Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Variable δ15N Diet-Tissue Discrimination Factors among Sharks: Implications for Trophic Position, Diet and Food Web Models

  • Jill A. Olin,

    Current address: Department of Oceanography and Coastal Sciences, Louisiana State University, Louisiana, United States of America

    Affiliation Great Lakes Institute for Environmental Research, University of Windsor, Windsor, Ontario, Canada

  • Nigel E. Hussey ,

    Affiliation Great Lakes Institute for Environmental Research, University of Windsor, Windsor, Ontario, Canada

  • Alice Grgicak-Mannion,

    Affiliation Great Lakes Institute for Environmental Research, University of Windsor, Windsor, Ontario, Canada

  • Mark W. Fritts,

    Current address: Illinois Natural History Survey, Illinois River Biological Station, Havana, Illinois, United States of America

    Affiliation Warnell School of Forestry and Natural Resources, University of Georgia, Athens, Georgia, United States of America

  • Sabine P. Wintner,

    Affiliations KwaZulu-Natal Sharks Board, Umhlanga Rocks, South Africa, Biomedical Resource Unit, University of KwaZulu-Natal, Durban, South Africa

  • Aaron T. Fisk

    Affiliation Great Lakes Institute for Environmental Research, University of Windsor, Windsor, Ontario, Canada

Variable δ15N Diet-Tissue Discrimination Factors among Sharks: Implications for Trophic Position, Diet and Food Web Models

  • Jill A. Olin, 
  • Nigel E. Hussey, 
  • Alice Grgicak-Mannion, 
  • Mark W. Fritts, 
  • Sabine P. Wintner, 
  • Aaron T. Fisk


The application of stable isotopes to characterize the complexities of a species foraging behavior and trophic relationships is dependent on assumptions of δ15N diet-tissue discrimination factors (∆15N). As ∆15N values have been experimentally shown to vary amongst consumers, tissues and diet composition, resolving appropriate species-specific ∆15N values can be complex. Given the logistical and ethical challenges of controlled feeding experiments for determining ∆15N values for large and/or endangered species, our objective was to conduct an assessment of a range of reported ∆15N values that can hypothetically serve as surrogates for describing the predator-prey relationships of four shark species that feed on prey from different trophic levels (i.e., different mean δ15N dietary values). Overall, the most suitable species-specific ∆15N values decreased with increasing dietary-δ15N values based on stable isotope Bayesian ellipse overlap estimates of shark and the principal prey functional groups contributing to the diet determined from stomach content analyses. Thus, a single ∆15N value was not supported for this speciose group of marine predatory fishes. For example, the ∆15N value of 3.7‰ provided the highest percent overlap between prey and predator isotope ellipses for the bonnethead shark (mean diet δ15N = 9‰) whereas a ∆15N value < 2.3‰ provided the highest percent overlap between prey and predator isotope ellipses for the white shark (mean diet δ15N = 15‰). These data corroborate the previously reported inverse ∆15N-dietary δ15N relationship when both isotope ellipses of principal prey functional groups and the broader identified diet of each species were considered supporting the adoption of different ∆15N values that reflect the predators’ δ15N-dietary value. These findings are critical for refining the application of stable isotope modeling approaches as inferences regarding a species’ ecological role in their community will be influenced with consequences for conservation and management actions.


Stable isotope analyses have proven to be a powerful tool for characterizing trophic relationships between predator and prey across ecosystems. An important parameter necessary to interpret stable isotope data in ecology is the diet-tissue discrimination factor (∆15N or ∆13C)—the enrichment in the heavy isotope in predators’ tissues relative to the prey consumed due to preferential assimilation of the heavy isotope and preferential excretion of the light isotope [1,2]. Following an influential paper by Post [3], the across-taxa mean enrichment of 3.4‰ remains the most broadly applied ∆15N value [4]. Controlled feeding experiments, however, have demonstrated that ∆15N values vary among species [5,6], with meta-analyses and experimental work empirically showing an inverse relationship between ∆15N and dietary δ15N values [7,8]. These data suggest that taxonomic classes containing a range of species that feed on diverse prey resources will exhibit intra-class variation in ∆15N, questioning the suitability of a single ∆15N value for characterizing predator-prey relationships [7,9].

Precise methods for characterizing the ecological role of organisms are imperative to the management of aquatic ecosystems [10]. This is especially relevant for marine predators that have undergone global declines in recent decades, as many species occupy a relatively high trophic position, and are therefore assumed to play an important role in structuring marine communities [1113]. The subclass Elasmobranchii (sharks, skates and rays) comprises ~450 species that inhabit a diverse range of habitats, from the deep ocean to exposed coral shelves. Species within this subclass range in size, from the dwarf lantern shark Etmopterus perryi (~17 cm) to the whale shark Rhincodon typus (~12 m) and feed on a broad prey base, from zooplankton to marine mammals. This considerable diversity together with the reported inverse ∆15N-dietary δ15N relationship, would suggest that variable diets among species within this subclass will result in variable mean dietary δ15N values (e.g., crustacean dominated diet vs. elasmobranch dominated diet) that would ultimately drive variable ∆15N values (Figure 1). To date, two experimentally derived ∆15N values are available for elasmobranchs; a ∆15N of 2.3‰ for large sharks fed a fish diet [14] and a ∆15N of 3.7‰ for a small shark fed a diet of squid [15]. These shark-specific ∆15N values are markedly different and when applied to dietary and/or food web analyses would result in vastly different inferences dependent on the ∆15N selected. This highlights the importance of considering the diet (including protein quality and quantity) of the consumer species examined to ensure the most accurate interpretation of stable isotope data for these purposes.

Figure 1. The predicted relationship between nitrogen stable isotope discrimination between predator and prey consumed (∆15N) and the prey stable nitrogen isotope composition (dietary-δ15N) estimates for each shark species based on the widely reported ∆15N-dietary δ15N relationship


Our objective was to conduct an assessment of a range of reported ∆15N values for describing the predator-prey relationships of four shark species that feed at different trophic levels (i.e., different mean δ15N dietary values) and in so doing, determine the most applicable surrogate ∆15N value for describing predator-prey relationships within this speciose group. Given the logistical and ethical complexities of controlled feeding experiments on large predators, field sampling to derive consumer and prey isotope values in conjunction with novel ellipse estimates were used to assess appropriate ∆15N values for four species that have well characterized diets from stomach content analysis; bonnethead (Sphyrna tiburo), Atlantic sharpnose (Rhizoprionodon terraenovae), bull (Carcharhinus leucas) and white (Carcharodon carcharias) shark. The broad range of predator-prey relationships represented by the trophic assessment of these four species presents the opportunity to assess whether species-specific diets that range from a low (i.e., bonnethead preying on crustaceans) to a high (i.e., white preying on elasmobranchs and marine mammals) δ15N value affect the choice of which ∆15N value to use.

