Insights into feeding interactions of shallow water cape hake (Merluccius capensis) and cape horse mackerel (Trachurus capensis) from the Northern Benguela (Namibia)

Regional Studies in Marine Science 34 (2020) 101071

Contents lists available at ScienceDirect

Regional Studies in Marine Science journal homepage: www.elsevier.com/locate/rsma

Insights into feeding interactions of shallow water cape hake (Merluccius capensis) and cape horse mackerel (Trachurus capensis) from the Northern Benguela (Namibia) Hendrina K. Kadila a,b , Dietlinde N. Nakwaya a , Mike Butler c , Johannes A. Iitembu a,b ,

∗

a

Department of Fisheries and Aquatic Sciences, University of Namibia, Hentiesbay, Namibia Sam Nujoma Marine and Coastal Resources Research Centre (SANUMARC), Hentiesbay, Namibia c iThemba LABS, Johannesburg, South Africa b

article

info

Article history: Received 3 November 2018 Received in revised form 12 January 2020 Accepted 12 January 2020 Available online 15 January 2020 Keywords: Stable isotopes Stomach contents Feeding interactions Northern Benguela ecosystem Shallow water cape hake Cape horse mackerel

a b s t r a c t Shallow water cape hake (SWCH) (Merluccius capensis) and cape horse mackerel (CHM) (Trachurus capensis) are ecologically and commercially important species in the northern Benguela ecosystem (Namibia). The understanding of their feeding interactions is however still limited. In this study, stable isotope measurements [carbon (δ 13C) and nitrogen (δ 15N)] of their muscles and stomach contents were used to understand their feeding interactions. Muscle tissues and stomach contents (n= 404), were collected during bottom trawl surveys in Namibian waters (November 2017). Results indicated that krill (Euphausiids) was the dominant prey in the diet of CHM and smaller SWCH, while the importance of CHM in the diet of larger SWCH was observed. Jacopever (Helicolenus dactylopterus) dominated the diet of SWCH that were larger than 51 cm. The diet compositions of the two species changed with latitude, an indication of the influence of prey availability. A potential for interspecific feeding competitions between the two species was observed as krill and anchovy were found as their common prey species. Significant differences were found in both δ 15N values and δ 13C values of the two species. A significant positive relationship between δ 13C values and size were observed for both species. A negative relationship between δ 15N values and size was observed in CHM. The length ranges of isotopic intersections were 34 to 36 cm for δ 15N and 32 cm to 40 cm for δ 13C, an indication that young SWCH possibly interacts more with larger CHM. Although there was niche overlap, a wider niche for SWCH than CHM was observed. This is the first study that has combined stable isotopes and stomach content analysis methodologies, to understand the feeding interaction of SWCH and CHM. © 2020 Elsevier B.V. All rights reserved.

1. Introduction Shallow water cape hake (SWCH) (Merluccius capensis) and cape horse mackerel (CHM) (Trachurus capensis) are commercially exploited fish species and key secondary consumers in the Northern Benguela ecosystem (Namibia) (Boyer and Hampton, 2001). The two species have overlapping depth distributions) with SWCH living in waters of 100–450 m (Burmeister, 2001; Jansen et al., 2015), while CHM lives in water depths of 0– 400 m (Axelsen et al., 2004). Both species have vertical diurnal movements between the pelagic and demersal components of the Benguela ecosystem (Pillar and Barange, 1998). Previous studies using stomach content analyses of the two species indicated that they both feed on small crustaceans and ∗ Corresponding author at: Department of Fisheries and Aquatic Sciences, University of Namibia, Hentiesbay, Namibia. E-mail address: [email protected] (J.A. Iitembu). https://doi.org/10.1016/j.rsma.2020.101071 2352-4855/© 2020 Elsevier B.V. All rights reserved.

small fish (Assorov and Kalinina, 1979; Konchina, 1986; Krzeptowski, 1982; Andronov, 1983). A predator–prey relationship between the two species, with hake feeding on horse mackerel, was also observed (Assorov and Kalinina, 1979; Konchina, 1986; Krzeptowski, 1982; Andronov, 1983). SWCH is considered as an opportunistic feeder, changing its preferred prey type with its local availability (Pillar and Wilkinson, 1995; Roel and Macpherson, 1988), while CHM is a facultative fish feeder whose diet consists mainly of krill (Krzeptowski, 1982; Andronov, 1983). Although stomach content analyses that have been used to study the feeding interactions of SWCH and CHM provided important dietary information, the method has inherent limitations due to several factors, including fast digestion rates and regurgitation of gastric contents during capture events (Varela et al., 2017). Several studies have recommended the usage of stable isotope analysis (δ 13C and δ 15N) to complement stomach content analyses, as it can provide time and space integrated dietary information (Varela et al., 2018). The usage of the traditional gut content approach along with stable isotope analysis is also

2

H.K. Kadila, D.N. Nakwaya, M. Butler et al. / Regional Studies in Marine Science 34 (2020) 101071

known to provide a better understanding of feeding interactions (Fanelli and Cartes, 2010; Soares et al., 2018). Additionally, recent developments have allowed, for the usage of stable isotope compositions to assess the trophic niche width of species (Bearhop et al., 2004; Jackson et al., 2011), a useful parameter to understand fish species’ feeding interactions. The two species have been important predators in the Benguela current system, but major changes that might affect their feeding interactions have occurred in the system. The observed changes in the Benguela current system include spatial changes in the distribution of small pelagic fish (Cury and Shannon, 2004; Heymans et al., 2004), and increases in biomass of other species such as jellyfish (Flynn et al., 2012; Roux et al., 2013) and gobies (Utne-Palm et al., 2010; Van Der Bank et al., 2011). Therefore, the observed SWCH-CHM feeding interactions in above-cited studies (Assorov and Kalinina, 1979; Konchina, 1986; Krzeptowski, 1982; Andronov, 1983) might have changed over time. This study was therefore aimed at investigating the feeding interactions of the SWCH and CHM using both their stomach contents and stable isotope measurements of their tissues. The specific objectives were (i) to compare the diets of the two species based on their stomach contents, (ii) to compare their isotopic signatures, and (iii) to investigate the possible overlap in their isotopic niches. Inferences on a hypothesis of altered SWCH-CHM feeding interactions as results of major changes observed in the Benguela current were also made. 2. Material and methods 2.1. Study area and field sampling Samples were collected during the annual monkfish biomass survey (11th–27th November 2017) off the coast of Namibia. The monkfish biomass survey was done to mainly estimate the biomass of the monkfish stock, the sampling, therefore, followed survey-predetermined stations. Sampling was done using an Albatross monkfish bottom trawl (head length 50.3 m, footrope 63.9 m and the vertical net opening 1.2–1.3 m) rigged with tickler chains along the footrope). The depth of sampling tows was between 100 and 800 m, at a speed of about three knots. SWCH and CHM were collected at 13 stations (the only stations in which the two species were present) from Conception Bay (24◦ S) to the Cunene River (17◦ S) (Fig. 1). Fish were identified and an equal number of individual fish of the two species were collected at each station. The information recorded was the species name, total length, and sex. A small piece of muscle [about 0.5 kg], was cut from the dorsal region of each fish and placed in Ziploc bags. Stomachs were collected from each fish and stored in separate bags. All collected samples were frozen at −20 ◦ C on the vessel until the end of the survey.

