Regional Studies in Marine Science 3 (2016) 262–272

Contents lists available at ScienceDirect

Regional Studies in Marine Science

journal homepage: www.elsevier.com/locate/rsma

Otolith fingerprints of the coral reef fish Stegastes fuscus in southeastBrazil: a useful tool for population and connectivity studiesFelippe Alexandre Daros a,b, Henry Louis Spach b, Alcides Nóbrega Sial c,Alberto Teodorico Correia a,b,d,∗

a Centro Interdisciplinar de Investigação Marinha e Ambiental (CIIMAR/CIMAR), Rua dos Bragas 289, 4050-123 Porto, Portugalb Universidade Federal do Paraná (UFPR), Campus Politécnico, Caixa Postal 19031, 81531-900 Curitiba, Brazilc Núcleo de Estudos Geoquímica – Laboratório de Isótopos Estáveis (NEG-LABISE), Departamento de Geologia, Universidade Federal de Pernambuco(UFPE), Caixa Postal 7852, 50670-000 Recife, Brazild Faculdade de Ciências da Saúde da Universidade Fernando Pessoa (FCS/UFP), Rua Carlos Maia 296, 4200-150 Porto, Portugal

h i g h l i g h t s

• Otolith chemical signatures of Stegastes fuscus in SE Brazil were investigated.• Data indicated high site fidelity and limited connectivity for adult damselfishes.• Existence of some fish group clustering was however recorded.

a r t i c l e i n f o

Article history:Received 13 August 2015Received in revised form26 November 2015Accepted 26 November 2015Available online 2 December 2015

Keywords:Brazilian damselfishSagittaeChemical compositionFish ecology

a b s t r a c t

Otolith fingerprinting is a useful tool in determining the population structure, movement patterns andconnectivity among fish habitats. Damselfish, Stegastes fuscus, is a highly abundant reef fish speciesin the Brazilian coasts. It has been assessed as least concern species according with the IUCN RedList. One hundred and twenty adults, ranging from 85 to 120 mm standard length, were collectedin April 2013 by spear fishing in six coastal islands located on Cananéia, Paranaguá, Guaratuba andBabitonga bays, southeast Brazil. Elemental and isotopic signatures of whole sagittae were determinedby inductively coupled plasma mass spectrometry and isotope ratio mass spectrometry, respectively.Element:calcium (Sr:Ca, Ba:Ca, Mn:Ca andMg:Ca) and isotopic ratios (δ18O and δ13C) were analyzed withunivariate and multivariate statistics to determine whether these fingerprints can be used to assess thedegree of separation between the individuals of these geographic locations. Whole otolith fingerprintsrepresentative of the fish entire life-history gave distinct small scale spatial signatures confirming thatS. fuscus is a sedentary reef species. Classification accuracy rate obtained from quadratic discriminantfunction analysis of whole otolith chemistry data was relatively high (71% of mean accuracy percentage).Furthermore canonical analysis of principal coordinates from otolith signatures showed the existenceof three regional groups probably a consequence of the similarity of the major estuarine systems thatcharacterize the environment of the nearby coastal islands where fish were collected. This study suggeststhat otoliths can be used to study the population structure and assess connectivity of the damselfish,providing new knowledge to adequately manage this species.

© 2015 Elsevier B.V. All rights reserved.

∗ Corresponding author at: Centro Interdisciplinar de Investigação Marinha eAmbiental (CIIMAR/CIMAR), Rua dos Bragas 289, 4050-123 Porto, Portugal. Tel.:+351 223 401 823.

E-mail address: atcorreia.ciimar@gmail.com (A.T. Correia).

http://dx.doi.org/10.1016/j.rsma.2015.11.0122352-4855/© 2015 Elsevier B.V. All rights reserved.

1. Introduction

Otoliths can be used as natural tags to reconstruct the en-vironmental life history experienced by fishes, since they aremetabolically inert structures, mineral material is deposited con-tinuously, and the uptake of elements into the growing structuresusually reflects the aquatic environmental proprieties (Campanaet al., 2000). Otolith fingerprinting is useful in determining the

F.A. Daros et al. / Regional Studies in Marine Science 3 (2016) 262–272 263

population structure, movement patterns and habitat connectivityamong fish population-units at spatial (Silva et al., 2011; Higginset al., 2013; Huijbers et al., 2013) and temporal scales (Hamer et al.,2003; Cook, 2011; Correia et al., 2014). It is also a well-establishedtechnique that can complement molecular techniques in resolv-ing population structure for marine fish with high gene flow, butthat live in habitats where local environmental water conditionspresent a unique geochemical signal (Bradbury et al., 2008; Smithand Campana, 2010; Correia et al., 2012). Moreover oxygen andcarbon stable isotope ratios have been also used successfully asnatural tags for fish population structure studies (Edmonds andFletcher, 1997; Gao et al., 2004; Correia et al., 2011). Furthermore,elemental or isotopic signatures of whole otoliths provide an envi-ronmental natural tag integrated over the fish’s entire life, i.e. frombirth until death (Campana et al., 2000).

Non-migratory and highly territorial fish species may preserveunique chemical signatures in otoliths corresponding of the sitewhere they lived (Kingsford and Gillanders, 2000; Lo-Yat et al.,2005). The damselfish Stegastes fuscus is an endemic Pomacentri-dae fish species (Floeter and Gasparini, 2000), particularly abun-dant in south Brazil (Ferreira et al., 2004) and is usually found inshallow coastalwaters (Hostim-Silva et al., 2006). S. fuscus is a non-migratory reef associated fish species which shows a high territo-rial feeding behavior (Alosh, 2003; Ferreira et al., 2004). It is con-sidered a key species on coastal reefs playing an important rolein the benthic communities where they live (Hixon and Brostoff,1983; Ferreira et al., 1998), but life history information is limitedpreventing an accurate assessment of the fisheries. S. fuscus is along-lived and slow-growing fish species, reaching about 15 yearsof age (Schwamborn and Ferreira, 2002). Individuals are sexuallymature with 6.2 and 7.0 cm of standard length for females andmales, respectively (Souza et al., 2007). Spawning takes place dur-ing the dry season mainly from September to February (Cananet al., 2011). S. fuscus uses the territory for nesting and males pro-tect its eggs from grazers (Thresher, 1991; Canan et al., 2011). Afterhatching, the pelagic larvae have a growing period of about threeweeks (Wellington and Victor, 1989). S. fucus is an endangered fishspecies of the coastal reefs because of the aquatic pollution andover-exploitation for ornamental fish-aquarium trade (Souza et al.,2007; Canan et al., 2011). To aid in the conservation and fisherymanagement there is an urgent need for more information aboutthe S. fuscus population structure, connectivity and coastal recruit-ment process.

The purpose of thisworkwas to investigate the otolith chemicalsignatures of S. fuscus at small spatial scales (15–50 km) allowing toassess the connectivity among fish groups, and touse the elementaland isotopic composition of otoliths to determine whether adultfish captured in six coastal islands in the South Brazil mayrepresent discrete population-units.

