Aquat. Living Resour. 21, 112 (2008)c EDP Sciences, IFREMER, IRD 2008DOI: 10.1051/alr:2008017www.alr-journal.org
Assessment of fish trophic status and relationships by stableisotope data in the coral reef lagoon of New Caledonia, southwestPacificLaure Carassou1,a, Michel Kulbicki1, Thomas J.R. Nicola2 and Nicholas V.C. Polunin2
1 Institute of Research for Development, BP A5, 98848 Nouma Cedex, New Caledonia2 Department of Marine Sciences and Coastal Management, University of Newcastle, Newcastle upon Tyne, UK
Received 25 September 2007; Accepted 20 February 2008
Abstract This study examines the trophic status and relationships of coral reef fish in the lagoon of New Caledonia,southwest pacific. The feeding habits of 34 fish species collected at three contrasted sites were first described using acompilation of gut contents observations and data from the literature. The carbon and nitrogen isotope signatures ofthese fish and of some of their potential ultimate food sources were also determined at each site. Despite some spatialvariations in the isotopic signatures of most food sources and fish trophic groups, the overall trophic structures of fishassemblages were similar at the three sites. Stable isotope data were then used to re-assign fish species to trophic groupsbased on the 15N signatures of fish and their food sources. Herbivorous fish species were clearly distinguished from theother trophic groups by their lower 15N signatures, consistent with an estimated trophic position of 2 for all speciesexamined. Scaridae were however characterized by relatively higher 13C and lower 15N, which is probably linked withthe role of detritus in their diet. The estimated trophic positions of planktivorous fish species were consistent with theirgut contents-based classification. Conversely, the isotopic signatures of carnivorous and piscivorous fish species largelyoverlapped, and their estimated trophic positions were much lower than expected. This suggests that these species feedover a broader range of trophic levels and food sources than implied by the gut contents observations, and indicatesthat their diet is partly omnivorous. Finally, the relationships between body mass and the isotopic signatures of four fishspecies were significant for at least one isotopic ratio for each species. Since ontogenetic variations and omnivorousdiets are difficult to assess with gut contents data only, stable isotopes revealed essential in estimating the actual trophicstatus and relationships characterizing the fish species under study.
Key words: Coral reef fish / Gut contents / Stable isotopes / Trophic position / Ontogenetic variability / Pacific Ocean
Rsum Utilisation des isotopes stables pour lestimation du rgime alimentaire et des relations trophiquesdes poissons du lagon de Nouvelle-Caldonie (Pacifique Sud-Ouest). Cette tude fournit une analyse du rgimealimentaire et des relations trophiques qui caractrisent les poissons du lagon de Nouvelle-Caldonie. Le rgime ali-mentaire de 34 espces, rcoltes dans trois stations contrastes, est dcrit par la mthode des contenus stomacaux enutilisant une large base de donnes locales, associe une tude bibliographique. Les isotopes stables du carbone etde lazote sont ensuite mesurs sur les poissons rcolts et sur diffrentes sources alimentaires potentielles prleves chaque station. Malgr des diffrences spatiales significatives pour la plupart des sources alimentaires et des groupestrophiques de poissons, la structure trophique globale des assemblages se rvle peu variable dune station lautre.Les donnes isotopiques sont ensuite utilises pour rassigner les espces de poissons diffrents groupes trophiques,sur la base du rapport entre le 15N des espces et de celui de leurs sources alimentaires. Les donnes isotopiquesconfirment les rsultats des analyses de contenus stomacaux pour les herbivores et les planctivores. Les Scaridae sontcependant caractriss par des valeurs particulirement leves en 13C et faibles en 15N, probablement lies unrgime alimentaire en partie dtritivore. A linverse, pour la plupart des espces carnivores et piscivores, les donnesisotopiques mettent en vidence une gamme de sources alimentaires plus diversifie que celle fournie par les tudesde contenus stomacaux, et un comportement alimentaire plutt omnivore. Enfin, la relation entre la masse corporelleet les ratios isotopiques, teste chez quatre espces, montre des variations ontogniques significatives pour au moins
a Corresponding author: firstname.lastname@example.org
Article published by EDP Sciences and available at http://www.alr-journal.org or http://dx.doi.org/10.1051/alr:2008017
2 L. Carassou et al.: Aquat. Living Resour. 21, 112 (2008)
un des isotopes tudis (carbone ou azote) pour les quatre espces, mais pas de changement de niveau trophique. Lesvariations ontogniques et les comportements alimentaires de type omnivore tant difficiles valuer par les seulesdonnes de contenus stomacaux, les isotopes stables se rvlent cruciaux pour lestimation des niveaux trophiques relsdes espces et la reprsentation des relations trophiques au sein des assemblages.