Fig. 1. Stations (squares) from which samples were collected during the monkfish biomass survey. The depth contours represent 200 m, 1000 m and 3000 m isobaths.

food in the sample (Hyslop, 1980; Bowen, 1996). The frequency of occurrence of prey was calculated to determine the most occurring food (Prey) item. The Index of Relative Importance (IRI) was calculated using the formula: IRI = ((%N + %W) * %F); where %N is percentage composition by number, %W is percentage composition weight of each prey and %F is frequency of prey occurrence (Pinkas et al., 1971; Cortés, 1997). An index of relative importance (IRI) (Hacunda, 1981) was calculated to determine the most important food item and to integrate the three parameters to eliminate any biases created once each method was analysed individually (Goldman and Sedberry, 2011). The percent IRI was calculated using the formula: %IRIi = (IRIi/Σ IRIi) × 100; where i represents prey item (Cortés, 1997).

2.2. Laboratory analyses 2.2.1. Stomach content analysis Each stomach was weighed and dissected. The contents were transferred to a petri dish, then wrapped with a paper towel to remove water. Prey were identified to the lowest taxonomic level by visual inspection of stomach contents according to FAO species identification guide for fishery purposes (Bianchi et al., 1999). Prey items were counted and weighed (to the nearest gram [g]). The contribution of each prey item was determined using frequency of occurrence calculated using the formula: %Fi = 100 ni/n, where Fi is frequency of occurrence of the i (prey item) food item in the sample; ni is the number of stomachs in which the prey item is found and n is the total number of stomachs with

2.2.2. Stable isotopes analysis A small piece of white muscle was oven-dried for 48 h at 60 ◦ C. The dried muscles were homogenized into powder using a mortar and pestle, and placed in small tin capsules of approximately 9 mm × 5 mm. Stable isotopes ratios of carbon and nitrogen were analysed at iThemba Laboratories (Johannesburg, South Africa) using a Flash HT Plus elemental analyser coupled to a Delta V Advantage isotope ratio mass spectrometer through a ConFloIV interface (equipment supplied by ThermoFisher, Bremen, Germany). The stable isotopic values were expressed in delta (δ ) notation relative to the standard for carbon (Pee Dee Belemnite) and nitrogen (N2) was done using the standard equation δ X = {(Rsample/Rstandard) −1} * 1000, where X is δ 13C or δ 15N; R is

H.K. Kadila, D.N. Nakwaya, M. Butler et al. / Regional Studies in Marine Science 34 (2020) 101071

the ratio of the heavy to light isotope for the sample (Rsample) and standard (Rstandard) in units of parts per thousand (per mille, h). Laboratory standards and blanks were run after every 24 unknown samples. Carbon and nitrogen isotope values were corrected against an in-house standard (Merck Gel) and a Urea Working Standard ((IVA Analyse Technik e.K., Meerbusch, Germany). Sample analytical precision was <0.17h for both δ 13C and δ 15N, respectively. Samples were not physically lipid extracted to avoid possible fractionation of δ 15N values (Sweeting et al., 2006; Post et al., 2007). All δ 13C values with C: N values greater than 3.5 the minimum limit for lipid extraction or correction (Post et al., 2007) were mathematically corrected using the normalization equation by Post et al. (2007): δ 13Cnormalized = δ 13Cuntreated − 3.32 + 0.99×C : N; where δ 13C untreated is the δ 13C of non-lipid extracted tissue, C: N is the mass ratio of carbon and nitrogen. 2.2.3. Statistical analyses The stable isotope measurements were tested for normality using the Kolmogorov Smirnov test. The data were not normally distributed, therefore Kruskal–Kruskal–Wallis Test was used to assess if there were significant differences in the stable isotope values (δ 15N and δ 13C) of SWCH and CHM. Linear regression analysis was performed to investigate the relationships, between size and isotopic measurement (δ 15N and δ 13C) of SWCH and CHM. Isotope-based metrics calculated included δ 15N range, indicating trophic diversity; δ 13C range representing the niche diversification at the base of a food web, total area (TA) of the convex hull standard ellipse areas and the standard ellipse area, corrected for small sample sizes. Differences in SEAC between species were estimated via Bayesian interference (SEAB ) (Jackson et al., 2011). All the statistical analyses were done in R [R Core Team (2018), Vienna, Austria] and SPSS. 3. Results 3.1. Diet compositions of shallow water cape hake (SWCH) A total of 202 stomachs of SWCH were collected and only 56 (27.72%) were not empty. The total length (TL) of fish individuals ranged from 21 cm to 68 cm. The diet compositions consisted of 13 different species, with krill being the dominant prey in terms of IRI (66.29%) while CHM dominated in terms of weight (Fig. 2). Diet compositions changed with increasing TL, where krill was the most frequently observed prey in ≤30 cm (F = 52.94%, 74.23%IRI) and 31–40 cm (F = 50%, 82.92%IRI) length classes. Cape horse mackerel was the most encountered at 41–50 cm length class (F = 28.57%, 47.54%IRI), while Jacopever dominated the ≥ 51 cm length class (F = 41.18%, 64.05%IRI) (Table 1). In terms of latitudes, krill was the most abundant prey item encountered at all latitude classes (Table 2). 3.2. Diet compositions of cape horse mackerel (CHM) A total of 202 stomachs of CHM were collected and only 74(36.63%) were not empty. The individual fish TL ranged from 19 cm to 40 cm. The diet consisted of three different species, with krill being the most abundant prey (98.70%W) and anchovy being the least encountered prey (0.39%W) (Fig. 3). In terms of length classes, the diet compositions changed with increasing TL; where krill was the most frequently observed prey (F = 94.55%, 99.90%IRI; F = 100%, 100%IRI) at both ≤30 cm and ≥31 cm (Table 3). In terms of latitudes, krill was the most prey item encountered at all latitude classes (Table 4).

