Marine Pollution Bulletin 76 (2013) 333–348

Contents lists available at ScienceDirect

Marine Pollution Bulletin

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

Impact evaluation of the industrial activities in the Bay of Bakar (AdriaticSea, Croatia): Recent benthic foraminifera and heavy metals

0025-326X/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.marpolbul.2013.09.039

⇑ Corresponding author. Tel.: +385 1 460 6116; fax: +385 1 460 6081.E-mail address: jelena.vidovic@geol.pmf.hr (J. Vidovic).

Adriana Popadic a, Jelena Vidovic b,⇑, Vlasta Cosovic b, Davorin Medakovic a, Matej Dolenec c, Igor Felja b

a Center for Marine Research Rovinj, Ruder Boskovic Institute, Giordano Paliaga 5, 52 210 Rovinj, Croatiab Department of Geology, Faculty of Science, University of Zagreb, Horvatovac 102a, 10 000 Zagreb, Croatiac Department of Geology, Faculty of Natural Sciences and Engineering, University of Ljubljana, Aškerceva 12, 1000 Ljubljana, Slovenia

a r t i c l e i n f o

Keywords:PollutionMajor, minor and trace elementsForaminiferaSedimentAdriatic Sea

a b s t r a c t

The Bay of Bakar is one of the most heavily polluted bays at the Eastern Adriatic. Three major industrialcompanies potentially endanger the bay. The concentration of major, minor and trace elements in surfacesediments from thirteen stations was discussed in relation to the sediment type and foraminiferal assem-blages. The distribution of major elements in the bay is influenced by geological nature of surroundings.Heavy metal distribution depends on pollution sources and on amount of mud fraction: fine-grained sed-iments are enriched by them in comparison with coarse-grained ones. Different sediment quality criteriacomplicate the pollution assessment in the bay. Heavy metal concentrations generally fall into alloweddepositional values for marine environments; only area in front of the coke plant and the City of Bakarharbor is heavily polluted. Stress-tolerant foraminiferal species dominate at stations with higher concen-trations of heavy metals and coarse-grained sediments consist of larger number of epifaunal taxa.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Anthropogenic activities like industry threaten the environmentwith the possibility of emission of various heavy metals, whosedistinctive feature is their non-biodegradability. After the dis-charge, they redistribute in water in particulate and dissolvedphases, precipitate or settle down in the sediment, accumulate inthe organisms and are consumed by them (Fichet et al., 1998).Through biogeochemical processes of desorption and resuspen-sion, they can be removed from the sediment and accordingly be-come the long-term source of the contamination (Fichet et al.,1998; Ouyang et al., 2006). The research on the Adriatic shelf re-vealed that the highest concentration of heavy metals is in thesouthern Adriatic and along the Italian coastal area, following thehighest concentrations of clay minerals (Vdovic et al., 1991). Con-trary to this, the Eastern Adriatic sediments have very high concen-tration of carbonates and consequently low heavy metal content(Dolenec et al., 1998; De Lazzari et al., 2004), with local anomaliesrelated to different types of anthropogenic activities (Ujevic et al.,2000; Mikulic et al., 2004, 2008; Lovrencic et al., 2005; Vreca andDolenec, 2005; Valkovic et al., 2007; Obho -daš and Valkovic,2010; Castelli and Kljajic, 2010; Cukrov et al., 2011; Obho -dašet al., 2006, 2012).

The distribution of benthic foraminifera in polluted marineenvironments has been the object of scientific interest for the last50 years (Resig, 1960; Watkins, 1961; Botlovskoy, 1965), oftenemphasizing that foraminifera are one of the most responsive indi-cators available for the environmental monitoring of the pollutionin the marine environments (Kramer and Botterweg, 1991).Numerous studies have been published in recent decades, focusingon the effect of different sources of pollution (urban organic waste,aquacultures, agricultural and industrial activities) on living fora-miniferal communities (e.g. Alve, 1995; Yanko et al., 1998; McG-ann et al., 2003; Armynot du Châtelet et al., 2004; Bergin et al.,2006; Nigam et al., 2006; Bouchet et al., 2007; Carnahan et al.,2009). The majority of studies in the Adriatic Sea dealing withforaminifera as bioindicators of various anthropogenic activitiesare from the Italian coast (e.g. Donnici et al., 1997; Coccioni,2000; Ferraro et al., 2006; Albani et al., 2007; Frontalini and Cocci-oni, 2008, 2012; Coccioni et al., 1997, 2009). There is only onestudy in the Eastern Adriatic that concerns the effect of pollutionon foraminiferal communities, investigating the impact of organicpollution coming from fish farming (Vidovic et al., 2009), followedby the updated and annotated list of determined foraminifera fromthe sea-bottom sediments (Cosovic et al., 2011). To the presentday, there is no documented study dealing with the effect of indus-trial activities and associated heavy metals on foraminiferal com-munities in the Eastern Adriatic, which draw our attention to theneed of carrying out such study.

334 A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348

The aim of this research is to identify benthic foraminiferalcommunity in the broader area of the Bay of Bakar and to studytheir abundance in relation with heavy metal concentrations inthe sediments. For that purpose we analysed: (1) the abundanceof major, minor and trace elements in polluted sediments, with aspecial regard to the trace elements, which are potentially toxiccontaminants (‘‘heavy metals’’); (2) the distribution and concen-tration of heavy metals coming from anthropogenic sources; (3)the composition, diversity and distribution of total foraminiferalassemblages in the surface sediments; (4) the correlation (if any)among fauna and chemical properties of the sediment, all withthe intention to make the assessment of industrial activities inthe bay and to contribute to present cognition of environmentalmonitoring tools and practice procedures.

2. Study area

The Bay of Bakar is located in the northern part of the EasternAdriatic coastal area, and is considered as one of the most heavilypolluted bays along the Croatian part of the coast. Geological fea-tures of the broader area around the Bay of Bakar include depositsranging from the lower Cretaceous to the top of the Quaternary:lower Cretaceous Limestone and breccias, upper Cretaceous lime-stones, dolomites and dolomite breccias, Eocene foraminiferallimestones and flysch deposits, Eocene–Oligocene limestone brec-cias and Quaternary diluvia deposits (Grimani et al., 1963; Šušnjaret al., 1970; Fig. 1). There are three major industrial companieswith associated facilities located in the broader area of the bay:Oil refinery INA Urinj, Bulk cargo terminal of the Port of Rijekaand the area of the former Coke plant Bakar. Besides mentioned

Fig. 1. Geological map of the study area wi

industries, the bay may be contaminated by discharges comingfrom small harbor and from domestic sewage.

The coastline of the City of Urinj due to its water depth is suit-able for tankers up to 200,000 dwt. Oil refinery does not producewastewaters, but possible oil spills represent potential danger forthe environment (Ðekic, 2005).

Bulk cargo terminal of the Port of Rijeka is intended for thetransshipment of iron ores and coal. The terminal is modernizedwith novel transshipment equipment that reduces the emissionof the dust (Dubrovic, 2001).

The coke plant in the City of Bakar began with the work in 1978and was closed in 1994. Within the coke plant 15 millions tones ofcoal had been processed, 11 tones of coke and 440,000 t of raw coaltar had been produced. During the activity of the coal plant, therehad been unchartered emissions of raw coal, oil and naphthalene.The degradation of the equipment and the parts of the coke plantlasted from 1994 to 2001, while the chimney was torn down in2005 (Nadilo and Sojcic, 2005).

Water well Dobrica is located near submarine spring whichcauses lower salinity in the neighbouring area. Similar environ-mental conditions are found at the rainfall drain and fresh waterspring in the City of Bakar.

Thirteen localities in the broader area of the Bay were selectedfor the study (Fig. 1, Table 1): two stations in the area of the Oilrefinery INA Urinj, the harbor of the City of Bakarac, the site withthe rainfall accumulation in the City of Bakarac, water well Dobri-ca, the site of INA Urinj Petroleum decanter, two stations in thearea of the coke plant, two stations at the Bulk cargo terminal,the rainfall drain in the City of Bakar and fresh water spring andsewage disposal in the City of Bakar. The reference station is lo-cated outside the bay.

th the positions of sampling locations.

Table 1Station numbers used in Fig. 1 with descriptions of associated anthropogenic activities and characteristics of sampling stations: water depth (m), sea water temperature (�C), pH,salinity (‰), bottom-water dissolved oxygen (ml/l) and oxygen saturation (%).

Stationnumber

Samplename

Location Water depth(m)

Temp.(�C)

pH Salinity(‰)

Dissolved oxygen(ml/l)

Oxygensaturation (%)

1. B1 Oil refinery INA Urinj 9 16 8.13 38.0 7.0 892. B2 Oil refinery INA Urinj 7 16 8.14 38.2 7.4 953. B3 Harbor of the City of Bakarac 9.5 17 8.15 38.2 7.2 944. B4 City of Bakarac (rainfall accumulation) 9.3 17 8.15 38.0 7.1 935. B5 Water well Dobrica 15.5 17 8.16 38.1 7.2 946. B6 INA Urinj (Petroleum decanter) 19.5 17 8.17 38.0 7.3 957. B7 Coke plant 17 16 8.16 38.3 7.1 918. B8 Bulk cargo terminal 24.5 17 8.16 38.4 7.1 1009. B9 Bulk cargo terminal 21 17 8.16 38.0 8.0 10410. B10 Coke plant (former coal and coke depot) 21 16 8.16 38.2 7.5 9611. B11 City of Bakar (rainfall drain) 10 17 8.15 38.1 7.4 9612. B12 Harbor in the City of Bakar (fresh water spring,

sewage)11 17 8.13 38.9 6.2 81

13. R1 Reference station 32 18 8.11 35.86 8.2 108

A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348 335

3. Methods

3.1. Sediment sampling

During the October, 2009, 13 stations in the broader area of theBay of Bakar were sampled. Samples were collected by scuba diverusing short PVC corers (maximum penetration depth 30 cm,75 mm in diameter). For the purpose of this study, only surfacesediments (0–2 cm) from the collected cores were analysed. Seatemperature and water depth were measured at each station dur-ing the sampling. Sediments were frozen immediately after sam-pling and preserved for foraminiferal, granulometrical andgeochemical analyses.

3.2. Granulometrical analysis

Granulometrical analysis was done in the laboratory of theDepartment of Geology, Faculty of science, University of Zagreb.Each sediment sample was treated with 30% hydrogen peroxideto remove organic matter prior to the grain size analysis. The grainsize of the sediment samples was analyzed by wet sieving, usingASTM standard stainless steel sieves with aperture sizes of 0.063,0.125, 0.250, 0.500, 1, 2 and 4 mm. Sediments were classifiedaccording to their gravel–sand–mud ratio (Folk, 1954). Sedimentfrom the reference station was analyzed only on sieves with aper-ture sizes of 0.063 and 2 mm.