2. Materials and methods

2.1. Study area

The study was performed in a coastal area of 140 km coveringthe southern region of São Paulo, the Paraná and the north of SantaCatarina (Fig. 1). It includes six reef islands in the nearby areaof four major estuarine complexes. Bom Abrigo Island (25°07′S,47°52′W) with an area of 1.54 km2 is located in São Paulo andit is 3.6 km far from mainland in the E-SE direction. FigueiraIsland (25°35′S, 48°03′W) is located in the ocean at 10.7 km fromthe restinga de Ararapira. Galheta Island (25°36′S; 48°19′W) islocated in the entrance to the Paranaguá Estuarine Complex (PEC)with a mean water depth ranging from 15 to 30 m. The CurraisArchipelago (25°44′S, 48°22′W) comprises three islands, 11 kmfrom the coast, with depths between 1.5 and 16 m. Itacolomis

Island (25°50′S, 48°24′W), consists of two small rocky islandslocated 13 km from the coast, with depths ranging from 3 to 17 m.These last four islands are located in Paraná State. In Santa Catarina,the Graças Archipelago (26°10′S, 48°29′W) consists of five islandsand six outcrops, about 3.5 km off the coast and about 37 km fromthe Itacolomis Island. The depth around the archipelago variesfrom1.5 to 18m. The islands substrate is composedmostly of smallrocks, which are usually covered by macroalgae, Palythoa sp. andZoanthus sp (Karmann et al., 1999; Daros et al., 2012; Passos et al.,2012). Cananéia bay is considered an estuarine complex located inthe south of São Paulo state (25°01′S, 47°55′W). It extends from theRibeira rivermouth until the Ararapira Channel with about 110 kmof length. It is composed by a system of lagoons and channels. Itshows characteristics of high salty concentration beaches and lowsalty concentration rivers that run intomangrove areas (Schaeffer-Novelli et al., 1990). The Paranaguá Estuarine Complex (PEC) has anarea of 612 km2 and a diversity of habitats, like tidal flats, channels,mangroves, salt-marshes, tidal creeks, estuarine beaches, rivers,and rock shores near the mouth of the estuary. The PEC, a partiallymixed estuary with semidiurnal tides and diurnal inequality, isconnected to the Cananéia Estuarine Complex, in the north, bythe Ararapira Channel and to the Atlantic Ocean, in the east, bySueste Channel and Galheta Channel. The climate of the regionis tropical with a mean annual rainfall of 2500 mm. The rainyseason typically starts at the end of spring and lasts until nearlythe end of summer. The dry season lasts from the end of autumnto the end of winter, but is interrupted by a short low-intensityrainy period that occurs at the beginning of winter (Lana et al.,2001). Guaratuba bay (25°52′S, 48°38′W) has an area of 50 km2

and a mean depth of 3 m. The bay receives most of its freshwaterinflow from the Cubatão and São João rivers, which contributewiththe larger part of the surrounding area’s freshwater runoff, whichattains up to 80m3 s−1. It is surrounded by seagrass andmangrovevegetation, which are more extensive along the northern marginof the estuary. The southern margin is largely occupied by urbansettlements and agricultural enterprises. The system has strongcurrents and semi-diurnal tides. The climate of this region isclassified as subtropical mesothermic with an average annualtemperature of 22 °C and a hot rainy summer season (Maroneet al., 2004). The Babitonga bay (26°13′S, 48°37′W) is a subtropicalestuary located on the northern of Santa Catarina. It occupies anarea of 130 km2 with a maximum depth of 28 m. The sedimentof the bay is mainly composed of sand of different textures. Theregion has a humid subtropical climate with well-distributed rainsduring the year and a drier winter. The estuary is subjected to amicrotidal system with an amplitude of 1.30 m. The south sideis extensively urbanized because of the cities of São Francisco doSul and Joinville. The northern margin have a great diversity ofenvironments, such as tidal flats, sandy beaches, salt marsh banksand mangrove forests interspersed with tidal channels and withthe outfall of small perennial rivers (IBAMA, 1998).

2.2. Fish sampling

S. fuscus individuals were sampled by spear fishing in rockyreefs, at a depth above 5 m, in six coastal islands in south Brazil(one in the São Paulo, four in Paraná and one in Santa Catarinastates) (Fig. 1). A total of 120 adults (20 per location), ranging from85 to 120 mm of standard length (average 100 ± 1 mm), werecollected in April 2013. In the laboratory, standard length (SL, mm)and mass (M, g) of fish were measured (Table 1). Sagittal otolithswere carefully extracted using plastic forceps to avoid metalliccontamination and stored dry in plastic vials for further chemicalanalysis.

264 F.A. Daros et al. / Regional Studies in Marine Science 3 (2016) 262–272

Table1

Geo

grap

hiclocatio

n,seaw

ater

tempe

ratures(

SST),sam

plesize

(n),stan

dard

leng

th(SL),m

ass(

M)a

ndotolith

elem

ental(un

detren

dedco

ncen

trations

,µelem

entg

−1calcium)a

ndisotop

ic(V

PDB,

h)finge

rprinting(m

ean

±SE

)of

S.fuscus.

Geo

grap

hiclocatio

nn

SST(°C)

SL(m

m)

M(g)

Sr/Ca

Ba/Ca

Mg/Ca

Mn/ca

δ18O

δ13C

Bom

Abrigo

25°7

.23′S–

048°

51.46′W

2023

.95

±0.08

100.15

±0.97

54.08

±1.84

1358

8±

868

121

±9

12.25

±0.42

2.18

±0.17

−0.98

±0.15

−5.82

±0.56

Figu

eira

25°2

1.41

′S–

048°

2.18

′W

2023

.67

±0.09

94.75

±1.25

44.60

±2.10

1295

8±

867

120

±8

14.65

±0.76

0.99

±0.05

−0.78

±0.19

−6.34

±0.69

Galhe

ta25

°35.07

′S–

048°

19.30′W

2023

.57

±0.07

105.85

±1.97

58.79

±2.45

1170

6±

378

89±

1016

.03

±0.60

1.08

±0.07

−1.24

±0.54

−5.98

±0.54

Currais

25°4

4.12

′S–

048°

22.02′W

2023

.45

±0.11

97.25

±1.07

47.37

±1.68

1318

3±

805

168

±12

14.30

±0.56

0.97

±0.04

−0.92

±0.15

−5.65

±0.51

Itaco

lomis

25°5

0.54

′S–

048°

24.50′W

2023

.52

±0.09

101.05

±1.10

61.68

±2.00

1445

3±

838

144

±8

14.17

±0.35

1.40

±0.05

−0.77

±0.16

−6.51

±0.64

Graças

26°1

0.77

′S–

048°

29.17′W

2023

.37

±0.09

100.70

±1.06

53.76

±1.60

1409

1±

514

91±

512

.63

±0.48

2.18

±0.13

−1.24

±0.30

−5.17

±0.44

Note:

island

sarelis

tedfrom

North

toSo

uth.

F.A. Daros et al. / Regional Studies in Marine Science 3 (2016) 262–272 265

Fig. 1. Map of the south Brazilian coast indicating the six sampling S. fuscus locations (Bom Abrigo, Figueira, Galheta, Currais, Itacolomis and Graças) and the associatedestuarine complexes (Cananéia Bay, Paranaguá Estuarine Complex, Guaratuba Bay and Babitonga Bay).

