lable at ScienceDirect

Journal of Environmental Radioactivity 155-156 (2016) 46e54

Contents lists avai

Journal of Environmental Radioactivity

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

Influence of submarine groundwater discharge on 210Po and 210Pbbioaccumulation in fish tissues

J. Garcia-Orellana a, b, *, E. L�opez-Castillo b, N. Casacuberta c, V. Rodellas b, d,P. Masqu�e a, b, e, f, G. Carmona-Catot g, M. Vilarrasa b, E. García-Berthou g

a Departament de Física, Universitat Aut�onoma de Barcelona, E-08193 Bellaterra, Spainb Institut de Ci�encia i Tecnologia Ambientals (ICTA), Universitat Aut�onoma de Barcelona, E-08193 Bellaterra, Spainc Laboratory of Ion Beam Physics, ETH Zürich, Otto-Stern-Weg, 5, CH 8093 Zürich, Switzerlandd Centre de Recherche et d’Enseignement de G�eosciences de l’Environment (CEREGE), Aix-Marseille Universit�e, 13545 Aix-en-Provence, Francee School of Natural Sciences, Edith Cowan University, 270 Joondalup Drive, Joondalup WA 6027, Australiaf Oceans Institute & School of Physics, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australiag GRECO, Institut d’Ecologia Aqu�atica, Universitat de Girona, E-17071 Girona, Spain

a r t i c l e i n f o

Article history:Received 29 October 2015Received in revised form8 January 2016Accepted 6 February 2016Available online xxx

Keywords:210Po210PbSGDFish tissuesIngestionMarsh

* Corresponding author. Departament de Física, Ucelona, E-08193 Bellaterra, Spain.

E-mail address: jordi.garcia@uab.cat (J. Garcia-Ore

http://dx.doi.org/10.1016/j.jenvrad.2016.02.0050265-931X/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

This study presents the results of the accumulation of 210Po and 210Pb in fish tissues and organs in abrackish-water marshland that is characterized by high concentrations of 222Rn and 226Ra supplied bysubmarine groundwater discharge (SGD). Tissues and organs from Cyprinus carpio, Chelon labrosus andCarassius auratus in the wetland were significantly enriched by both 210Pb and 210Po (up to 55 and 66times, respectively) compared to blanks. The major input route of 210Pb and 210Po into the fish bodyseems to be through ingestion, due to the high levels of 210Pb and 210Po found in the gut content as wellas in organs involved in digestion and metabolism (i.e. gut, kidney and hepatopancreas). Results showedthat 210Po was more accumulated in all fish tissues and organs except for the spine, which showed ahigher affinity for 210Pb, due to its capacity to replace Ca from apatite in bones. Over all the variablesanalyzed, fish tissues/organs and, secondarily, fish species were the most important factors explainingthe concentration of radionuclides, whereas fish length and the sampling location played a minor role.The relationship of the two radionuclides varied markedly among tissues and their concentration levelswere only correlated in gills, gut and, marginally, in spines. In general, the highest values of 210Pb and210Po concentrations in tissues were found on C. labrosus tissues rather C. auratus and C. carpio. This studydemonstrates that inputs of natural radionuclides supplied by SGD to coastal semi-enclosed areas (suchas marshlands, lagoons or ponds) may significantly increase the contents of 210Pb and 210Po in fish tis-sues/organs. Thus, this study represents one of the first evidences of direct ecological effects derivedfrom SGD.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Existing literature on interactions between radionuclides andaquatic biota is usually focused on artificial radionuclides releasedby nuclear accidents (Fleishman, 1973; Koulikov and Ryabov, 1992;Sax�en and Koskelainen, 2000) and on the contribution of radio-nuclides to the human dose due to their accumulation in somemarine organisms (i.e. Cherry and Heyraud, 1982). Naturally

niversitat Aut�onoma de Bar-

llana).