3.3. Foraminiferal analyses

After the unfreezing, samples were washed on the 63 l sieveand dried at 50 �C. Each sample was split by Reich microsplitterin aliquots of at least 300 specimens that were subject of furtheranalysis. Individuals were identified under binocular magnifier fol-lowing the generic classification of Loeblich and Tappan (1987) andCimerman and Langer (1991). An estimation of the species diver-sity was based on diversity indices: Dominance D, Shannon–Wie-ner index H, Evenness e, Equitability index J, Fisher a index.Diversity indices were computed using PAST program (Hammeret al., 2001). Qualitative (genera and species determination) andquantitative (number of individuals, absolute and relative abun-dance of species, dominant species, foraminiferal density per 1 g,percentage of damaged and deformed individuals) analyses of fora-miniferal assemblages were performed. Considering foraminiferaldensity per 1 g, three categories were recognized: low (up to1000 tests), medium (1000–10,000) and high (greater than10,000 tests). Foraminiferal species were interpreted according to

their habitat, with special regard to epiphytic species, which ap-pear to be less sensitive to the acute toxicity of Cd, Cu and Hg(Bresler and Yanko, 1995). Finally, Yanko et al., 1999 (followedby Armynot du Châtelet et al., 2004; Luciani, 2007; Romanoet al., 2008; Coccioni et al., 2009) summarized the list of generawith known tolerance to pollution (Ammonia, Bolivina, Cribroelphi-dium, Haynesina, Elphidium, Ammobaculutes and Trochammina),according to which stress tolerant taxa in this study weredetermined.

3.4. Geochemical analyses

Samples of sea bottom water were sampled with plastic bottles.Sea water was taken in order to determine its chemical parame-ters: pH, salinity, dissolved oxygen, oxygen saturation. Givenparameters were measured at the Croatian National Institute ofPublic Health in Rijeka, Croatia.

In order to determine the elemental composition of the sedi-ment, bulk samples were analysed using Handheld Thermo Scien-tific Niton XL3t + GOLDD 900S-He energy-dispersive X-rayfluorescence (EDXRF) analyzer. Approximately 3 g of each samplewas placed into stainless capsules and with special hand-heldpress containing stainless hammer were pressed into pills and in-serted into the opening of the analyzer. The upper surface of eachcompressed sample was analysed. Obtained results represent theaverage elemental composition of the entire sample. During theanalysis two modules were used: »Mining« module for the mea-surement of main elements and »Soil« module for the detectionof heavy metals. While using »Mining« module, helium was addedinto the analyzer, which allowed better detection of light elements(Mg, Si, Al, P). The measurement time for each sample was 210 swith »Mining« module and 180 s with »Soil« module. To monitorthe precision and accuracy of the method, 6 international referencestandards were measured (NIST-1d: limestone, NIST-1633a: coalfly ash, NIST-88b: dolomitic limestone, BHVO-1: basalt, STM-1:syenite, SGR-1: Green River shale), both at the beginning and atthe end of the analyses. The results were subsequently recalculatedaccording to the values for each reference standard.

3.5. Statistical analyses

Statistical analyses were performed using Past program (Ham-mer et al., 2001). Prior to analyses an additive logarithmic transfor-mation log (x + 1) was preformed on standardized data (relativeabundances of foraminiferal species) in order to reduce the impor-tance of extreme values, to reduce the contribution of common

Table 2Percentages of sediment fractions of each sample with sediment type classification according to Folk (1954).

B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 R1U mm % % % % % % % % % % % % %

�2 >4 0.26 0.22 0.04 0 5.9 0 5.07 53.94 4.32 43.34 0 0.17�1 2–4 1.4 4.46 0.06 0 8.96 0.52 8.33 9.64 8.8 10.52 0.1 0.3 7.50 1–2 9.28 36.34 0.24 0.06 4.04 2.7 10.93 4.28 22.52 6.56 0.23 0.81 0.5–1 35.56 53 0.56 0.18 5 9.82 13.43 3.5 20.04 4.36 1 2.72 0.25–0.5 23.64 3.84 3 2.02 5.44 23.04 12.57 2.72 12.8 4.36 1.7 43 0.125–0.25 18.52 0.08 16.42 68.62 12.6 38.02 15.97 4.48 9.78 4.52 3.9 9.834 0.063–0.125 7.18 0.02 47.2 24.22 26.36 19.08 16.4 5.36 7.4 4.4 25.27 29.73 83.8– <0.063 4.16 2.04 32.48 4.9 31.7 6.82 17.3 16.08 14.34 21.94 67.8 52.47 8.7

Sed. type (g)S (g)S mS S gmS S gmS msG gmS msG sM sM gmS

336 A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348

species, to enhance the contribution of the rare species and to nor-malize the data (Krebs, 1998).

Cluster analysis is technique for identifying groups and sub-groups in a multivariate dataset, based on a given distance or sim-ilarity measure (Hammer and Harper, 2006). Ward’s method(Dillon and Goldstein, 1984) was used for measuring similarity be-tween the clusters.

Principal component analysis (PCA) and detrended correspon-dence analysis (DCA) are projections of multivariate dataset downto a few dimensions, in order to visualize geographical trends,groupings and underlying environmental gradients from taxo-nomic counts in a number of samples (Hammer and Harper, 2006).

Correlation is compatibility in varying of two or more variables.The strength of correlation is measured using the correlation coef-ficient ‘‘r’’, also known as Pearson’s r or the product-moment cor-relation coefficient (Hammer and Harper, 2006). Values ofcorrelation coefficient (r) range from zero (indicating no correla-tion between variables) to ±1, pointing to complete positive or neg-ative correlation (Hammer and Harper, 2006). Correlation waspreformed for component and axis from multivariate plots (PCA,DCA) which carry the highest percentage of variance (component1 and axis 1). Correlation with sediment type was performed usingthe percentage of fine-grained fraction (mud) in the sediment.

3.6. Generating spatial distribution maps

Geochemical maps of selected elements were generated in theArcGIS extension of the Geostatistical Analyst. Inverse DistanceWeighting (IDW), a deterministic spatial interpolation techniquewas used to present the concentration of a particular element asaccurately as possible. The interpolation was preformed on 12nearest input sample points. The results of the chemical analyseswere divided into classes. Spatial distribution of every particularclass in the map was represented as a polygon in a correspondingcolor, with blue nuances for low and red for high concentrations.Map with distribution of mud share in the bay was generated inthe same way, with light brown nuances for low and dark brownfor high mud shares.

4. Results

4.1. Sedimentological analysis

Sediment types are determined using Folk (1954) classification,based on shares of gravel, sand and silt in samples. The grain sizeanalysis of the sediments showed that the structure of the samplesmostly consisted of sand. Samples B1 and B2 are classified asslightly gravelly sand, B3 as muddy sand, B4 and B6 were deter-mined as sand, B5, B7, B9 and R1 as gravelly muddy sand, B8 andB10 as muddy sandy gravel and B11 and B12 as sandy mud(Table 2). Due to fact that there is no evident decrease in grain size

with water depth, factors like location of sampling site and waterenergy are of importance.

4.2. Foraminiferal assemblages

Total foraminiferal assemblages in the Bay of Bakar consisted ofbenthic species, with high percentage of individuals belonging tosuborders Miliolina and Rotaliina. Planktonic specimens were notfound in examined samples (Table 3).

The assemblages in the sediments near the Oil refinery INA Ur-inj (Stations B1, B2 and B6) were dominated by epiphytic species:station B1 by Peneroplis pertusus, Cibicides advenum, Elphidium cri-spum, Rosalina sp., Quinqueloculina sp. and Asterigerinata mariae, to-gether making 32% of the community; station B2 with epiphyticQuinqueloculina sp., E. crispum, Sorites orbiculus, Neoconorbina ter-quemi, Rosalina floridana and Cycloforina contorta, making withinfaunal Ammonia convexa 40% of the assemblage; station B6,where Asterigerinata mamilla, A. mariae, N. terquemi and A. convexamade 40% of the community. This group of samples is character-ized by the dominance of miliolids (from 21% to 43%) and minimalcontribution of stress-tolerant taxa (9–11%). The abundance offoraminiferal tests varied from the lowest to the medium values(68–9185 specimens per g), while quantity of mechanically dam-ages tests reached up to one third of the foraminiferal assemblage(Fig. 2). Station B1 had 3.2% of deformed individuals (Table 4).

At station B3, located in the vicinity of the harbor of the City ofBakarac, Textulariids made up 12% of the assemblage; stress-toler-ant taxa made up 38%, while smaller miliolid taxa were subordi-nate and made up 8%. The assemblage was dominated by stress-tolerant species, with Ammonia tepida as the largest contributor(13%). Other dominant species were: Aubignyna perlucida (9%),Elphidium granosum (7%), Eggerelloides scaber (6%), Haynesinadepressula (6%), A. mariae (4%) and Elphidium incertum (4%). Almost16% of all tests had a kind of damage marks and there were no sin-gle deformed specimen in this assemblage. Foraminiferal densitywas around 1000 individuals per g of sediment (Fig. 2; Table 4).

The assemblage near the rainfall accumulation in the City ofBakarac (station B4) was dominated by A. tepida, which made36% of the assemblage, while the rest of the community was com-posed of species whose abundance is 4% or less. Around 17% offoraminiferal tests were damaged and further 3.2% of tests weredeformed. Foraminiferal density was very low at this station(Fig. 2; Table 4).

Water well Dobrica (station B5) had the community composedpredominantly of epiphytic species: Textularia conica (13%), Textu-laria bocki (5%), A. mamilla (6%), N. terquemi (5%), Elphidium sp. (4%),Rosalina floridana (4%) with one epifaunal species Epistominella ex-igua (4%). Among 59 identified species, around 8.6% of them weredamaged and foraminiferal density was around 1000 individualsper g of sediment (Fig. 2; Table 4).

The assemblage at the station B7, located near the coke plant,was dominated by epiphytic species (62%), among which smaller

Table 3Relative abundance of foraminiferal species for stations B1–B12 and for the reference station (R1).