2.3. Otolith elemental analysis

Prior to the chemical analyses, the left sagittae were cleanedin an ultrasonic bath for 5 min in ultrapure water (Milli-Q-Water)followed by immersion in 3% analytical grade hydrogen peroxide(H2O2, Fluka TraceSelect) for 15 min to remove any adherentbiological tissues. Otoliths were decontaminated by immersion in1% nitric acid (HNO3, Fluka TraceSelect) solution for 10 s followedby a double-immersion in ultrapure water (Milli-Q-Water) for5min (Rooker et al., 2001). Otoliths were stored in new, previouslydecontaminated, FalconTM tubes, where they were allowed to airdry in a laminar flow fume hood (Patterson et al., 1999). Chemicalcomposition of whole otoliths was determined using solution-based inductively coupled plasma mass spectrometry (SB-ICP-MS). Decontaminated otoliths were weighed on an analyticalbalance (0.0001 g) and dissolved for 15 min in 10% ultrapureHNO3 to a final volume of 3 ml (Silva et al., 2011). SB-ICP-MSanalyses were made using a double focusing magnetic sector fieldinstrument ICP-SF-MS (Thermo ICP-MS x series, Thermo ElectronCorporation). This instrument was equipped with a compactdouble-focusing magnetic sector mass spectrometer of reversedNier–Johnson geometry. All measurements were made at amedium resolution setting (m/1m = 4000) to avoid false readings

from spectral interferences. The instrument was equipped with amicro flow nebulizer (PFAAR35-1-C1E, Glass Expansion), operatedin the self aspirating mode (sample uptake rate ∼0.93 L min−1).Quantification of trace elements was based on the externalcalibration method, preparing multi-element standards thatcontained the elements of interest in the expected concentrationrange. To minimize the effect of any plasma fluctuations ordifferent nebulizer aspiration rates between the samples, 115In ofa known concentration was added to all samples and standards asan additional internal standard. Concentrations were calculated bylinear interpolation (sum of least squares) based on normalizationwith the internal standard, and on calibration curves made fromsingle element standards (Merck KGaA) covering the individualexpected concentration ranges. A calibration was made at thebeginning of each session. The matrix of both the blank and thestandard solutions was 1% HNO3. A preliminary analysis wasmadeto determine the most likely elements (44Ca, 88Sr, 137Ba, 26Mg,55Mn, 7Li, 60Ni, 54Fe, 208Pb and 66Zn) to serve as environmentalindicators, taking into account the metallic elements alreadyreported in the otoliths of congener species (Stegastes nigricans: Lo-Yat et al., 2005; Stegastes partitus: Chittaro et al., 2006; Chittaro andHogan, 2013), but excluding elements under strictly physiologicalregulation (Campana, 2005). Five elements (44Ca, 88Sr, 137Ba, 26Mg

266 F.A. Daros et al. / Regional Studies in Marine Science 3 (2016) 262–272

and 55Mn) were consistently detectable in whole otoliths andwere used for further SB-ICP-MS analysis. Otolith samples wereanalyzed in random order to avoid possible sequence effects andan otolith certified reference material (FEBS-1) was analyzed foraccuracy quality control (Sturgeon et al., 2005). With regard to theanalytical accuracy, the elemental concentrations determined inFEBS-1 were within the certified values, with a value of recoverybetween 78% and 96%. Precision of replicate analyses of individualelements ranged between 1% and 5% relative standard deviation(RSD). The limits of detection were calculated from the individualcalibration curves using the three sigma criteria andwere (in ppb):44Ca (30000), 88Sr (20), 137Ba (5), 26Mg (5) and 55Mn (1).

2.4. Otolith isotopic analysis

For carbon and oxygen isotopic otolith analysis (δ18O and δ13C),carbon dioxide (CO2) was extracted from powdered carbonates(whole right sagittae) in a high vacuum line after reaction with an-hydrous orthophosphoric acid (H3PO4, Sigma–Aldrich) for 12 h at25 °C (Craig, 1957). The released CO2 was analyzed for carbon andoxygen isotopes in a double inlet, triple collector (SIRA III) massspectrometer, using Borborema Skarn Calcite (BSC) as the refer-ence gas. BSC was calibrated against NBS-18, NBS-19 and NBS-20.The precision of analysis was better than 0.1h based on multipleanalyses of the internal standard. The results are expressed in thenotation h (per mil) in relation to international Vienna Pee-DeeBelemnite (VPDB) scale (Epstein et al., 1953).

2.5. Seawater surface temperatures

Seawater surface temperatures (SST) for each fishing area (4 ×

4 km = 16 km2) were obtained from an open access his-torical database of the weekly mean sea surface temperatures(http://modis-ocean.gsfc.nasa.gov/) providedby theNational Aero-nautics and Space Administration (NASA). SSTs were calculated foreach individual calibrated to the estimated age of fish to repre-sents its lifetime. Individual annual fish age was obtained usingthe inverse function of the von Bertalanffy growth curve (VBGC)(Mackay and Moreau, 1990) taking into account the parameter es-timates of the VBGC for the species (Humann, 1999: L∞ = 15 cm;Schwamborn and Ferreira, 2002: K = 0.19 y − 1 and t0 = −1).Standard length (SL)was previously transformed to fork length (FL)(FL = 0.28 + 1.12 SL) (Schwamborn and Ferreira, 2002).

2.6. Data analysis

SB-ICP-MS concentrations of trace elements, originally in µgelement L−1 solution, were transformed to µg element g−1 otolithand then to µg element g−1 calcium. Raw data for each elementwere checked for normality, homoscedasticity and homogeneityof variance–covariance matrices prior to statistical analysis. Theseassumptions were met after log 10 transformation (log Sr, logMn and log Mg). Although there were no significant differencesin the mean lengths of fish among locations (One-Way ANOVA,n = 120, F = 8.53, p < 0.05), we tested for relationships betweenelemental concentration and fish size (expressed as otolith mass)with analysis of covariance (ANCOVA, otolith mass as co-variate).Otolith elemental concentrations were significantly correlatedwith otolith mass for all elements, with the exception of Ba (r2 =

0.02, n = 120, p = 0.098). Sr presented a positive relationship(r2 = 0.28, n = 120, p < 0.05), which was the opposite to Mg(r2 = 0.06, n = 120, p < 0.05) and Mn (r2 = 0.05, n = 120, p <0.05) that showed a very weak negative relationships, althoughsignificant. To ensure that differences in fish size among samplesdid not confound any site-specific differences in otolith chemistry,concentrations of elements were weight-detrended by subtraction

of the product of the commonwithin-group linear slopemultipliedby the otolith mass from the observed concentration (Campanaet al., 2000).

δ13C and δ18Ovalueswere analyzed by an analysis of covariance(ANCOVA). Otolith mass was considered to be a proxy for age andwas used as the covariate in ANCOVA. Location was treated as afixed factor. δ18O values were plotted against the individual SSTsaveraged over the entire life of the fish. The relationship betweenδ18O values of otolith carbonate and SSTs was explored assumingthat the isotopic signature of their carbonate otoliths would becorrelated with the water temperature where they resided dueto temperature dependent fractionation of 18O /16 O during theprecipitation of the otolith carbonate. It was also assumed thattherewas no significant variation in the isotopic composition of theambient seawater across the oceanic areas of interest (Correia et al.,2011). Furthermore the potential temporal mismatch between SSTand oxygen isotopic signatures due to the fact that most accretionoccurred in the juvenile phase and proportionally less from olderages, was minimized since fish were from a similar range lengthand no significant SL differences existed among locations.