occurring radionuclides of the 238U and the 232Th series (i.e. 226Ra,210Pb, 210Po, 228Ra) can also accumulate in biota and therefore needto be studied to assess their potential toxicological significance aswell as for the artificial radioactivity (Shaheed et al., 1997; Fisheret al., 2013). In particular, 210Po and 210Pb from the 238U decaychain have attracted the attention of scientists because of theirrelatively high concentrations in marine organisms in comparisonwith those in terrestrial organisms (Carvalho, 2011; Fowler, 2011).210Po and, to a lesser extent 210Pb, are of special interest becausethey can contribute to the human internal radiation dose, mainlythrough seafood consumption (ICRP, 1979). In particular, 210Po(alpha-emitting radionuclide with T1/2 ¼ 138 d) is considered the

J. Garcia-Orellana et al. / Journal of Environmental Radioactivity 155-156 (2016) 46e54 47

most important source of internal radiation dose from naturalsources to marine organisms (Cherry and Shannon, 1974; Cherryand Heyraud, 1982; Fowler, 2011). 210Pb is produced by the decayof 222Rn gas emanating from the Earth's soil into the atmosphere,which in turn decays into 210Po. There is extensive literature aboutthe interaction of 210Po and 210Pb with biota in the marine envi-ronment (e.g. Carvalho, 2011), but only few data is availableregarding accumulation of 210Pb and 210Po in freshwater ecosys-tems (e.g. Brown et al., 2011), and even less in brackish environ-ments (e.g. NKS, 2009). Aside from U mining (e.g. Clulow et al.,1998; Skipperub et al., 2013), other processes can also introducesignificant amounts of natural radionuclides into terrestrial aquaticenvironments. For instance, submarine groundwater discharge(SGD) has been recently shown to significantly increase the naturalradioactivity levels in coastal lagoons (Garcia-Orellana et al., 2013).

Bioaccumulation of radionuclides in the tissues or organs oforganisms from aquatic environments depends on the chemicalcharacteristics of the radionuclide and its speciation in water orsediment, as well as on biological processes e including rates ofuptake fromwater or diet, excretion, andmetabolic transformation.These in turn may be influenced directly by the species-specificecophysiology of the organism, which is of course affected by arange of biological, physical and chemical factors, including itshabitat and feeding habits (Stewart et al., 2008). For instance, 210Pois mainly incorporated into biota (e.g. fish) due to ingestion by fishof organic particles that have adsorbed 210Po dissolved in water(Carvalho and Fowler, 1994). Less information is available about210Pb. Neither 210Pb nor 210Po have any known biological function;hence, organisms would not actively be ‘seeking’ to incorporatethem through enzymatic action or through specific membranechannels or other transport mechanisms, as is the case withessential metals (Williams, 1981; Simkiss and Taylor, 1995).

In this study, we selected a well-characterized coastal wetlandin the Mediterranean Sea (Peníscola marsh), where high concen-trations of natural radionuclides in both wetland waters and sedi-ments have been documented (Rodellas et al., 2012; Garcia-Orellana et al., 2013). The high levels of natural radioactivity inthis wetland, with concentrations of 222Rn and 226Ra dissolved inwater of up to 500 kBq m�3 and 1500 Bq m�3, respectively, weremainly attributed to the inputs of radionuclides supplied by sub-marine groundwater discharge (Rodellas et al., 2012; Garcia-Orellana et al., 2013). This wetland is thus an ideal environmentto study the accumulation of 210Po and 210Pb in fishes living inbrackish water, and to obtain some first evidences of direct bio-logical effects derived from the supply of radionuclides from sub-marine groundwater discharge. The main objective of this study isto provide new data of 210Pb and 210Po bioaccumulation in differentfish species and tissues/organs (i.e. kidney, muscle, gut, gills, spine,and hepatopancreas) from samples collected in brackish water of acoastal marshland.

