0.92 16.64

6.58 37.50 11.18 2.63 10.53 7.24 12.50 2.63 6.58

1.64 5.75 1.48 0.26 2.11 0.79 1.11 0.58 0.53

1.16 2.21 0.90 0.07 0.79 0.23 0.22 0.10 0.47

0.15 2.43

7.52 1.50 5.26 0.75 5.26 8.27 5.26

1.25 0.25 0.58 0.08 1.50 1.83 1.00

1.06 0.17 0.45 0.21 0.31 0.35 0.22

1.50

0.67

0.05

1.50 9.77

0.17 2.42

0.08 0.33

1.97

0.16

3.70

0.06

0.66 0.66

0.05 0.05

2.94 0.64

0.66 59.21

0.05

0.12

54.14 6.02 0.75 1.50 15.04 6.02 6.02 0.75 27.82 28.57

10.34 1.08 0.08 0.17 2.84 1.58 0.83 0.08 3.67

43.66 2.07 1.70 0.56 20.67 4.44 2.39 1.28 10.55

Harpacticoida Gammaridea total Ampeliscidae Eusiridae Leucothoidae Melitidae Oedicerotidae Phoxocephalidae Gammaridea sp.A Gammaridea sp.B Unidentified Total Total Total Total Aegaeon lacazei Alpheoidea Crangonidae Palaemonidae Palaemon serenus Leptochela sydniensis Unidentified Total Galathea australiensis Munida sp. Munidopsis sp. Uropthchus sp. Galatheoidea Paguroidea Total Ebalia spp. Halicarcinides sp. Liocancinus corrugatus Majidae Megalopa sp.A Megalopa sp.B Portunidae Trapeziidae Unidentified Unidentified Total Total Serolidae Cirolanidae Sphaeromatidae Isopoda sp.A Unidentified Unidentified Unidentified Total Callionymidae Gobidae Lophonectes gallus Moridae Ophichthidae Scorpaenidae Sillago flindersi Unidentified Gravels

1.13 15.29 0.03 0.84

0.24

69.78

0.01 0.05

<0.01 1.28 0.00 9.94

2.16

49.67

<0.01 0.10 0.20

0.29

31.55

306

J.M. Park et al. / Fisheries Research 186 (2017) 301–310

Fig. 3. Size-related variation in dietary composition for a) Lepidotrigla mulhalli and b) L. vanessa by%IRI.

Table 2 Mean squares (MS), pseudo-F ratios, components of variation (COV) and significance levels (P) for a series of PERMANOVA tests, employing Bray-Curtis similarity matrices derived from the mean percentage mass contributions of the various prey taxa to the stomach contents for Lepidotrigla mulhalli and L. vanessa. Source

df

MS

Pseudo-F

COV

P

Species Size class Species × size class Residual

1 1 1 52

15604 1184 923 540

28.890 2.193 1.718

31.677 6.551 7.188 23.240

0.001 0.058 0.151

higher than those for size class and its interaction. Results of two-way crossed ANOSIM tests also showed that the dietary compositions of the two species were significantly different (R¯ = 0.880, P = 0.001), but not with size class (R¯ = 0.006, P = 0.457). As the twoway ANOSIMs detected no size-related difference for either species, one-way ANOSIM were performed on size data for each species separately, to test whether any intraspecific difference occurred. Diets between juvenile and adult groups of L. mulhalli did not differ significantly (R¯ = −0.104, p = 0.733), but was significant for the diet of L. vanessa (R¯ = 0.242, p = 0.002). SIMPER analysis indicated that dietary dissimilarity between species ranged from 50.74% to 58.27%, and that ten prey categories contributed more than 90% to the dissimilarity between the two species. The most-observed prey items that contributed to the inter-specific dissimilarity were teleosts, carids and crabs (Table 3). Diets of L. mulhalli contained consistently greater volumes of crabs, while consistently greater volumes of teleosts distinguished diets of L. vanessa from those of L. mulhalli (Table 3). The main typifying prey taxa of L. mulhalli were similar between juvenile and adult groups (i.e. crabs and cumaceans), whereas those of L. vanessa typically progressed from crabs to teleosts with increasing body size, with the former and latter prey taxa becoming prevalent in the diet of the smaller and larger individuals, respectively (Table 3). Differences in dietary composition between the two species were also illustrated by spatial nMDS (Fig. 4). As with the other analyses, nMDS ordination depicted a clear visual difference in the diet between species. Dietary samples for L. mulhalli and L. vanessa on the nMDS ordination plot displayed discrete groups of samples with the former and the latter species lying left and right side of the plot, respectively. In terms of size classes, the points for both juveniles and adults of L. mulhalli were interspersed throughout the nMDS plot. For L. vanessa, the points of juveniles lie on the upper left of the plot while those for the large size class lie at the lower left of the plot.