One-way analysis of variance (ANOVA) was used to exploreindividual elemental fingerprint differences between locations. Ifsignificant differences were found, this was followed by a Tukeypost hoc test. Multivariate analysis of variance (MANOVA) wasused to explore multi-elemental fingerprints and detect differ-ences in the multi-elemental otolith composition from differentlocations. For the MANOVA, we reported the approximate F-ratiostatistic for the most robust test of multivariate statistics (Pillai’strace). Post-hoc multivariate pairwise comparisons between lo-cations were performed using the Hotelling T -square test. Multi-element compositions of otoliths were analyzed with a QuadraticDiscriminant Function Analysis (QDFA). QDFA was used to visual-ize spatial differences and to examine the re-classification accuracysuccess of fishes to this original location. Cross-validations wereperformed using jackknifed (‘‘leave one out’’) procedures (Correiaet al., 2014). The correlation matrix from the elemental and iso-topic data set was analyzed by a Canonical Analysis of principalCoordinates (CAP) based on Euclidian distances (Spearman correla-tion of 55%) (Anderson andWillis, 2003) and the results presentedin a two-dimensional biplot (Lo-Yat et al., 2005).

The statistical analyses were performed using Systat (version13.0) and PRIMER 6 + PERMANOVA softwares. The statisticallevel of significance (α) was 0.05. Data are presented as meanvalues ± standard errors.

3. Results

All element:Ca ratios (Sr, Ba, Mg and Mn) and isotopic ratios(δ18O and δ13C) differed significantly among islands (ANOVA, p <0.05). Galheta island showed the lowest value of Sr:Ca ratio (Tukeytest, p < 0.05) (Fig. 2(A)). This island also recorded the lowestvalue for Ba:Ca ratio although not significantly different from BomAbrigo, Figueira and Graças (Tukey tests, p > 0.05) (Fig. 2(B)).Galheta also exhibited the highest value of Mg:Ca ratio compar-atively to the other islands, with exception of Itacolomis (Tukeytests, p < 0.05) (Fig. 2(C)). Figueira presented the lowest value ofMn:Ca ratio comparatively to the other islands with exception ofCurrais (Tukey tests, p < 0.05) (Fig. 2(D)). Figueira and Itacolomisdid not show any univariate pairwise differences, with exceptionof the Mn:Ca ratio (Tukey tests, p < 0.05) (Fig. 2(D)).

Themean otolith isotopic ratios obtained from the six samplinglocations ranged from −8.81h to −8.07h and from −1.64h to−0.33h for δ13C and δ18O, respectively. For δ18O Bom Abrigo,Galheta and Graças islands showed the lowest mean values (Tukeytests, p < 0.05) (Fig. 2(E)). Itacolomis showed the highest valuefor δ13C, but not significantly different from Figueira (Tukey tests,

F.A. Daros et al. / Regional Studies in Marine Science 3 (2016) 262–272 267

Fig. 2. Elemental and isotopic concentrations (mean ± SE) observed in the whole otoliths of S. fuscus from fishes collected in South Brazil. Elemental and isotopicconcentrations are given in µg element g−1 calcium and VPDB (h), respectively. The locations marked with the same letter above the error bars are not significantlydifferent from each other (P > 0.05).

p < 0.05) (Fig. 2(F)). δ13C showed no significant relationshipswith otolith mass (Fig. 3(A)) and 35% of the sum of squares wasexplained by location (Table 2). For δ18O, 44% of the sum of

squares was explained by location and although otolith mass wassignificant, it only explained less than 1% of the sum of squares(Fig. 3(B)) (Table 2). No relationship was found between otolith

268 F.A. Daros et al. / Regional Studies in Marine Science 3 (2016) 262–272

Fig. 3. δ13C (A) and δ18O (B) versus otoliths mass for all data.

Fig. 4. δ18O versus SST for all S. fuscus individuals.

δ18O signatures and the individual averaged SSTs experienced byfish (Fig. 4). The bi-plot using the both isotopes ratios suggests thatthe isotopic signatures overlap for all locations, with exception ofGraças in which individuals appeared to have a more site specificsignal (Fig. 5).

MANOVA indicated a significant difference in the multi-elemental signatures of the whole otoliths (Pillai Trace, F30,565 =

11.059, p < 0.05). All pairwise comparisons gave significantdifferences between locations, with exception of Figueira andItacolomis (Hotelling’s T -Square Trace, F = 13.428, p = 0.103).For each pair of groups, Figueira and Itacolomis (F6,109 = 2.140)and Figueira and Graças (F6,109 = 58.035), showed the lowestand highest Mahalanobis distances (Between Group F-Matrix),respectively.

F.A. Daros et al. / Regional Studies in Marine Science 3 (2016) 262–272 269

Fig. 5. δ13C versus δ18O for sagittal otolith carbonate from S. fuscus.

Fig. 6. Canonical variate plots displaying spatial differences inmulti-elemental tagsof S. fuscus whole otoliths from South Brazil. Bom Abrigo Island (•), Figueira Island(◦), Galheta Island (�), Currais Arch. (+), Itacolomis Island (N) and Graças Arch. (⋆).Ellipses represent 95% confidence intervals around the data and symbols representindividual fish.

Table 2ANCOVA for δ13C and for δ18O values of otolith.

Source DF SS MS F-ratio P-value

δ13C

Location 5 22.056 4.411 12.647 0.000Otolith weight 1 1.322 1.322 3.789 0.054Error 113 39.414 0.349Total 119 62.792

δ18O

Location 5 3.629 0.726 20.968 0.000Otolith weight 1 0.733 0.733 21.171 0.000Error 113 3.911 0.035Total 119 8.273

QDFA plot showed a separation among the coastal islands basedon the elemental and isotopic composition of otoliths, althoughsome overlap was evident for some regions namely for Figueira,Currais and Itacolomis (Fig. 6). Jackknifed classification accuracywas moderate to high ranging from 55% (Itacolomis) to 80% (BomAbrigo and Currais), and showed an overall mean of 71% (Table 3).

Fig. 7. Canonical analysis of principal coordinates (CAP) plot of the elemental andisotopic otolith signatures from S. fuscus (n = 120) collected in the southeast coastof Brazil. Ellipses are 95% confidence limits for the three groups defined by the CAP.

The CAP identified threemain groups in two-dimensional spacecomposed by individuals from Galheta (Group 1), Bom Abrigo andGraças (Group 2), and Currais Figueira and Itacolomis (Group 3).The vectors for Mn:Ca and δ13Cwere aligned with the group 2, andBa:Ca and δ18O vectors with group 3 (Fig. 7).

4. Discussion

In the present work, the chemical composition of the wholeotoliths of S. fuscus adults was assessed in a small scale geographicarea ranging from ten to hundreds kilometers. It has been basedon trace elements commonly found at informative levels inprevious studies with Stegastes spp (Lo-Yat et al., 2005; Chittaroet al., 2006; Chittaro and Hogan, 2013). The hereby obtainedconcentrations are within the reported values for other marinespecies (Albuquerque et al., 2012;Higgins et al., 2013; Correia et al.,2014). Similarly, patterns of connectivity among populations of S.partitus in the west Caribbean Sea were assessed using chemicalsignatures recorded in the otolith’s edge, both at small and large

270 F.A. Daros et al. / Regional Studies in Marine Science 3 (2016) 262–272

Table 3Jackknife classification matrix of S. fuscus adults based on whole otolith’s used in quadratic discriminant function analysis.