2. Experimental

2.1. Sampling site

The Peníscola marsh is awetland (105 ha) located in the SpanishMediterranean coast (Fig. 1a) and is considered a Special ProtectionArea (SPA) by the European Union Legislation, with several pro-tected fish species. Although this wetland is partially isolated fromthe sea by a sandy barrier, it is drained by three main channels thatconverge before flowing to the sea through a narrow outlet(Fig. 1b). The Peníscola marsh is recharged by groundwater inputsconverging in the wetland from four different hydrogeological flowpaths (Rodellas et al., 2012; Zarroca et al., 2014) (Fig. 1b): (i) ashallow and horizontal flow of fresh groundwater related to the

detrital Vinar�osePeníscola coastal plain, (ii) a shallow, local andhorizontal flow from the karstic Irta Range system, (iii) an upwardregional flow associated with the deep and confined karstic aquiferof El Maestrat, and (iv) seawater intrusion. The upward regionalwater flow associated to the deep Maestrat aquifer is characterizedby high temperatures (>40 �C), high salinities (>20) and high 226Raactivities (up to 4500 Bq m�3) (Rodellas et al., 2012; Zarroca et al.,2014). Consequently, even though groundwater from the deepMaestrat aquifer represents a minor contribution to the wetlandwater budget (10e15%), this hydrogeological unit is the mainsource of Ra isotopes to the Peníscola marsh (Rodellas et al., 2012).Deep groundwater (already enriched in Ra isotopes), advectedthrough marsh sediments, is mixed with seawater intrusion andfresh groundwater from the local shallow aquifers(Vinar�osePeníscola and Irta Range) and is discharged into thePeníscola marsh as brackish groundwater, developing severalspring pools in topographical depressed areas (Zarroca et al., 2014).As marsh sediments have also high 226Ra activities (up to800 Bq kg�1), 226Ra-rich brackish groundwater becomes highlyenriched in 222Rn during its advection through wetland sediments(Rodellas et al., 2012). Thus, brackish groundwater seeping throughmarsh sediments (40e60% of water inputs) is considered to be theprimary source of 222Rn and 226Ra and therefore 210Pb and 210Po (20and 5.7 Bq m�3, respectively) to wetland waters (Rodellas et al.,2012; Garcia-Orellana et al., 2013).

2.2. Sample collection

Sampling stations were distributed considering the different Raisotopes and 222Rn concentrationsmeasured in the Peníscolamarshby Rodellas et al. (2012). ST1 was located where the presence offresh groundwater was related to the detrital Vinar�osePeníscolacoastal plain. ST2 was located in the area where water flow wasmainly associated with the deep and confined karstic aquifer of ElMaestrat. Finally, ST3 was located at the outlet of the Peníscolamarsh, where waters from different origins were well mixed. Fishsamples were collected in the Peníscola marsh during differentcampaigns in 2010. The location of sampling stations is shown inFig. 2. Individuals of Chelon labrosus (n ¼ 5), Carassius auratus(n ¼ 7) and Cyprinus carpio (n ¼ 4) were captured in May 2010 atST1, ST2 and ST3. One individual per stationwas selected accordingto its size and weight. C. labrosus is an autochthonous and catad-romous specie, while C. auratus and C. carpio are introduced spe-cies. Although the three species have rather omnivorous andopportunistic diets, and they all have a lifespan longer than 10years, they differ considerably: C. auratus feeds on detritus andbenthic invertebrates, similarly to C. carpio that often consumeshard material such as molluscs and plant seeds, whereas C. labrosusconsumes more live plant material, including microalgae and fila-mentous algae (García-Berthou, 2001; Kottelat and Freyhof, 2007).Fish sample features (sex, length, sampling area, weight and dry-wet weight ratio) of the analyzed species are described in Table 1.