3

3.3. Isotopic characteristics A total of 404 samples that consisted of SWCH (n = 202) and CHM (n = 202) were analysed for the stable isotope ratios of nitrogen and carbon. CHM TL ranged from 19 cm to 40 cm, while SWCH TL ranged from 20 cm to 68 cm. The δ 15N values for SWCH had the highest mean δ 15N value (11.17h) that ranged from 9.31 h to 13.34h, while that of CHM ranged 8.75 h to 13.07h. The δ 13C values for SWCH ranged from −17.26h to −14.24h, while that of CHM ranged from −17.28 to −14.64h. CHN had the most negative average δ 13C value (−17.10h) than SWCH (−16.10h). CHM (4.73) had a higher mean C: N ratio than SWCH (3.97). A significant difference between the δ 13C of SWCH and CHM (H = 20.39, P < 0.001) was observed. There was also a significant difference between the δ 15N of SWCH and CHM (H = 31.04, P < 0.001). In terms of the relationship between the isotopic values and TL, SWCH showed a stronger relationship between size and δ 13C (P < 0.001, R2 = 0.38) compared to CHM (P < 0.001, R2 = 0.02). The length of intersection in term of δ 13C between the two species was between 32 cm to 40 cm (Fig. 4). There was a positive relationship (P < 0.001, R2 = 0.01) between δ 15N values and TL of SWCH while CHM had a negative relationship (P < 0.001, R2 = 0.08). The length of the intersection for δ 15N values of the two species was 34 cm to 36 cm (Fig. 5). A substantial overlap in the isotopic niches of the two species was observed, with SWCH having a wider niche than CHM (Fig. 6). The range of δ 13C and δ 15N for CHM was higher than for SWCH (Table 5). SWCH had a larger TA of 7.76 than for CHM. 4. Discussion This study aimed at investigating the feeding interactions of SWCH and CHM off Namibia, using stomach content and stable isotope analyses. Stomach content analyses showed a higher diversity of prey in SWCH than in CHM. The diets of smaller SWCH (30–40 cm) were dominated by krill, while that of middle-sized (41–50 cm) and larger ones (≥51 cm) were dominated by CHM and Jacopever (Helicolenus dactylopterus), respectively. The diet compositions of the two species changed with both length and latitude, an indication of the influence of prey availability. The diet compositions of both species are similar to those observed in a previous study using stomach content analyses (Assorov and Kalinina, 1979; Konchina, 1986; Krzeptowski, 1982; Andronov, 1983). SWCH had a higher mean δ 15N value (11.17h) than CHM (10.67h), an indication that it fed at relatively higher trophic positions. In terms of δ 13C values, SWCH had lower average δ 13C values (−16.10h) than CHM (−17.10h), an indication of the influence of benthic and pelagic food sources in their respective diets. Significant differences in food sources (δ 13C) and feeding positions (δ 15N) of both species were also observed. A significant positive relationship was observed between δ 13C and size (total length (cm)) in both species, indicating the influence of growth on their feeding strategies. For δ 15N and size, CHM had a negative relationship, an indication that they feed on more 15N-depleted sources as they grow, which differ from what was observed in SWCH. Although SWCH had a wider trophic niche than for CHM, there was substantial overlap in their niches. Overall these results indicate that the two species have multiple trophic interactions as shallow hake may be dependent on horse mackerel as one of the prey, while the two species are possibly competing for smaller prey like krill and anchovy. A high number of empty stomachs were observed in both species, perhaps because most sampling stations were in deep waters and fish from deeper water are more prone to stomach eversion (Payne et al., 1987; Punt et al., 1992). The observation

4

H.K. Kadila, D.N. Nakwaya, M. Butler et al. / Regional Studies in Marine Science 34 (2020) 101071

Fig. 2. Diet compositions of shallow water cape hake (M. capensis) by weight percentages of all the prey items consumed based on stomach contents analysis. Table 1 Frequency of prey occurrence (%F ), diet composition by number (%N), diet composition by weight (%W) and index of relative importance (IRI) of SWCH prey by length classes. Prey items

≤30 cm (n = 61)

31–40 cm (n = 66)

41–50 cm (n = 49)

≥51 cm (n = 26)

%F

%N

%W

%IRI

%F

%N

%W

%IRI

%F

%N

%W

%IRI

%F

%N

%W

%IRI

Krill (Euphausia hanseni)

52.94

62.96

16.43

74.23

50.0

81.71

4.06

82.92

19.05

79.67

3.06

39.59

17.65

9.52

0.14

2.65

Prawns (Parapenaeus longirostris)

0.00

0.00

0.00

0.00

12.50

3.66

2.36

1.46

9.52

3.25

1.67

1.18

5.88

4.76

0.23

0.46

Grenadier (Nezumia micronychodon)

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

5.88

9.52

7.32

1.54

Cape horse mackerel (Trachurus capensis)

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

28.57

8.94

57.30

47.54

29.41

23.81

44.64

31.30

Jacopever (Helicolenus dactylopterus)

5.88

1.85

2.66

0.47

4.17

1.22

1.18

0.19

14.29

3.25

12.99

5.83

41.18

52.38

47.66

64.05

Goby (Sufflogobius bibarbatus)

23.53

11.11

31.40

17.67

8.33

3.66

2.36

0.97

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Shortnose greeneye (Chlorophthalmus agassizi)

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

9.52

1.63

12.33

3.34

0.00

0.00

0.00

0.00

Squat lobster (Galathea squamifera)

0.00

0.00

0.00

0.00

4.17

1.22

1.18

0.19

9.52

1.63

3.66

1.26

0.00

0.00

0.00

0.00

Cephalopods (Class)

5.88

20.37

45.89

6.88

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Deep water cape hake (Merluccius paradoxus)

0.00

0.00

0.00

0.00

4.17

1.22

11.43

1.02

4.76

0.81

4.83

0.67

0.00

0.00

0.00

0.00

Shallow water cape hake (Merluccius capensis)

0.00

0.00

0.00

0.00

8.33

3.66

76.07

12.84

4.76

0.81

4.16

0.59

0.00

0.00

0.00

0.00

Anchovy (Engraulis capensis)

5.88

1.85

1.69

13.47

4.17

2.44

0.79

0.26

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

White mussel

5.88

1.85

1.93

0.39

4.17

1.22

0.55

0.14

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

100.00

100.00

100.00

100.00

100.00

100.00

100.00

100.00

100.00

100.00

100.00

100.00

Total

of more empty stomachs has also been documented by previous studies in both hakes (Payne et al., 1987; Mahe et al., 2007) and horse mackerel (Borges and Gordo, 1991). Empty stomachs in fish may be consequences of autecological factors, individual

fish health, environmental conditions or sampling artefacts (Vinson and Angradi, 2011). Generally, some species may be more vulnerable to fishing gears when they have empty stomachs because they are extra active (Vinson and Angradi, 2011). An increased number of empty stomachs has also been attributed to

H.K. Kadila, D.N. Nakwaya, M. Butler et al. / Regional Studies in Marine Science 34 (2020) 101071

5

Table 2 Percentages of frequency of occurrence (%F ), diet composition by number (%N), diet composition by weight (%W) and index of relative importance (%IRI) of prey items found in stomachs of shallow water cape hake (SWCH) by Latitude classes (◦ S). 17◦ 00–20◦ 59

Prey items

21◦ 00–25◦ 00

%F

%N

%W

%IRI

%F

%N

%W

%IRI

Anchovy (Engraulis capensis)

2.00

1.11

0.15

0.07

3.45

1.00

0.17

0.07

Krill (Euphausia hanseni)

22.00

65.00

1.60

41.64

58.62

84.00

3.69

89.07

White mussel

0.00

0.00

0.00

0.00

6.90

2.00

0.54

0.30

Cephalopods (class)

2.00

6.11

1.41

0.43

0.00

0.00

0.00

0.00

Shortnose greeneye (Chlorophthalmus agassizi)