Table 3 Prey taxa identified by SIMPER for typifying (grey boxes) the dietary composition of each size class in each species of Lepidotrigla mulhalli and L. vanessa, and distinguishing (unshaded boxes) among the dietary compositions of four species-size groupings in the northeastern Tasmanian waters, Australia. *Indicates that the percentage contribution of a prey taxa is greater for the size class in the vertical column than in the horizontal row. Species/size class

Lepidotrigla mulhalli

L. vanessa

Lepidotrigla mulhalli

L. vanessa

Juvenile

Adult

Juvenile

Crabs Cumaceans Gammarids Teleosts Anomurans Carids Crabs*

Crabs Teleosts Anomurans Carids

Juvenile Crabs Cumaceans Euphausia Adult Cumaceans* Anomurans Juvenile Teleosts Carids Anumurans Crabs* Euphausia* Teleosts Adult Carids Crabs*

Adult

Teleosts Anomurans Teleosts Carids Carids Crabs Crabs* Gammarids Carids Cumaceans*

Fig. 4. nMDS ordination of the dietary composition constructed from Bray–Curtis similarity matrices that employed the mass contributions of diet between the two species of gurnards in northeastern Tasmanian waters, Australia (triangle, Lepidotrigla mulhalli; square, L. vanessa; open symbol, juvenile; closed symbol, large size class).

3.4. Stable isotope signatures A total of 43 white muscle tissue samples (21 for L. mulhalli and 22 for L. vanessa) were analysed. The ␦13 C values ranged

J.M. Park et al. / Fisheries Research 186 (2017) 301–310 Table 4 Result of a two-way ANOVA testing the effects of species and size class on carbon (␦13 C) and nitrogen (␦15 N) stable isotopes of Lepidotrigla mulhalli and L. vanessa. Source ␦13 C Species Size class Species × size class Error ␦15 N Species Size class Species × size class Error

df

MS

F

P

1 1 1 39

0.385 0.107 0.023 0.213

1.805 0.504 0.110

0.187 0.482 0.742

1 1 1 39

0.534 4.108 0.398 0.266

2.008 15.458 1.498

0.164 <0.001 0.228

Fig. 5. Biplot of carbon (␦13 C) and nitrogen (␦15 N) stable isotope ratios (mean ± SD) for adult and juvenile of L. mulhalli and L. vanessa.