Materials and Methods

Ethics Statement

All large sharks, rays and dolphins sampled in South Africa were found deceased on capture in beach protection nets and no animals were sacrificed for this study. All live net-caught sharks and rays were tagged and released according to the KwaZulu-Natal (KZN) Sharks Board tag and release protocols. Cape fur seals were sampled from animal strandings and regional culls undertaken by Marine and Coastal Management (MCM) at the Department of Environmental Affairs and Tourism, South Africa. Permission to use white shark and cape fur seal muscle tissue samples was granted by the KZN Sharks Board and MCM, respectively and samples were exported in strict accordance with the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) requirements (CITES South African Permit No. 106704 and 106627). All sharks, fish and invertebrates sampled in Georgia and additional sharks, dolphins, fish and invertebrates from South Africa are non-threatened species and are not listed on CITES and therefore export permits were not required. Sampling was undertaken under the authority of the Marine Fisheries Section, Coastal Resources Division, Department of Natural resources, Georgia, MCM, Cape Town, South Africa and the Oceanographic Research Institute, Durban, South Africa. All vertebrate work was conducted in accordance with the University of Windsor’s Animal Use and Care Guidelines (AAUP #07-13) and respective policies of the Department of Natural Resources, Georgia, the KZN Sharks Board, MCM, South Africa and the Oceanographic Research Institute, South Africa.

Sample collection

Predator (four focal shark species) and prey species (e.g., invertebrates, fishes, mammals) with the exception of cape fur seals (see below) were sampled from two study sites, the estuaries of coastal Georgia, USA and from the continental shelf of KwaZulu-Natal (KZN), South Africa. All sampling of predators and prey was conducted over the same time frames. In Georgia, invertebrates and teleosts were sampled via otter trawl during scientific research cruises, and bonnethead and Atlantic sharpnose sharks were sampled from incidental captures from shallow water bottom-set longlines (see 16). In KZN invertebrates and teleosts were sampled from three main sources: (i) by-catch in the shallow water prawn trawl fishery, (ii) organized spear fishermen catches and (iii) organized recreational fishermen/scientific catches. All elasmobranch species including bull and white sharks and delphind species were sampled from incidental captures in beach protection nets along the KZN coast [17]; cape fur seals were sampled from animal strandings and regional culls at colonies around the Eastern and Western Cape where white sharks are known to seasonally reside [1820]. Immediately following capture, a muscle tissue sample was excised from prey species and stored frozen (-20°C) as follows: from the claws and tails of invertebrates, from the dorsal section anterior to the dorsal fin in teleosts and sharks, from the mid-wing section of batoids, and from the mid-dorsal region of seals and delphinids. For the four focal shark species, all individuals were measured (TL–total or PCL–precaudal length) and white muscle tissue was excised anterior to the first dorsal fin and stored frozen (-20°C).

Stomach Content and Stable Isotope Data

To characterize the diet of the four focal species, we used a combination of published literature and stomach contents sampled from animals included in this study. For bonnethead and Atlantic sharpnose sharks, stomach content data were quantified following standard methods (see 21,22, M. Fritts unpublished data). For bull and white sharks, quantified stomach content data from South African animals were taken from Cliff and Dudley [23] and Hussey et al. [24], respectively. The importance of (i) functional prey groups (e.g., crustacean, mollusk, teleost, elasmobranch, mammal) and (ii) individual prey items (identified to species level where possible) to the diet of each shark was assessed using four commonly applied dietary indices: (i) %W, the weight contribution of a prey item, (ii) %F, the frequency of occurrence of each prey item, (iii) %N, the numbers of each prey item and (iv) %IRI, the relative importance of the prey item. Specifically, %W was calculated as the weight of each prey item divided by the total weight of prey items from an individual stomach, %F was calculated as the number of stomachs containing a prey item divided by the total number of stomachs containing prey and %N was calculated as the number of each prey item divided by the total number of prey items. The index of relative importance is a compound index that incorporates the previous three indices, expressed as IRI = %F (%N + %W). This product is then expressed as a percentage (%IRI) by dividing the total IRI for each prey item by the total IRI for all prey items [22]. From the complete stomach content data (for complete diet analysis see Cliff and Dudley [23]; Hussey et al. [24]; M. Fritts unpublished data), the five most commonly occurring prey items in the diet of each shark [referenced as principal prey (PP) hereafter] were determined based on the highest percent contribution by the combination of %W, %F, %N and %IRI for each prey item (Table 1).

Sphyrna tiburo Rhizoprionodon terraenovaeCarcharhinus leucasCarcharodon carcharias
TL = 39–47 cmTL = 34–49 cmPCL = 145–208 cmPCL = 185–234.9 cm
Common dolphin (Delphinus delphis)
Seal (Pinnipedia)
Guitarfish (Rhinobatidae)
Dusky/Sharpnose shark (Carcharhinidae)6.22.383.3a
Bull/Blue/Honeycomb stingray (Dasyatidae)
Dusky shark(Carcharhinus obscurus)10.92.618.120.9a
Milk shark(Rhizoprionodon acutus007A)
Spotted eagle ray (Aetobatus narinari)
Spot (Leiostomus xanthurus)2.31.315.31.1a
Star drum (Stellifer lanceolatus)2.31.313.71b
Spotted grunter (Pomadasys commersonnii )
Blue crab (Callinectes sp.)22.511.518.212.2a
Lady crab (Ovalipes ocellatus)7.54.613.72.5b
Hermit crab (Pagurus sp.)104.651.7d6.85.372.4d
Shrimp (Penaeidae)156.910.44.7c18.210.715.313.2c
Mantis shrimp (Stomatopoda)104.66.32e
Squid sp. (Teuthoidea)2.31.311.30.8e

Table 1. Summarized functional prey groups and the principal prey (PP) of bonnethead Sphyrna tiburo, Atlantic sharpnose Rhizoprionodon terraenovae, bull Carcharhinus leucas, and white Carcharodon carcharias shark identified from stomach content data (see 9,23 M. Fritts unpublished data).