Real location Predicted location %CorrectBom Abrigo Figueira Galheta Currais Itacolomis Graças

Bom Abrigo 16 0 2 0 0 2 80Figueira 0 11 1 3 5 0 55Galheta 0 1 15 2 2 0 75Currais 0 2 1 16 1 0 80Itacolomis 1 6 0 1 12 0 60Graças 5 0 0 0 0 15 75Total 22 20 19 22 20 17 71

regional scales, suggesting a substantial self-recruitment in someareas (Chittaro and Hogan, 2013). Furthermore, whole chemicalotolith analysis in S. nigricans from the French Polynesia showedsignificant differences among nearby reefs (distances less than200 m) suggesting the effectiveness of this tool to study theconnectivity among populations at small scale distances (Lo-Yatet al., 2005). Furthermore cross validation showed that, in average,71% of all fish were successfully assigned to the region of origin.The hereby results suggest a high regional site fidelity in S. fuscusadults and a low degree of mixing of water masses. It also meansthat connectivity scales for adult S. fuscus in this region appear tobe very limited (15–50 km), and probably these individuals couldbe considered discrete population-units for fishery conservationpurposes. High resolution (82%) of sites less than 10 km apart usingtrace elements for a reef-associated fish has been already reported(Lara et al., 2008).

In the present study all individuals were captured in shallowcoastal waters (less than 4 m) and therefore no variation was ex-pected to occurs due to the collection water depth (Kingsford andGillanders, 2000). However, variation in water chemistry wouldbe expected to occur in coastal environments due to the pres-ence of estuaries, human activities, pollution and freshwater runoffregimes (Kingsford and Gillanders, 2000; Forrester and Swearer,2002; Hamer et al., 2003). The hereby sampling area represents adistance of 140 kmbetween themost distant islands (i.e., fromBomAbrigo toGraças). This area is under the direct influence of fourma-jor estuaries located in the southeast Brazilian coast (Diegues andRosman, 1998). The climate is under the influence of the subtrop-ical type characterized by well-distributed rains through the yearand a drier winter (Peel et al., 2007), withmean annual rainfall andtemperature of 2500 mm and 22 °C, respectively (IPARDES, 1989).The principal estuary, Paranaguá Estuarine Complex, has approx-imately 550 km2 of flooded surface area (Noernberg et al., 2006).Cananéia and Babitonga bays have about 150 km2 flooded surfacearea (Miyao et al., 1986; IBAMA, 1998) while Guaratuba, the small-est estuary, has only 50 km2 flooded surface area (Mizerkowskiet al., 2012). Human activity is greater in Paranaguá and BabitongaBays comparatively to Guaratuba and Cananéia Bays, mainly be-cause of the shelter industries, ports and the existence of a highpopulation density (IPARDES, 1989; IBAMA, 1998; Romero et al.,2010).

Three groups were somewhat distinguished by the visualinspection of the QDFA and CAP plots: the Galheta island is thenearest of the coast, being located in the outfall of the ParanaguáEstuarine Complex (PEC); Bom Abrigo and Graças, located about6.5 km of the Cananéia and Babitonga bays respectively; andFigueira, Currais and Itacolomis islands, located approximately9 km of coast, with a distance ranging between 13 and 33 kmfrom the nearest bay. Individuals collected in Galheta Island, nearthe mouth of PEC, presented in their otoliths low levels of Srand Ba, and high levels of Mg. The low Sr concentration couldbe related with the output of the estuarine freshwater, since itis well known the positive relationship between Sr and watersalinity (Webb et al., 2012). The recorded freshwater runoff is 57,

75 and 80 m3 s−1 for Babitonga (DNIT, 2004), Cananéia (Tundisiand Tundisi, 2001) and Guaratubá (Mizerkowski et al., 2012) bays,respectively. Paranaguá Bay has however amean annual river freshwater input up to 200 m3 s−1 (Marone et al., 2005). Althoughall bays are considered euhaline areas, its salinity is relativelylower comparatively to the shallow coastal waters of the Braziliansouth-eastern platform (34.0 psu: Brandini, 1988). Paranaguá bay,for instance, could have a salinity as lower as 24.0 in the outerarea (Passos et al., 2013). However, it is difficult to identify thecause behind the high content of Mg in Galheta individuals.Although it was been recently found a positive relationship amongMg:Ca ratios in otoliths and water temperature (Barnes andGillanders, 2013), the reported SSTs of the sampling locations, atleast considering the mean of the last ten years, are relativelysimilar (∼23 °C). It means that this difference is probably relatedwith other abiotic or physiological processes (Woodcock et al.,2012). Located in the internal coastal shelf of the Paraná, Figueira,Currais and Itacolomis islands, did not suffer the influence of theestuarine plumes, and the coastal waters are mixed by winds andcurrents (Brandini et al., 2007; Nemes and Marone, 2013). Alsointeresting is the fact that both Bom Abrigo and Graças, locatedin front of Cananéia and Babitonga bays respectively, presentedhigh concentrations of Mn. Both areas have mangroves and saltmarshes that are rich environments in manganese, iron and sulfur.Furthermore a geochemistry study showed that Mn in Cananéiamangroves soils tend to be mobilized and lost (Otero et al., 2009).

δ18O in otoliths is deposited in equilibrium with ambientwater δ18O, but is influenced by temperature and salinity. Thefractionation factor of δ18O in otoliths is temperature-dependent(Thorrold et al., 1997) and the negative linear relationship betweenwater temperature and δ18O in otoliths has been validated forseveral species (Høie et al., 2003). Considering that the observedaveraged water temperature range within the study area is only ofabout 1 °C, as reported by a previous study (Brandini et al., 2007),this would correspond to about only 0.2h difference in the oxygenisotopic ratio (Høie et al., 2004). Salinity is also commonly usedas a proxy for water δ18O values through the fractional amountof runoff-sourced freshwater namely in the shallow coastal, andthe δ18O profiles otoliths can reflect seasonal freshwater input(Matta et al., 2013). This suggests that the small δ18O differencesfound in individuals from different islands are probably relatedwith the environmental salinity (Elsdon and Gillanders, 2002). InBom Abrigo, Galheta and Graças islands the mean values of δ18Owere significantly lower than the other locations. These islands arelocated in proximity of an estuary, where the salinity is low dueto freshwater runoff. The input of freshwater from estuaries waslikely an important factor in explaining the relatively low otolithsδ18O values (Correia et al., 2011).

For δ13C the results did not shown any general trend, beingdifficult to suggest factorswhich could have influenced the presentdata. It is well known that δ13C values measured in fish otolithscan be influenced by ontogenetic changes in trophic levels thatcomprise the fish diet and metabolism (Schwarcz et al., 1998;Gao and Beamish, 2003; Gao et al., 2004), but may also reflect

F.A. Daros et al. / Regional Studies in Marine Science 3 (2016) 262–272 271

geographic variation in the δ13C of the DIC of the ambient water(Thorrold et al., 1997; Patterson, 1999; Solomon et al., 2006).

The incorporation of trace elements into otoliths is a complexprocess poorly understood but mainly influenced by abiotic (salin-ity, temperature, water chemistry) and biotic (age, growth, physi-ology and diet) factors (Webb et al., 2012; Woodcock et al., 2012;Barnes and Gillanders, 2013). The incorporation of Ba and Sr intootoliths is known to be influenced by salinity ambient concentra-tions (Elsdon and Gillanders, 2003; Martin and Wuenschel, 2006;Walther and Thorrold, 2006), although temperature and growthrate may also affect the incorporation of the former metallic el-ements in otoliths (Townsend et al., 1992; Sadovy and Severin,1994; Elsdon and Gillanders, 2004). For Mn and Mg there is noclear relationship between their concentration in otoliths and insurrounding water, and variation appears to be related to unclearendo and exogenous process (Martin andWuenschel, 2006; Hamerand Jenkins, 2007; Woodcock et al., 2012). Moreover several stud-ies have shown that stable oxygen ratios (δ18O) in otoliths canbe used as a proxy of the ambient sea temperature (Radtke et al.,1996; Thorrold et al., 1997; Høie et al., 2004), while carbon isotoperatios (δ13C) are mainly influenced by fish metabolism and feed-ing regime (Kalish, 1991; Schwarcz et al., 1998; Høie et al., 2003),and by the dissolved inorganic carbon in the water (Thorrold et al.,1997; Patterson, 1999; Solomon et al., 2006). The main result fromthis study is that isotopic and elemental values of adult otolithswere slight different among sampling areas, with some clusteringof locations, that cannot be related to distance or SST, but to otheruninvestigated underlying factors. Nonetheless local year-0 fluc-tuations in both temperature and/or salinity, as well as multipleanthropogenic influences, which complicate chemical interpreta-tion of otolith signatures, cannot also be excluded and need fur-ther investigation. Besides the exogenous variation of the otolithchemistry driven by the water proprieties of the study area, otherendogenous factors such as sex, growth and gonadal maturationof fish, cannot be excluded. However, knowledge of the primarycauses behind the incorporation of the metallic elements in thearagonite matrix is facultative for stock discrimination purposes(Thresher, 1999).

