SPATIAL VARIATION IN OTOLITH CHEMISTRY OF EPINEPHELUS MERRA IN BAAATOLL, MALDIVES BY GERALDINE CRIQUET,1 LAURENT VIGLIOLA,2AND PASCALE CHABANET 1 INTRODUCTION The particular geography of Maldives archipelago induces that the economy is essentially dependant on marine resources. Despite the fact that pelagic resources are the most exploited, reeffin fishery strongly increased due to tourism and export industry development in the last decades (Sattar et al this issue). Although there are no clear ., signs of overexploitation and decline of fish population overall, Sattar et al. (this issue) have concerns for some groupers species in some locations. Chabanet et al. (this issue) also report no sighting of sharks in Baa for instance. Thus, Maldivian government and stakeholders remain vigilant, and several MPAs were established in various atolls. The ww m goal ofthe Atoll Ecosystem Conservation Project (http:// w.biodiversitv. v/aec/) aims to establish a network of conservation for Baa atoll for instance. Anthropogenic activities have strongly affected coral reef ecosystems worldwide and Marine Protected Areas (MPAs) are increasingly considered as an effective tool for fishery management (Roberts et al 2001) and biodiversity conservation (Micheli ., et al., 2004). To design MPAs and soundly manage fish stocks, it is vital to obtain information on connectivity between populations, migration and movement, and describe A population dynamics. variety of methods have been used to assess movements between populations such as population genetic (Bradbury et al 2008), hydrodynamic and ., lagrangian dispersal models (Cowen et al 2000), external tags (Willis et al 2001, ., ., Sattar et al this issue), acoustic telemetry (Topping et al 2005) or transgenerational ., ., marking of embryonic otolith (Thorrold et al 2006). Recently, elemental composition ., of fish otoliths has been successfully used to examine connectivity between populations (Milton and Cheney, 2001; Chittaro et al 2006). Elemental chemistry is used because ., trace elements from the environment are incorporated into the otolith during growth and the otolith is metabolically inert (Campana, 1999). According to Hamer et al;(2003), the success of using otolith chemistry to measure connectivity would be dependant on a detectable level of chemical variation at biologically relevant spatial scales. The present study investigated whether otoliths ofEpinephelus merra collected from different coral reefs in Baa Atoll could be discriminated on the basis of multielemental chemistry. The study aimed at evaluating the potential of otolith microchemistry technique to study connectivity among reefs at Baa Atoll, and brings on MPAs the long run key data for selection. TRD Reunion, UMR 227, BP 50172, 97492 Ste Clotilde cedex, La Reunion UMR 2IRD Noumea, 227, BPA5, 98800 Noumea cedex, New-Caledonia 202 MATERIAL AND METHODS Study Site and Sampling The Republic of Maldives is an archipelago of 26 atolls located in central Indian Ocean. Baa Atoll, circa 40 km long and wide, is located in the western side ofthe Maldives archipelago and ranges from 4°49’N and 5°23'N and 73°06’E (Fig. 1). The survey presented here took part during a biodiversity census survey conducted in May-June 2009, on lagoon and outer slope habitats in Baa Atoll (Fig. 1). For this study, a total of 93 E. merra were collected at 6 stations: stations 3 and 28 (lagoon reef flat), 6 and 21 (lagoon reef slope), 17 (pass slope) and 20 (reef flat under oceanic influence). Fish were stored in ice immediately after capture and dissected within one hour of collection. BAA ATOLL o 10 Km Zl Indian Ocean Figure 1 Map ofBaaAtoll and position ofE. merra sampling stations. . 203 Laboratory Analysis mm For each individual, total length (TL) was measured to the nearest and sagittal otoliths were extracted with acid-washed plastic tweezers, cleaned of adhering tissues in ultrapure water and stored dry in acid-washed eppendorftubes. At the lab, otoliths were cleaned of organic material by soaking in an equal-volume mixture of 30% ultrapure H,0 and 0. 