B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 R1

Lagenammina sp. 2 1 1Reophax scorpiurus 1Eggerella sp. 2 1Eggerelloides scaber 6 4 1 1Textularia sp. 1 1 1 1Textularia agglutinans 1 1 1Textularia bocki 1 2 3 5 2 1 1 1 2 1Textularia conica 1 2 13 3 6 14 2 10 4 2 2Vertebralina striata 1Adelosina sp. 1 1Adelosina cliarensis 1Adelosina dubia 2 2A. mediterranensis 2Adelosina longirostra 1Spiroloculina ornata 1Biloculinella labiata 1Cycloforina contorta 4 2 1 1 1 2Cycloforina villafranca 1Miliolinella sp. 1 1 1Miliolinella subrotunda 1 1 1 3 2 3 1 2 2Pseudotriloculina rotunda 1 1Quinqueloculina sp. 5 7 3 4 3 3 2 5 2 3 2 5 5Quinqueloculina lata 1 2 1 1Q. annectens 2 1 3Q. berthelotiana 2 1 1 1 2 1 1 2Q. bicarinata 1 3 3 1 5 3 2 3 8Q. bidentata 1Quinqueloculina bosciana 1 2 1 1 3 1Quinqueloculina bradyana 1Quinqueloculina jugosa 2Quinqueloculina laevigata 2Q. lamarckiana 2 2Q. nodulosa 1Quinqueloculina parvula 1Quinqueloculina seminula 1 2 1 1 1 1 10 1Quinqueloculina stalkeri 1Q. viennensis 1 1 1Sigmoilinita costata 2 1 3 2 1 1 2 2Siphonaptera sp. 1 1 1Siphonaptera agglutinans 1 1 1Siphonaptera aspera 4Siphonaptera dilatata 1 1Siphonaptera irregularis 1 1Triloculina sp. 1 1 1 1 1 2 1 2 1Triloculina adriatica 1Triloculina marioni 1 3 1 2 2 1 1Triloculina tricarinata 1 1 1 2 1Peneroplis sp. 3 1Peneroplis pertusus 8 2 1 3Sorites orbiculus 5Fissurina sp. 1Lagena sp. 1 1Lagena doveyensis 1Lenticulina cultrata 1Globulina gibba 1Bolivina sp. 3Bolivina pseudoplicata 1 1 1 1 1 1Brizalina sp. 1Brizalina spathulata 2 4 2Brizalina variabilis 2 1 3 4 3 4G. subglobosa 1 3Stainforthia fusiformis 1Bulimina elongata 2 1 1 2 1Bulimina inflata 1Bulimina marginata 2 2 2 11 7 6 1Reussella spinulosa 2 2 2 3 2 3 1 1Conorbella pulvinata 1Gavelinopsis praegeri 1 2 2 1 4 1Neoconorbina sp. 1 1Neoconorbina terquemi 3 4 2 5 11 13 3 12 1 12Rosalina sp. 5 1 1 1 1 3 3 1 2Rosalina bradyi 2 3 2 1 3 3 2 2 2 6Rosalina floridana 3 4 2 4 1 2 3 3 1 2 3Rosalina globularis 2

(continued on next page)

A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348 337

Table 3 (continued)

B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 R1

Rosalina macropora 1 1Epistominella exigua 3 4 1 14 1 17 11 19 1Cibicides sp. 3 1 1 1 2 2 3 1Cibicides advenum 5 3 2 3 2 1 1 1 4Cibicides boueanum 1Lobatula lobatula 3 2 1 1 2 1 1 2 1 5P. mediterranensis 1 1 1 1 2 1 3 2Asterigerinata sp. 1 1Asterigerinata mamilla 1 2 6 16 17 4 8 3 4Asterigerinata mariae 4 3 4 3 7 5 1 3 3 7Haynesina depressula 2 6 1 1 1 3 1 2 8 10 3Haynesina germanica 2 1 1 3 1Nonionella turgida 1 2 4 2Nonionoides turgida 1 1Aubignyna perlucida 9 2 1 1 4 4Buccella frigida 2Buccella granulata 3 1 3 4 1Elphidium sp. 2 3 2 1 4 3 3 2 4 4 6 1 3Elphidium aculeatum 1 1Elphidium advenum 1 2Elphidium complanatum 2 1 2 3 3 6Elphidium crispum 5 7 1 1 3 1 1Elphidium cuvilleri 2 1 3 3 2 2 1 1Elphidium excavatum 2Elphidium gerthi 3 3 1Elphidium granosum 1 7 2 2 2 4 1 8 1Elphidium incertum 4 3 1 2 1 4 2Elphidium pauciloculum 1 3Elphidium pulvereum 1 1 1Elphidium punctatum 1 2 2 1 1 1 1 1Ammonia sp. 1 3 2 2 1Ammonia beccarii 1 2Ammonia convexa 3 9 1 6 2 4 3Ammonia tepida 1 3 13 36 3 3 3 2 10 7Eponides sp. 1

Fig. 2. Foraminiferal density (no of individuals per 1 g) compared to the mud sharein the sediment.

338 A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348

miliolids contributed with 21%. The most abundant species were:A. mamilla and N. terquemi, which made up 30% of the community,

while T. conica and A. mariae made further 11%. Stress-tolerant spe-cies made 15% of the assemblage. One third of all tests were frag-mented. Foraminiferal density was around 4700 individuals per gof sediment (Fig. 2; Table 4).

The assemblage in front of the bulk cargo terminal (B8 and B9)had up to 10% of stress-tolerant species and 16–36% of epiphytictaxa. Again species T. conica and E. exigua together made 28% ofthe community at the station B8 and adjoining genus Quinquelocu-lina made further 10%. At station B9, the most abundant specieswere N. terquemi and A. mamilla, making together 20% of the com-munity. Station B8 had approximately 600 individuals and stationB9 1500 individuals per g of sediment (Fig. 2). The percentage ofdeformed individuals varied from 10 to 26, and the percentage ofdeformed specimens was negligible (Table 4).

Former coal and coke depot was monitored at station B10. Spe-cies E. exigua made 17%, Bulimina marginata 11% and T. conica 10%of the assemblage. Low density of foraminiferal tests and negligibleamount of deformed tests characterize this foraminiferal group(Fig. 2; Table 4).

Station B11 is located near the rainfall drain in the City of Bak-arac. Stress-tolerant species made up one third of the assemblage:E. exigua (11%), A. tepida (10%), H. depressula (8%), E. granosum (8%),B. marginata (7%) and Elphidium sp. (6%), while miliolids made uponly 9%. The abundance of foraminiferal tests per gram was veryhigh (11,913 individuals) and around 13% of specimens were de-formed (Fig. 2; Table 4).

Fresh water spring and sewage disposal area in the harbor ofthe City of Bakar were studied at station B12. Once more, epifaunalspecies prevailed in the community: E. exigua, Quinqueloculina sp.,Quinqueloculina seminula and Quinqueloculina bicarinata, makingtogether 42% of the assemblage. H. depressula, A. tepida and B. mar-ginata made further 23%. Around 17% of broken tests were found in

Table 4No of taxa, no of individuals, diversity indices, foraminiferal density per 1 g and percentage od damaged and deformed specimens for stations B1–B12 and for the reference stationR1.

B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 R1

Taxa_S 53 55 42 48 59 43 41 50 61 50 38 37 47Individuals 286 276 270 229 268 337 287 287 318 307 289 348 228Dominance_D 0.03 0.04 0.05 0.15 0.04 0.06 0.06 0.06 0.04 0.06 0.06 0.08 0.05Shannon_H 3.7 3.6 3.3 2.9 3.6 3.3 3.2 3.3 3.6 3.2 3.1 2.9 3.4Evenness_e 0.7 0.7 0.7 0.4 0.6 0.6 0.6 0.6 0.6 0.5 0.6 0.5 0.6Equitability_J 0.9 0.9 0.9 0.8 0.9 0.9 0.9 0.8 0.9 0.8 0.8 0.8 0.9Fisher_alpha 19.1 20.6 13.9 18.5 23.3 13 13.1 17.5 22.4 16.9 11.7 10.4 17.9Density per 1 g 1462 68 1011 216 1261 9185 4733 632 1464 610 11,933 15,909 1028Damaged (%) 32.1 22.1 15.9 17.4 8.58 23.7 27.5 10.1 26.1 13.6 13.4 16.9 18.4Deformed (%) 3.2 1 0 3.2 1 0 0 0.7 0 0.4 0.7 2.1 1.3

A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348 339

this assemblage and only 2% of individuals were deformed. Forami-niferal density was very high, making almost 16,000 individualsper g of sediment (Fig. 2; Table 4).

The reference station, located outside the Bay of Bakar, had theassemblage with prevailing epiphytic species: N. terquemi, Rosalinabradyi, Lobatula Lobatula, A. mariae, A. mamilla, C. advenum, Quinqu-eloculina sp. and Elphidium complanatum, which together made 47%of the community. Around 18% of tests were fragmented and fora-miniferal density was 1028 individuals per g of sediment (Fig. 2;Table 4).

Diversity indices were more or less monotonous through allsamples (Table 4). Fisher alpha was very high, ranging from 10.4to 22.4. Shannon index was high and also equable through moni-tored area, ranging from 2.9 to 3.7. Equitability index showed thelowest variation (0.8–0.9). Evenness index was equable at themajority of stations, with slightly lower value at the station B4(0.4). Finally, dominance was generally low (0.03–0.08), whereasstation B4 had somewhat higher value (0.15).

4.3. Geochemical analyses

The abundance of major, minor and trace elements was exam-ined in sediments from the Bay of Bakar. The spatial distributionsof the elements, together with the results of the international ref-erence standards measured at the beginning and at the end of theanalyses, are presented in Table 5, while correlation coefficientsare given in Table 6. Spatial distribution maps were generated forelements which showed significant changes in their concentrationsacross the bay: Al, As, Ca, Cr, Cu, Fe, K, Mn, Ni, Pb, Rb, Si, Sr, Th, Ti, V,Zn and Zr (Figs. 3 and 4).

Silicon concentrations showed great variation among differentlocalities in the bay. Samples B1, B2 and R1 had the lowest concen-tration (0.79–1.98%), followed by samples B6, B7, B8 and B9, withconcentrations ranging from 8.28% to 9.92%. The highest concen-trations were measured in samples B3, B4, B5, B10, B11 and B12,with concentrations ranging from 12.61% to 22.58%.

The lowest concentrations of calcium were recorded in samplesB3, B4, B5 and B11, varying from 8.23% to 10.38%. In samples B10and B12 Ca concentrations were slightly higher (13.85–15.15%)and the highest values were obtained in samples B1, B2, B6, B7,B8, B9 and R1 (21.32–36.79%).

Concentrations of magnesium slightly varied among the sam-ples, ranging from 1.27% to 1.70%, with the exception of the sampleB4 where 4.42% of Mg was measured. Aluminum and kalium con-centrations in the sediments were low; aluminum ranged from0.20% to 5.69% and kalium from 0.08% to 1.32%. Iron concentrationswere low in samples B1, B2 and R1, ranging from 0.19% to 0.31%. Inall other samples, the abundance of Fe was higher than 1%, withmaximal value of 4.65%. Sediment concentrations of manganesevaried throughout the bay, ranging from 224.00 to 724.54 ppm.