4.00

1.11

5.50

0.76

0.00

0.00

0.00

0.00

Goby (Sufflogobius bibarbatus)

12.00

5.00

1.41

2.22

0.00

0.00

0.00

0.00

Deep water cape hake (Merluccius paradoxus)

2.00

0.55

2.15

0.16

3.45

1.00

7.09

0.48

Shallow water cape hake (Merluccius capensis)

2.00

0.55

5.72

0.36

6.90

3.00

34.47

4.48

Jacopever (Helicolenus dactylopterus)

18.00

7.78

32.82

21.08

10.34

3.00

12.25

2.73

Cape horse mackerel (Trachurus capensis)

20.00

8.33

46.64

31.72

3.45

1.00

24.44

1.52

Common Atlantic Grenadier (Nezumia micronychodon)

0.00

0.00

0.00

0.00

3.45

2.00

15.40

1.04

Prawns (Parapenaeus longirostris)

10.00

2.78

0.74

1.02

3.45

1.96

17.11

0.30

Squat lobster (Galathea squamifera)

6.00

1.67

1.86

0.54

0.00

0.00

0.00

0.00

100.00

100.00

100.00

100.00

100.00

100.00

Total

Table 3 Frequency of prey occurrence (%F ), diet composition by number (%N), diet composition by weight (%W) and index of relative importance (IRI) expressed as a percent of prey items found in stomachs of cape horse mackerel (CHM) by length classes. Prey items

Krill (Euphausia hanseni)

≤30 (N = 160)

≥31 (N = 42)

% F

%N

%W

%IRI

%F

%N

%W

%IRI

94.55

94.41

97.19

99.90

100.00

100.00

100.00

100.00

Amphipods (order)

1.82

5.37

1.96

0.07

0.00

0.00

0.00

0.00

Anchovy (Engraulis capensis)

3.64

0.22

1.05

0.03

0.00

0.00

0.00

0.00

100.00

100.00

100.00

100.00

100.00

100.00

Total

Table 4 Frequency of prey occurrence (%F ), diet composition by number (%N), diet composition by weight (%W) and index of relative importance (%IRI) of prey items found in stomachs of cape horse mackerel (CHM) by Latitude classes (17◦ S–25◦ S). Prey items

17◦ 00′ –20◦ 59

21◦ 00′ –25◦ 00

% F

%N

%W

%IRI

%F

%N

%W

%IRI

Krill (Euphausia hanseni)

96.55

99.87

99.56

99.93

95.74

79.50

90.77

99.37

Amphipods (order)

0.00

0.00

0.00

0.00

3.45

20.50

9.23

0.63

Anchovy (Engraulis capensis)

4.26

0.13

0.44

0.07

0.00

0.00

0.00

0.00

100.00

100.00

100.00

100.00

100.00

100.00

Total

Fig. 3. Diet composition of cape horse mackerel (CHM) by weight percentages of all the prey items consumed based on stomach contents analysis.

6

H.K. Kadila, D.N. Nakwaya, M. Butler et al. / Regional Studies in Marine Science 34 (2020) 101071

Fig. 4. The relationship between δ 13 C and total length (cm) of shallow water Cape hake [SWCH] and Cape horse mackerel [CHM].

Fig. 5. The relationship between δ 15 N and total length (cm) of shallow water Cape hake [SWCH] and Cape horse mackerel [CHM].

H.K. Kadila, D.N. Nakwaya, M. Butler et al. / Regional Studies in Marine Science 34 (2020) 101071

7

Table 5 The carbon-13 range (δ 13 CR), nitrogen-15 range (δ 15 NR), total area of the convex hull (TA) and standard ellipse area (SEAc) of SWCH and CHM.

δ 13 CR

δ 15 NR

TA

SEAC

δ 15 N (±SD)

δ 13 C (±SD)

Shallow water Cape hake (SWCH) (n = 202)

−3.00

4.03

7.76

1.24

11.17 ± 0.73

−15.48 ± 0.54

Cape horse mackerel (CHM) (n = 202)

−2.63

4.33

7.27

1.24

10.67 ± 0.89

−15.74 ± 0.56

Species

Fig. 6. The isotopic niches comparisons of shallow water Cape hake [SWCH] and Cape horse mackerel [CHM] as depicted by convex hull (polygons) and SEAc (ellipses).

the sampling method, as bottom trawling is known to cause fish to regurgitate during capture because of the pressure gradients (Vignon and Dierking, 2011). The diet composition of SWCH was different from that of CHM as most important prey consumed by SWCH were bony fishes. The increased prey diversity observed in SWCH prey can be because it is an opportunistic feeder and has been observed as changing its preferred prey type based on prey local availability and abundance (Pillar and Wilkinson, 1995). In this study, cape horse mackerel was observed as a dominant prey of the middle-sized SWCH, an indication of the importance of CHM to the diet of SWCH. Because the stomach contents were only collected at stations where both SWCH and CHM were present, the availability and abundance of CHM at these stations might have led to CHM being more available to SWCH. The importance of CHM in the diet of this species can also be linked to vertical diurnal movements of both species between the pelagic and demersal components of the Benguela ecosystem (Pillar and Barange, 1998; Iilende et al., 2001), making it a more available prey to SWCH. Other important SWCH prey observed in this study included small crustaceans, which corresponds to the findings of Roel and Macpherson (1988). However, the above finding differed from that of Assorov and Kalinina (1979), whose study documented that SWCH fed mainly on the young of their own species (cannibalism) and gobies. The cannibalistic nature of SWCH feeding has also been observed in other studies (Punt et al., 1992), with Macpherson and Gordoa (1994) study suggesting that SWCH tends to prefer the young of their own over other prey. The differences in the observations from different diet studies can be related to fluctuations in the prey populations over time (Mehl, 1986) or local availability and abundance of these prey. One peculiar prey, white mussel, was observed in the diet of SWCH in this study but was not observed in any of the previous studies (Roel and Macpherson, 1988; Pillar and Wilkinson, 1995). The presence of white mussels in the diet of SWCH can be an