Data are percent frequency of occurrence (%F), number (%N), weight (%W) and the index of relative importance (%IRI). TL represents total length; PCL represents precaudal length. Subscripts represent PP species in Figure 4.
Note: Principal prey (PP) do not contribute 100% to the total of each functional prey group. A component of the diet of each of each shark species consists of unidentified species (see 9,23).
Download CSV

Stable carbon and nitrogen isotopes were analyzed in muscle tissue of the four predators and the determined PP species. Additional prey species identified in the diet of each species, where samples were available from respective study regions and time periods, were also analyzed. To ensure no bias of the maternal isotope signature, only bonnethead (n = 11; 39–47 cm TL) and Atlantic sharpnose (n = 6; 34–49 cm TL) sharks that were characterized as juveniles, yet older than umbilical scar stage 5 were used [19]. To avoid complications of ontogenetic diet shifts, defined size classes of bull (n = 11) and white (n = 16) sharks were selected; sub-adults/adults ranging in length from 145–208 cm PCL and sub-adults ranging in length from 185–235 cm PCL, respectively. For complete stable isotope sample preparation, lab analyses and analytical precision, see Olin et al. [16] and Hussey et al. [25]. Briefly, muscle tissues were sub-sampled from all shark and prey species (~1.0 g), freeze-dried for 48 h, ground and lipid extracted by twice agitating the ground tissue in 2:1 chloroform/methanol solution for 24 h and decanting the solvent (modified method outlined in [26]). The relative abundances of carbon (13C/12C) and nitrogen (15N/14N) were determined on ~0.5–1.0 mg sub-samples on a Thermo Finnigan DeltaPlus mass spectrometer (Thermo Finnigan, San Jose, CA, USA) coupled with an elemental analyzer (Costech, Valencia, CA, USA). Analytical accuracy was 0.14‰ for δ15N data and 0.05‰ for δ13C data based on a single run of NIST standard sucrose (NIST 8542; n = 13) and ammonium sulfate (NIST 8547; n = 13).

To examine whether ∆15N values varied among species and were dependent on the diet-δ15N value for a particular shark species, a range of ∆15N values (and associated ∆13C) were selected from the published literature (Table 2). The δ15N and δ13C data for each shark species were then adjusted using each of the four chosen ∆15N and ∆13C pairs from Table 2, by subtracting the specific ∆15N and ∆13C value from each individual predator’s δ15N and δ13C value, respectively. For each shark species, four δ15N vs. δ13C bi-plots were then constructed for each ∆15N- and ∆13C-adjusted shark isotope data and overlaid over the raw prey stable isotope data. Raw shark isotope data were also plotted over raw prey isotope data. This enabled a visual assessment of the ∆15N and ∆13C-adjusted shark isotope values, i.e., it would be expected that the best fit ∆15N and ∆13C-adjusted shark isotope values would fall within the range of the PP it consumes (Figure 2), following Philips and Gregg [27] and Fry [28].

Diet-Tissue Discrimination Factor ± SD (‰)
0.9 ± 0.32.3 ± 0.2[14]1
0.8 ± 0.12.8 ± 0.1[7]2
0.4 ± 1.33.4 ± 1.0[3]2
1.7 ± 0.53.7 ± 0.4[15]1

Table 2. Diet-tissue discrimination factor (∆13C and ∆15N are the changes in isotope ratios (δ13C and δ15N) between predator and prey, calculated as δbXpredator - δbXprey = ∆bXpredator-prey) derived from controlled feeding experiments specific to elasmobranchs1 and those commonly accepted from the literature2.

Download CSV
Figure 2. Illustration of the expected relationship between stable isotope values of a predator and its’ prey in mixing space [27,28], employing the Bayesian approach of Jackson et al. [29], centered on multivariate ellipse based metrics.

In choosing discrimination factor (∆15N and ∆13C) values, it would be expected that the δ15N and δ13C values of the predator after adjustment to specific ∆15N and ∆13C values should overlay or fall within the range of δ15N values of the PP it consumes (see Table 2), indicating a best-fit scenario between predator and prey [ellipses represent prey (black) and predator (gray) respectively]. Black points represent δ13C and δ15N values of a predator, gray points (light and dark) represent adjusted-δ13C and adjusted-δ15N values with two different ∆15N and ∆13C values. White shapes represent mean (± variance) of prey species.

Data Analysis

To determine whether the δ15N value of the diet was different among the four predators, an analysis of variance (ANOVA) followed by a Tukey’s HSD post-hoc analysis was performed on the δ15N data for the five-combined PP of each shark. To then quantify the best-fit ∆15N for each shark species, standard ellipses around the isotopic values of PP and ∆15N and ∆13C-adjusted predator combination were created using a Bayesian approach centered on multivariate ellipse-based metrics (SIBER–Stable Isotope Bayesian Ellipses; [29]). To reduce the influence of isotopic variability associated with variable sample sizes, we grouped PP into functional prey groups according to Table 1. This grouping then allowed for qualitative comparisons among predator and prey isotope and stomach content contributions. The adopted SIBER metric analyses were adapted from community–level metrics developed originally by Layman et al. [30] based on the mean δ13C and δ15N of all species in a community. Briefly, the standard ellipse represents a set of bivariate data calculated from the variance and covariance of the x and y data. The ellipse contains approximately 40% of the data and therefore represents the core niche or dietary isotopic space [31]. This approach generates standard ellipse areas (SEA), which are designed to identify differences in isotopic space between characteristic members of a population and account for populations with different sample sizes [29]. Following Jackson et al. [29], SEA was graphically expressed using a corrected SEAC measurement. The area of overlap between predator and functional prey group ellipses was then calculated using the Stable Isotope in R package (SIAR; [32]) in R 2.13.0 [33]. Our expectation was that the greatest overlap between predator and PP ellipses best characterizes the predator-prey relationship and as a result, the most suitable ∆15N for each species. In addition, the overlap of predator ellipse with functional prey group ellipses would rank in terms of importance to diet from stomach content data, providing further confidence in the selection of the most suitable ∆15N value. The above analyses were repeated, using isotopic values from a broader range of prey species identified in the diet of each shark, i.e., those species identified from stomach contents with available isotopic values (number of additional prey species: bonnethead = 11; Atlantic sharpnose = 15; bull = 41; and white = 25). This enabled a comparison of ellipse overlap between predator vs. functional prey groups (based on PP) and predator vs. broader diet for each species to illustrate the importance of considering the number of dietary items and their relative importance to diet for confidence in this approach.