Titanium abundance in sediments from the Bay of Bakar variedin great extent. Stations B1, B2, B6 and R1 had the lowest abun-dance of Ti, ranging from 0.01% to 0.03%; concentration values atstations B4, B7 and B9 ranged between 0.14% and 0.18%, while sta-tions B3, B5, B8, B10, B11 and B12 had the highest abundance of Ti,varying from 0.23% to 0.30%. Concentration of phosphorus wasvery low in the majority of samples, ranging from 0.02% to 0.17%and was not registered in samples B3, B4, B5 and B10.

The concentrations of heavy metals in sediments from the Bayof Bakar were as following: As (8.37–29.38 ppm), Cu (14.51–135.02 ppm), Ni (28.50–62.56 ppm), Pb (8.25–123.48 ppm), Rb(5.41–64.92 ppm), Sr (52.52–1960.40 ppm), Th (3.36–9.81 ppm),Zn (14.02–269.92 ppm), Zr (11.96–150.67 ppm), Cr (61.98–103.71 ppm), V (28.57–111.81 ppm) and Sb (2.42–3.29 ppm).

4.4. Statistical analyses

Cluster analyses were performed both for foraminiferal assem-blages and for measured elements. In first case, analysis was doneby using relative abundances of foraminiferal species, while in thesecond, elemental concentrations were used. Clustering procedurewas done according to Ward’s method. Analysis comprised all sam-ples and was based on relative abundance of species. Dendrogramobtained by cluster analysis of foraminiferal assemblages revealedtwo major clusters (Fig. 5). Cluster 1 comprised samples B3, B5, B8,B10, B11 and B12, while cluster 2 included samples B1, B2, B4, B6,B7, B9 and R1. Clustering of elemental concentrations divided sta-tions into two groups: group 1, with stations B3, B5, B8, B9, B10,B11, B12, and group 2, with stations B1, B2, B4, B6, B7 and R1(Fig. 6).

Principal component analysis (PCA) was done for each stationwith a view to determine differences in assemblages, to identifyvarieties in community based on such differences, and to definewhich parameter is controlling foraminiferal community composi-tion. PCA grouped samples similar as cluster analysis; samples cor-responding with cluster 1 were located at the left side of thediagram (negative values of component 1), while samples corre-sponding to the cluster 2 were located towards higher values ofcomponent 1 (Fig. 7). Two principal components together ex-plained 43.6% of the data variance. Detrended correspondenceanalysis revealed same grouping of the samples (Fig. 7).

In order to reveal inherent environmental factors representedby components in PCA and axes in DCA, measured elements werecorrelated with numerical values of component 1 and axis 1(Table 6). Finally, all elements were mutually correlated in orderto interpret data in terms of their provenance (source rock oranthropogenic origin), similarity or discrepancy of their chemicalbehavior, association with particular minerals (e.g. clay minerals)or sediment fraction and relation with other elements. Obtainedcorrelation coefficients are presented in Table 6.

Table 5The elemental composition of the samples and international reference standards. The explanation of the abbreviations: NIST-1d: limestone, NIST-1633a: coal fly ash, NIST-88b: dolomitic limestone, BHVO-1: basalt, STM-1: syenite, SGR-1: Green River shale.

Units B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 R1 NIST-1d NIST-1633a NIST-88b BHVO-1 STM-1 SGR-1

Si % 1.98 0.79 22.58 18.76 20.60 8.74 8.28 9.31 9.92 12.61 18.14 13.25 1.84 23.81 23.11 1.89 0.69 27.61 14.11Al % 0.52 0.20 3.23 2.68 4.91 1.13 2.39 4.87 3.11 5.69 4.00 3.73 0.45 10.31 17.72 0.36 0.25 9.49 2.99Fe % 0.31 0.19 1.52 1.20 2.96 1.01 2.21 4.65 2.96 4.65 3.12 4.06 0.26 8.54 9.36 0.20 0.17 3.67 2.06Ca % 34.38 36.79 9.21 9.90 10.38 25.44 24.24 21.32 22.68 15.15 8.23 13.85 34.62 7.28 0.99 38.17 21.56 0.64 5.84K % 0.12 0.08 0.83 0.61 1.32 0.27 0.56 0.88 0.75 1.30 1.11 0.86 0.08 0.42 1.98 0.12 0.08 3.56 1.57Mg % 1.42 1.27 1.65 4.42 1.52 1.43 1.61 1.70 1.37 1.26 1.15 1.57 1.53 4.66 0.35 – 14.32 – 2.41Cl % 2.09 0.84 1.70 1.75 2.09 2.13 4.11 3.37 2.40 1.53 2.38 2.49 0.00 37.45 16.43 1088.37 0.75 59.09 –P % 0.04 0.03 – – – 0.05 0.02 0.06 0.09 – 0.05 0.17 0.03 0.06 0.15 – – – 0.15S % 0.33 0.23 0.15 0.08 0.28 0.30 0.38 0.34 0.37 0.32 0.43 0.75 0.32 – 0.23 0.08 0.14 – 1.53As ppm 10.21 9.61 8.37 8.54 18.31 22.05 29.38 14.56 24.50 20.47 10.10 23.72 8.44 – 151.86 – – 6.65 56.23Cu ppm 17.19 14.51 18.64 17.28 30.47 29.04 26.85 29.67 22.44 135.02 56.54 94.19 27.68 128.72 120.22 12.02 14.18 37.59 59.66Mn ppm 627.72 240.73 383.95 224.00 724.54 394.08 485.57 422.45 496.49 542.00 375.90 260.23 317.07 1348.94 208.25 326.05 221.91 1646.50 177.05Ni ppm 43.22 47.34 39.83 28.50 51.55 41.13 51.94 59.87 48.72 62.56 48.71 42.71 43.26 76.51 73.50 38.25 33.30 33.70 16.95Pb ppm 8.26 8.25 14.87 8.68 22.21 28.08 21.33 22.47 19.53 38.33 47.64 123.48 12.35 10.77 77.33 6.24 5.10 19.24 36.88Rb ppm 8.81 5.41 36.55 24.59 64.92 17.34 30.15 55.52 35.08 71.30 63.18 56.03 10.59 9.35 137.15 4.67 4.06 118.54 79.30Sr ppm 1397.53 1532.70 108.37 52.52 303.96 1211.48 1202.39 329.71 1155.38 260.34 173.23 226.50 1960.40 394.83 830.85 240.39 61.46 729.42 383.21Th ppm 3.39 – 5.92 3.36 7.17 4.10 5.17 8.84 6.41 9.81 9.58 8.38 – 2.99 26.00 – – 31.82 3.92Zn ppm 26.12 14.02 47.57 30.18 58.32 62.65 45.81 59.82 44.17 269.92 184.94 272.63 23.27 64.06 211.86 40.56 9.76 258.20 71.10Zr ppm 33.91 – 149.05 57.11 120.20 26.15 57.33 52.15 50.08 71.82 150.67 110.91 11.96 150.67 233.43 12.66 3.35 1236.59 36.35Cr ppm – – 76.86 – 70.24 – – 66.22 85.18 103.71 92.83 61.98 – 226.11 188.57 – – 36.05 34.68Ti % 0.03 0.01 0.24 0.17 0.28 0.08 0.14 0.23 0.18 0.30 0.30 0.24 0.02 1.55 0.91 0.02 0.01 0.10 0.16V ppm – – 38.34 41.23 68.64 – 28.57 63.10 51.56 111.81 81.71 77.84 – 268.38 304.98 – – – 158.84Ag ppm 11.24 11.26 5.85 7.40 8.42 9.31 9.79 10.24 8.90 11.76 0.00 6.97 0.00 0.00 0.00 0.00 0.00 0.00 –Sb ppm 3.29 3.24 3.01 2.80 2.60 2.81 2.73 2.91 2.90 2.81 2.47 2.42 0.00 0.00 0.00 0.00 0.00 0.00 –Sn ppm 2.04 2.23 1.57 1.71 1.47 1.87 1.58 2.07 1.94 2.45 1.47 2.19 0.00 0.00 0.00 0.00 0.00 0.00 –

340A

.Popadicet

al./Marine

PollutionBulletin

76(2013)

333–348

Table 6correlation coefficient ‘‘r’’: compatibility between numerical values of Axis 1 from multivariate analyses and elements and mutual compatibility between elements.