indication of the opportunistic feeding nature of hake, but it can also be that it was consumed by its prey, as the shell is normally not fully broken down by the digestion process. Diet compositions for SWCH changed with length, an indication of ontogenic trophic shift of this species which was reported in other studies (e.g. Assorov and Kalinina, 1979; Roel and Macpherson, 1988; Pillar and Wilkinson, 1995; Mahe et al., 2007) and confirmed by stable isotopes results (Iitembu et al., 2012). Krill was found to be one of the crustacean constituents of SWCH, especially at a small length. Confirming the finding of Roel and Macpherson (1988) that observed that the importance of krill as food diminishes with the size of the predator, while the importance of fish in the diet increases with hake size. Similarly, Pillar and Wilkinson (1995) observed ontogenic trophic shift in diet of this species from larvae to adult, as their preferences switch from small copepods to larger prey items such as krill, mesopelagic and pelagic fish (mainly anchovy in the diet of smaller (<30 cm) and when adult, to demersal fish (mostly other hake). The shift in diet was also observed by Assorov and Kalinina (1979) where fish of the family myctophidae and hake were the most abundant in the hake diet with the proportion of myctophids decreasing with hake length while that of hake increased. In general, an ontogenic trophic shift in SWCH could be related to ontogenetic development particularly an increase in mouth size and mobility (Mahe et al., 2007). The ontogenic trophic shift observed can also be influenced by the preference of prey that has a higher energy content than crustaceans (common in non-specialist fish) (Juanes et al., 2002). Krill was the most prey item encountered by SWCH at all latitude classes, an indication of the importance of this species to their diet. Globally, krill is one of the species that occur in higher abundances (Huenerlage and Buchholz, 2013), which can be the explanation of SWCH diet dominance observed in this study. There was also a difference in diversity of the prey with latitudes, with higher diversity observed at 17–21◦ S than 21–25◦ S latitude classes. The higher diversity of prey at lower latitudes can be related to a general pattern of species richness that increases with decreasing latitude (Kaufman, 1995). The overall pattern for species richness tends to increase from polar to tropical regions (Willig et al., 2003; Hillebrand, 2004), irrespective of the taxonomic affiliation of organisms (e.g. mammals, fishes, insects, and plants) or geographic site in which they occur (e.g. Africa, South America, and the Atlantic Ocean) (Willig et al., 2003). The diet of CHM consisted of three different species, with krill observed as the major prey and anchovy was the least prey. Konchina (1986) observed that the diet of horse mackerel included pteropods, crustaceans (copepods, krill, hyperiids, and decapods), with krill and shrimps being the main food items. Boyer and Hampton (2001) reported that 95% of the diet of adult CHM fish comprised of krill and shrimps. The above results indicate that krill is one of the key prey of horse mackerel. Similar to SCWH the diet compositions of CHM also changed with increasing size, with larger CHM feeding exclusively on krill, which agrees with the finding by Andronov (1983). The above is possibly an influence of fish growth, where small fish are known to be mostly zooplanktophagous, while large specimens are mainly ichthyophagous (Cabral and Murta, 2002; Jardas et al., 2004). Konchina (1986) observed that 21–26 cm (2 years old) horse mackerel were feeding mostly on krill and lanternfishes, this was

8

H.K. Kadila, D.N. Nakwaya, M. Butler et al. / Regional Studies in Marine Science 34 (2020) 101071

linked to the fact that in the northern part of the area, horse mackerel fed on mesopelagic migratory animals while moving towards the surface at night. Konchina (1986) findings reported that these species feed predominantly on zooplankton for up to two years of age. CHM has also been considered as a facultative fish feeder whose diet consists mainly of krill (Andronov, 1983; Krzeptowski, 1982). In term of isotopic measurement of their tissues, CHM had lower average mean δ 15N value than SWCH. As the δ 15N values of organisms reflect the trophic level at which they are feeding (Peterson and Fry, 1987), this indicating that SWCH species fed at a higher trophic level than CHM. These results are also supported by the stomach content results that showed the diet of SWCH had more fish prey than CHM. There were also statistically significant differences in the δ 15N values of the two species, which showed that these species do not feed at the same trophic position. The isotope-based trophic level was also observed to be different by Erasmus (2015). Both species had a positive linear relationship between δ 15N and TL (cm) which is an indication of a change in feeding habits with growth (Araújo et al., 2011). Positive relationships between total length and trophic positions are common in several marine organisms (Deudero et al., 2004), and have been reported in hake (Iitembu et al., 2012), sharks (Estrada et al., 2006) and shrimps (Endjambi et al., 2015). Although there was a significant difference in δ 15N values, the two species appear to be feeding at the same trophic positions at the length of <34 cm and >36 cm which was the length of δ 15N values intersection. This is an indication that young SWCH interacts more with older CHM as the age range of 34–36 cm CHM is 6– 7 years (Kirchner et al., 2010), while SWCH of the same length is about 1–2.5 years old (Wilhelm et al., 2017). At these length ranges, the two species’ diets are dominated by krill, indicating the possibility of interspecific feeding competition between these species. The δ 15N range for CHM was wider than for SWCH, an indication that it fed on prey from various trophic levels. This, however, contradicts the finding of the stomach content analysis that observed increased prey diversity in SWCH. Perhaps due to differences between trophic positions of CHM prey were higher compared to the SWCH prey. CHM was more highly enriched in δ 13C. Considering that δ 13C is used as proxy sources of dietary carbon (DeNiro and Epstein, 1978; Peterson and Fry, 1987), this showed that CHM may be more dependent on pelagic prey, which is generally more enriched in δ 13C than demersal prey (Davenport and Bax, 2002; France, 1995). SWCH which had a lower average δ 13C enrichment value (−15.48h), suggests had increased proportions of demersal prey in its diet. The stomach content results confirm this as the diet of CHM was found to consist of krill, anchovy, and amphipods, which are all pelagic species. The differences in δ 13C values have also been attributed to the differences between coastal and offshore waters, with organisms that feed close to the shore having higher δ 13C values compared to those that feed offshore (Kurle and Worthy, 2001). In most coastal habitats, pelagic phytoplankton yield more negative δ 13C values than alternative carbon sources (Bird et al., 2018). As two species have overlapping distribution (Burmeister, 2001) and the samples in this study where collected at the same station where both species where caught, the inshore–offshore differences might not be significant enough to influence the δ 13C differences observed. A positive linear relationship was observed between δ 13C values and TL of both species. Both species are considered to have a demersal component of their stocks, with juveniles being pelagic (Jansen et al., 2016; Kirchner et al., 2010) suggesting the main differences occur in mostly larger individuals. The length of intersection in term of δ 13C was 32–40 cm, which is within that observed for δ 15N. CHM had a wider carbon range (CR) (δ 13CR’)

than SWCH, an indication of its prey might have been dependent on multiple basal resources. The isotopic niche which has been used as a proxy species’ trophic niche (Newsome et al., 2012; Pagell and Shevchenko, 2014) was wider for SWCH than CHM. Larger trophic niche for SWCH can be an indication of a broader trophic diversity, while narrower trophic niche for SWCH represents a lower trophic diversity or a more specialized niche. The niche overlap observed, however, shows the two species might be dependent on the same resources at some stage of their life, especially at their length of carbon and nitrogen value intersections. The total area (TA) which represents the total amount of niche area occupied by a species (Layman et al., 2007) was also larger for SWCH than CHM. The larger TA is possibly a reflection of the high diversity of prey in the SWCH diet than in CHM. In terms of standard ellipse area (SEAc), which is a measure of the mean core population isotopic niche (Jackson et al., 2011), these two species also had overlapping SEAc, an indication of their overlapping feeding ranges. Overall, the present study showed that SWCH and CHM have multiple trophic interactions. Their feeding interactions may be one of the factors that regulate their population dynamics. Additionally, because younger SWCH possibly interact more with older CHM, there is a possibility that the CHM fishing has a direct influence on SWCH recruitment. Future studies should consider ascertaining the degree to which these feeding interactions influence the population dynamics of the two species. Although the two fisheries (SWCH and CHM) are still managed as single species, the finding of this study suggests the need for consideration of their feeding interaction in their management approaches. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. CRediT authorship contribution statement Hendrina K. Kadila: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Data curation, Writing original draft, Writing - review & editing. Dietlinde N. Nakwaya: Writing - review & editing, Supervision. Mike Butler: Resources, Data curation, Writing - review & editing. Johannes A. Iitembu: Conceptualization, Methodology, Validation, Formal analysis, Resources, Writing - review & editing, Supervision, Funding acquisition. Acknowledgements We thank researchers (November 2017 Monkfish survey team, led by Ms. Ester Nangolo (Cruise leader)) at the Ministry of Fisheries and Marine Resources (MFMR) for their support during fieldwork. We are also thankful to P. Aitembu, L. Impinge, T. Haipinge for helping with sample preparation, Mike Butler and Osborne Malinga for analyses of stable isotopes samples at iThemba LABS (Johannesburg, South Africa). Financial support for this study was provided by the SANUMARC community trust fund to the University of Namibia, Namibia as part of MSc scholarship to Ms. HK Kadila.