from −19.33 to −18.19‰ (mean ± SD = −18.80 ± 0.33‰) for L. mulhalli and −19.05 to −17.61‰ (−18.44 ± 0.41‰) for L. vanessa. ␦15 N ranged from 11.83 to 14.32‰ (13.43 ± 0.57‰) for L. mulhalli and 12.81 to 14.29‰ (13.56 ± 0.52‰) for L. vanessa. A two-way ANOVA reveled no significant differences of ␦13 C values between species, size classes, and the interaction between the two (Table 4). ␦15 N values were significantly influenced by size, but no significant differences occurred between species or species × size (Table 4). Adults of both species were significantly enriched in 15 N relative to juveniles (two-way ANOVA, p < 0.001, Fig. 5), whereas ␦13 C were not different significantly between adults and juveniles of both species (two-way ANOVA, p = 0.482, Fig. 5). Regression analysis indicated no significant rate of increase in ␦13 C with standard length in both species (p > 0.05). However, significant positive relationships between ␦15 N and total length were evident in both species with individuals becoming progressively enriched in ␦15 N as fish size increased (␦15 N = 9.719 + 0.023TL, r2 = 0.690, p < 0.001 for L. mulhalli; ␦15 N = 11.071 + 0.012TL, r2 = 0.323, p = 0.006 for L. vanessa). When ␦15 N is plotted against total length separately in each size class for both species, ␦15 N of adult L. mulhalli was only significantly increased with increasing total length (p < 0.006, Fig. 6). 3.5. Trophic position Both species exhibited relatively similar mean dietary trophic positions between secondary and tertiary consumers (Fig. 7). L. mulhalli exhibited a lower average trophic position by dietary data,

307

with values of 3.45 and 3.48, while L. vanessa averaged values of 3.76 and 3.87 for juveniles and adults, respectively. In terms of mean trophic position estimated by ␦15 N values, adults of both species showed higher trophic positions than those of their juvenile counterparts, but indicated similar trophic positioning between adults of both species. The mean isotopic trophic position for adult L. mulhalli was estimated as a much higher value than that estimated by its diet, but mean trophic positions for other individuals were plotted closely to 1:1 line, suggesting that diet and trophic positioning were closely aligned (Fig. 7).

4. Discussion This study describes the dietary habits and trophic ecology of two Lepidotrigla species inhabiting northeastern Tasmanian waters. Although two species of gurnards generally consumed similar types of prey (i.e. benthic crustaceans), results of both stomach contents and stable isotope analyses provide evidence of diet and trophic niche partitioning. Such partitioning indicates a low diet overlap and low competition for food resources between the two species. Stomach content analyses showed differences in dietary composition with a higher proportion of benthic crustaceans (i.e. crabs, amphipods, cumaceans, caridean shrmips and anomurans) in the diets of L. mulhalli and higher contributions of both benthic crustaceans and teleosts for L. vanessa diets. Among prey taxa, teleosts were the key prey that contributed to a high dissimilarity between L. mulhalli and L. vanessa regardless of their size. In terms of isotope signatures, although L. vanessa showed slightly enriched carbon and nitrogen isotopes than L. mulhalli, however, no significant differences of both ␦13 C and ␦15 N were detected between the species. In support of the present study, several studies also have reported dietary difference and food resource partitioning (Platell and Potter, 1999; Terrats et al., 2000; Lopez-Lopez et al., 2011), and differences in isotope signatures among Triglidae species (Davenport and Bax, 2002). Similar diets have been recorded previously for other Triglidae fishes included Trigla lucerna and Aspitrigla obscura in the western Mediterranean (Morte et al., 1997), L. modesta and L. papilio in the south-western Australia (Platell and Potter, 1999), Aspitrigla cuculus, L. cavillone and Trigloporus lastoviza in the eastern Mediterranean (Terrats et al., 2000), C. obscurus and C. lastoviza in Tunisian waters (Boudaya et al., 2007), and Chelidonichthys spinosus and L. guentheri in Korean waters (Huh et al., 2007; Baeck et al., 2011). Triglidae fishes hunt benthic prey by exploring the substratum using six modified pectoral fin rays, which are equipped with chemosensors; a characteristic of gurnard species (Roberts, 1978; Finger, 1982). Teleosts, however, are also considered to be a fundamental feeding resource, at least for some gurnard species (Colloca et al., 1994; Moreno-Amich, 1992, 1994). Coleman and Mobley (1984) and Bulman et al. (2001) also found a higher contribution of teleosts in the diets of L. vanessa in a study of demersal fishes in a different area off the south-eastern Australian shelf. Different feeding behaviors and/or strategies may be responsible for the resource partitioning observed in this study. Diet diversity and niche breadth were considerably higher for L. vanessa, indicating that this species has a more generalist (less discriminate) feeding strategy than L. mulhalli. The greater diversity in diet may increase their competitive advantage, while less diversity exhibits more limited use of a resource. Therefore, a generalist can be influenced by a diverse prey assemblage and a specialist influenced by only a sub-set (Jiang and Morin, 2005). However, due to the absence of data on prey availability, and on the diets of both species occurring separately, the indices of diet diversity and niche breadth must be interpreted with caution (Hurlbert 1978; Feinsinger et al., 1981; Knickle and Rose, 2014).

