Diet composition of New Zealand’s endemic rough skate, Zearaja nasuta

ABSTRACT

Diet analysis can provide an insight into the structure and function of an ecosystem, and can be used in ecosystem-based frameworks to inform management and conservation decisions. Diet composition of rough skate (Zearaja nausta) from the south east coast of New Zealand, was investigated for the first time. We examined 35 stomachs from three trawls between March and June 2017. Prey importance was assessed by a prey specific index of relative importance (PSIRI). This population of rough skates (n = 32) was found to be specialised feeders, primarily preying on one species, Nectocarcinus antarcticus. This study provides the first insight into rough skate trophic interactions in the surrounding marine community.

KEYWORDS:

• Feeding habits
• stomach content analysis
• elasmobranch
• predator
• PSIRI

Introduction

The rough skate, Zearaja nasuta (Müller & Henle 1841) is the most common endemic skate in New Zealand waters and an increasingly important commercial fish species (Ministry for Primary Industries 2017). Managed in New Zealand’s Quota Management System (QMS) with a second skate species, smooth skate, Dipturus innominatus, rough skate reported landings have increased from 325 to 1,553 tonne between 1996–97 and 2015–16 (Ministry for Primary Industries, 2017). A recent qualitative risk assessment of New Zealand elasmobranchs suggested rough skates are among the most vulnerable species to overexploitation, due to its high range overlap with commercial fishing grounds, being caught for more than 300 days a year and a lack of information pertaining to ecological interactions (Ford et al. 2018). Recently, regional fisheries management organisations have recognised the value of diet studies as one of the main components supporting ecosystem-based fisheries management frameworks (e.g. Juan-Jordá et al. 2018). Diet descriptions can also provide insight into the structure and functioning of ecological communities by quantifying interactions between a predator and its prey (Caddy and Sharp 1986; Cortés 1999). Therefore, understanding feeding patterns and the ecological role played by respective species in ecosystems are imperative to develop smarter management strategies and make smarter conservation decisions (De Ruiter et al. 2005). No dietary information is available for rough skates in any location surrounding New Zealand and without information pertaining to trophic interactions and feeding ecology, the ecological role of this species remains unclear.

In this short communication, we describe the diet composition of rough skates from one geographical area over a three-month period using an index of relative importance. Although we are limited by the spatio-temporal scale of this study, the information is relevant in providing the first diet composition description for this species and first insight into rough skate trophic interactions with the surrounding marine community.

Materials and methods

Specimen collection and dissection

Rough skates were collected in waters around Nugget Point, off the eastern coast of New Zealand (approximately 46.4481°S, 169.8147°E) by a commercial bottom trawler on three occasions between March and June 2017. Collections occurred between <1–20 km offshore at a depth of approximately 40 metres.

Stomach analysis

The stomach state was recorded, being either empty or not empty. Any empty stomachs were excluded from the dietary analysis. Stomach contents were rinsed, counted, and weighed to the nearest gram. Contents were examined and prey items were identified to the lowest taxonomic level possible using reference guides (e.g. Naylor et al. 2005; Tracey et al. 2005; McMillan, Francis, et al. 2011; McMillan, Griggs, et al. 2011). Whenever fragments of prey items were found, the number of individuals was taken as the smallest possible number of individuals from which the fragments could have originated.

To determine the importance of each prey in the diet, the prey-specific index of relative importance (%PSIRI) (Brown et al. 2012) was used according to the equation: %PSIRI = 0.5 x %O x (%P_Ni + %P_Wi), where %O is the percent frequency of occurrence (the number of stomachs containing prey i divided by the total number of stomachs, n), and %P_Ni and %P_Wi are the prey-specific abundances by number and weight, respectively. Prey-specific abundance was calculated with the equation: %P_Ai = , where %A_ij is the abundance (by number %P_Ni or weight %P_Wi) of prey i in stomach sample j and n_i is the number of stomachs containing prey i. The %PSIRI is a modification of the index of relative importance (IRI) (Pinkas et al. 1971).

A cumulative prey curve was computed using the software EstimateS version 9.1.0 (Colwell 2013) to determine whether our sample size was sufficient to recover all prey items comprising rough skate diet. Diversity settings were set to 100 randomizations and rarefaction curve was extrapolated a further 32 samples. Adequacy of the number of stomachs sampled was assessed using the slope of the linear regression through the last five sub-samples or through five consecutive sub-samples until the slope reached 0.05 (considered adequately close to the asymptote as per Bizzaro et al. 2009). The feeding strategy was identified by plotting the prey-specific abundance by number (%P_Ni) of the three main prey groups (Malacostraca, Actinopterygii and Cephalopoda) against the mean %O (Costello 1990; Amundsen et al. 1996). In this method, any prey located close to 100% P_Ni and 100% O represents specialisation.

