Fisheries Research 179 (2016) 104–114 Contents lists available at ScienceDirect Fisheries Research journal homepage: www.elsevier.com/locate/fishres The usefulness of otolith chemistry to determine the life history of the honeycomb grouper around Reunion Island (SW Indian Ocean) A. Riou a,b , G. Bareille c,∗ , F. Morat d , K. Pothin b , N. Bru e , P. Chabanet a a UMR IRD-UR-CNRS ENTROPIE, Laboratoire d’Excellence Labex CORAIL, Institut de Recherche pour le Développement, CS 41095, Ste-Clotilde, 97495 Réunion Island, France b Réserve Naturelle Marine de la Réunion, 39 rue du Lagon, Lot Dayot1, 97434, La Saline Les Bains, Reunion, France c LCABIE, UMR 5254 IPREM UPPA/CNRS, Hélioparc Pau Pyrénées, 2 Avenue du Président Angot, F-64053 PAU Cédex 9, France d Irstea-Centre d’Aix en Provence, 3275 Route de Cézanne, CS 40061, 13182 Aix en Provence Cedex 5, France e Laboratoire de Mathématiques et de leurs Applications de Pau (LMAP), UMR CNRS 5142, F-64600 Anglet, France a r t i c l e i n f o Article history: Received 17 July 2015 Received in revised form 3 February 2016 Accepted 9 February 2016 Keywords: LA-ICP-MS Barium Epinephelus merra Pelagic larval phase Reef a b s t r a c t Due to the relative geographical isolation of Reunion Island the grouper Epinephelus merra is assumed to be self-recruited but almost nothing is known about its larval history. We used elemental composi- tion of the otoliths (Ba, Sr, Mn, Mg, Na) of fifty-eight one year-old groupers collected from four main coral reefs on the west coast of Reunion Island, to determine environmental variations during their early life history. Hierarchical clustering analysis of the otolith chemical composition of core and early larval phase, allowed the identification of three groups of larvae. Larvae from the first group (cluster 1) were born and dispersed in Ba-enriched water bodies, while the second and third groups included larvae that had crossed Ba-poor water masses. Larvae of group 1 were primarily found in the southern sector (74%) decreasing northward, while group 3 were more abundant in the northern sector (56%) and cluster 2 showed a scattered distribution. Such opposite spatial distributions might suggest that clusters 1 and 3 originated from opposite dispersal kernels, south and north respectively. It is possible however that larvae from both groups came from the same spawning ground but represented different cohorts that expe- rienced changing geochemical conditions either at the spawning site or within the surrounding pelagic environment, over the course of the breeding season. Both scenarios on the origin and dispersion may thus suggest significant influence of hydrographic features causing or preventing larvae from becoming displaced far from their natal area. © 2016 Elsevier B.V. All rights reserved. 1. Introduction The pelagic phase of many reef fish, either as eggs or larvae, enables larvae to be dispersed over vast distances by ocean cur- rents (Doherty et al., 1985; Shulman and Bermingham, 1995), thus creating an open population with the arrival of recruits from local or non-local sources (Roughgarden et al., 1988; Caley et al., 1996). Understanding larval dispersal patterns and the recruitment vari- ability is thus fundamental to further understanding the population dynamics and implementing effective management of marine har- vested species (Cowen and Sponaugle, 2009; Sale et al., 2010). ∗ Corresponding author. Fax: +33 559 407 781. E-mail addresses: antoine.riou@ird.fr (A. Riou), gilles.bareille@univ-pau.fr (G. Bareille), fabien.morat@irstea.fr (F. Morat), karine.pothin@reservemarinereunion.fr (K. Pothin), noelle.bru@univ-pau.fr (N. Bru), pascale.chabanet@ird.fr (P. Chabanet). Identifying the dispersal history of settled reef fish is partic- ularly challenging due to the difficulties in tracking them during their pelagic larval stage. Assigning geographical origin and migra- tory pathways to the larvae of a recruitment site requires that potential sources of interest are known (i.e., spawning aggrega- tions). However, this is not always the case, especially in the marine environment as complete sampling is not realistic and spawn- ing grounds are not well documented for all marine fish species (Campana et al., 2000). During their dispersal stage, larvae can be expected to inhabit various heterogeneous environments for extended periods, whether due to food and/or physicochemical conditions (temper- ature, salinity, water chemistry composition). Variable conditions experienced by larvae in the pelagic environment may thus influ- ence their phenotypic variability (size, morphology, pelagic larval duration, growth rate, etc.,) as well as the content of chemi- cal elements archived in the daily incremental growth of their otoliths. This aragonite structure from the inner ear of fish, grows http://dx.doi.org/10.1016/j.fishres.2016.02.007 0165-7836/© 2016 Elsevier B.V. All rights reserved.