308

J.M. Park et al. / Fisheries Research 186 (2017) 301–310

Fig. 6. Relationships between ␦15 N and total length (mm) for juvenile (open symbol) and adult (closed symbol) Lepidotrigla mulhalli (a) and L. vanessa (b).

Fig. 7. Comparison of mean trophic position (±SD) estimates of the species included in this study, calculated using dietary and d15 N methods. The diagonal line represents the 1:1 line.

Based on stomach contents and carbon isotope signatures (i.e. ␦13 C values), both gurnards fed mainly in coastal benthic environments. Marine and coastal ecosystems in this region are supported by a range of primary producers including macroalgae, benthic algae, seagrasses, and phytoplankton, as well as by the recycling of dissolved carbon (Fry and Sherr, 1984). In the present study, the ␦13 C values of L. mulhalli and L. vanessa were consistent with those of benthic fishes off southeastern Australia that consume benthic invertebrates (Bulman et al., 2001; Devenport and Bax, 2002). Stable nitrogen isotopes (i.e. ␦15 N values) of L. mulhalli and L. vanessa placed within the range of benthic secondary and/or tertiary consumers in the southeast Australian marine ecosystem (Davenport and Bax, 2002). Based on stomach content analyses, L. vanessa exhibited a higher dietary trophic position than L. mulhalli, because L. vanessa preyed more on teleosts than L. mulhalli, and fish feeders tend to have higher trophic levels than crustacean feeders (Stergiou and Karpouzi, 2002; Cresson et al., 2014). In contrast to the partitioning observed via stomach analyses, however, 15 N signatures were similar between species suggesting a lack of difference in trophic positioning (Minagawa and Wada, 1984; Post, 2002). These results suggest that despite short-term differ-

ences in dietary composition between L. mulhalli and L. vanessa, they occupy similar trophic positions within the coastal Tasmanian ecosystem, although adults of both species occupied higher trophic position than smaller conspecifics. Stable isotope and stomach content analyses can produce different results because of the different temporal aspects of the two methodologies. Stomach content analysis identifies prey that were recently consumed (i.e., on the order of hours; Hyslop, 1980), whereas stable isotopes provide a more time-integrated dietary representation on a scale of weeks to months (Buchheister and Latour, 2010). Thus inconsistencies of trophic position estimated by the two methodologies suggests that 1) the two species have consumed similar prey items for the past few months but had recently consumed different prey and/or 2) although L. vanessa preyed more on teleosts, ␦15 N values of the teleosts and crustaceans are similar. However, it is difficult to distinguish between these two explanations in the absence of isotopic values for the specific prey items. Ontogenetic diet changes are common in fish species and are usually related to maximizing energy intake (Gerking, 1994), and these changes allow larger individuals to do not directly compete with smaller conspecific individuals for food resources or habitats (Langton, 1982; Chizinski et al., 2007; Barnes et al., 2011). Generally, ontogenetic feeding behavior can be attributed to the ability of large predators to prey upon larger prey and/or to consume various types of prey. Although there was no significant difference in diet for L. mulhalli as they grew, the dietary composition significantly changed from small to large L. vanessa. Stable isotope analyses also show that ␦15 N values for adults of both species were enriched relative to juveniles of conspecifics, indicating that juveniles fed on prey with lower ␦15 N values (e.g. small crustaceans), while adult fishes preferred to feed on larger prey with higher ␦15 N values (e.g. teleosts). The results indicate that both species undergo a dietary ontogenetic shift from lower trophic levels to higher trophic levels as size increases. For L. vanessa, this could partly reflect a sizerelated change in teleosts consumption in the diets of this species, which attained a larger body size than L. mulhalli. Additionally, in terms of diversity and niche breadth, larger L. mulhalli and L. vanessa displayed higher values of diet diversity and niche breadth, suggesting that they exploited a broader range of prey. Lopez-Lopez et al. (2011) also recorded an increase in teleosts in stomach contents of Eutrigla gurnardus and Aspitrigla cuculus with increasing body size. In support of the present study, many Triglidae fishes also showed ontogenetic changes in diet (e.g. Ross, 1978; Platell and Potter, 1999; Huh et al., 2007; Baeck et al., 2011). This study gives important insights into the diet and trophic ecology of two gurnards, L. mulhalli and L. vanessa in northeastern