Si Al Fe Ca K Mg Cl P S As Cu Mn Ni Pb Rb Sr Th Zn Zr Cr Ti V Ag Sb Sn mud

PCA Comp 1 �0.7 �0.84 �0.79 0.83 �0.84 0.07 �0.24 �0.23 �0.41 �0.07 �0.6 0.02 �0.26 �0.6 �0.89 0.85 �0.9 �0.81 �0.83 �0.65 �0.88 �0.86 0.6 0.74 0.16 �0.84DCA Axis 1 �0.68 �0.77 �0.75 0.81 �0.78 0.03 �0.27 �0.35 �0.5 �0.11 �0.6 0.1 �0.15 �0.7 �0.83 0.82 �0.88 �0.84 �0.82 �0.6 �0.84 �0.82 0.65 0.78 0.16 �0.86Si 1 0.63 0.32 �0.97 0.73 0.33 �0.03 �0.22 �0.16 �0.16 0.15 0.01 �0.21 0.17 0.6 �0.86 0.5 0.35 0.85 0.64 0.9 0.55 �0.65 �0.6 �0.6 0.6Al 0.63 1 0.9 �0.74 0.96 �0.04 0.22 �0.01 0.2 0.2 0.6 0.3 0.53 0.34 0.96 �0.78 0.89 0.63 0.6 0.55 0.81 0.94 �0.24 �0.6 �0.04 0.54Fe 0.32 0.9 1 �0.5 0.81 �0.21 �0.39 0.32 0.51 0.41 0.69 0.17 0.66 0.54 0.9 �0.57 0.91 0.73 0.4 0.35 0.58 0.89 �0.12 �0.61 0.2 0.52Ca �0.97 �0.74 �0.5 1 �0.81 �0.31 �0.04 0.09 �0.01 0.08 �0.34 0.05 0.11 �0.33 �0.72 0.93 �0.65 �0.54 �0.85 �0.64 �0.95 �0.7 0.67 0.72 0.47 �0.67K 0.73 0.96 0.81 �0.81 1 �0.08 0.13 �0.07 0.18 0.16 0.58 0.31 0.45 0.33 0.96 �0.78 0.87 0.64 0.73 0.72 0.89 0.93 �0.38 �0.67 �0.19 0.65Mg 0.33 �0.04 �0.21 �0.31 �0.08 1 �0.08 �0.25 �0.46 �0.3 �0.25 �0.4 �0.64 �0.2 �0.2 �0.36 �0.27 �0.25 �0.08 �0.3 0.16 �0.07 �0.06 �0.04 �0.2 �0.3Cl �0.03 0.22 �0.39 �0.04 0.13 �0.08 1 0.25 0.39 0.57 �0.07 0.2 0.29 0.16 0.21 0.02 0.37 0.08 0.08 0.02 �0.02 0.01 �0.06 �0.37 �0.3 0.16P �0.22 �0.01 0.32 0.09 �0.07 �0.25 0.25 1 0.86 0.38 0.24 �0.3 �0.03 0.78 0.01 0.06 0.24 0.46 0.02 �0.13 �0.12 0.12 �0.14 �0.34 0.33 0.35S �0.16 0.2 0.51 �0.01 0.18 �0.46 0.39 0.86 1 0.55 0.51 �0.03 0.27 0.88 0.37 �0.01 0.47 0.68 0.2 0.24 0.06 0.35 �0.16 �0.54 0.27 0.56As �0.16 0.2 0.41 0.08 0.16 �0.3 0.57 0.38 0.55 1 0.32 0.26 0.37 0.38 0.22 0.24 0.28 0.26 �0.12 �0.01 �0.18 0.19 0.27 �0.42 0.12 0.03Cu 0.15 0.6 0.69 �0.34 0.58 �0.25 �0.07 0.24 0.51 0.32 1 0.04 0.46 0.65 0.68 �0.42 0.67 0.87 0.28 0.38 0.45 0.77 �0.01 �0.48 0.5 0.46Mn 0.01 0.3 0.17 0.05 0.31 �0.4 0.2 �0.3 �0.03 0.26 0.04 1 0.49 �0.27 0.27 0.13 0.27 �0.12 0.1 0.38 0.03 0.14 0.27 0.06 �0.16 �0.05Ni �0.21 0.53 0.66 0.11 0.45 �0.64 0.29 �0.03 0.27 0.37 0.46 0.49 1 0.05 0.55 0.03 0.54 0.3 �0.01 0.22 0.08 0.49 0.3 �0.07 0.34 0.13Pb 0.17 0.34 0.54 �0.33 0.33 �0.2 0.16 0.78 0.88 0.38 0.65 �0.27 0.05 1 0.49 �0.36 0.52 0.84 0.4 0.34 0.33 0.5 �0.31 �0.7 0.24 0.67Rb 0.6 0.96 0.9 �0.72 0.96 �0.2 0.21 0.01 0.37 0.22 0.68 0.27 0.55 0.49 1 �0.74 0.93 0.77 0.68 0.68 0.83 0.95 �0.35 �0.71 �0.05 0.7Sr �0.86 �0.78 �0.57 0.93 �0.78 �0.36 0.02 0.06 �0.01 0.24 �0.42 0.13 0.03 �0.36 �0.74 1 �0.66 �0.57 �0.76 �0.53 �0.93 �0.74 0.53 0.6 0.23 �0.6Th 0.5 0.89 0.91 �0.65 0.87 �0.27 0.37 0.24 0.47 0.28 0.67 0.27 0.54 0.52 0.93 �0.66 1 0.78 0.65 0.56 0.74 0.89 �0.38 �0.67 �0.03 0.7Zn 0.35 0.63 0.73 �0.54 0.64 �0.25 0.08 0.46 0.68 0.26 0.87 �0.12 0.3 0.84 0.77 �0.57 0.78 1 0.55 0.54 0.63 0.79 �0.44 �0.76 0.2 0.79Zr 0.85 0.6 0.4 �0.85 0.73 �0.08 0.08 0.02 0.2 �0.12 0.28 0.1 �0.01 0.4 0.68 �0.76 0.65 0.55 1 0.85 0.88 0.6 �0.77 �0.62 �0.53 0.87Cr 0.64 0.55 0.35 �0.64 0.72 �0.3 0.02 �0.13 0.24 �0.01 0.38 0.38 0.22 0.34 0.68 �0.53 0.56 0.54 0.85 1 0.88 0.6 �0.77 �0.62 �0.54 0.87Ti 0.9 0.81 0.58 �0.95 0.89 0.16 �0.02 �0.12 0.06 �0.18 0.45 0.03 0.08 0.33 0.83 �0.93 0.74 0.63 0.88 0.88 1 0.6 �0.77 �0.62 �0.53 0.86V 0.55 0.94 0.89 �0.7 0.93 �0.07 0.01 0.12 0.35 0.19 0.77 0.14 0.49 0.5 0.95 �0.74 0.89 0.79 0.6 0.6 0.6 1 �0.3 �0.66 0.09 0.64Ag �0.65 �0.24 �0.12 0.67 �0.38 �0.06 �0.06 �0.14 �0.16 0.27 �0.01 0.27 0.3 �0.31 �0.35 0.53 �0.38 �0.44 �0.77 �0.77 �0.77 �0.3 1 �0.3 0.09 0.64Sb �0.6 �0.6 �0.61 0.72 �0.67 �0.04 �0.37 �0.34 �0.54 �0.42 �0.48 0.06 �0.07 �0.7 �0.71 0.6 �0.67 �0.76 �0.62 �0.62 �0.62 �0.66 �0.3 1 0.33 �0.76Sn �0.6 �0.04 0.2 0.47 �0.19 �0.2 �0.3 0.33 0.27 0.12 0.5 �0.16 0.34 0.24 �0.05 0.23 �0.03 0.2 �0.53 �0.54 �0.53 0.09 0.09 0.33 1 �0.32mud 0.6 0.54 0.52 �0.67 0.65 �0.3 0.16 0.35 0.56 0.03 0.46 �0.05 0.13 0.67 0.7 �0.6 0.7 0.79 0.87 0.87 0.86 0.64 0.64 �0.76 �0.32 1

A.Popadic

etal./M

arinePollution

Bulletin76

(2013)333–

348341

Fig. 3. Distribution of Al, As, Ca, Cr, Cu, Fe, K, Mn and Ni in the surface sediments of the Bay of Bakar.

342 A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348

5. Discussion

5.1. Major elements

The broader area of the bay is predominantly composed of lime-stones, with sporadic occurrences of flysch deposits (Fig. 1). Suchcarbonate nature of surrounding rocks explains very low concen-trations of Si in sediments near Oil refinery (stations B1 and B2),INA Urinj petroleum decanter (B6), coke plant (B7), bulk cargo ter-minal (B8 and B9) and the reference station (R1) and is supportedwith very strong negative correlation between Si and Ca(r = �0.97). At the other hand, stations B3, B4, B5, B10, B11 andB12, located near flysch deposits, show higher concentrations ofSi (12.61–21.55%). Very strong positive correlation of Si with Kand Al (r = 0.73 and 0.63 respectively), also point to clayey prove-nance of the material.

When observing Ca concentrations in the sediments, similargrouping of stations can be made: first group (B1, B2, B6, B7, B8,B9, R1), impoverished with Si, is enriched with Ca (up to 36.79%),while second group (B3, B4, B5, B10, B11, B12) comprises sedi-ments with higher Si concentrations and impoverished with Ca,all of which reflect the geology of the bay area. Ca concentrationshave very strong positive correlation with Sr (r = 0.97), suggestingtheir possible substitution in carbonate rocks. Furthermore, Ca isnegatively correlated not only to Si, but to other elements comingfrom silicaceous materials (K, Al) and associated trace elements(Rb, Sr, Th, Zn, Zr, Cr, Ti, V). Similar concentrations and distributiontrends are observed in Maslinica and Punat Bay, areas in the Northand Central Adriatic polluted with shipyard activities, untreatedwaste disposals and antifouling paints (Mikulic et al., 2004, 2008).

Concentrations of magnesium slightly varied among the sam-ples, ranging from 1.27% to 1.70%, with the exception of the sample

Fig. 4. Distribution of Pb, Rb, Si, Sr, Th, Ti, V, Zn and Zr in the surface sediments of the Bay of Bakar.

A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348 343

B4 where 4.42% of Mg was measured. Similar concentrations aremeasured in the Eastern Adriatic coastal sediments (Zvab-Rozicet al., 2012), as well as in the Central Adriatic surficial sediments(Dolenec et al., 1998).

Aluminum and potassium are two of the major constituents ofclay minerals, feldspars, micas, pyroxenes and amphiboles. Car-bonate rocks and sediments typically contain <2% of K and <2.5%of Al2O3 (De Vos and Tarvainen, 2006). The majority of measuredK and Al values fall into this ranges, similar as in sediments fromother Adriatic studies (Dolenec et al., 1998; Valkovic et al., 2007;Mikulic et al., 2008; Obho -daš and Valkovic, 2010; Zvab-Rozicet al., 2012). The correlation between K and Al concentration inthe bay is very strongly positive (r = 0.96), suggesting their similardepositional characteristics. The stations B3, B4, B5, B8, B9, B10,

B11 and B12 make the exception of this trend, with slightly in-creased Al concentrations (up to 5.69%), which correspond to in-creased values of heavy metals.

The concentration of iron at stations B1, B2 (oil refinery) and R1is lower than 0.5%, which is expected for the carbonate geologicalprovenance (De Vos and Tarvainen, 2006). Similar values are mea-sured in sediments from Maslinica Bay, polluted with sewagewaste and trace elements released from antifouling paints (Mikulicet al., 2008). All other stations in the Bay of Bakar had values reach-ing 4.65%, which are higher than those measured in the EasternAdriatic harbors, marinas and marine service areas (Obho -dašet al., 2006; Valkovic et al., 2007; Obho -daš and Valkovic, 2010).Such high values are measured in the Kaštela Bay, another heavilypolluted (industrial and domestic wastewater) area (Lovrencic

Fig. 5. Dendrogram obtained by cluster analysis of foraminiferal assemblages.

Fig. 6. Dendrogram obtained by cluster analysis of elemental concentrations.Fig. 7. Ordination diagrams obtained by detrended correspondence (A) andprincipal component (B) analyses.

344 A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348

et al., 2005). The distribution of Fe is very strongly correlated withthe distribution of heavy metals, suggesting their similar prove-nance and depositional characteristics.

5.2. Minor elements

Concentrations of Mn varied from 224 to 724.54 ppm and assuch were similar as measured Adriatic offshore and nearshore lev-els (Vukadin et al., 1982; Sondi et al., 1994; Dolenec et al., 1998;Ujevic et al., 2000; Valkovic et al., 2007; Obho -daš et al., 2010;Zvab-Rozic et al., 2012).

Titanium abundances in the sediments from the Bay of Bakar variedfrom 0.01% to 0.3%. Elevated values, similar to those measured in har-bors, marinas and marine service areas in the broader area of NorthernAdriatic (Valkovic et al., 2007), are registered at stations B3, B4, B5, B10,B11 and B12. Moreover, these values are similar to concentrations of Timeasured in sediments of Raša River and Estuary, as well as in lime-stones that are present around the Raša Estuary (Sondi et al., 1994),suggesting its possible natural origin in the Bay of Bakar.