5. Conclusion

The present study suggests, for the first time, that unique chem-ical signatures in otoliths of S. fuscus can occur at local scales insome coastal environments in South Brazil, with some clusteringthat cannot be directly related to distance and water tempera-tures. However since S. fuscus appear to be a site-attached and non-mobile fish and given that the only opportunity for dispersal is dur-ing the three week larval period, both traits may prove useful forfuture fish connectivity studies within reef scales. This study alsoshowed that S. fuscus individuals in the region cannot be consid-ered a single uniform population, which can raises the question ofwhatwould be the consequence of the disproportionate removal ofthe individuals from one island, as results of the over-exploitationfor ornamental fish-aquarium purposes, in terms of the regionaldynamics for the species. By comparing the larval core (natal) sig-natures of individuals of the same cohort using laser-based analy-sis it would be possible to assess the contribution of each island, asspawning or nursery area, to the overall adult population. Answer-ing this relevant ecological question is essential for the effectiveconservation of the species.

Acknowledgments

This research was supported by the CNPq (National Council ofTechnological and Scientific Development)—473181/2012-6 and401190/2014-5 and by the Strategic Funding UID/Multi/04423/

2013 through national funds provided by FCT—Foundation forScience and Technology and EuropeanRegional Development Fund(ERDF), in the framework of the programme PT2020. Felippe Darosbenefited from a Brazilian Ph.D. Grant (CAPES—BEX 1906/13-5). Also a special thanks to all anonymous referees for theirsuggestions and comments that allowed to improve an earlier draftof this manuscript.

References

Albuquerque, C.Q.,Miekeley, N.,Muelbert, J.H.,Walther, B.D., Jaureguizar, A.J., 2012.Estuarine dependency in a marine fish evaluated with otolith chemistry. Mar.Biol. 159, 2229–2239.

Alosh, K., 2003. Reproductive parameters of female Hawaiian damselfish Dascyllusalbisella with comparison to other tropical and subtropical damselfishes. Mar.Biol. 143, 803–810.

Anderson, M.J., Willis, T.J., 2003. Canonical analysis of principal coordinates: auseful method of constrained ordination for ecology. Ecology 84, 511–525.

Barnes, T.C., Gillanders, B.M., 2013. Combined effects of extrinsic and intrinsicfactors on otoliths chemistry: implications for environmental reconstructions.Can. J. Fish. Aquat. Sci. 70, 1159–1166.

Bradbury, I.R., Campana, S.E., Bentzen, P., 2008. Estimating contemporary earlylife-history dispersal in an estuarine fish: integrating molecular and otolithselemental approaches. Mar. Ecol. 17, 1438–1450.

Brandini, F.P., 1988. Hydrography, phytoplankton biomass and photosynthesis inshelf and oceanic waters off southeastern Brazil during autumn (May/June,1983). Bolm. Inst. Oceanogr. 36, 63–72.

Brandini, F.P., Silva, A.S., Silva, E.T., Kolm, H., 2007. Sources of nutrients and seasonaldynamics of chlorophyll in the inner shelf off Paraná State—South Brazil Bight.J. Coast. Res. 23, 1131–1140.

Campana, S.E., 2005. Otolith elemental composition as a natural marker of fishsticks. In: Cadrin, S,X, Friedland, K.D.,Waldaman, J.R. (Eds.), Stock IdentificationMethods: Applications in Fishery Science. Elsevier Academic Press, California,pp. 227–245.

Campana, S.E., Chouinard, G.A., Hanson, J.M., Frechet, A., Brattey, J., 2000. Otolithelemental fingerprints as biological tracers of fish stocks. Fish. Res. 46, 343–357.

Canan, B., Pessoa, E.K.R., Volpato, G.L., Araujo, A., Chellappa, S., 2011. Feeding andreproductive dynamics of the damselfish, Stegastes fuscus, in the coastal reefsof Northeastern Brazil. Anim. Biol. J. 2, 113–126.

Chittaro, P.M., Hogan, J.D., 2013. Patterns of connectivity among populations of acoral reef fish. Coral Reefs 32, 341–354.

Chittaro, P.M., Hogan, J.D., Gagnon, J., Fryer, B.J., Sale, P.F., 2006. In situ experimentof ontogenetic variability in the otoliths chemistry of Stegastes partitus. Mar.Biol. 149, 1227–1235.

Cook, G.S., 2011. Changes in otoliths microchemistry over a protracted spawningseason influence assignment of natal origin. Mar. Ecol. Prog. Ser. 423, 197–209.

Correia, A.T., Barros, F., Sial, A.N., 2011. Stock discrimination of European conger eel(Conger conger L.) using otoliths stable isotope ratios. Fish. Res. 108, 88–94.

Correia, A.T., Hamer, P., Carocinho, B., Silva, A., 2014. Evidence for meta-populationstructure of Sardina pilchardus in the Atlantic Iberian waters from otolithselemental signatures of a strong cohort. Fish. Res. 149, 76–85.

Correia, A.T., Ramos, A.A., Barros, F., Silva, G., Hamer, P., Morais, P., Cunha, R.L.,Castilho, R., 2012. Population structure and connectivity of the Europeanconger eel (Conger conger) across the Northeastern-Atlantic and Western-Mediterranean: integrating molecular and otolith elemental approaches. Mar.Biol. 159, 1509–1525.

Craig, H., 1957. Isotope standard for carbon and oxygen and correction factors formass spectrometry analysis of carbon dioxide. Geochim. Cosmochim. Acta 12,133–149.

Daros, F.A., Bueno, L.S., Vilar, C.C., Passos, A.C., Spach, H.L., 2012. Checklist of rockyreef fishes from the Currais Archipelago and Itacolomis Island, Paraná state,Brazil. Check List 8, 349–354.

Diegues, A.C., Rosman, P.C.C., 1998. Caracterização dos ativos ambientais em áreasselecionadas da zona costeira brasileira. Ministério do Meio Ambientec, dosRecursos Hídricos e da Amazônia Legal, Brasília, 136 p.

DNIT, 2004. Departamento Nacional de Infra-Estrutura de Transportes. RelatórioFinal dos Estudos Ambientais da Baía da Babitonga—Canal do Linguado. 149 pp.

Edmonds, J.S., Fletcher,W.J., 1997. Stock discrimination of pilchards Sardinops sagaxby stable isotope ratio analysis of otolith carbonate. Mar. Ecol. Prog. Ser. 152,241–247.