2.3. Pretreatment

After collection, individuals were dissected and subsamples ofmuscle, gills, spine, gonads, kidney, hepatopancreas and gut wereobtained. A portion of each tissue/organ was obtained and subse-quently deposited in a Petri dish, weighted, dried at 60 �C for 24 hand weighted again to determine the dry and wet weight differ-ence. In the case of gut, its content was squeezed, weighted anddeposited in a Petri dish. Then, the gut was washed with water inorder to remove any remaining content. Blanks of C. labrosus andC. carpio were collected in November 2011 in an area free of sub-marine groundwater discharge influence, 300 km north of the

Fig. 1. (a) Location of the Peníscola marsh in the Mediterranean Sea Basin. (b) Aerial image of the Peníscola marsh, where the wetland channels and spring pools are highlighted inblue and the flow direction is also indicated. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2. Sampling sites in the studied area. Water stations results are showed in Garcia-Orellana et al. (2013).

Table 1Features of the fish samples analyzed.

Species Station Sex Size (cm) Weight (g)

Cyprinus carpio ST1 M 26 295ST2 F 38 481ST3 M 21 156

Chelon lebrosus ST1 M 30 244ST2 M 33 356ST3 F 32 312

Carassius auratus ST1 F 19 136ST2 F 21 161ST3 F 20 125

J. Garcia-Orellana et al. / Journal of Environmental Radioactivity 155-156 (2016) 46e5448

Peníscola Marsh, at Empuriabrava (42.2525 ºN 3.0936 ºE). Blankswere pretreated the same as the rest of the samples.

2.4. 210Pb and 210Po determination

Between 0.5 and 1 g of dry and homogenized sample of fishtissues/organs were digested in HNO3 and H2O2 in open vesselsafter adding a known amount of 209Po (0.703 ± 0.014 Bq mL�1) as ayield tracer. After digestion, the solutionwas evaporated to drynessin order to remove the HNO3, and then subsequently converted to ahydrochloric form by adding 2e3 mL of concentrated HCl andevaporated to dryness. This step was repeated three times. 210Poand 209Po from the solution were spontaneously deposited onto asilver disc and suspended in the sample solution by means of anylon thread taped to the beaker. One face of the disc was lacqueredwith urethane in order to deposit Po isotopes onto a unique diskface. The 209Po and 210Po isotopes were measured using PIPS a-spectrometers (CANBERRA, Model PD-450.18 AM). After plating,the remaining solution was stored for 6 months to allow ingrowthof 210Po from 210Pb, and then the 210Po platting step was repeated.Appropriate ingrowth and decay corrections were applied to

J. Garcia-Orellana et al. / Journal of Environmental Radioactivity 155-156 (2016) 46e54 49

calculate the activities of 210Pb and 210Po on the sampling date(Rigaud et al., 2013). The analytical quality of radio-analyticalmeasurements was ensured by analysing 5 samples of certifiedreference materials (IAEA e 414) composed of mixed fish speciescollected in the eastern Irish Seawith an information value for 210Poof 2.1 ± 0.4 Bq kg�1. Mean value of 5 replicates was2.0 ± 0.3 Bq kg�1.

2.5. Statistical methods

We analyzed the contribution of tissue/organ, fish species, fishlength, and site on 210Pb and 210Po fish concentrations through ananalysis of covariance, ANCOVA (Sokal and Rohlf, 2012) using fishlength as covariate and the remaining sources of variation as cat-egorical factors. The effects of the same three categorical factors onthe relationship between these two radionuclides were also testedwith ANCOVA. All these ANCOVAs included the interactions be-tween the covariate and factors, which test whether the slopes arehomogeneous (García-Berthou and Moreno-Amich, 1993).Although P values are reported, we relied on Akaike InformationCriteria (AIC) for model selection (Burnham and Anderson, 2002)because of the low sample sizes. AIC is a measure based on infor-mation theory and maximum likelihood that considers the numberof predictors (parsimony) in addition to goodness-of-fit, and is usedto compare candidate models describing the same data with thebest fitting model having the lowest AIC (Burnham and Anderson,2002). We analyzed the concentrations in gut contents separatelywith a linear model without covariates. All statistical analyses wereperformed with the R environment, v. 3.2.3 (R Core Team, 2015),including the MuMin package (Barto�n, 2015).