1 mol L'1 ultrapure NaOH for 1 hour. Each otolith was then ultrasonicated 0 for five minutes and rinsed with ultrapure water and individually soaked in five separated 5-min baths of ultrapure water. After the fifth bath, otoliths were air-dried in a HEPA- hltered class 100 laminar flow hood. Whole otoliths were placed on double-sided tape just before laser ablation inductively coupled plasma mass spectrometry (LA-ICPMS) analysis (Warner et al., 2005). Ten otoliths per station were randomly chosen (except for station N=3) and 6, ablated with the LA-ICPMS at the edge of otolith postrostrum. All analyzes were (UMR conducted at the University ofMontpellier 2 5243 Geosciences), using a 193 nm Excimer Laser System (CompEx 102, LambdaPhysiks) coupled to a Element XR sector field ICPMS 5thermoFisher). For all otoliths, the laser beam diameter was set at 5 pm and the laser was operated with a repetition rate of 4 Hz at 15 J.cnr2 Helium 1 . was used as the ablation gas, to enhance sensitivity and reduce particle condensation on the surface. An Argon gas flow was then admixed to the laser-generated aerosol, prior to introduction into the ICPMS for elemental analysis. The instrument was operated in low mass resolution and calcium was used as an internal standard using a stoichiometric A value of 56% CaO. standard reference glass material (NIST 612) was used to calibrate analyzes and to control for instrumental drift. Apart from Ca, 20 elements were Mn measured: Li, B, Mg, Ti, Cr, Co, Ni, Cu, Zn, Rb, Sr, Mo, Cd, Sn, Ba, Ce, Pb, Th and U. Calculation of drift and limits of detection (LOD) were made off-line using the Glitter software. Elements for which 25% ofthe measures were below LOD irrespective of fish origin were removed from further analysis. Remaining elements were: Li, B, Mg, Mn, Co, Ni, Zn, Sr, Cd, Sn and Ba. Data were expressed as ratios to calcium and element LOD concentration below were set to zero. Statistical Analysis We used discriminant function analyses (DFAs) to examine spatial patterns of multielemental chemical signature of otoliths. Element ratios that contributed most to the discrimination among sites (Sr/Ca, Mg/Ca, Mn/Ca and Ba/Ca) were further analyzed by ANOVAs non-parametric (Kruskall-Wallis) in order to investigate spatial differences. 204 RESULTS AND DISCUSSION There was no significant difference in fish total length between sampling stations (Table 1). DFA successfully reclassified 98% ofthe individuals (Cohen-kappa test) and some spatial structure was apparent (Fig. 2). The first discriminant function (Wilk’s X < 0.001, P < 0.001) distinguished fishes from lagoon reef flat (stations 3 and 28) to individuals from reefflat under oceanic influence (station 20) and from lagoonal and pass slope (station 21, 17 and 6). The second discriminant function (Wilk's X = 0.009, P < 0.001) separated fishes from reef slope into two groups: on one hand fishes from lagoon reef slope (station 21) and on the other hand fishes from outer reef pass slope (station 17) and from lagoon reef slope (station 6). Mg Greatest contributors to the discrimination among sites were Sr, Mn, and Ba and their ratios to calcium are given for each station in Table 2. Significant differences in concentration among sites were observed for Mg/Ca and Ba/Ca (P < 0.05) but not for Sr/ Ca and Mn/Ca. Fish otoliths from lagoon reef flat (station 3 and 28) were characterized by a higher Mg/Ca and Ba/Ca ratios than the others. The same pattern was lightly observed for Sr/Ca and Mn/Ca.. ANOVA Table 1. Results of on fish total length. Mean size (cm) and standard error are given for each station. Stations Mean size df F P All stations 15.69±0.204 5 0.385 0.858 3 15.06±0.502 6 15.86±1.004 17 15.96±0.494 20 15.73±0.435 21 15.74±0.437 28 15.79±0.456 Station 021 3 06 28 017 n Group 20 Centroid Figure 2. Discriminant function analysis (DFA) achieved with the multielemental chemical signature (Li, B, Mg, Mn, Co, Ni, Zn, Sr, Cd, Sn and Ba) ofE. merra otoliths from different stations ofBaaAtoll. Squares represent reefflat and circles represent reef slope. 