Elsdon, T.S., Gillanders, B.M., 2002. Interactive effects of temperature and salinity onotolith chemistry: challenges for determining environmental histories of fish.Can. J. Fish. Aquat. Sci. 59, 1796–1808.

Elsdon, T.S., Gillanders, B.M., 2003. Relationship between water and otolithelemental concentration in juvenile black bream, Acanthopagrus butcheri. Mar.Ecol. Prog. Ser. 260, 263–272.

Elsdon, T.S., Gillanders, B.M., 2004. Fish otoliths chemistry influenced by exposureto multiple environmental variables. J. Exp. Mar. Biol. Ecol. 313, 269–284.

Epstein, S., Buchsbaum, R., Lowenstam, H.A., Urey, H.C., 1953. Revised carbonatewater isotopic temperature scale. Bull. Geol. Soc. Am. 64, 1315–1326.

Ferreira, C.E.L., Floeter, S.R., Gasparini, J.L., Ferreira, P.B., Joyeux, J.C., 2004. Trophicstructure patterns of Brazilian reef fishes: a latitudinal comparison. J. Biogeogr.31, 1093–1106.

Ferreira, C.E.L., Gonçalves, J.E.A., Coutinho, R., Peret, A.C., 1998. Herbivory by thedusky damselfish Stegastes fuscus (Cuvier, 1830) in a tropical rocky shore:effects on the benthic community. J. Exp. Mar. Biol. Ecol. 229, 241–264.

272 F.A. Daros et al. / Regional Studies in Marine Science 3 (2016) 262–272

Floeter, S.R., Gasparini, J.L., 2000. The southwestern Atlantic reef fish fauna:composition and zoogeographic patterns. J. Fish Biol. 56, 1099–1114.

Forrester, G.E., Swearer, S.E., 2002. Trace elements in otoliths indicate the use ofopen-coast versus bay nursery habitats by juvenile California halibut.Mar. Ecol.Prog. Ser. 241, 201–213.

Gao, Y., Beamish, R.J., 2003. Stable isotope variations in otoliths of Pacific halibut(Hippoglossus stenolepis) and indications of the possible 1990 regime shift. Fish.Res. 60, 393–404.

Gao, Y., Joner, S.H., Svec, R.A., Weinberg, K.L., 2004. Stable isotopic comparisonin otoliths of juvenile sablefish (Anoplopoma fimbria) from waters off theWashington and Oregon coast. Fish. Res. 68, 351–360.

Hamer, P.A., Jenkins, G.P., 2007. Comparison of spatial variation in otolith chemistryof two fish species and relationships with water chemistry and otolith growth.J. Fish Biol. 71, 1035–1055.

Hamer, P.A., Jenkins, G.P., Gillanders, B.M., 2003. Otolith chemistry of juvenilesnapper Pagrus auratus in Victorian waters: natural chemical tags and theirtemporal variation. Mar. Ecol. Prog. Ser. 263, 261–273.

Higgins, R., Isidro, E., Menezes, G., Correia, A.T., 2013. Otolith elemental signaturesindicate population separation in deep-sea rockfish, Helicolenus dactylopterusand Pontinus kuhlii, from the Azores. J. Sea Res. 83, 202–208.

Hixon, M.A., Brostoff, W.N., 1983. Damselfish as keystone species in reverse:intermediate disturbance and diversity of reef algae. Science 220, 511–513.

Høie, H., Folkvord, A., Otterlei, E., 2003. Effect of somatic and otolith growth rate onstable isotopic composition of early juvenile cod (Gadus morhua L.) otoliths. J.Exp. Mar. Biol. Ecol. 289, 41–58.

Høie, H., Otterlei, E., Folkvord, A., 2004. Temperature-dependent fractionation ofstable oxygen isotopes in otoliths of juvenile cod (Gadus morhua L.). ICES J. Mar.Sci. 61, 243–251.

Hostim-Silva, M., Bertoncini, A.A., Machado, L.F., Gerhardinger, L.C., Daros, F.A.,Barreiros, J.P., Godoy, E.A.S., 2006. Peixes de costão rochoso de Santa Catarina:I. Arvoredo, first ed. Universidade do Vale do Itajaí, Itajaí.

Huijbers, C.M., Nagelkerken, I., Debrot, A.O., Jongejans, E., 2013. Geographiccoupling of juvenile and adult habitat shapes spatial population dynamics ofcoral reef fish. Ecology 94, 1859–1870.

Humann, P., 1999. Reef Fish Identification: Florida, Caribbean, Bahamas. Jack-sonville. NewWorld Publication, Inc., Florida.

IBAMA, 1998. Proteção e controle de ecossistemas costeiros: manguezal da Baíada Babitonga. Instituto Brasileiro do Meio Ambiente e dos Recursos NaturaisRenováveis, Brasília, 146 p.

IPARDES, 1989. Zoneamento do litoral paranaense. Instituto Paranaense deDesenvolvimento Econômico e Social. Secretaria de Estado do Planejamento eCoordenação Geral. Fundação Édson Vieira, Curitiba, 175 p.

Kalish, J.M., 1991. 13C and 18O isotopic disequilibria in fish otoliths: metabolic andkinetic effects. Mar. Ecol. Prog. Ser. 75, 191–203.

Karmann, I., Neto, C.M.D., Weber, W., 1999. Caracterização litológica e estruturaldas rochas metassedimentares do conjunto insular cardoso, sul do Estado deSão Paulo. Rev. Bras. Geociênc. 29, 157–162.

Kingsford,M.J., Gillanders, B.M., 2000. Variation in concentrations of trace elementsin otoliths and eye lenses of a temperate reef fish, Parmamicrolepis, as a functionof depth, spatial scale, and age. Mar. Biol. 137, 403–414.

Lana, P.C., Marone, E., Lopes, R.M., Machado, E.C., 2001. The subtropical estuarinecomplex of Paranaguá Bay, Brazil. In: Seeliger, U., Kjerfve, B. (Eds.), CoastalMarine Ecosystems of Latin America. Springer-Verlag, Berlin, pp. 131–145.

Lara, M.R., Jones, D.L., Chen, Z., Lamkin, J.T., Jones, C.M., 2008. Spatial variation ofotolith elemental signatures among juvenile gray snapper (Lutjanus griseus)inhabiting southern Florida waters. Mar. Biol. 153, 235–248.

Lo-Yat, A., Meekan, M., Munksgaard, N., Parry, D., Planes, S., Wolter, M., Carleton,J., 2005. Small-scale spatial variation in the elemental composition of otolithsof Stegastes nigricans (Pomacentridae) in French Polynesia. Coral Reefs 24,646–653.

Mackay, D., Moreau, J., 1990. A note on the inverse function of the von Bertalanffygrowth function. Fishbyte 1, 23–31.

Marone, E., Machado, E.C., Lopes, C.R.M., Silva, E.T., 2005. Land-ocean fluxes in theParanaguá Bay Estuarine System. Braz. J. Oceanogr. 53, 169–181.

Marone, E., Noernberg, M.A., Dos Santos, I., Lautert, L.F., Andreoli, O.R., Buba, H.,Fill, H.D., 2004. Hydrodynamics of Guaratuba Bay, PR, Brazil. J. Coast. Res. 39,1879–1883.

Martin, G.B., Wuenschel, M.J., 2006. Effect of temperature and salinity on otolithselement incorporation in juvenile gray snapper Lutjanus griseus. Mar. Ecol. Prog.Ser. 324, 229–239.