H.K. Kadila, D.N. Nakwaya, M. Butler et al. / Regional Studies in Marine Science 34 (2020) 101071

References Andronov, V.N., 1983. Feeding of cape horse mackerel (Trachurus trachurus capensis) and cape hake (Merluccius capensis) off Namibia in January 1982. Collect. Sci. Pap. Int. Comm. South East. Atl. Fish. 10, 1–8. Araújo, M.S., Bolnick, D.I., Layman, C.A., 2011. The ecological causes of individual specialisation. Ecol. Lett. 14, 948–958. Assorov, V.V., Kalinina, M.I., 1979. Some peculiarities of the feeding habits of cape hake and South African deep-water hake. Collect. Sci. Pap. Int. Comm. South East. Atl. Fish. 6, 219–227. Axelsen, B.E., Krakstad, J.O., Bauleth-D’Almeida, G., 2004. Aggregation dynamics and behaviour of the cape horse mackerel (Trachurus capensis) in the Northern Benguela—implication for acoustic abundance estimation. In: Sumaila, U.R., Boyer, D., Skogen, M.D., Steinshamn, S.I. (Eds.), Namibia’s Fisheries. Eburon Academic Publishers, Delft, pp. 135–164. 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. Bianchi, G., Carpenter, K.E., Roux, J., Molloy, F.J., Boyer, D., Boyer, H.J., 1999. FAO species identification field guide for fishery purposes. In: The Field Guide to the Living Marine Resources of Namibia. FAO, Rome. Bird, C.S., Veríssimo, A., Magozzi, S., Abrantes, K.G., Aguilar, A., Al-Reasi, H., Barnett, A., Bethea, D.M., Biais, G., Borrell, A., Bouchoucha, M., Boyle, M., Brooks, E.J., Brunnschweiler, J., Bustamante, P., Carlisle, A., Catarino, D., Caut, S., Cherel, Y., Chouvelon, T., Churchill, D., Ciancio, J., Claes, J., Colaço, A., Courtney, D.L., Cresson, P., Daly, R., De Necker, L., Endo, T., Figueiredo, I., Frisch, A.J., Hansen, J.H., Heithaus, M., Hussey, N.E., Iitembu, J., Juanes, F., Kinney, M.J., Kiszka, J.J., Klarian, S.A., Kopp, D., Leaf, R., Li, Y., Lorrain, A., Madigan, D.J., Maljković, A., Malpica-Cruz, L., Matich, P., Meekan, M.G., Ménard, F., Menezes, G.M., Munroe, S.E.M., Newman, M.C., Papastamatiou, Y.P., Pethybridge, H., Plumlee, J.D., Polo-Silva, C., Quaeck-Davies, K., Raoult, V., Reum, J., Torres-Rojas, Y.E., Shiffman, D.S., Shipley, O.N., Speed, C.W., Staudinger, M.D., Teffer, A.K., Tilley, A., Valls, M., Vaudo, J.J., Wai, T.C., Wells, R.J.D., Wyatt, A.S.J., Yool, A., Trueman, C.N., 2018. A global perspective on the trophic geography of sharks. Nat. Ecol. Evol. 2, 299–305. http://dx.doi.org/10.1038/s41559-0170432-z. Borges, M.F., Gordo, L.S., 1991. Spatial distribution by season and some biological parameters of horse mackerel (Trachurus trachurus L.) in the Portuguese continental waters (Division IXa). ICES CM 1991/H: 54. p. 8. Bowen, S.H., 1996. Quantitative description of the diet. Fisheries Techniques, second ed. American Fisheries Society, Bethesda, Maryland, pp. 513–532. Boyer, D.C., Hampton, I., 2001. An overview of the living marine resources of Namibia. South Afr. J. Mar. Sci. 23, 5–35. http://dx.doi.org/10.2989/ 025776101784528953. Burmeister, L., 2001. Depth-stratified density estimates and distribution of the cape hake Merluccius capensis and M. paradoxus off Namibia deduced from survey data, 1990–1999. South Afr. J. Mar. Sci. 23, 347–356. http://dx.doi. org/10.2989/025776101784528863. Cabral, H.N., Murta, A.G., 2002. The diet of blue whiting, hake, horse mackerel and mackerel off Portugal. J. Appl. Ichthyol. 18, 14–23. http://dx.doi.org/10. 1046/j.1439-0426.2002.00297.x. 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. Cury, P., Shannon, L., 2004. Regime shifts in upwelling ecosystems: Observed changes and possible mechanisms in the northern and southern Benguela. Prog. Oceanogr. 60, 223–243. http://dx.doi.org/10.1016/j.pocean.2004.02.007. 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. DeNiro, M.J., Epstein, S., 1978. Influence of diet on the distribution of carbon isotopes in animals. Geochim. Cosmochim. Acta 42, 495–506. http://dx.doi. org/10.1016/0016-7037(78)90199-0. Deudero, S., Pinnegar, J.K., Polunin, N.V.C., Morey, G., Morales-Nin, B., 2004. Spatial variation and ontogenic shifts in the isotopic composition of Mediterranean littoral fishes. Mar. Biol. 145, 971–981. http://dx.doi.org/10.1007/ s00227-004-1374-y. Endjambi, T., Iitembu, J.A., Nashima, F.P., 2015. Trophic relationships of three deep-sea crustacean species (Aristeus varidens, Plesionika martia and Funchalia woodwardi) off Namibia. Int. Sci. Technol. J. Namibia 7, 3–13. Erasmus, V.N., 2015. Trophic Relationships of Prey Species (Etrumeus whiteheadi, Myctophidae and Sepia) and the Predators (Merluccius capensis, Merluccius paradoxus, Lophius vomerinus and Trachurus capensis) Off Namibia (MSc thesis). The University of Namibia, Namibia, http://hdl.handle.net/11070/ 1534. Estrada, J.A., Rice, A.N., Natanson, L.J., Skomal, G.B., 2006. Use of isotopic analysis of vertebrae in reconstructiong ontogenetic feeding ecology in white sharks. Ecology 87, 829–834. http://dx.doi.org/10.1890/0012-9658(2006) 87[829:UOIAOV]2.0.CO;2.