In terms of %W, %F, %N and %IRI crustaceans were the principal diet component of the bonnethead, teleosts for the Atlantic sharpnose, elasmobranchs and teleosts for the bull shark, and elasmobranch and marine mammals for the white shark (Table 1). The mean diet δ15N value for each shark species based on PP ranged from low in the bonnethead shark (8.8 ± 0.9‰; Figure 3), to high in the white shark (15.1 ± 1.7‰; Figure 3), and differed significantly among the four species (F3,322 = 177.4, P < 0.0001; Figure 3).

Figure 3. Box plots representing the δ15N values of all of the PP derived from stomach content data of the bonnethead Sphyrna tiburo, Atlantic sharpnose Rhizoprionodon terraenovae, bull Carcharhinus leucas, and white Carcharodon carcharias shark.

The bonnethead shark exhibited the most enriched raw δ15N values relative to PP, while the white shark exhibited the least enriched raw δ15N values relative to PP (Figure 4a, f, k, p). For the bonnethead shark that had the lowest δ15N diet value, the largest ∆15N of 3.7‰ produced both the greatest visual and ellipse overlap between predator and the functional prey group identified from stomach contents (Figure 4e; Table 3). The most suitable ∆15N for the Atlantic sharpnose based on ellipse overlap was lower than the bonnethead, a ∆15N value of 3.4‰ (Figure 4i; Table 3), while 2.3‰ provided the highest measure of overlap for the bull shark (Figure 4l; Table 3). For the white shark, a ∆15N of 2.3‰ produced the greatest area of overlap between predator and functional prey group ellipses but observation of the data suggested that the ∆15N for this highly carnivorous species is likely lower (Figure 4p, q; Table 3). The greatest area of overlap between ∆15N-adjusted predator values and functional prey group isotope values for all four shark species generally correspond with what we expected based on the contributions estimated from stomach content analyses. For example, teleosts followed by crustaceans and mollusks contributed the highest combined percentage to the diet of the Atlantic sharpnose (%W = 58.5, 28.2 and 11.3, respectively; Table 1), which corresponded to the calculated ellipse area overlap between the Atlantic sharpnose and the functional prey groups using the ∆15N value of 3.4‰ (Table 3). These results were further supported when including stable isotope values for additional prey items identified from the broader diet with the exception of the bonnethead shark (Figure 4b, e; Table 3).

Figure 4. Dual-plot of individual predator (■) and mean (± SD) δ13C and δ15N values of the PP for each predator ((a), (f), (k), (p); see Table 1).

Standard ellipse areas corrected for sample size (SEAc) of sharks (solid black) and PP functional prey groups (Crustacean, dashed light gray; Mollusk, dotted light gray; Teleost, dashed dark gray; Elasmobranch, solid dark gray; Mammal solid light gray), and the broader diet (dotted black) following Jackson et al. [29]. Note different scales on the x- and y-axes in each species.

15N ValuesPredator (isotopic area)Functional Prey Group (isotopic area)Overlap Predator-Functional Prey Group (area)
Sphyrna tiburo
2.3 ± 0.20.301.662.290.000.16
2.8 ± 0.10.301.662.290.000.03
3.4 ± 1.00.301.662.290.000.00
3.7 ± 0.40.301.662.290.270.01
Rhizoprionodon terraenovae
2.3 ± 0.20.901.
2.8 ± 0.10.901.
3.4 ± 1.00.901.
3.7 ± 0.40.901.
Carcharhinus leucas
2.3 ± 0.20.841.880.960.451.890.
2.8 ± 0.10.841.880.960.451.890.002.06-181.14-182.39-18
3.4 ± 1.00.841.880.960.451.890.007.40-181.12-171.40-17
3.7 ± 0.40.841.880.960.451.896.70-
Carcharodon carcharias
2.3 ± 0.20.430.691.572.860.340.000.40
2.8 ± 0.10.430.691.572.860.060.000.05
3.4 ± 1.00.430.691.572.861.61-160.005.20-18
3.7 ± 0.40.430.691.572.860.000.001.09-18

Table 3. Calculated overlap of SIBER ellipses between (i) isotopic values of the PP functional prey groups (see Table 1) and (ii) the isotopic values of the broader diet of each shark derived from stomach content analyses, and the ∆15N- and ∆13C-adjusted isotopic values (see Table 2) for each shark (see Figure 4) following Jackson et al. [29].

The extent of overlap ranges from 0 to 1, with values closer to 1 representing more overlap. Bold values highlight the greatest overlap between predator and designated prey group.
Note: Functional prey groups are noted as follows: Crust = Crustacean; Moll = Mollusk; Elasmo = Elasmobranch; Mamm = Mammals. The broader diet (All) consists of prey species identified from stomach contents with available isotopic values from the same sampling location (# of species included: bonnethead = 11; Atlantic sharpnose = 15; bull = 41; and white = 25).
Download CSV


Selection of appropriate ∆15N values for interpreting stable isotope data remains a controversial point (see 7,14,34,35). The implication of using inappropriate ∆15N values when calculating trophic position or undertaking mixing model analyses will result in inaccurate inferences regarding diet reconstruction and more critically, the role of an organism within aquatic food webs [36]. Species-specific ∆15N values have been recognized [37,38] and are particularly important given the advanced isotope modeling approaches currently being adopted to examine complex feeding behavior at both the population and individual level [30,39]. These data find that the greatest area of ellipse overlap between predator and prey was a result of selecting different ∆15N values that were dependent on the diet δ15N value, indicating that ∆15N values for species within the subclass Elasmobranchii vary greatly. Although considerable progress in stable isotope controlled feeding studies has been made [5,6,8], the ∆15N-diet δ15N relationship has largely been ignored based on the fact that the mechanism driving the relationship remains unknown. Despite this, our results further corroborate this relationship and future experimental analyses should focus on identifying the mechanism driving this relationship. Whether the relationship is driven by protein quantity and quality [36,40] or the variable kinematics of 14N and 15N routing and incorporation in animal tissues [37] remains unclear, but questions the use of a single ∆15N value for this subclass.