J.M. Park et al. / Fisheries Research 186 (2017) 301–310

Tasmanian waters, Australia. Stomach content and stable isotope analyses indicated that both species are associated with the benthic food web, feeding mainly on abundant benthic and benthopelagic species. Although the two species fed upon benthic crustaceans, teleosts also were frequently consumed by larger L. vanessa. While no ecologically relevant dietary differences were observed in ␦13 C values, those of ␦15 N yielded isotopic difference between juveniles and adults of both species and supported the relative food source. The use of diet analyses in combination with isotopic analyses has greatly increased the rigor and resolution of our results. A lack of samples collected over time, however, limits our ability to describe the absolute diets of these species. Such studies of dietary habits enhance our understanding of prey-predator and predator–predator relationships in benthic ecosystems, and act as baseline studies for future research on demersal fish assemblages in the southeastern Australian waters. Acknowledgements We thank Amanda Dehmlow, Amelia Turcato, Emma Filpi, Gemma Dixon, Jacqui Brenden, Michael Davis and Rebecca Parlett for help with dissections. This research was funded by the Department of Biological Sciences at Macquarie University and by the University of Newcastle. We are indebted to the crew of the RV Bluefin for their assistance in sample collection. Samples were collected as by-catch under the University of Tasmania Animal Ethics Permit A0011023. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.fishres.2016.10. 005. References Anderson, M.J., Gorley, R.N., Clarke, K.R., 2008. PERMANOVA for PRIMER: Guide to Software and Statistical Methods PRIMER-E. Plymouth Marine Laboratory, Plymouth, UK. Baeck, G.W., Huh, S.H., Choi, H.C., Park, J.M., 2011. Feeding habits of the redbanded searobin Lepidotrigla guentheri in the Coastal Waters off Gori, Korea. Korean J. Fish. Aquat. Sci. 44, 372–377. Barnes, L.M., Leclerc, M., Gray, C.A., Williamson, J.E., 2011. Dietary niche differentiation of five sympatric species of Platycephalidae. Environ. Biol. Fish. 90, 429–441, http://dx.doi.org/10.1007/S10641-010-9752-4. Bearhop, S., Adams, C.E., Waldron, S., Fuller, R.A., MacLeod, H., 2004. Determining trophic niche width: a novel approach using stable isotope analysis. J. Anim. Ecol. 73, 1007–1012, http://dx.doi.org/10.1111/j. 0021-8790.2004.00861.x. Boudaya, L., Neifar, L., Taktak, A., Ghorbel, M., Bouain, A., 2007. Diet of Chelidonichthys obscurus and Chelidonichthys lastoviza (Pisces: triglidae) from the gulf of gabes (Tunisia). J. Appl. Ichthyol. 23, 646–653, http://dx.doi.org/10. 1111/j. 1439-0426.2007.00861.x. Buchheister, A., Latour, R.J., 2010. Turnover and fractionation of carbon and nitrogen stable isotopes in tissues of amigratory coastal predator, summer flounder (Paralichthys dentatus). Can. J. Fish. Aquat. Sci. 67, 445–461, http://dx. doi.org/10.1139/F09-196. 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. Mar. Freshwater Res. 52, 537–548, http://dx.doi.org/10.1071/MF99152. Chizinski, C.J., Huber, C.G., Longoria, M., Pope, K.L., 2007. Intraspecific resource partitioning by an opportunistic strategist, inland silverside Menidia beryllina. J. Appl. Ichthyol. 23, 147–151, http://dx.doi.org/10.1111/j. 1439-0426.2006. 00811.x. Clarke, K.R., Gorley, R.N., 2006. PRIMER V6: User Manual/tutorial. Primer-E Ltd, Plymouth. Coleman, N., Mobley, M., 1984. Diets of commercially exploited fish from Bass Strait and adjacent Victorian waters: south-eastern Australia. Aust. J. Mar. Freshwater Res. 35, 549–560. Colloca, F., Ardizzone, G.D., Gravina, M.F., 1994. Trophic ecology of gurnards (Pisces: triglidae) in the central Mediterranean Sea. Mar. life 4, 45–57. Cortés, E., 1997. A critical review of methods of studying fish feeding based on analysis of stomach contents: application to elasmobranch fishes. Can. J. Fish. Aquat. Sci. 54, 726–738, http://dx.doi.org/10.1139/f96-316. Cresson, P., Ruitton, S., Ourgaud, M., Harmelin-Vivien, M., 2014. Contrasting perception of fish trophic level from stomach content and stable isotope