Concentration of phosphorus was very low or negligible in themajority of samples, similar as in other Adriatic sediments (Dole-nec et al., 1998; Zvab-Rozic et al., 2012).

5.3. Heavy metals

The majority of heavy metals (Cu, Pb, Rb, Th, Zn, Zr, Cr, Ti, V) arestrongly related with Al (r = 0.55–0.94) and K (r = 0.64–0.96),

indicating their close association with clay minerals. Furthermore,they are significantly mutually correlated, which points to theirsimilar provenance and depositional characteristics. Their distribu-tion in the Bay of Bakar coincides with the sediment type (theshare of silt-clay fraction): the highest concentrations of heavymetals together with the highest concentration of Fe and Al aremeasured in sediments containing from 31.7% to 67.8% of mud(stations B3, B5, B10, B11 and B12). All above mentioned revealthe fact that the clay minerals are the main carriers of these ele-ments in the bay.

Cu, Zn, As and Pb are the highest heavy metal pollutants at theEastern Adriatic Sea (Obho -daš et al., 2010). Their distribution inmarine sediments is related primarily with the usage of anti-foul-ing paint and consequent bioaccumulation in the marine organ-isms. Additional sources of these elements are sewage disposalsand agricultural activities (ibidem).

The concentrations of Cu are generally low (up to 56.54 ppm)and as such are similar to the values previously reported in theAdriatic sediments (Vukadin et al., 1982; Zvonaric and Stegnar,1987; Dolenec et al., 1998; Vreca and Dolenec, 2005; Obho -dašet al., 2006, 2010; Castelli and Kljajic, 2010; Obho -daš and Valkovic,2010; Zvab-Rozic et al., 2012) and coastal areas in the NorthernAdriatic (Valkovic et al., 2007). Only at stations B10 and B12 con-centrations reached 135 ppm. Such high values can be attributedto the monitored activities: coal and coke depot (B10) and sewagedisposal in the harbor of the City of Bakar (B12). Likewise,estuarine sediments in Raša River Mouth had 145 ppm, which is

A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348 345

explained with anthropogenic origin of Cu and its adsorption onthe surface of the clay minerals (Sondi et al., 1994). The sedimentsfrom the Kaštela Bay, City of Rijeka harbor, Maslinica and PunatBay, as well as sediments from the Krka River Estuary also hadelevated values of Cu, coming from domestic sewage, industrialsources, antifouling copper paints and combustion of leadedgasoline (Vukadin et al., 1982; Prohic and Juracic, 1989; Mikulicet al., 2004, 2008; Cukrov et al., 2011). Much higher values of Cu(up to 12,640 ppm) are measured in harbors, marinas and marineservice areas along the Kvarner Bay (Valkovic et al., 2007).

The abundance of Zn is also generally low (up to 62.65 ppm),like in various beaches and bays along the Eastern Adriatic(Obho -daš and Valkovic, 2010). Stations B10, B11 and B12 makethe exception with values reaching 272.63 ppm. Zn concentrationsdemonstrate high positive correlation with Fe, which can lead toconclusion that Fe oxides carry Zn into the bay (Prohic and Juracic,1989). Zn concentrations are also high in the Kaštela Bay (Vukadinet al., 1982), in different Adriatic harbors (Zvonaric and Stegnar,1987; Mikulic et al., 2004, 2008; Obho -daš et al., 2010; Cukrovet al., 2011) and in the Northern Adriatic offshore sediments,where they have been introduced from the coastal and hinterlandarea and are strongly influenced by the pollution coming from theriver Po and other northern Italian and Croatian rivers along whichagricultural, industrial and mining activities take place (Dolenecet al., 1998). It has to be noticed that very high values of Zn are re-ported in harbors and marine areas along the Eastern Adriatic(Valkovic et al., 2007; Obho -daš and Valkovic, 2010).

Arsenic concentrations vary from 8.37 to 29.38 ppm throughoutthe bay. Sediments at stations B5, B6, B7, B9, B10 and B12 have thehighest concentrations and these values are similar or even higherto those measured in the Kaštela Bay and Rijeka harbor (Vukadinet al., 1982; Cukrov et al., 2011). Anthropogenic sources of As comefrom industrial activities like metal production, chemical industry,coal burning and from agriculture (Obho -daš et al., 2010), which canexplain these elevated values of As in the Bay of Bakar. Such con-clusions are supported by mean values of As in the Adriatic harbors(29.9 ppm) and near the cities, where it reaches 45.1 ppm (Valk-ovic et al., 2007; Obho -daš et al., 2010). Very high values of As(up to 524 ppm) are measured in harbors and marine areas ofthe Kvarner Bay (Valkovic et al., 2007; Obho -daš and Valkovic,2010).

The concentration of Pb in bay sediment is generally low, reach-ing of up to 47.64 ppm, similar to various beaches and bays alongthe Eastern Adriatic (Obho -daš and Valkovic, 2010), and lower thanthe mean values measured in Adriatic harbors (473 ppm), cities(100.3 ppm), estuarine sediments (132 ppm), and marine serviceareas (367 ppm) (Sondi et al., 1994; Valkovic et al., 2007; Mikulicet al., 2004, 2008; Obho -daš et al., 2010; Cukrov et al., 2011). Also,low values of Pb are measured in the Kaštela Bay, which is knownto be heavily contaminated by urban and industrial wastewaters(Ujevic et al., 2000). Only station B12 stands out with maximumof 123.48 ppm (sewage disposal of the harbor in the City of Bakar),which is still within the range of allowed depositional values forthe marine environments (Obho -daš et al., 2010). In the Bay of Ba-kar, Pb seems to be associated mostly with Zn (r = 0.84).

Strontium has interesting distribution; stations near Oil refin-ery INA Urinj (B1, B2), petroleum decanter of INA Urinj (B6), cokeplant (B7), bulk cargo terminal (B9) and the reference stationhave high concentrations of Sr (from 1155.38 to 1960.40 ppm).The sediments at these stations are enriched in Ca, which, to-gether with high Sr concentrations suggest possible substitutionwith Ca ions in surrounding carbonates. In the Adriatic nearshoreand offshore sediments, as well as in other Adriatic bays, similarvalues are measured (Dolenec et al., 1998; Valkovic et al., 2007;Obho -daš and Valkovic, 2010; Obho -daš et al., 2010; Zvab-Rozicet al, 2012).

There are several sediment quality criteria in use around theworld, all of which generally set two threshold levels, one belowwhich effects rarely occur and one above which effects are likelyto occur (Burton, 2002). Furthermore, EU Action Standards forthe deposition in marine environments also use different criteria(Action Level 1 and Action Level 2) and vary among Europeancountries (Obho -daš et al., 2010). Even in the same group of criteria(e.g. solely in the group that defines the concentration belowwhich effects rarely occur), there are still variations among thestandards (Burton, 2002).

In the Bay of Bakar, heavy metal concentrations are generallylow and fall below the majority of defined threshold levels (Burton,2002; Obho -daš et al., 2010). However, there is sporadic elementalincrease mostly located at stations B10, B11 and B12, which re-quired more detailed comparison with selected criteria (Thresholdeffect level – TEL, Toxic effect threshold – TET and Probable effectslevel – PEL), all specified in Burton (2002).

According to TEL, below which the effects rarely occur, Cu, Znand Pb concentrations mostly fall into allowed depositional values,with the exception at stations B10, B11 and B12, whereas As, Niand Cr are above the threshold values. According to TET, abovewhich effects are likely to occur, concentrations of Pb, Zn, Cr andNi fall into allowed depositional values, whereas those of Cu areabove the threshold only at stations B10 and B12, and As at sta-tions B5, B6, B7, B9, B10 and B12. Finally, according to PEL, all Cuand Zn concentrations are below given thresholds; Pb is abovethe threshold only at station B12, Cr at stations B10 and B11 andAs at stations B5, B6, B7, B9, B10 and B12. At majority of stationsin the bay and even at the reference station, Ni concentrationsare above this threshold.

The association of heavy metals (Cu, Pb, Rb, Th, Zn, Zr, Cr, Ti andV) is related with the distribution of clay minerals (K and Al) andthe share of the mud fraction in the sediments. Such trend isclearly visible in cluster analysis of elemental concentrations(Fig. 6): this elemental association has higher values at stationsB3, B5, B8, B9, B10, B11 and B12, and as such in great extent corre-sponds to grouping of samples obtained by statistical analyses offoraminiferal assemblages (multivariate and cluster analyses).

5.4. Foraminiferal assemblages

Foraminiferal assemblages were statistically analysed in orderto discover (possible) grouping of communities from different sta-tions. Cluster analysis revealed two major clusters, one comprisingstations B3, B5, B8, B10, B11 and B12 and the second including sta-tions B1, B2, B4, B6, B7, B9 and the reference station (Fig. 5). Suchgrouping was confirmed by multivariate analyses (Fig. 7). Interest-ingly, these results, obtained solely on relative abundance of spe-cies, strongly correspond to the results of geochemical analyses.Foraminiferal cluster 1 actually comprises stations with increasedconcentrations of heavy metals. Such conclusions are supportedwith high correlation between multivariate axes and concentra-tions of heavy metals (Table 6).

Species that dominate in the cluster 1 assemblage seem to bepredominantly stress-tolerant: Ammonia tepida, Ammonia convexa,Haynesina depressula, Epistominella exigua and Bulimina marginata,with several epifaunal species: Asterigerinata mamilla, Quinquelocu-lina seminula, Siphonaperta aspera, Textularia conica and Neoconor-bina terquemi (Table 3).

Ammonia and Haynesina group are described as stress-tolerant(Coccioni et al., 1997; Yanko et al., 1999; Armynot du Châteletet al., 2004; Luciani, 2007; Romano et al., 2008; Coccioni et al.,2009). In particular, A. tepida is tolerant to the increased valuesof heavy metal pollution (Jorissen, 1988; Coccioni et al., 1997;Armynot du Châtelet et al., 2004) and its abundance increases withthe increase of Ni, Cr, Cd, As and Hg concentrations (Frontalini and

346 A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348

Coccioni, 2008). H. depressula is also very tolerant to heavy metalpollution (Coccioni et al., 1997). E. exigua and B. marginata preferenvironments enriched with organic matter (Sun et al., 2006;Eberwein and Mackensen, 2006; Vidovic et al., 2009), but thisstudy reveals that they can also tolerate slightly enhanced heavymetal concentrations. The appearance of trochospiral and epifau-nal species in this association is very likely related to sandy andgravely sediment fraction, which is more favorable to their habitat.