Matta, M.E., Orland, I.J., Ushikubo, T., Helser, T.E., Black, B.A., Valley, J.W., 2013.Otolith oxygen isotopes measured by high-precision secondary ion massspectrometry reflect life history of a yellowfin sole (Limanda aspera). RapidCommun. Mass Spectrom. 27, 691–699.

Miyao, S.Y., Hishihara, L., Sarti, C.C., 1986. Características físicas e químicasdo sistema estuarino-lagunar de Cananéia-Iguapê. Bolm. Inst. Oceanogr. 34,23–36.

Mizerkowski, B.D., Machado, E.C., Brandini, N., Nazario, M.G., Bonfim, K.V., 2012.Environmental water quality assessment in Guaratuba Bay, state of Paraná,Southern Brazil. Braz. J. Oceanogr. 60, 109–115.

Nemes, D.D.,Marone, E., 2013. Caracterização das ondas de superfície na plataformainterna do Estado do Paraná, Brasil. Biol. Parana. Geociênc. 68, 12–25.

Noernberg, M.A., Lautert, L.F.C., Araújo, A.D., Marone, E., Angelotti, R., Netto Jr.,J.P.B., Krug, L.A., 2006. Remote sensing and GIS integration for modelling theParanaguá Estuarine Complex, Brazil. J. Coast. Res. 39, 1627–1631.

Otero, X.L., Ferreira, T.O., Huerta-Díaz, M.A., Partiti, C.S.M., Souza Jr., V., Vidal-Torrado, P., Macías, F., 2009. Geochemistry of iron and manganese in soils andsediments of a mangrove system, Island of Pai Matos (Cananéia—SP, Brazil).Geoderma 148, 318–335.

Passos, A.C., Contente, R.F., Abbatepaulo, F.V., Spach, H.L., Vilar, C.C., Joyeux, J.C.,2013. Analysis of fish assemblages in sectors along a salinity gradient based onspecies, families and functional groups. Braz. J. Oceanogr. 61, 251–264.

Passos, A.C., Contente, R.F., Araujo, C.C.V., Daros, F.A., Spach, H.L., Abilhôa, V., Fávaro,L.F., 2012. Fishes of Paranaguá Estuarine Complex, South West Atlantic. BiotaNeotrop. 12, 226–238.

Patterson,W.P., 1999. Oldest isotopically characterized fish otoliths provide insightto Jurassic continental climate of Europe. Geology 27, 199–202.

Patterson, H.M., Thorrold, S.R., Shenker, J.M., 1999. Analysis of otolith chemistryin Nassau grouper (Epinephelus striatus) from the Bahamas and Belize usingsolution-based ICPMS. Coral Reefs 18, 171–178.

Peel, C.M., Finlayson, B.L., McMahon, T.A., 2007. Update world map of the Köppen-Geiger climate classification. Hydrol. Earth Syst. Sci. 11, 1633–1644.

Radtke, R.L., Showers, W., Moksness, E., Lenz, P., 1996. Environmental informationstored in otoliths: insights from stable isotopes. Mar. Biol. 127, 161–170.

Romero, A.F., Riedel, P.S., Milanelli, J.C.C., 2010. Carta de sensibilidade ambiental aoóleo do Sistema Estuarino-lagunar de Cananéia-Iguapê, litoral sul de São Paulo.Rev. Bras. Cartogr. 62, 229–238.

Rooker, J.R., Zdanowicz, V.S., Secor, D.H., 2001. Chemistry of tuna otoliths:assessment of base composition and postmortem handling effects. Mar. Biol.139, 35–43.

Sadovy, Y., Severin, K., 1994. Elemental patterns in Red Hind (Epinephelus guttatus)otoliths from Bermuda and Puerto Rico reflect growth rate, not temperature.Can. J. Fish. Aquat. Sci. 51, 133–141.

Schaeffer-Novelli, Y., Mesquita, H.S.L., Cintrón-Molero, G., 1990. Cananéia lagoonestuarine system, São Paulo, Brazil. Estuaries 13, 193–203.

Schwamborn, H.L.S., Ferreira, B.P., 2002. Age structure and growth of thedusky damselfish, Stegastes fuscus, from Tamandaré reefs, Pernambuco, Brazil.Environ. Biol. Fish. 63, 79–88.

Schwarcz, H.P., Gao, Y., Campana, S., Browne, D., Brand, U., 1998. Stable carbonisotope variations in otolith Atlantic cod (Gadus morhua). Can. J. Fish. Aquat.Sci. 55, 1798–1806.

Silva, D.M., Santos, P., Correia, A.T., 2011. Discrimination of Trisopterus luscus stocksin northern Portugal using otoliths elemental fingerprints. Aquat. LivingResour.24, 85–91.

Smith, S.J., Campana, S.E., 2010. Integrated stock mixture analysis for continuousand categorical data, with application to genetic–otolith combinations. Can. J.Fish. Aquat. Sci. 67, 1533–1567.

Solomon, C.T., Weber, P.K., Cech, J.J., Imgram, B.L., Conrad, M.E., Machavaram, M.V.,Pogodina, A.R., Franklin, R.L., 2006. Experimental determination of the sourcesof otolith carbon and associated isotopic fractionation. Can. J. Fish. Aquat. Sci.63, 79–89.

Souza, L.L.G., Chellappa, S., Gurgel, H.C.B., 2007. Biologia reprodutiva do peixe-donzela, Stegastes fuscus Cuvier, em arrecifes rochosos no nordeste do Brasil.Rev. Bras. Zool. 24, 419–425.

Sturgeon, R.E., Willie, S.N., Yang, L., Greenberg, R., Spatz, R.O., Chen, Z., Scriver, C.,Clancy, V., Lam, J.W., Thorrold, S., 2005. Certification of a fish otolith referencematerial in support of quality assurance for trace element analysis. J. Anal. At.Spectrom. 20, 1067–1071.

Thorrold, S.R., Campana, S.E., Jones, C.M., Swart, P.K., 1997. Factors determiningδ13C and δ18O fractionation in aragonite otoliths of marine fish. Geochim.Cosmochim. Acta 61, 2909–2919.

Thresher, R.E., 1991. Geographic variability in the ecology of coral reef fishes:evidence, evolution and possible implications. In: Sale, P.F. (Ed.), The Ecologyof Fishes on Coral Reefs. Academic Press, San Diego, pp. 401–436.

Thresher, R.E., 1999. Elemental composition of otoliths as a stock delineator infishes. Fish. Res. 43, 165–204.

Townsend, D.W., Radtke, R.L., Corwin, S., Libby, D.A., 1992. Strontium: calciumratios in juvenile Atlantic herring, Clupea harengus L. otoliths as a function ofwater temperature. J. Exp. Mar. Biol. Ecol. 160, 131–140.

Walther, B.D., Thorrold, S.R., 2006. Water, not food, contributes the majority ofstrontium and barium deposited in the otoliths of amarine fish. Mar. Ecol. Prog.Ser. 311, 125–130.

Webb, S.D., Woodcock, S.H., Gillanders, B.M., 2012. Sources of otoliths barium andstrontium in estuarine fish and the influence of salinity and temperature. Mar.Ecol. Prog. Ser. 453, 189–199.

Wellington, G.M., Victor, B.C., 1989. Planktonic larval duration of one hundredspecies of Pacific and Atlantic damselfishes (Pomacentridae). Mar. Biol. 101,557–567.

Woodcock, S.H., Munro, A.R., Crook, D.A., Gillanders, B.M., 2012. Incorporation ofmagnesium into fish otoliths: Determining contribution from water and diet.Geochim. Cosmochim. Acta 94, 12–21.