Variation in ∆15N values dependent on the consumers’ dietary-δ15N value within this subclass of species would indicate this result is ubiquitous across a broad range of taxa in both aquatic and terrestrial ecosystems. These results were consistent across shark species when comparing both isotope data of functional prey groups and the broader identified diet, the latter with the exception of the bonnethead shark. This inconsistency in ∆15N values of the bonnethead resulted from the inclusion of teleost prey species with higher δ15N values (i.e., anchovies) in the broader diet, prey species that are minor contributors to overall diet. The bonnethead shark is known to primarily feed on crustaceans with minimal contribution from teleost prey (see 41 and references therein). Our overall results were also consistent with other studies that have reported greater isotopic discrimination between predator and prey when a low δ15N diet is assimilated [7,40]. Indeed, Kim et al. [42], observed a higher ∆15N value in a controlled laboratory study on leopard sharks (Triakis semifasciata) following a diet switch from squid (~3.7‰; δ15N of diet = ~13.3‰) to tilapia (~5.5‰; δ15N of diet = ~7.9 ‰). Similarly, Blanchet-Aurigny et al. [43] observed higher ∆15N values for brittle stars Ophiocomina nigra and Ophiothrix fragilis fed a diet of mussels and/or macroalgae (2-4‰) compared to a diet of fish (~-1‰). This demonstrates that ∆15N values can vary within a species dependent on the dietary-δ15N value the animal consumes, in agreement with previous work [6,44]. If ∆15N values vary within a species, dependent on the dietary-δ15N value, it would be expected that ∆15N values will vary among species within a subclass that have diverse diets as our data indicate.

Compound specific stable isotope analysis of amino acids (AA-CSIA) has proven to be a complimentary method for estimating ∆15N values within organisms (trophic enrichment factor —TEF). Initial experimental work suggested a TEF value of 7.6‰ was universal across species and taxa [45], similar to the widely applied value of 3.4‰ used in bulk stable isotope analysis [3]. Recently published data on marine mammals fed high δ15N diets exhibited TEFs that were lower than previously observed (4.3 vs. 7.6‰) [46]. These AA-CSIA data are consistent with the relationship observed in bulk stable isotope analysis [7,8,14] and suggest that TEF estimates may vary dependent on the diet δ15N value. Knowledge of diet ascertained through stomach content data remains critical for guiding isotopic assumptions on ∆15N values and thereby improving the efficiency of yielded results from modeling exercises, for example mixing models [47,48].

Estimating trophic position and the application of mixing models for dietary reconstruction are particularly vulnerable to the basic assumption of which ∆15N value to use [6,49], with consequences not only for ecological investigations but for management decisions [49]. This point is pertinent given the promoted use of stable isotopes to investigate threatened and endangered species [49]. Specific to the white shark, stable isotopes have recently been used to derive data on feeding behavior, movement and their role in food webs, overcoming the logistical difficulties and conservation concerns of studying these large predators. For example, Carlisle et al. [39] developed a novel isotope mixing model integrating both movement data and tissue turnover rates to estimate the relative importance of different focal habitat areas and associated prey to white shark diet, and to elucidate migratory behavior. Whereas Kim et al. [50] generated ontogenetic isotopic profiles of individual white sharks using serial sampling of vertebrae. In addition, a number of studies have presented trophic position estimates for a range of elasmobranch species based on stable isotopes (e.g., [5153]). The integrity of data interpretation in all the above studies and the broader isotopic literature, which may ultimately influence management decisions such as conservation status listing (i.e., CITES) and the designation of critical habitat and/or marine protected areas for predator and/or prey, is dependent on assumptions over ∆15N values. This is of particular importance given the recognized use of food web models by resource managers that may incorporate parameter estimates from stable isotope analyses (e.g., [54]). Future modeling exercises focused on the trophic ecology of these large marine predators using isotopic data should consider evaluating the sensitivity of predictions generated from models to different fractionation assumptions (e.g., [55]).

It is possible that several factors may have biased our observed ∆15N-diet δ15N results. These include (i) the use of stomach content data for ranking the importance of prey species, (ii) movement of predators between isotopically distinct food webs, (iii) ontogenetic diet shifts, and, (iv) temporal variability in prey and baseline isotopic values. Stomach content data is often criticized for only providing a ‘snapshot’ of a species diet [56], biases largely resulting from varying rates of prey digestion, empty stomachs and limited sampling in spatial and temporal scope. However stomach content analysis remains one of the most important measures of an animal’s feeding ecology providing data on the actual species consumed [57] and when coupled with stable isotopes analysis have resulted in significant ecological insights, for example, reef community connectivity [55], sized-based interactions [52] and niche overlap [58] that would have been difficult to infer using a single approach. The high degree of isotopic ellipse overlap between predators and prey, ranked by importance from stomach content data, would indicate our dual approach was valid. Sharks are known to be mobile predators that can feed over variable spatial and temporal domains [39]. For this reason we targeted specific size classes of our focal species to avoid confounding ontogenetic effects. For example, juvenile bonnethead and Atlantic sharpnose sharks have relatively small home-ranges, largely inhabiting coastal bays and estuaries [59,60] and sub-adult/adult bull sharks show high site fidelity [61], with only occasional large scale movements in coastal habitats [62,63]. In the case of sub-adult white sharks used in this study, individuals make directed seasonal movements between two distinct regions, the Western Cape and KZN. Considering this point, prey species were sampled in KZN and cape fur seals, the principal diet item from the Cape region were sampled, accounting for potential spatial bias in prey isotope values. Moreover, isotopic variability across space and time might be expected to enhance variation within functional prey groups (i.e., larger ellipse areas) and therefore lead to greater overlap between predator and prey. This was not observed in our analyses. Finally, integration of prey isotopes values in shark muscle tissue is on the order of <12 months (e.g., [34,64]). Consequently shark muscle tissue isotope values represent foraging over multiple seasons and locations, thereby minimizing the influence of these variables on the results of our analysis.