Biodiversity indices of cluster 1 samples point to normal marineconditions and high biodiversity of the community, which leads toconclusion that the influence of this heavy metal enrichment onforaminiferal community is small. Such conclusion is supportedwith the very low percentage of deformed specimens (0–3.2%),which usually appear in heavy polluted areas (Yanko et al., 1998;Romano et al., 2008). Furthermore, the density and species rich-ness should decrease in areas of elevated heavy metal contamina-tions (Sharifi et al., 1991; Armynot du Châtelet et al., 2004;Frontalini and Coccioni, 2012), but this is not the case in the Bayof Bakar (Fig. 2). Moreover, the density is higher in sediments withthe highest concentrations of heavy metals (with the exception ofstation B10), which affirms that pollution is not affecting forami-niferal community in any significant extent; rather it is mud shareand associated organic matter and nutrient abundance that controlforaminiferal density in the bay.

Stations comprised with cluster 2 by their geochemical charac-teristic can be described as more carbonate, and as such comprisesediments with lower heavy metal concentrations. Those are B1,B2, B4, B6, B7, B9, together with the reference station. Assemblageis dominated by epifaunal species: Peneroplis pertusus, Cibicidesadvenum, Lobatula Lobatula, Rosalina bradyi, A. mamilla, N. terquemiand T. conica. The dominance of epifaunal species is related to thetype of substrate: all comprised stations have sandy substrate.Additionally, station B4 (rainfall accumulation near the City of Bak-arac) has unusually high appearance of A. tepida, making even 36%of the community. It is unclear which factor caused such signifi-cant appearance of A. tepida, is it fresh water coming from rainfallaccumulation and/or high titanium concentration (2474.79 ppm),possibly natural in origin, since the limestones can contain tita-nium minerals (Sondi et al., 1994). Foraminiferal density per 1 gin this cluster varies in great extent, reflecting different nutritionalcharacteristics between localities (Fig. 2, Table 4). Biodiversityindices point to normal marine conditions and high biodiversityof the community. The percentage of morphological deformitiesof foraminiferal tests is low (maximal values up to 3.2%, Table 4).Greater proportion of mechanically damaged tests (up to 32% atstation B1, Table 4) may be attributed to the water depth (influ-ences of waves, tides).

Following characteristics of foraminiferal assemblages in theBay of Bakar were observed: 1. the highest densities of tests insediments coincide with the highest concentrations of heavy met-als, suggesting that pollution is not affecting foraminiferal com-munity in any significant extent; 2. drop of foraminiferal testconcentrations only at station B10 proved that increased concen-tration of Cu (135 ppm) lead to lowering of foraminiferal diversity(Frontalini and Coccioni, 2012); 3. species known to be the mosttolerant to the heavy metal pollution: Haynesina germanica, Elphi-dium excavatum and Ammonia beccarii (Sharifi et al., 1991; Debe-nay et al., 2001; Armynot du Châtelet et al., 2004) are of limiteddistribution within the studied samples; 4. A. tepida (the mostabundant species within genus Ammonia), is confirmed to be tol-erant to heavy metal pollution; 5. E. exigua and B. marginata alsotolerate environments with slightly enhanced heavy metal con-centrations; 6. greater occurrences of trochospiral and epifaunaltests in foraminiferal assemblages are very likely related to sandyand gravely sediment fraction, which is more favorable to theirhabitat.

6. Conclusions

The distribution of metals and benthic foraminifera were deter-mined in surface sediments collected from thirteen stations alongthe Bay of Bakar (northern part of the Eastern Adriatic Sea). Thedistribution of major elements in the surface sediments from theBay of Bakar revealed that both concentration and distribution ofthese elements are strongly influenced by geological provenanceof the material. Broader area of the bay is composed of carbonates,with minor appearance of flysch deposits which causes inverselyproportional behavior between Ca and all other major elements:Si, K, Al, Mg, Fe. Strontium concentrations follow the ones of cal-cium which suggest their possible substitution in surroundingcarbonates.

The highest concentrations of heavy metals (Cu, Pb, Rb, Th, Zn,Zr, Cr, Ti and V) were found in sediments with significant amountof fine-grained fraction and with high content of clay minerals.Consequently, their concentration and distribution in the bay de-pend not only on potential pollution sources, but also on amountof clay minerals that carry and accumulate heavy metal associa-tion. Heavy metal concentrations throughout the bay generally fallinto allowed depositional values defined for marine environments,but depending on sediment quality criteria, stations B10, B11 andB12 have Cu, Zn, Pb and Cr concentrations above given thresholds.According to the toxic effect threshold (TET) and probable effectslevel (PEL) criteria, arsenic concentrations are above the thresholdat several stations in the bay. Finally, nickel concentrationsthroughout the bay are elevated only according to the threshold ef-fect level (TEL) criteria. All above mentioned points to the com-plexity of defining the extent of the pollution in the bay, whereasthe only certitude is that the maritime area in front of the cokeplant and the City of Bakar harbor (stations B10, B11 and B12) isthe most heavily polluted part of the bay.

Foraminiferal assemblages correspond to geochemical charac-teristics of the sediment, but the type of substrate controls thecommunity as well. Stress-tolerant species (Ammonia and Haynesi-na group) prevailed in fine-grained sediments with increased con-centrations of heavy metals. This assemblage is characterized byhigh biodiversity, low percentage of deformed specimens, highspecies richness and the highest foraminiferal densities, with theexception of station B10 where the decrease of foraminiferal den-sity is recorded. The second foraminiferal assemblage consists oflarger number of epifaunal taxa that thrived on sandy and gravelysediments with lower heavy metal concentrations.

Finally, species known to be the most tolerant to the heavy me-tal pollution are of limited distribution throughout the bay,whereas E. exigua and B. marginata are confirmed to be tolerantto the increased values of heavy metal concentrations.

Acknowledgements

This research was carried out within projects ‘‘Biomineraliza-tion processes of marine organisms’’ and ‘‘Recent sediments andfossil environments of Adriatic coastal zone’’, both supported bythe Croatian Ministry of Science and Education. We would like toexpress our gratitude to dipl. ing. Zeljko Dedic for helping us gen-erate spatial distribution maps and to Robert Košcal for the art-work. We would like to thank to the anonymous reviewers forhelping us improve the contents of the manuscript.

References

Albani, A.D., Serandrei Barbero, R., Donnici, S., 2007. Foraminifera as ecologicalindicators in the Lagoon of Venice, Italy. Ecol. Ind. 7, 239–253.

Alve, E., 1995. Benthic foraminiferal responses to estuarine pollution: a review. J.Foramin. Res. 25, 190–203.

A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348 347

Armynot du Châtelet, E., Debenay, J.-P., Soulard, R., 2004. Foraminiferal proxies forpollution monitoring in moderately polluted harbors. Environ. Pollut. 127, 27–40.

Bergin, F., Kucuksezgin, F., Uluturhan, E., Barut, I.F., Meric, E., Avsar, N., Nazik, A.,2006. The response of benthic foraminifera and ostracoda to heavy metalpollution in Gulf of Izmir (Eastern Aegean Sea). Estuar. Coast. Shelf Sci. 66, 368–386.

Botlovskoy, E., 1965. Los foraminiferos Recientes (Recent foraminifera). EditorialUniversitaria de Buenos Aires, Argentina, 510 pp.

Bouchet, V.M.P., Debenay, J.P., Sauriau, P.G., Radford-Knoery, J., Soletchnik, P., 2007.Effects of short-term environmental disturbances on living benthic foraminiferaduring the Pacific oyster summer mortality in the Marennes-Oléron bay(France). Mar. Environ. Res. 64, 358–383.

Bresler, V., Yanko, V., 1995. Acute toxicity of heavy metals for benthic epiphyticforaminifera Pararotalia spinigera (LeCalvez) and influence of seaweed-derivedDOC. Environ. Toxicol. Chem. 14, 1687–1695.

Burton, G.A., 2002. Sediment quality criteria in use around the world. Limnology 3,65–75.

Carnahan, E.A., Hoare, A.M., Hallock, P., Lidz, B.H., Reich, C.D., 2009. Foraminiferalassemblages in Biscayne Bay, Florida, USA: responses to urban and agriculturalinfluence in a subtropical estuary. Mar. Pollut. Bull. 59, 221–233.

Castelli, A., Kljajic, Z., 2010. Heavy metals distribution in marine sediments of theEast Adriatic Sea. Rapports et procés-verbaux des réunions Commissioninternationale pour l’exploration scientifique de la Mer Méditerranée 39, 231.

Cimerman, F., Langer, M.R., 1991. Mediterranean foraminifera. Razred zanaravoslovne vede, classis IV: Historia Naturalis, dela opera 30. Slovenskaakademija, Ljubljana, 119 pp.

Coccioni, R., Gabbianelli, G., Gentiloni Silverj, D., 1997. Benthic foraminiferalresponse to heavy metal pollution in the Goro Lagoon (Italy). First InternationalConference on Applications of Micropaleontology in Environmental Sciences,June 15–20, Tel Aviv, Israel, Abstract Volume, 47–48.

Coccioni, R., 2000. Benthic foraminifera as bioindicators of heavy metal pollution: acase study from the Goro Lagoon (Italy). In: Martin, R.E. (Ed.), EnvironmentalMicropaleontology: The Application of Microfossils to Environmental Geology.Kluwer Academic/Plenum, New York, pp. 1–100.

Coccioni, R., Frontalini, F., Marsili, A., Mana, D., 2009. Benthic foraminifera and traceelement distribution: a case-study from the heavily polluted lagoon of Venice(Italy). Mar. Pollut. Bull. 59, 257–267.

Cukrov, N., Franciškovic-Bilinski, S., Hlaca, B., Barišic, D., 2011. A recent history ofmetal accumulation in the sediments of Rijeka harbor, Adriatic Sea, Croatia.Mar. Pollut. Bull. 2011, 154–167.

Cosovic, V., Zavodnik, D., Borcic, A., Vidovic, J., Deak, S., Moro, A., 2011. A checklist ofForaminifera of the Eastern Shelf of the Adriatic Sea. Zootaxa 3035, 1–56.

Debenay, J.P., Tsakiridis, E., Soulard, R., Grossel, H., 2001. Factors determining thedistribution of foraminiferal assemblages in Port Joinville Harbor (Ile d’Yeu,France): the influence of pollution. Mar. Micropaleontol. 43, 75–118.

De Lazzari, A., Rampazzo, G., Pavoni, B., 2004. Geochemistry of sediments in theNorthern and Central Adriatic Sea. Estuar. Coast. Shelf Sci. 59, 429–440.

De Vos, W., Tarvainen, T., 2006. Geochemical Atlas of Europe, Part 2. Interpretationof Geochemical Maps, Additional Tables, Figures, Maps and RelatedPublications. Geological Survey of Finland, Espoo.