205 Table 2. Mean and standard error oftrace element ratios for each station. Stations 3 6 17 20 21 28 Sr/Ca 1.26±3.45 l.lliO.05 4.61i2.34 3.28i2.12 5.76i3.02 7.14i3.04 1 Ao Mg/Ca 11.71±3.94 0.12i0.02 2.98±1.69 2.36i2.16 5.06i3.20 24.13ill.14 Mn/Ca 8.91i2.98 0.08i0.03 3.82i2.48 2.15i2.02 4.95i3.07 5.63i2.68 Ba/Ca 0.035i0.001 0.006i0.001 0.02liO.004 0.010i0.002 0.012i0.003 0.074i0.045 Our results showed that E. merra captured in Baa Atoll different habitats may be We differentiated based on their otolith chemistry. found a clear discrimination between otoliths of fishes from lagoon reef flat stations and lagoonal-pass slope stations, with higher trace element ratios for the first. These findings could derive from contrasting differences in environmental conditions ofthose stations. In fact, some environmental characteristics such as water temperature and salinity are likely to have direct effects on the concentration of some elements (Sr) in the water that affects otolith elemental composition (Gillanders and Kingsford, 2000). In the present study, we have no data on the environmental characteristics of each station (water temperature, salinity and chemistry) to support this hypothesis. Endogenous factors (diet, stress, ontogenetic effects...) can also influence otolith elemental composition, but we minimized some of these effects by analyzing otoliths of fishes with similar sizes. The analysis of multielemental composition along E. merra otolith growth axis (juvenile part to adult part) may be conducted to determine the origin of fishes and to reconstruct their movement’s history within Baa Atoll. However, temporal stability of elemental signature of otoliths (Gillanders and Kingsford, 2003; Ruttenberg et al., 2008) ofE. merra at Baa Atoll should first be investigated. In conclusion, multielemental composition of fish otoliths showed a good potential to measure connectivity between populations within an atoll at local spatial scale, and the technique may be useful on the long run for the design of MPAs and the management of fisheries. ACKNOWLEDGEMENTS This study was funded by the French Fondation pour la Recherche sur la Biodiversite, Program “Biodiversity of Indian Ocean Islands” and by the Atoll Ecosystem We Conservation Project. thank Fionel Bigot for his help for capture offishes, Olivier Bruguier for the use of ICP-MS and Audrey Darnaude and Franck Ferraton for the help for the preparation of samples. We acknowledge Serge Andrefouet and Shiham Adam at the Marine Research Center for the organization ofthe surveys and the management of the project. 206 REFERENCES Bradbury, I.R., S.E. Campana, and P. Bentzen 2008. Estimating contemporary early life-history dispersal in an estuarine fish: integrating molecular and otolith elemental approaches. Molecular Ecology 17: 1438-1450. Campana, S.E. 1999. Chemistry and composition of fish otoliths: pathways, mechanisms and applications. Marine Ecology Progress Series 188: 263-297. Chittaro, P.M., P. Usseglio, B.J. Fryer and P.F. Sale 2006. Spatial variation in otolith chemistry ofLutjanus apodus at Turneffe Atoll, Belize. Estuarine Coastal and ShelfScience 67: 673-680. Cowen, R.K., K.M.M. Lwiza, S. Sponaugle, C.B. Paris and D.B. Olson 2000. Connectivity ofmarine populations: open or closed? Science 287: 857-859. Gil2l0a0n3d.ers, B.M. and M.J. Kingsford 2000. Elemental fingerprints of otoliths offish may distinguish estuarine “nursery” habitats. Marine Ecology Progress Series 201: 273-286. Gillanders, B.M. and M.J. Kingsford Spatial variation in elemental composition of otoliths of species of fish (family Sparidae). Estuarine Coastal and ShelfScience 57: 1049-1064. Hamer, P, G.P Jenkins and B.M. Gillanders 2003. Otolith chemistry ofjuvenile snapper Pagrus auratus in Victorian waters; natural chemical tags and their temporal variation. Marine Ecology Progress Series 263: 261-273. Micheli, F., B.J. Halpern, L.W. Botsford and R.R. Warner 2004. Trajectories and correlates of community change in no-take marine reserves. Ecological Applications 14: 1709-1723. Milton D.A. and S.R. Cheney 2001. Can otolith chemistry detect the population structure ofthe shad hilsa Tenualosa ilishal Marine Ecology Progress Series 222: 239-251. Roberts, C.M., J.A. Bohnsack, F.R. Gell J.P. Hawkins and R. Goodridge , 2001. Effects of marine reserves on adjacent fisheries. Science 294: 1920-1923. Topping D.T., C.G. Lowe and J.E. Caselle Home 2005. range and habitat utilization of adult California sheephead Semicossyphus pulcher (Labridae), in a temperate no-take marine reserve. Marine Ecology 147: 301-311. Warner, R.R., S.E. Swearer, J.E. Caselle, M. Sheeby, G. Paradis 2005. Natl trace-elemental signatures in otolith of an open coast fish. Limnology and Oceanography 50: 1529-1542. Willis, T.J., D.M. Parsons and R.C. Babcock 2001. Evidence for long-term site fidelity of a snapper Pagrus auratus within a (. ) marine reserve. New Zealand Journal ofMarine and Freshwater Research 35: 581-590.