309

analyses: a Mediterranean artificial reef experience. J. Exp. Mar. Biol. Ecol. 452, 54–62, http://dx.doi.org/10.1016/j.jembe.2013.11.014. Davenport, S.R., Bax, N.J., 2002. A trophic study of a marine ecosystem off southeastern Australia using stable isotopes of carbon and nitrogen. Can. J. Fish. Aquat. Sci. 59, 514–530, http://dx.doi.org/10.1139/f02-031. Duarte, L.O., Garcia, C.B., 1999. Diet of the mutton snapper Lutjanus analis (Cuvier) from the gulf of Salamanca, Colombia, Caribbean Sea. Bull. Mar. Sci. 65, 453–465. Elmqvist, T., Folke, C., Nyström, M., Peterson, G., Bengtsson, J., Walker, B., Norberg, J., 2003. Response diversity, ecosystem change, and resilience. Front. Ecol. Environ. 1, 488–494, http://dx.doi.org/10.1890/15409295(2003)001[0488:RDECAR]2.0.CO;2. Feinsinger, P., Spears, E.E., Poole, R.W., 1981. A simple measure of niche breadth. Ecology 62, 27–32. Ferry, L.A., Cailliet, G.M., 1996. Sample size and data analysis: are we characterizing and comparing diet properly? In: MacKinlay, D., Shearer, K. (Eds.), International Congress on the Biology of Fishes. University of California, San Francisco, California, pp. 71–80. Finger, T.E., 1982. Somatotopy in the representation of the pectoral fin and free fin rays in the spinal cord of the sea robin. Biol. Bull. 163, 154–161. Froese, R., Pauly, D., 2015. FishBase. World Wide Web electronic publication (accessed 27.10.15) http://www.fishbase.org/. Fry, B., Sherr, E.B., 1984. ␦13 C measurements as indicators of carbon flow in marine and freshwater ecosystems. Contrib. Mar. Sci. 27, 13–47. Fry, B., 2006. Stable Isotope Ecology. Springer-Verlag, New York. Gerking, S.D., 1994. Feeding Ecology of Fish. Academic Press, San Diego. Gomon, D.M.F., Bray, D.J., Kuiter, R.H., 2008. Fishes of Australia’s southern coast. New Holland. Greenstreet, S.P., Rogers, S.I., 2006. Indicators of the health of the North Sea fish community: identifying reference levels for an ecosystem approach to management. ICES J. Mar. Sci. 63, 573–593, http://dx.doi.org/10.1016/j.icesjms. 2005.12.009. Huh, S.H., Park, J.M., Baeck, G.W., 2007. Feeding habits of bluefin searobin (Chelidonichthys spinosus) in the coastal waters off Busan. Korean J. Ichthyol. 19, 51–56. Hurlbert, S.H., 1978. The measurement of niche overlap and some relatives. Ecology 59, 67–77. Huveneers, C., Otway, N.M., Gibbs, S.E., Harcourt, R.G., 2007. Quantitative diet assessment of wobbegong sharks (genus orectolobus) in new south wales, Australia. ICES J. Mar. Sci. 64, 1272–1281, http://dx.doi.org/10.1093/icesjms/ fsm111. Hyslop, E.J., 1980. Stomach contents analysis—a review of methods and their application. J. Fish Biol. 17, 411–429. Jiang, L., Morin, P.J., 2005. Predator diet breadth