The structure and functioning of fish assemblages relyin part on the trophic status and relationships characterizingfish species. The understanding of fish trophic networks hasthus become increasingly important for the establishment ofecologically based management programmes (Kulbicki et al.2005). In ecosystems characterized by a very high species di-versity, such as coral reefs, information on the diet of fishspecies remains incomplete and difficult to assess. Fish dietsare often studied using gut contents, which give no informationabout the origin of the food sources consumed and are repre-sentative only of the food ingested by fish at the time of sam-pling (Pinnegar and Polunin 1999). Conversely, the analysis ofthe stable isotope signatures of prey and predators provides in-formation about the food-web structure and energy flow overlong time periods (Pinnegar and Polunin 2000; Phillips andGregg 2003). Indeed, the ratio of 13C/12C, noted 13C, in-creases by less than 1.0% on average per trophic level (Post2002), and can thus be used to discriminate between the car-bon sources used by consumers (Peterson 1999). The ratio of15N/14N, noted 15N, increases from diet to consumers witha highly variable enrichment level, comprised between 2 and5% (Olive et al. 2003), a mean enrichment of 3.4% beinggenerally admitted (Post 2002). Despite its variability, the 15Nenrichment is consistent throughout successive trophic levels,and can thus provide a measure of trophic position (Post 2002).
In New Caledonia, southwest Pacific, the lagoon extendson an area of 19 000 km2, thus providing a wide range of envi-ronmental conditions, from highly productive coastal ecosys-tems, influenced by terrigeneous inputs (Tenrio et al. 2005)with high chl a concentrations (Pinazo et al. 2004) and zoo-plankton densities (Champalbert 1993), to barrier reef ecosys-tems influenced by oligotrophic oceanic water flows (Pinazoet al. 2004). This strong environmental variability and theproximity of New Caledonia to the coral reefs biodiversitycenter located in the China Sea-Philippines-Indonesia regionresult in highly diverse shorefish assemblages in the lagoon,with a total of 1694 species (Fricke and Kulbicki 2006). How-ever, little is known about the trophic relationships that supportthis highly diverse shorefish fauna, since the few studies avail-able have only considered specific trophic groups or only usedgut contents data (Grimaud and Kulbicki 1998; Bozec et al.2005; Kulbicki et al. 2005).
The principal aim of the present study was to assess the fishtrophic status and relationships in the lagoon of New Caledo-nia through the use of stable isotope data, while taking intoaccount the high spatial heterogeneity of the lagoon.
This study was therefore conducted in four stages. The fish species collected at three sites, chosen for their
contrasted environmental conditions, were assigned to trophicgroups, relying on previous gut contents data and on a synthe-sis from the literature.
The isotopic signatures of those fish and of some of theirpotential food sources were assessed at each site, and com-pared between sites, in order to determine whether the trophicfunctioning of fish assemblages varied spatially. The fish species were re-assigned to trophic groups based
on stable isotope analyses, in order to check whether any po-tential food sources had been neglected by previous gut con-tents analyses. The relationships between 13C and 15N signatures and
body mass were tested for four fish species, so as to detect ifan ontogenetic variability occurred in their diet.
2 Material and methods
2.1 Sampling sites
Observations from hydrodynamic 3D models developedat the Institute of Research for the Development (IRD) wereused in order to select three sampling sites where water col-umn conditions were representative of the contrasts in watercolumn productivity characterizing the lagoon of New Cale-donia (see Ouillon et al. 2005; Jouon et al. 2006 for recentapplications of the models). The three chosen sites were sep-arated by distances of 11 to 14 km, and included: a fringingreef site (Site 1: 2010.0S 16620.3E), an intermediate reefsite (Site 2: 2019.2S 16623.5E) and a barrier reef site (Site3: 2212.3S 16617.7E), lying approximately 20 km from thecoast (Fig. 1).
All samples were collected within a two month period,from July to August 1996, in order to reduce the temporalvariations in the isotopic signatures of the organisms (Owens1987). Except for spanish mackerel Scomberomorus commer-son, provided by the South Pacific Commiss