Results

Sample description

In total, 35 rough skate stomachs were collected, of which three (8.6%) were empty. The number of skates collected per sampling date ranged from 10 to 14 individuals. Twelve specimens were male (TL 56.0 to 85.0 cm) and 23 were female (TL 71.0 to 94.0 cm).

Diet composition %P_Ni and %P_Wi

In total, 13 prey species were identified, including five species of Malacostraca, six species of Actinopterygii, and two species of Cephalopoda (Table 1). In terms of %PSIRI, rough skates fed primarily on Malacostraca (85.2%), while Actinopterygii (13.0%) and Cephalopoda (1.9%) had a much lower contribution to the diet. Additionally, Malacostraca dominated the diet in terms of %P_Ni (95.2%), %O (93.8%), and %P_Wi (86.5%). Among Malacostraca, one species of crab, Nectocarcinus antarcticus contributed the most to the diet: %P_Ni (87.8%), %O (84.4%), %P_Wi (79.2%), and %PSIRI (70.5%) (Table 1). The second most important prey species to the diet of rough skate was another species of Malacostraca, Pterygosquilla armata, with a %PSIRI of 5.8%. Actinopterygii accounted for considerably less of the overall diet with %P_Ni, %O, and %P_Wi of 29.9%, 31.3%, and 51.0%, respectively (Table 1). Cephalopoda contributed less to the overall diet of rough skate in terms of all dietary indices (%P_Ni = 8.3%, %O = 9.4%, %P_Wi = 31.6%).

Table 1. Diet composition of Z. nasuta off the coast of Nugget Point in terms of frequency of occurrence (%O), per cent prey-specific abundance by number (%P_Ni), prey-specific abundance by weight (%P_Wi), and prey-specific index of relative importance (%PSIRI), where N is number of prey and n_i is number of stomachs containing specific prey.

Prey taxa	N	Weight (g)	ni	%O	%PNi	%PWi	%PSIRI
Malacostraca	248	1011.2	30	93.75	95.18	86.50	85.16
N. antarcticus	217	883.3	27	84.38	87.84	79.22	70.48
Pandalidae Gen. sp. (unidentified)	22	12.6	4	12.50	49.17	26.21	4.71
P. armata	7	37.7	7	21.88	24.32	28.98	5.83
Leptomithrax garricki	1	35.5	1	3.13	100.00	100.00	3.13
Neommatocarcinus huttoni	1	42.1	1	3.13	16.67	48.39	1.02
Actinopterygii	13	820.5	10	31.25	29.09	51.01	12.97
Sprattus antipodum	4	11.6	2	6.25	8.33	16.09	0.76
Genypterus blacodes	2	107.8	2	6.25	56.25	87.11	4.48
Peltorhamphus novaezeelandiae	2	323.2	1	3.13	50.00	94.67	2.26
Trachurus novaezelandiae	2	303.6	2	6.25	55.56	68.99	3.89
Macruronus novaezelandiae	1	64.5	1	3.13	14.29	50.87	1.02
Unidentified teleost sp.	2	9.8	2	6.25	7.69	10.10	0.56
Cephalopoda	3	45.1	3	9.38	8.25	31.64	1.87
Octopoda spp.	2	39.9	2	6.25	8.81	38.63	1.48
Cranchiidae spp.	1	5.2	1	3.13	7.14	17.66	0.39

Feeding strategy

Overall, rough skates show specialisation on Malacostraca with %P_Ni of 95.2% and %O of 93.8% (Figure 1). Few skates preyed on both Cephalopoda and Actinopterygii, respectively, represented by P_Ni of less than 30% for both groups.

Figure 1. Graphical representation of the feeding strategy of rough skates. Prey specific abundance (%P_Ni) plotted against percentage frequency of occurrence (%O).

Adequacy of sample size

Our cumulative prey curve failed to reach asymptote at 32 stomachs and 13 prey items. Rarefaction curve estimated that between 14 and 15 prey items would comprise rough skate diet at 55 stomachs samples (when our projected curve would reach a slope of 0.0475).

Discussion

Our study shows that off the shore of Nugget Point and between the months of March and June, New Zealand’s endemic rough skate has a specialised diet, primarily consuming one species of crab, N. antarcticus. A crustacean dominated diet is a trait shared by many other small bodied skates (e.g. Ebert et al. 1991; Walmsley-Hart et al. 1999; Braccini and Perez 2005; San Martin et al. 2007; Carrier et al. 2012). All other prey items contributed little to the overall diet, possibly a result of opportunistic or incidental prey capture.