influences the relative importance of bottom-up and top-down control of prey biomass and diversity. Am. Nat. 165, 350–363, http://dx.doi.org/10.1086/428300. Knickle, D.C., Rose, G.A., 2014. Dietary niche partitioning in sympatric gadid species in coastal Newfoundland: evidence from stomachs and CN isotopes. Environ. Biol. Fish. 97, 343–355, http://dx.doi.org/10.1007/s10641-013-0156-0. Krebs, C.J., 1989. Ecological Methodology. Harper & Row Publishers, New York, NY. Langton, R.W., 1982. Diet overlap between Atlantic cod, Gadus morhua, silver hake Merluccius bilinearis and fifteen other northwest Atlantic finfish. Fish. Bull. 80, 745–759. Lin, H.J., Kao, W.Y., Wang, Y.T., 2007. Analyses of stomach contents and stable isotopes reveal food sources of estuarine detritivorous fish in tropical/subtropical Taiwan. Estuar. Coast. Shelf Sci. 73, 527–537, http://dx.doi. org/10.1016/j.ecss.2007.02.013. Linke, T., 2011. Trophic Interactions Among Abundant Members of the Fish Fauna in a Permanently-open and a Seasonally-open Estuary in South-western Australia Doctoral Dissertation. Murdoch University, Perth. Lopez-Lopez, L., Preciado, I., Velasco, F., Olaso, I., Gutiérrez-Zabala, J.L., 2011. Resource partitioning amongst five coexisting species of gurnards (Scorpaeniforme: triglidae): Role of trophic and habitat segregation. J. Sea Res. 66, 58–68, http://dx.doi.org/10.1016/j.seares.2011.04.012. Marshall, A.D., Kyne, P.M., Bennett, M.B., 2008. Comparing the diet of two sympatric urolophid elasmobranchs (Trygonoptera testacea Müller & Henle and Urolophus kapalensis Yearsley & Last): evidence of ontogenetic shifts and possible resource partitioning. J. Fish Biol. 72, 883–898, http://dx.doi.org/10. 1111/j. 1095-8649.2007.01762.x. May, J.L., Maxwell, J.G.H., 1986. Field Guide to Trawl Fish from the Temperate Waters of Australia. CSIRO, Division of Fisheries Research. Micheli, F., Halpern, B.S., 2005. Low functional redundancy in coastal marine assemblages. Ecol. Lett. 8, 391–400. Minagawa, M., Wada, E., 1984. Stepwise enrichment of 15 N along food chains: further evidence and the relation between ␦15 N and animal age. Geochim. Cosmochim. Acta 48, 1135–1140. Moreno-Amich, R., 1992. Feeding habits of red gurnard, aspitrigla cuculus (L. 1758) (Scorpaeniformes, triglidae), along the catalan coast (northwestern mediterranean). Hydrobiologia 228, 175–184. Moreno-Amich, R., 1994. Feeding habits of grey gurnard, Eutrigla gurnardus (L., 1758) along the Catalan coast (northwestern Mediterranean). Hydrobiologia 273, 57–66. Morte, M.S., Redon, M.J., Sanz-Brau, A., 1997. Trophic relationships between two gurnards Trigla lucerna and Aspitrigla obscura from the western Mediterranean. J. Mar. Biol. Assoc. UK 77, 527–537.