Dillon, W.R., Goldstein, M., 1984. Multivariate Analysis: Methods and Application.John Wiley & Sons, New York.

Dolenec, T., Faganeli, J., Pirc, S., 1998. Major, minor and trace elements in surficialsediments from the open Adriatic Sea: a regional geochemical study. Geol.Croatica 51, 59–73.

Donnici, S., Serandrei Barbero, R., Taroni, G., 1997. Living benthic foraminifera in theLagoon of Venice (Italy): population dynamics and its significance.Micropaleontology 43, 440–454.

Dubrovic, E. (Ed.), 2001. Rijecka luka. Povijest – izgradnja – promet (The Port ofRijeka. History – Construction – Transport – in Croatian). Museum of the City ofRijeka, Rijeka.

Ðekic, V., 2005. Prerada nafte u Rijeci 1882–2005 (Oil refining in Rijeka 1882–2005– In Croatian). INA, Rijeka.

Eberwein, A., Mackensen, A., 2006. Regional primary productivity differences offMorocco (NW-Africa) recorded by modern benthic foraminifera and their stablecarbon isotopic composition. Deep-Sea Res. I 53, 1379–1405.

Ferraro, L., Sprovieri, M., Alberico, I., Lirer, F., Prevedello, L., Marsella, E., 2006.Benthic foraminifera and heavy metals distribution: a case study from theNaples Harbor (Tyrrhenian Sea, Southern Italy). Environ. Pollut. 142, 274–284.

Fichet, D., Radenac, G., Miramand, P., 1998. Experimental studies of impacts ofharbor sediments resuspension to marine invertebrates larvae: bioavailabilityof Cd, Cu, Pb and Zn and toxicity. Mar. Pollut. Bull. 36, 509–518.

Folk, R.L., 1954. The distinction between grain size and mineral composition insedimentary rock nomenclature. J. Geol. 62, 344–356.

Frontalini, F., Coccioni, R., 2008. Benthic foraminifera for heavy metal pollutionmonitoring: a case study from the central Adriatic Sea coast of Italy. Estuar.Coast. Shelf Sci. 76, 404–417.

Frontalini, F., Coccioni, R., 2012. The response of benthic foraminiferal assemblagesto copper exposure: a pilot mesocosm investigation. J. Environ. Prot. 3, 342–352.

Grimani, I., Šušnjar, M., Bukovac, J., Milan, A., Nikler, L., Crnolatac, I., Šikic, D.,Blaškovic, I., 1963. Tumac za list Crikvenica 1:100 000, L33–102 (Explanatorynotes for sheet Crikvenica - In Croatian). Savezni geološki zavod Beograd, 47 pp.

Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: Paleontological StatisticsSoftware Package for Education and Data Analysis. Palaeontol. Electron. 4, 1–9.

Hammer, Ø., Harper, D.A.T., 2006. Paleontological Data Analysis. BlackwellPublishing Ltd, Oxford.

Jorissen, F.J., 1988. Benthic foraminifera from the Adriatic Sea: principles ofphenotypic variation. Utrecht Micropaleontol. Bull. 37, 1–174.

Kramer, K.J.M., Botterweg, J., 1991. Aquatic biological early warning system: anoverview. In: Jeffrey, D.W., Mudden, B. (Eds.), Bioindicators and EnvironmentalManagement. Academic Press, London, pp. 95–126.

Krebs, C.J., 1998. Ecological Methodology. Addison Wesley Longman, Menlo Park,California, 620 pp.

Luciani, V., 2007. Test abnormalities in benthic foraminifera and heavy metalpollution at the Goro lagoon (Italy): a multi – year history. Geophys. Res. Abstr.9, 09765.

Loeblich Jr., A.R., Tappan, H., 1987. Foraminiferal Genera and Their Classification.Van Nostrand Reinhold, New York, 970 pp + 847 pls.

Lovrencic, I., Orešcanin, V., Barišic, D., Mikelic, D., Rozmaric Macefat, M., Lulic, S.,Pavlovic, G., 2005. Characterisation of tenorm and sediments of Kaštela Bay andthe influence of tenorm on the quality of sediments. Global NEST J. 7, 188–196.

McGann, M., Alexander, C.R., Bay, S.M., 2003. Response of benthic foraminifers tosewage discharge and remediation in Santa Monica bay, California. Mar.Environ. Res. 56, 299–342.

Mikulic, N., Orešcanin, V., Legovic, T., Zugaj, R., 2004. Estimation of heavy metals(Cu, Zn, Pb) input into Punat Bay. Environ. Geol. 46, 62–70.

Mikulic, N., Orešcanin, V., Elez, L., Pavicic, L., Pezelj, D., Lovrencic, I., Lulic, I., 2008.Distribution of trace elements in the coastal sea sediments of Maslinica Bay.Environ. Geol. 53, 1413–1419.

Nadilo, B., Sojcic, M., 2005. Rušenje armiranobetonskog dimnjaka koksare u Bakru(Removal of the reinforced-concrete stack from the Bakar coke plant - InCroatian). Gra -devinar 57, 859–947.

Nigam, R., Saraswat, R., Panchang, R., 2006. Application of foraminifers inecotoxicology: retrospect, perspect and prospect. Environ. Int. 32, 273–283.

Obho -daš, J., Kutle, A., Valkovic, V., 2006. Concentrations of some elements in thecoastal sea sediments: bays with marinas. J. Radioanal. Nucl. Chem. 270, 75–85.

Obho -daš, J., Valkovic, V., 2010. Contamination of the coastal sea sediments by heavymetals. Appl. Radiat. Isot. 68, 807–811.

Obho -daš, J., Valkovic, V., Kutle, A., 2010. Atlas sedimenata obalnog podrucja i otokahrvatskog dijela Jadranskog mora (Atlas of Adriatic coastal sea sediments - inCroatian). Udruga Lijepa naša, Zagreb, 231 pp.

Obho -daš, J., Valkovic, V., Matjacic, L., Na -d, K., Sudac, D., 2012. Evaluation ofelemental composition of sediments from the Adriatic Sea by using EDXRFtechnique. Appl. Radiat. Isot. 70, 1392–1395.

Ouyang, T.P., Zhu, Z.Y., Kuang, Y.Q., Huang, N.S., Tan, J.J., Guoa, G.Z., Gu, L.S., Sun, B.,2006. Dissolved trace element in river water: spatial distribution and theinfluencing factor, a study for the Pearl River Delta Economic Zone, China.Environ. Geol. 49, 733–742.

Prohic, E., Juracic, M., 1989. Heavy metals in sediments - problems concerningdetermination of anthropogenic influence: study in the Krka River Estuary,Eastern Adriatic Coast, Yugoslavia. Environ. Geol. Water Sci. 13, 145–151.

Resig, J.M., 1960. Foraminiferal ecology around ocean outfalls of southern California.In: Person, E. (Ed.), Disposal in the Marine Environment. Pergamon Press,London, pp. 104–121.

Romano, E., Bergamin, L., Finoia, M.G., Carboni, M.G., Ausili, A., Gabellini, M., 2008.Industrial pollution at Bagnoli (Naples, Italy): Benthic foraminifera as a tool inintegrated programs of environmental characterization. Mar. Pollut. Bull. 56,439–457.

Sharifi, A.R., Croudace, T.W., Austin, R.L., 1991. Benthic foraminiferids as pollutionindicators in Southampton Water, southern England, UK. J. Micropalaeontol. 10,109–113.

Sondi, I., Juracic, M., Prohic, E., Pravdic, V., 1994. Particulates and environmentalcapacity for trace metals: a small river as a model for a land-sea transfersystem: Raša River Estuary. Sci. Total Environ. 155, 173–185.

Sun, X., Corliss, B.H., Brown, C.W., Schowers, W.J., 2006. The effect of primaryproductivity and seasonality on the distribution of deep-sea benthicforaminifera in the North Atlantic. Deep-Sea Res. I 53, 28–47.

Šušnjar, M., Bukovac, I., Nikler, R., Crnolatac, I., Milan, A., Šikic, D., Grimani, I., Vulic,Z., Blaškovic, I., 1970. Osnovna geološka karta SFRJ, List Crikvenica 1:100 000,L33–102 (Basic geological map of SFRY, 1:100000, Crikvenica sheet - inCroatian). Institut za geološka istrazivanja Zagreb, Savezni geološki zavod,Beograd.

Ujevic, I., Odzak, N., Baric, A., 2000. Trace metal accumulation in different grain sizefractions of the sediments from a semi enclosed bay heavily contaminated byurban and industrial wastewaters. Water Res. 34, 3055–3061.

Valkovic, V., Obho -daš, J., Crnjar, M., 2007. Concentration of some elements in theAdriatic coastal sea sediments. Case study: the Kvarner Bay. X-Ray Spectrom.36, 11–19.

Vdovic, N., Bišcan, J., Juracic, M., 1991. Relationship between specific surface areaand some chemical and physical properties of particulates: study in thenorthern Adriatic. Mar. Chem. 36, 317–328.

Vidovic, J., Cosovic, V., Juracic, M., Petricioli, D., 2009. Impact of fish farming onforaminiferal community, Drvenik Veliki Island, Adriatic Sea, Croatia. Mar.Pollut. Bull. 58, 1297–1309.

Vreca, P., Dolenec, T., 2005. Geochemical estimation of copper contamination in thehealing mud from Makirina Bay, central Adriatic. Environ. Int. 31, 53–61.

Vukadin, I., Stegnar, P., Smodiš, B., 1982. Fate and distribution of toxic heavy metalsin sediments and organisms of the Kaštela Bay. Acta Adriat. 23, 307–312.

Watkins, J.G., 1961. Foraminiferal ecology around the Orange County, California,ocean sewer outfall. Micropaleontology 7, 199–206.

348 A. Popadic et al. / Marine Pollution Bulletin 76 (2013) 333–348

Yanko, V., Ahmad, M., Kaminski, M., 1998. Morphological deformities of benthicforaminiferal test in response to pollution by heavy metals: implications forpollution monitoring. J. Foramin. Res. 28, 177–200.

Yanko, V., Arnold, A.J., Parker, W.C., 1999. Effects of marine pollution on benthicforaminifera. In: Sen Gupta, B.S. (Ed.), Modern Foraminifera. Kluwer AcademicPublishers, pp. 217–238.

Zvonaric, T., Stegnar, P., 1987. Total mercury, cadmium, copper, zinc and arseniccontents in surface sediments from the coastal region of the central Adriatic.Acta Adriat. 28, 65–71.

Zvab-Rozic, P., Dolenec, T., Bazdaric, B., Karamarko, V., Kniewald, G., Dolenec, M.,2012. Major, minor and trace element content derived from acquaculturalactivity of marine sediments (Central Adriatic, Croatia). Environ. Sci. Pollut. Res.19, 2708–2721.