3. Results & discussion

3.1. Variation of 210Pb and 210Po concentrations with fish speciesand tissues/organs

The two ANCOVAs (linear models) that best explained the 210Pband 210Po concentrations in fish according to AIC weights includedthe main effects and also fish length e tissue/organ interactions(Table 2). These models, which explained most of the variation(>66%, Table 2), showed several significant sources of variationdespite the low sample size and indicated that: i) tissues andsecondarily fish species were the most important factors to explainthe concentration of 210Pb and 210Po; ii) site, fish length and in-teractions had less influence to the variation of 210Pb and 210Po

Table 2Analysis of covariance of the effects of fish tissues/organs, fish species, length, andsite on 210Pb and 210Po fish concentrations. The degrees of freedom (df), Sum ofsquares (SS), and P values are shown. Significant P values (P < 0.10) are bolded. Theexplained variation (R2

adj) was 0.667 for 210Pb and 0.946 for 210Po.

Source of variation df 210Pb 210Po

SS P SS P

Fish length (L) 1 0.685 0.171 0.070 0.179Tissue 6 21.150 0.043 13.591 0.007Species 2 1.921 0.140 1.811 0.018Site 2 0.055 0.851 0.804 0.040Tissue � L 6 0.158 0.962 0.726 0.128Species � L 2 0.963 0.245 0.185 0.155Tissue � Species 11 1.231 0.711 0.412 0.352Site � L 1 0.092 0.524 0.164 0.090Tissue � Site 12 0.517 0.941 0.413 0.376Tissue � Species � L 11 0.806 0.835 0.295 0.449Tissue � Site � L 4 0.206 0.842 0.234 0.236Residuals 2 0.312 0.034

concentrations; iii) 210Pb depended mostly on tissue/organ,whereas 210Po varied more with sources of variation. Table 3 showsthe 210Pb and 210Po concentrations in C. auratus, C. carpio, C. lab-rosus, and the fresh and dry weight ratios (fw:dw). The concen-trations of 210Pb and 210Po in tissues/organs of C. carpio, C. labrosusand C. auratus from Peníscola marsh waters showed a range of0.6e280 Bq kg�1 (f.w.) of 210Pb, and of 4.4e1600 Bq kg�1 (f.w.) for210Po. Blanks showed 210Pb and 210Po concentrations in C. carpioand C. labrosus in a range of 0.06e15 Bq kg�1 (f.w.) and of0.3e19 Bq kg�1 (f.w.), respectively. The average concentration of210Pb and 210Po in C. labrosus and C. carpio blanks were up to twoorders of magnitude lower than those values obtained in Peníscolamarsh.

Mean 210Pb concentrations were higher in C. labrosus tissues/organs than in C. auratus and C. carpio. The 210Pb concentrationmeasured in C. labrosus ranged from 1.3 to 280 Bq kg�1 (f.w.),compared to those measured in C. auratus and C. carpio that rangedfrom 0.6 to 130 Bq kg�1 (f.w.) and 1.3e120 Bq kg�1 (f.w.), respec-tively. Similar 210Po concentrations were measured between tis-sues/organs from different individuals with values from 5.5 to1600 Bq kg�1 (f.w.) in C. labrosus, 4.4e1100 Bq kg�1 (f.w.) inC. carpio, and 21e1200 Bq kg�1 (f.w.) in C. auratus. The higher 210Pband 210Po concentrations in C. labrosus compared to the two otherspecies might be related to the different diet of the C. labrosusspecies, which is mainly based on live plant material. Although210Pb and 210Po concentrations in gut contents did not seem to varywith fish species or site (linear models and AIC), herbivorous dietsare often compensated for lower caloric values with higher foodingestion, possibly explaining the higher concentrations in tissues/organs.