9

Fanelli, E., Cartes, J.E., 2010. Temporal variations in the feeding habits and trophic levels of three deep-sea demersal fishes from the western Mediterranean Sea, based on stomach contents and stable isotope analyses. Mar. Ecol. Prog. Ser. 402, 213–232. http://dx.doi.org/10.3354/meps08421. Flynn, B.A., Richardson, A.J., Brierley, A.S., Boyer, D.C., Axelsen, B.E., Scott, L., Moroff, N.E., Kainge, P.I., Tjizoo, B.M., Gibbons, M.J., 2012. Temporal and spatial patterns in the abundance of Jellyfish in the Northern Benguela upwelling ecosystem and their link to Thwarted Pelagic Fishery recovery. Afr. J. Mar. Sci. 34, 131–146. http://dx.doi.org/10.2989/1814232X.2012.675122. France, R.L., 1995. Carbon-13 enrichment in benthic compared to planktonic algae: foodweb implications. Mar. Ecol. Prog. Ser. 124, 307–312. http://dx. doi.org/10.3354/meps124307. Goldman, S.F., Sedberry, G.R., 2011. Feeding habits of some demersal fish on the Charleston Bump off the southeastern United States. ICES J. Mar. Sci. 68, 390–398. Hacunda, J.S., 1981. Trophic relationships among demersal fishes in a coastal area of the Gulf of Maine. US. Fish. Bull. 79, 775–788. Heymans, J.J., Shannon, L.J., Jarre, A., 2004. Changes in the northern Benguela ecosystem over three decades: 1970s, 1980s, and 1990s. Ecol. Model. 172, 175–195. http://dx.doi.org/10.1016/j.ecolmodel.2003.09.006. Hillebrand, H., 2004. On the generality of the Latitudinal diversity gradient. Am. Nat. 163, 192–211. http://dx.doi.org/10.1086/381004. Huenerlage, K., Buchholz, F., 2013. Krill of the northern Benguela Current and the Angola–Benguela frontal zone compared: Physiological performance and short-term starvation in Euphausia hanseni. J. Plankton Res. 35, 337–351. http://dx.doi.org/10.1093/plankt/fbs086. Hyslop, E.J., 1980. Stomach contents analysis: a review of methods and their application. J. Fish. Biol. 17, 411–429. http://dx.doi.org/10.1111/j.1095-8649. 1980.tb02775.x. Iilende, T., Stromme, T., Johnsen, E., 2001. Dynamics of the pelagic component of Namibian hake stocks. South Afr. J. Mar. Sci. 23, 337–346. http://dx.doi. org/10.2989/025776101784529024. Iitembu, J.A., Miller, T.W., Ohmori, K., Kanime, A., Wells, S., 2012. Comparison of ontogenetic trophic shift in two hake species, Merluccius capensis and Merluccius paradoxus, from the Northern Benguela Current ecosystem (Namibia) using stable isotope analysis. Fish. Oceanogr. 21, 215–225. http: //dx.doi.org/10.1111/j.1365-2419.2012.00614.x. Jackson, A.L., Inger, R., Parnell, A.C., Bearhop, S., 2011. Comparing isotopic niche widths among and within communities: SIBER - Stable Isotope Bayesian Ellipses in R. J. Anim. Ecol. 80, 595–602. http://dx.doi.org/10.1111/j.13652656.2011.01806.x. Jansen, T., Kainge, P., Singh, L., Wilhelm, M., Durholtz, D., Strømme, T., Kathena, J., Erasmus, V., 2015. Spawning patterns of shallow-water hake (Merluccius capensis) and deep-water hake (M. paradoxus) in the Benguela Current Large Marine Ecosystem inferred from gonadosomatic indices. Fish. Res. 172, 168–180. http://dx.doi.org/10.1016/j.fishres.2015.07.009. Jansen, T., Kristensen, K., Kainge, P., Durholtz, D., Strømme, T., Thygesen, U.H., Wilhelm, M.R., Kathena, J., Fairweather, T.P., Paulus, S., Degel, H., Lipinski, M.R., Beyer, J.E., 2016. Migration, distribution and population (stock) structure of shallow-water hake (Merluccius capensis) in the Benguela Current Large Marine Ecosystem inferred using a geostatistical population model. Fish. Res. 179, 156–167. http://dx.doi.org/10.1016/j.fishres.2016.02.026. Jardas, I., Šantić, M., Pallaoro, A., 2004. Diet composition and feeding intensity of horse mackerel, Trachurus trachurus (Osteichthyes: Carangidae) in the eastern Adriatic. Mar. Biol. 144, 1051–1056. http://dx.doi.org/10.1007/s00227-0031281-7. Juanes, F., Buckel, J.A., Scharf, F.S., 2002. Feeding ecology of piscivorous fishes. In: Handbook of Fish Biology and Fisheries, Vol. 1. http://dx.doi.org/10.1002/ 9780470693803.ch12. Kaufman, D.M., 1995. Diversity of new world mammals: Universality of the latitudinal gradients of species and bauplans. J. Mammal. 76, 322–334. Kirchner, C., Bauleth-D’Almeida, G., Wilhelm, M.R., 2010. Assessment and management of cape horse mackerel Trachurus capensis off Namibia based on a fleet-disaggregated age-structured production model. Afr. J. Mar. Sci. 32, 525–541. http://dx.doi.org/10.2989/1814232X.2010.538144. Konchina, Y.V., 1986. Distribution and feeding of the South African horse mackerel and hake in the Namibian shelf waters. Collect. Sci. Pap. Int. Comm. South East. Atl. Fish. 13, 7–18. Krzeptowski, M., 1982. Trophic relationship between horse mackerel (Trachurus trachurus Castelnau) and cape hake (Merluccius capensis Castelnau) off Namibia. Collect. Sci. Pap. Int. Comm. South East. Atl. Fish. 9, 111–119. Kurle, C.M., Worthy, G.A.J., 2001. Stable isotope assessment of temporal and geographic differences in feeding ecology of northern fur seals (Callorhinus ursinus) and their prey. Oecologia 126, 254–265. http://dx.doi.org/10.1007/ s004420000518. Layman, C.A., Quattrochi, J.P., Peyer, C.M., Allgeier, J.E., 2007. Niche width collapse in a resilient top predator following ecosystem fragmentation. Ecol. Lett. 10, 937–944. http://dx.doi.org/10.1111/j.1461-0248.2007.01087.x. Macpherson, E., Gordoa, A., 1994. Effect of prey densities on cannibalism in cape hake (Merluccius capensis) off Namibia. Mar. Biol. 119, 145–149. http: //dx.doi.org/10.1007/BF00350116.