310

J.M. Park et al. / Fisheries Research 186 (2017) 301–310

O’Shea, O.R., Thums, M., Van Keulen, M., Kempster, R.M., Meekan, M.G., 2013. Dietary partitioning by five sympatric species of stingray (Dasyatidae) on coral reefs. J. Fish Biol. 82, 1805–1820, http://dx.doi.org/10.1111/jfb.12104. Pauly, D., Froese, R., Sa-a, P.S., Palomares, M.L., Christensen, V., Rius, J., 2000. TrophLab Manual. ICLARM, Manila, Philippines. Pinkas, L., Oliphant, M.S., Iverson, I.L.K., 1971. Food habits of albacore, bluefin tuna and bonito in California waters. Fish. Bull. 152, 1–105. Pinnegar, J.K., Polunin, N.V.C., 1999. Differential fractionation of 13 C and 15 N among fish tissues: implications for the study of trophic interactions. Funct. Ecol. 13, 225–231, http://dx.doi.org/10.1046/j. 1365-2435.1999.00301.x. Platell, M.E., Potter, I.C., 1999. Partitioning of habitat and prey by abundant and similar-sized species of the Triglidae and Pempherididae (Teleostei) in coastal waters. Estuar. Coast. Shelf Sci. 48, 235–252, http://dx.doi.org/10.1006/ecss. 1998.0419. Platell, M.E., Potter, I.C., 2001. Partitioning of food resources amongst 18 abundant benthic carnivorous fish species in marine waters on the lower west coast of Australia. J. Exp. Mar. Biol. Ecol. 261, 31–54, http://dx.doi.org/10.1016/S00220981(01)00257-X. Post, D.M., 2002. Using stable isotopes to estimate trophic position: models, methods, and assumptions. Ecology 83, 703–718, http://dx.doi.org/10.1890/ 0012-9658. Richards, W.J., Jones, D.L., 2002. Preliminary classification of the gurnards (Triglidae: scorpaeniformes). Mar. Freshwater Res. 53, 274–282, http://dx.doi. org/10.1071/MF01128. Roberts, S.C., 1978. Biological and Fisheries Data on Northern Searobin Prionotus Carolinus (Linnaeus) (No. 13). US National Marine Fisheries Service.

Ross, S.T., 1977. Patterns of resource partitioning in searobins (Pisces:Triglidae). Copeia 1977, 561–571. Ross, S.T., 1978. Trophic ontogeny of the leopard searobin, Prionotus scitulus (Pisces:Triglidae). Fish. Bull. 76, 225–234. Ross, S.T., 1986. Resource partitioning in fish assemblages: a review of field studies. Copeia 1986, 352–388. Rowling, K., Hegarty, A.M., Ives, M., 2010. Status of Fisheries Resources in NSW 2008/09., pp. 237–239. Stergiou, K.I., Karpouzi, V.S., 2002. Feeding habits and trophic levels of Mediterranean fish. Rev. Fish Biol. Fish. 11, 217–254, http://dx.doi.org/10. 1023/A:1020556722822. Terrats, A., Petrakis, G., Papaconstantinou, C., 2000. Feeding habits of Aspitrigla cuculus (L., 1758) (red gurnard) lepidotrigla cavillone (Lac., 1802) (large scale gurnard) and trigloporus lastoviza (Brunn., 1768) (rock gurnard) around cyclades and dodecanese islands (E. mediterranean). Mediterranean Mar. Sci. 1, 91–104, http://dx.doi.org/10.12681/mms.280. Tuuli, C.D., De Mitcheson, Y.S., Liu, M., 2011. Reproductive biology of the greyfin croaker Pennahia anea in the northern South China Sea. Ichthyol. Res. 58, 302–309, http://dx.doi.org/10.1007/s10228-011-0228-0. 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. Mar. Biol. 144, 439–448, http://dx.doi.org/10.1007/ s00227-003-1218-1.