210Pb and 210Po showed a wide range of concentrationsdepending on the analyzed fish tissues/organs. 210Po concentra-tions in fish tissues/organs were higher than those for 210Pb in allthe tissues/organs analyzed, except for gills. 210Pbwas concentratedin tissues such as spine and gills with concentrations of 60e280 and10e260 Bq kg�1 (f.w.), respectively. The lowest accumulations of210Pb were found in muscle and gonads, with concentrations of0.6e3 and 1e19 Bq kg�1 (f.w.), respectively. On the other hand,210Po was concentrated in tissues/organs such as gut and kidneywith concentrations ranging from 210 to 1010 and180e680 Bq kg�1 (f.w.), respectively. The lowest concentrationswere measured in spine and muscle, with concentrations rangingfrom 34 to 200 and 4.4e120 Bq kg�1 (f.w.), respectively. However,the highest concentration of both radionuclides was measured inthe gut content, ranging from 4.2 to 110 Bq kg�1 (f.w.) for 210Pb, andfrom 380 to 1600 Bq kg�1 (f.w.) for 210Po. Thus, the high levels of210Pb and 210Po found in gut content, as well as in the organsinvolved in digestion and metabolism (i.e. gut, kidney and hepa-topancreas), indicate that the major input route of 210Pb and 210Pointo the fish body is ingestion, as was already observed in otherstudies (e.g. Carvalho, 2011).

3.2. Relationship of 210Pb and 210Po concentrations

The ANCOVA of the effects of tissue/organ, fish species and 210Pbconcentration on 210Po concentration, selected according to AICweights (Table 4), showed that: i) the three main variables (tissue/organ, fish species and 210Pb concentration) were necessary toexplain the variation of 210Po, though some were not formallysignificant; and ii) fish tissue/organ was the most important sourceof variation. The relationship between the two radionuclides thatdepend on fish tissue/organ was only significant for gills and gut(P < 0.05), and marginally spines (P ¼ 0.088). The different tissue/organ accumulation in both radionuclides responds to theirdifferent behaviour and affinity for each tissue/organ, which can be

Table 3Concentrations of 210Pb and 210Po (in Bq kg�1 fresh weight, f.w.) in tissues/organs from C. carpio, C. labrosus and C. auratus and the ratio between fresh and dry weight (fw:dw).

Site 1 Site 2 Site 3 Blank

F:D 210Pb 210Po F:D 210Pb 210Po F:D 210Pb 210Po F:D 210Pb 210Po 210Po/210Pb

(Bq kg�1) (Bq kg�1) (Bq kg�1) (Bq kg�1)