10

H.K. Kadila, D.N. Nakwaya, M. Butler et al. / Regional Studies in Marine Science 34 (2020) 101071

Mahe, K., Amara, R., Bryckaert, T., Kacher, M., Brylinski, J.M., 2007. Ontogenetic and spatial variation in the diet of hake (Merluccius merluccius) in the Bay of Biscay and the Celtic Sea. ICES J. Mar. Sci. 64, 1210–1219. http: //dx.doi.org/10.1093/icesjms/fsm100. Mehl, J.B., 1986. Acoustic resonance frequencies of deformed spherical resonators. II. J. Acoust. Soc. Am. 79, 278–285. http://dx.doi.org/10.1121/1. 393568. Newsome, S.D., Yeakel, J.D., Wheatley, P.V., Tinker, M.T., 2012. Tools for quantifying isotopic niche space and dietary variation at the individual and population level. J. Mammal. 93, 329–341. http://dx.doi.org/10.1644/11MAMM-S-187.1. Pagell, M., Shevchenko, A., 2014. Why research in sustainable supply chain management should have no future. J. Supply Chain Manage. 50, 44–55. http://dx.doi.org/10.1111/jscm.12037. Payne, A.I.L., Rose, B., Leslie, R.W., 1987. Feeding of hake and a first attempt at determining their trophic role in the South African west coast marine environment. South Afr. J. Mar. Sci. 5, 471–501. http://dx.doi.org/10.2989/ 025776187784522667. Peterson, B.J., Fry, B., 1987. Stable isotopes in ecosystem studies. Annu. Rev. Ecol. Syst. 18, 293–320. http://dx.doi.org/10.1146/annurev.es.18.110187.001453. Pillar, S.C., Barange, M., 1998. Feeding habits, daily ration and vertical migration of the cape horse mackerel off South Africa. South Afr. J. Mar. Sci. 19, 263–274. http://dx.doi.org/10.2989/025776198784126683. Pillar, S.C., Wilkinson, I.S., 1995. The diet of cape hake Merluccius capensis on the south coast of South Africa. South Afr. J. Mar. Sci. 15, 225–239. http://dx.doi.org/10.2989/02577619509504845. Pinkas, L., Oliphant, M., Iverson, L., 1971. Food Habits of Albacore Bluefin Tuna and Bonito in California Waters. University of California, https://escholarship. org/uc/item/7t5868rd. Post, D.M., Layman, C.A., Arrington, D.A., Takimoto, G., Quattrochi, J., Montaña, C.G., 2007. Getting to the fat of the matter: Models methods and assumptions for dealing with lipids in stable isotope analyses. Oecologia 152, 179–189. http://dx.doi.org/10.1007/s00442-006-0630-x. Punt, A.E., Leslie, R.W., Du Plessis, S.E., 1992. Estimation of the annual consumption of food by cape hake Merluccius capensis and M. paradoxus off the South African west coast. South Afr. J. Mar. Sci. 12, 611–634. http: //dx.doi.org/10.2989/02577619209504728. R Core Team, 2018. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, URL https://www.Rproject.org/. Roel, B.A., Macpherson, E., 1988. Feeding of Merluccius capensis and M. paradoxus off Namibia. South Afr. J. Mar. Sci. 6, 227–243. http://dx.doi.org/10.2989/ 025776188784480546.

Roux, J.-P., van der Lingen, C.D., Gibbons, M.J., Moroff, N., Shannon, L.J., Smith, A.D., Cury, P.M., 2013. Jellyfication of marine ecosystems as a consequence of overfishing small pelagic fish: lessons from the benguela. Bull. Mar. Sci. 89, 249–284. http://dx.doi.org/10.5343/bms.2011.1145. Soares, L.S.H., Arantes, L.P.L., Lamas, R.A., Lima, F.A., Pucci, M.C.J., RossiWongtschowski, C.L.D.B., 2018. Fish feeding interactions in a subtropical coastal system in the southwestern Atlantic. Ocean Coast. Manag. 0–1. http://dx.doi.org/10.1016/j.ocecoaman.2018.04.002. Sweeting, C.J., Polunin, N.V.C., Jennings, S., 2006. Effects of chemical lipid extraction and arithmetic lipid correction on stable isotope ratios of fish tissues. Rapid Commun. Mass Spectrom. 20, 595–601. http://dx.doi.org/10. 1002/rcm.2347. Utne-Palm, A.C., Salvanes, A.G.V., Currie, B., Kaartvedt, S., Nilsson, G.E., Braithwaite, V.A., Stecyk, J.A.W., Hundt, M., Bank, M. Van Der, Flynn, B., Sandvik, G.K., Klevjer, T.A., Sweetman, A.K., Brüchert, V., Pittman, K., Peard, K.R., Lunde, I.G., Strandaba, R.A.U., Gibbons, M.J., 2010. Trophic structure and community stability in an overfished ecosystem. Science 329 (80), 333–336. http://dx.doi.org/10.1126/science.1190708. Van Der Bank, M.G., Utne-Palm, A.C., Pittman, K., Sweetman, A.K., Richoux, N.B., Brüchert, V., Gibbons, M.J., 2011. Dietary success of a ‘‘new’’ key fish in an overfished ecosystem: Evidence from fatty acid and stable isotope signatures. Mar. Ecol. Prog. Ser. 428, 219–233. http://dx.doi.org/10.3354/meps09078. Varela, J.L., Intriago, K.M., Flores, J.C., Lucas-Pilozo, C.R., 2017. Feeding habits of juvenile yellowfin tuna (Thunnus albacares) in Ecuadorian waters assessed from stomach content and stable isotope analysis. Fish. Res. 194, 89–98. http://dx.doi.org/10.1016/j.fishres.2017.05.017. Varela, L.J., Sorell, J.M., Macías, D., Goñi, N., Arrizabalaga, H., Medina, A., 2018. New insight into the trophic biology of age-0 Atlantic bluefin tuna in the western Mediterranean using stomach content and stable isotope analyses. Fish. Res. 208, 274–285. http://dx.doi.org/10.1016/j.fishres.2018.08.004. Vignon, M., Dierking, J., 2011. Prey regurgitation and stomach vacuity among groupers and snappers. Environ. Biol. Fish. 90, 361–366. http://dx.doi.org/10. 1007/s10641-010-9746-2. Vinson, M.R., Angradi, T.R., 2011. Stomach emptiness in fishes : Sources of variation and study design implications stomach emptiness in fishes : Sources of variation and study. Rev. Fish. Sci. 19, 63–73. http://dx.doi.org/ 10.1080/10641262.2010.536856. Wilhelm, M.R., Moloney, C.L., Paulus, S., Roux, J.P., 2017. Fast growth inferred for northern Benguela shallow-water hake Merluccius capensis using annual survey- and monthly commercial length-frequency distributions. Fish. Res. 193, 7–14. http://dx.doi.org/10.1016/j.fishres.2017.03.001. Willig, M.R., Kaufman, D.M., Stevens, R.D., 2003. Latitudinal gradients of biodiversity: Pattern, process, scale, and synthesis. Annu. Rev. Ecol. Evol. Syst. 34, 273–309. http://dx.doi.org/10.1146/annurev.ecolsys.34.012103.144032.