In conclusion, variable ∆15N values provided a more accurate representation of the predator-prey isotopic relationship for sharks rather than the application of a single ∆15N for species that feed across different trophic levels or dietary δ15N. Controlled feeding experiments to derive experimental ∆15N values for large predators are challenging due to the ethical issues of feeding animals a single diet item and therefore require in the vast majority of cases, researchers to compromise and apply a ∆15N value based on studies of either ecologically or taxonomically similar species [49]. Moreover, most organisms such as sharks feed on multiple prey items. Consequently experimental derived values based on a single diet item could be inaccurate or misleading. Given the strength of empirical work to date, we suggest that evaluation of a range of ∆15N values is a necessary consideration for future trophic and diet reconstruction analyses and is essential for improving the understanding of a species’ ecological role in its’ environment and to enhance associated conservation and management actions. With the recent advancements in mathematical modeling in Bayesian frameworks [65,66], whereby the use of stable isotopes to examine predator-prey relationships has become more precise and approaches more refined, the novel application of SIBER that incorporates predator and prey isotopic variability combined with stomach content data provided a greater level of confidence, in the context of this analysis, in choosing an appropriate ∆15N value.


Thanks to Carolyn Belcher and KwaZulu-Natal Sharks Board laboratory staff for their assistance with sample collection and to Sandra Ellis, Anna Hussey and Tom Maddox for stable isotope preparation and analysis.

Author Contributions

Conceived and designed the experiments: JAO NEH ATF. Performed the experiments: JAO NEH MWF SPW. Analyzed the data: JAO NEH. Contributed reagents/materials/analysis tools: AGM MWF SPW ATF. Wrote the manuscript: JAO NEH ATF.