The rough skates examined in this study have a continuous feeding strategy. This is based on observations of few empty stomachs and different levels of prey digestion (Pers Obs.). Continuous feeding strategies are evident in many skate species (e.g. Wetherbee et al. 1990, 2004; Ebert et al. 1991; Muto et al. 2001; Motta 2004; Wilga et al. 2012) and is possibly explained by the abundance of readily available prey living on the surface and subsurface layer of the benthos (Wetherbee et al. 1990, 2004; Jacobsen and Bennett 2013).

A small sample size and large variation in food categories can alter the dietary importance of each prey items (Hyslop 1980). The minimum number of stomachs required for a precise description of a predator’s diet should be determined. For rough skates, 55 stomachs are required to reach a predicted stable prey diversity asymptote suggesting our sample size does not capture the true diversity of rough skate diet. Our study does, however, allow the first insight into the diet composition of rough skates off the coast of Nugget point between autumn and winter months for 2017. Additional sampling in other years or other months would be necessary to make broader generalisations and to reach a sufficient sample size, although we would expect only one or two additional prey items to be recorded by increased sampling.

Given that our samples were obtained from a commercial fisherman, it was not possible to investigate ontogenetic shifts in feeding habits of rough skates, as our sample showed a significant size bias towards larger adult specimens. In many skate species, as they grow, their diet tends to shift towards larger and heavier prey items (e.g. Wetherbee et al. 2004; Treloar et al. 2007; Lucifora et al. 2009; Carrier et al. 2012; Forman and Dunn 2012; Kadri et al. 2013). Typically, skates feed on crustaceans and move to a more teleost-based diet as they grow (e.g. Treloar et al. 2007; Carrier et al. 2012; Forman and Dunn 2012). We would expect rough skate to undergo ontogenetic shift in diet as they grow and mature, but this will need to be investigated further.

The dietary information presented in this study provides a snapshot of the diet composition of rough skates between the autumn and winter months of a given year. It is very likely that rough skates will exhibit seasonal variation in feeding ecology, such that reflects environmental changes that influence the distribution and abundance of prey items. Feeding patterns will differ at other times of the year as found in other skate species (Motta 2004). However, the benthic invertebrate fauna and its seasonal patterns off the southeast coast of New Zealand are poorly studied with only brief reports on the communities from these areas (e.g. Probert et al. 1979). N. antarcticus is an endemic species that is common and abundant at depths of 20–60 m (Main 1974), but whose relative and seasonal abundances have yet to be assessed.

Rough skates are an increasingly important resource within New Zealand’s commercial fisheries (Ministry for Primary Industries 2017). According to the most recent fisheries report, it is not yet known if the current catch levels or total allowable catch are sustainable (Ministry for Primary Industries 2017). This information lacks as a result of under researched ecological parameters pertaining to population productivity and life history characteristics. Dietary studies are an important step in supporting ecosystem-based fisheries management (Brodeur et al. 2017; Livingston et al. 2017; Rohan and Buckley 2017). Understanding basic dietary composition and trophic interactions of a predator and its prey within a localised environment may help prevent the collapse of, or reduce the risks of, local extirpation of these species while keeping fishing levels at a sustainable level.

In summary, the rough skate population located off Nugget Point, around the east coast of the South Island of New Zealand, during the autumn and winter months is a specialised and continuous feeder that primarily feeds on crustaceans, and in particular on a single species of crab, N. antarcticus. These results suggest that rough skates might play a unique functional role in New Zealand’s benthic food web due to its strong predation interaction with N. antarcticus. Despite rough skate’s potential food web importance and vulnerability to overexploitation (Ford et al. 2018), its life history characteristics and overall ecological role is still poorly known within this environment, however we now have basic information pertaining to diet composition and trophic interactions of this species.

Acknowledgements

The authors are grateful to Gavin Heineman for collecting the skate specimens, to Fatima Jorge for guidance with the analyses and to Olwyn Friesen, Brandon Ruhele and Bronwen Presswell for help with the skate dissections. We also thank members of the Evolutionary and Ecological Parasitology Research Group for their comments on the early daft of this manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Jerusha Bennett http://orcid.org/0000-0003-1037-894X

Haseeb S. Randhawa http://orcid.org/0000-0001-6262-970X

Additional information

Funding

This work was financially supported through a University of Otago MSc student research budget provided by the Zoology Department and Ecology Degree Programme.