  1. 1. DeNiro MJ, Epstein S (1978) Influence of diet on the distribution of carbon isotopes in animals. Geochim Cosmochim Acta 42: 495–506. doi:10.1016/0016-7037(78)90199-0.
  2. 2. DeNiro MJ, Epstein S (1981) Influence of diet on the distribution of nitrogen isotopes in animals. Geochim Cosmochim Acta 45: 343–351.
  3. 3. Post DM (2002) Using stable isotopes to estimate trophic position: models, methods and assumptions. Ecology 83: 703–718. doi:10.1890/0012-9658(2002)083[0703:USITET]2.0.CO;2.
  4. 4. Madigan DJ, Litvin SY, Popp BN, Carlisle AB, Farwell CJ et al. (2012) Tissue turnover rates and isotopic trophic discrimination factors in the endothermic teleost, pacific bluefin tuna (Thunnus orientalis). PLOS ONE 7(11): e49220. doi:10.1371/journal.pone.0049220. PubMed: 23145128.
  5. 5. Robbins CT, Felicetti LA, Sponheimer M (2005) The effect of dietary protein quality on nitrogen isotope discrimination in mammals and birds. Oecologia 144: 534–540. doi:10.1007/s00442-005-0021-8. PubMed: 15800751.
  6. 6. Lecomte N, Ahlstrøm Ø, Ehrich D, Fuglei E, Ims RA et al. (2011) Intrapopulation variability shaping isotope discrimination and turnover: experimental evidence in arctic foxes. PLOS ONE 6: e21357. doi:10.1371/journal.pone.0021357. PubMed: 21731715.
  7. 7. Caut S, Angulo E, Courchamp F (2009) Variation in discrimination factors (Δ15N and Δ13C): the effect of diet isotopic values and applications for diet reconstruction. J Appl Ecol 46: 443–453. doi:10.1111/j.1365-2664.2009.01620.x.
  8. 8. Dennis C, MacNeil MA, Rosati J, Pitcher T, Fisk AT (2010) Diet tissue discrimination factors are inversely related to δ15N and δ13C values of food for fish under controlled conditions. Rap Commun. J Mass Spectrom 24: 3515–3520.
  9. 9. Hussey NE, MacNeil MA, Fisk AT (2010a) The requirement for accurate diet-tissue discrimination factors for interpreting stable isotopes in sharks. Hydrobiologia 654: 1–5. doi:10.1007/s10750-010-0361-1.
  10. 10. Branch TA, Watson R, Fulton EA, Jennings S, McGilliard CR et al. (2010) The trophic fingerprint of marine fisheries. Nature 468: 431–435. doi:10.1038/nature09528. PubMed: 21085178.
  11. 11. Myers RA, Baum JK, Shepherd TD, Powers SP, Peterson CH (2007) Cascading effects of the loss of apex predatory sharks from a coastal ocean. Science 315: 1846–1850. doi:10.1126/science.1138657. PubMed: 17395829.
  12. 12. Heithaus MR, Frid A, Wirsing AJ, Worm B (2008) Predicting ecological consequences of marine top predator declines. Trends Ecol Evol 4: 202–210.
  13. 13. Worm B, Davis B, Kettemer L, Ward-Paige CA, Chapman D et al. (2013) Global catches, exploitation rates, and rebuilding options for sharks. Mar Policy 40: 194-204. doi:10.1016/j.marpol.2012.12.034.
  14. 14. Hussey NE, Brush J, McCarthy ID, Fisk AT (2010b) δ15N and δ13C diet–tissue discrimination factors for large sharks under semi-controlled conditions. Comp Biochem Physiol A 155: 445–453. doi:10.1016/j.cbpa.2009.09.023.
  15. 15. Kim SL, Casper D, Galván-Magaña F, Ochoa-Díaz R, Hernández-Aguilar S et al. (2011) Carbon and nitrogen discrimination factors for elasmobranch soft tissues based on a long- term controlled feeding study. Environ Biol. Fish 95: 37–52.
  16. 16. Olin JA, Hussey NE, Fritts MW, Heupel MR, Simpfendorfer CA et al. (2011) Maternal meddling in neonatal sharks: implications for interpreting stable isotopes in young animals. Rap Commun. J Mass Spectrom 25: 1008–1016.
  17. 17. Dudley SFJ, Cliff G, Zungu MP, Smale MJ (2005) Sharks caught in protective gill nets off KwaZulu-Natal, South Africa. 10. The dusky shark, Carcharhinus obscurus (Lesueur 1818). SA J Mar Sci 27: 107–127. doi:10.2989/18142320509504072.
  18. 18. Martin RA, Hammerschlag N, Collier RS, Fallows C (2005) Predatory behaviour of white sharks (Carcharodon carcharias) at Seal Island, South Africa. J Mar Biol Assoc UK 85: 1121–1135. doi:10.1017/S002531540501218X.
  19. 19. Cliff G, Dudley SFJ, Davis B (1989) Sharks caught in the protective gill nets off Natal, South Africa. 2. The great white shark Carcharodon carcharias Linnaeus. SA J Mar Sci 8: 131–144. doi:10.1287/mksc.8.2.131.
  20. 20. Cliff G, Dudley SFJ, Fry MR (1996) Catches of white sharks in KwaZulu-Natal, South Africa and environmental influences. In: AP KlimleyDG Ainley; White Great Sharks: The Biology of Carcharodon carcharias. . San Diego, California: Academic Press . Pp. 351–362.
  21. 21. Hyslop EJ (1980) Stomach content analysis: a review of methods and their application. J Fish Biol 17: 411–429. doi:10.1111/j.1095-8649.1980.tb02775.x.
  22. 22. Cortés E (1999) Standardized diet compositions and trophic levels of sharks. ICES J Mar Sci 56: 707–717. doi:10.1006/jmsc.1999.0489.
  23. 23. Cliff G, Dudley SFJ (1991) Sharks caught in the protective gill nets off Natal, South Africa. 4. The bull shark Carcharhinus leucas Valenciennes. SA. J Mar Sci 1: 253–270.
  24. 24. Hussey NE, McCann H, Cliff G, Dudley SFJ, Wintner SP et al. (2012a) Size-based analysis of diet and trophic position of the white shark Carcharodon carcharias in South African waters. In: ML Domeier. Global Perspectives on the Biology and Life History of the White Shark. Boca Raton, FL: CRC Press. pp. 27–48.
  25. 25. Hussey NE, Olin JA, Kinney MJ, McMeans BC, Fisk AT (2012b) Lipid extraction effects on stable isotope values (δ13C and δ15N) of elasmobranch muscle tissue. J Exp Mar Biol Ecol 434–435: 7–15.
  26. 26. Bligh EG, Dyer WJ (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37: 911–917. doi:10.1139/o59-099. PubMed: 13671378.
  27. 27. Phillips DL, Gregg JW (2003) Source partitioning using stable isotopes: coping with too many sources. Oecologia 136: 261–269. doi:10.1007/s00442-003-1218-3. PubMed: 12759813.
  28. 28. Fry B (2006) Stable Isotope Ecology. New York: Springer-Verlag. 308 pp.
  29. 29. Jackson AL, Parnell AC, Inger R, Bearhop S (2011) Comparing isotopic niche widths among and within communities: SIBER–Stable Isotope Bayesian Ellipses in R. J Anim Ecol 80: 595–602. doi:10.1111/j.1365-2656.2011.01806.x. PubMed: 21401589.
  30. 30. Layman CA, Allgeier JE (2012) Characterizing trophic ecology of generalist consumers: a case study of the invasive lionfish in The Bahamas. Mar Ecol Prog Ser 448: 131–141. doi:10.3354/meps09511.
  31. 31. Jackson MC, Donohue I, Jackson AL, Britton JR, Harper DM et al. (2012) Population-level metrics of trophic structure based on stable isotopes and their application to invasion ecology. PLOS ONE 7(2): e31757. doi:10.1371/journal.pone.0031757. PubMed: 22363724.
  32. 32. Parnell A, Jackson AL (2011) Stable Isotope Analysis in R (SIAR).
  33. 33. R Development Core Team (2011) R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.
  34. 34. Logan JM, Lutcavage ME (2010) Stable isotope dynamics is elasmobranch fishes. Hydrobiologia 644: 231–244. doi:10.1007/s10750-010-0120-3.
  35. 35. Perga ME, Grey J (2010) Laboratory measures of isotope discrimination factors: comments on Caut et al. (2008, 2009). J Appl Ecol 47: 942–947. doi:10.1111/j.1365-2664.2009.01730.x.
  36. 36. Prado P, Carmichael RH, Watts SA, Cebrian J, Heck KL (2012) Diet-dependent δ13C and δ15N fractionation among sea urchin Lytechinus variegates tissues: implications for food web models. Mar Ecol Prog Ser 462: 175–190. doi:10.3354/meps09786.
  37. 37. Gannes LZ, O’Brien DM, Martínez del Rio C (1997) Stable isotopes in animal ecology: assumptions, caveats and a call for more laboratory experiments. Ecology 78: 1271- 1276. doi:10.1890/0012-9658(1997)078[1271:SIIAEA]2.0.CO;2.
  38. 38. Martínez del Rio C, Wolf N, Carleton SA, Gannes Z (2009) Isotopic ecology ten years after a call for more laboratory experiments. Biol Rev 84: 91–111. doi:10.1111/j.1469-185X.2008.00064.x. PubMed: 19046398.
  39. 39. Carlisle AB, Kim SL, Semmens BX, Madigan DJ, Jorgensen SJ et al. (2012) Using stable isotope analysis to understand the migration and trophic ecology of northeastern Pacific white sharks (Carcharodon carcharias). PLOS ONE 7: e30492. doi:10.1371/journal.pone.0030492. PubMed: 22355313.
  40. 40. Robbins CT, Felicetti LA, Florin ST (2010) The impact of protein quality on stable nitrogen isotope ratio discrimination and assimilated diet estimation. Oecologia 162: 571–579. doi:10.1007/s00442-009-1485-8. PubMed: 19898979.
  41. 41. Bethea DM, Hale L, Carlson JK, Cortés E, Manire CA et al. (2007) Geographic and ontogenetic variation in the diet and daily ration of the bonnethead shark, Sphyrna tiburo, from the eastern Gulf of Mexico. Mar Biol 152: 1009–1020. doi:10.1007/s00227-007-0728-7.
  42. 42. Kim SL, Martínez del Rio C, Casper D, Koch PL (2012) Isotopic incorporation rates for shark tissues from a long-term captive feeding study. J Exp Biol 215: 2495–2500. doi:10.1242/jeb.070656. PubMed: 22723489.
  43. 43. Blanchet-Aurigny A, Guillou M, Pernet F, Gaffet JD, Dubois SF (2012) Tissue-diet discrimination factors of isotopic ratios (Δδ13C and Δδ15N) in two brittle star species: effect of reproductive state, diet and tissue composition. J Exp Mar Biol Ecol 426: 68- 77.
  44. 44. Milakovic B, Parker KL (2011) Using stable isotopes to define diets of wolves in northern British Columbia, Canada. J Mammal 92: 295–304. doi:10.1644/10-MAMM-A-038.1.
  45. 45. Chikaraishi Y, Ogawa NO, Kashiyama Y, Takano Y, Suga H et al. (2009) Determination of aquatic food-web structure based on compound-specific nitrogen isotopic composition of amino acids. Limnol Oceanog Meth 7: 740–750. doi:10.4319/lom.2009.7.740.
  46. 46. Germain LR, Koch PL, Harvey J, McCarthy MD (2013) Nitrogen isotope fractionation in amino acids from harbor seals: implications for compound-specific trophic position calculations. Mar Ecol Prog Ser 482: 265–277. doi:10.3354/meps10257.
  47. 47. Logan JM, Rodríguez-Marín E, Goñi N, Barreiro S, Arrizabalaga H et al. (2011) Diet of young Atlantic bluefin tuna (Thunnus thynnus) in eastern and western Atlantic foraging grounds. Mar Biol 158: 73–85. doi:10.1007/s00227-010-1543-0.
  48. 48. Heithaus MR, Vaudo JJ, Kreicker S, Layman CA, Krützen M et al. (2013) Apparent resource partitioning and trophic structure of large-bodied marine predators in a relatively pristine seagrass ecosystem. Mar Ecol Prog Ser 481: 225–237. doi:10.3354/meps10235.
  49. 49. Bond AL, Diamond AW (2011) Recent Bayesian stable-isotope mixing models are highly sensitive to variation in discrimination factors. Ecol. App 21: 1017–1023.
  50. 50. Kim SL, Tinker MT, Estes JA, Koch PL (2012) Ontogenetic and among-individual variation in foraging strategies of Northeast Pacific white sharks based on stable isotope analysis. PLOS ONE 7: e45068. doi:10.1371/journal.pone.0045068. PubMed: 23028766.
  51. 51. Estrada JA, Rice AN, Lutcavage ME, Skomal GB (2003) Predicting trophic position in sharks of the north-west Atlantic Ocean using stable isotopes. J Mar Biol Assoc UK 83: 1347–1350. doi:10.1017/S0025315403008798.
  52. 52. Hussey NE, Dudley SFJ, McCarthy ID, Cliff G, Fisk AT (2011) Stable isotope profiles of large marine predators: viable indicators of trophic position, diet and movement in sharks? Can J Fish Aquat Sci 68: 2029–2045. doi:10.1139/f2011-115.
  53. 53. Drymon JM, Powers SP, Carmichael RH (2012) Trophic plasticity in the Atlantic sharpnose shark (Rhizoprionodon terraenovae) from the north central Gulf of Mexico. Environ Biol Fishes 95: 21-35. doi:10.1007/s10641-011-9922-z.
  54. 54. Milessi AC, Danilo C, Rodríguez-Graña L, Conde D, Sellanes J et al. (2010) Trophic mass- balance model of a subtropical coastal lagoon, including a comparison with a stable isotope analysis of the food-web. Ecol Modell 221: 2859–2869. doi:10.1016/j.ecolmodel.2010.08.037.
  55. 55. McCauley DJ, Young HS, Dunbar RB, Estes JA, Semmens BX et al. (2012) Assessing the effects of large mobile predators on ecosystem connectivity. Ecol. App 22: 1711–1717.
  56. 56. Pinnegar JK, Polunin NVC (1999) Differential fractionation of δ13C and δ15N among fish tissues: implications for the study of trophic interactions. Func Ecol 13: 225–231. doi:10.1046/j.1365-2435.1999.00301.x.
  57. 57. Baker R, Buckland A, Sheaves M (2013) Fish gut content analysis: robust measures of diet composition. Fish Fisher. doi:10.1111/faf.12026.
  58. 58. Vaudo JJ, Heithaus MR (2011) Dietary niche overlap in a nearshore elasmobranch mesopredator community. Mar Ecol Prog Ser 425: 247–260. doi:10.3354/meps08988.
  59. 59. Heupel MR, Simpfendorfer CA, Collins AB, Tyminski JP (2006) Residency and movement patterns of bonnethead sharks, Sphyrna tiburo, in a large Florida estuary. Environ Biol. Fish 76: 47–67. doi:10.1007/s10641-006-9007-6.
  60. 60. Carlson JK, Heupel MR, Bethea DM, Hollensead LD (2008) Coastal habitat use and residency of juvenile Atlantic sharpnose sharks (Rhizoprionodon terraenovae). Estuar Coasts 31: 931–940. doi:10.1007/s12237-008-9075-2.
  61. 61. Brunnschweiler JM, Barnett A (2013) Opportunistic visitors: long-term behavioural response of bull sharks to food provisioning in Fiji. PLOS ONE 8: e58522. doi:10.1371/journal.pone.0058522. PubMed: 23516496.
  62. 62. Carlson JK, Ribera MM, Conrath CL, Heupel MR, Burgess GH (2010) Habitat use and movement patterns of bull sharks Carcharhinus leucas determined using pop-up satellite archival tags. J Fish Biol 77: 661–675. PubMed: 20701646.
  63. 63. Hammerschlag N, Luo J, Irschick DJ, Ault JS (2012) A comparison of spatial and movement patterns between sympatric predators: bull sharks (Carcharhinus leucas) and Atlantic tarpon (Megalops atlanticus). PLOS ONE 7: e45958. doi:10.1371/journal.pone.0045958. PubMed: 23049904.
  64. 64. MacNeil MA, Drouillard KG, Fisk AT (2006) Variable uptake and elimination of stable nitrogen isotopes between tissues in fish. Can J Fish Aquat Sci 63: 345–353. doi:10.1139/f05-219.
  65. 65. Moore JW, Semmens BX (2008) Incorporating uncertainty and prior information into stable isotope mixing models. Ecol Lett 11: 470–480. doi:10.1111/j.1461-0248.2008.01163.x. PubMed: 18294213.
  66. 66. Parnell AC, Inger R, Bearhop S, Jackson AL (2010) Source partitioning using stable isotopes: coping with too much variation. PLOS ONE 5: e9672. doi:10.1371/journal.pone.0009672. PubMed: 20300637.
  67. 67. Overmyer JP, MacNeil MA, Fisk AT (2008) Fractionation and metabolic turnover of carbon and nitrogen stable isotopes in black fly larvae. Rap Commun. J Mass Spectrom 22: 694- 700.