Cyprinus carpioMuscle 4.3 3.0 ± 0.2 4.4 ± 0.6 5.2 1.5 ± 0.2 8.8 ± 0.6 5.3 1.3 ± 0.1 7.8 ± 0.5 5.3 0.2 ± 0.1 0.3 ± 0.1 2.3 ± 0.8Spine 1.9 119 ± 6 51 ± 13 1.5 n.m. n.m. 2.1 60 ± 3 34 ± 4 2.1 12 ± 4 5 ± 2 0.4 ± 0.2Gonads 4.8 2.9 ± 0.2 20.0 ± 1.0 4.6 4.5 ± 0.6 251 ± 11 4.9 3.6 ± 0.3 80 ± 4 4.0 2.3 ± 0.8 5.8 ± 1.3 2.5 ± 1.0Gills 5.4 31 ± 2 44 ± 5 6.4 12.8 ± 0.8 54 ± 3 6.3 10.3 ± 0.6 37 ± 2 5.6 1.0 ± 0.2 1.8 ± 0.2 1.8 ± 0.4Kidney 5.2 17.5 ± 1.2 190 ± 12 5.3 11.1 ± 0.7 676 ± 23 5.6 4.2 ± 0.4 180 ± 10 5.3 1.0 ± 0.4 10.9 ± 0.9 10 ± 4Hepatopancreas 5.4 24 ± 2 131 ± 7 5.2 5.1 ± 0.5 224 ± 9 5.2 5.6 ± 0.7 56 ± 3 6.3 0.6 ± 0.2 4.3 ± 0.3 7 ± 2Gut 7.8 5.0 ± 0.5 255 ± 13 5.0 7.6 ± 0.4 755 ± 26 7.4 2.5 ± 0.2 205 ± 9 6.3 0.3 ± 0.1 10.7 ± 0.6 37 ± 13(Gut content) 5.3 39 ± 2 881 ± 41 5.9 13.8 ± 0.7 1111 ± 40 5.2 38 ± 2 629 ± 28 5.3 0.6 ± 0.1 2.9 ± 0.2 4.6 ± 0.7Chelon labrosusMuscle 4.5 1.3 ± 0.1 5.5 ± 0.4 4.2 1.5 ± 0.2 8.8 ± 0.7 3.4 2.0 ± 0.3 10.0 ± 0.6 4.3 0.06 ± 0.04 0.6 ± 0.1 9.3 ± 6.3Spine 1.3 283 ± 11 198 ± 32 1.5 273 ± 16 93 ± 20 2.1 183 ± 7 173 ± 20 1.6 15 ± 6 7 ± 5 0.5 ± 0.4Gonads 1.5 19 ± 2 88 ± 8 4.1 11.0 ± 1.8 102 ± 6 4.9 3.3 ± 0.3 81 ± 3 1.4 5 ± 2 3 ± 2 0.6 ± 0.5Gills 4.6 75 ± 3 81 ± 8 4.8 110 ± 5 139 ± 10 4.0 261 ± 11 208 ± 13 3.3 2.7 ± 0.5 6.0 ± 0.6 2.2 ± 0.5Kidney 4.5 16.6 ± 1.3 269 ± 16 5.1 52 ± 3 220 ± 14 3.7 145 ± 7 386 ± 17 4.0 1.0 ± 0.2 15.5 ± 1.0 16 ± 3Hepatopancreas 4.4 12.0 ± 0.9 255 ± 15 4.0 37 ± 2 289 ± 13 3.2 40.6 ± 1.5 195 ± 9 3.2 0.7 ± 0.2 8.8 ± 0.5 13 ± 4Gut n.m. n.m. 3.1 9.6 ± 0.7 392 ± 16 3.3 60 ± 3 1011 ± 40 3.7 0.6 ± 0.1 10.7 ± 0.7 17 ± 4(Gut content) 5.3 17.2 ± 1.0 882 ± 44 4.0 106 ± 5 377 ± 17 5.6 57 ± 3 1565 ± 60 4.3 4.3 ± 0.5 18.9 ± 1.1 4.4 ± 0.6Carassius auratusMuscle 5.1 0.6 ± 0.1 21.1 ± 1.3 4.8 1.8 ± 0.2 20.7 ± 1.3 4.8 1.4 ± 0.3 115 ± 3Spine 1.4 97 ± 4 89 ± 12 1.5 133 ± 7 39 ± 11 1.8 116 ± 6 132 ± 11Gonads 4.5 1.1 ± 0.1 127 ± 5 5.1 5.6 ± 0.8 67 ± 5 3.6 4.0 ± 0.4 391 ± 17Gills 5.1 21.0 ± 1.0 154 ± 7 4.5 24.0 ± 1.3 119 ± 7 5.7 34.6 ± 1.9 132 ± 8Kidney 5 10.0 ± 1.0 591 ± 26 5.1 15 ± 2 328 ± 17 4.7 16 ± 3 591 ± 20Hepatopancreas 5.2 4.5 ± 0.7 239 ± 12 5.7 7.0 ± 1.0 83 ± 6 5.6 5.3 ± 1.6 367 ± 13Gut 8.4 4.4 ± 0.6 532 ± 22 1.5 15.1 ± 1.7 264 ± 15 9.1 10.3 ± 0.9 645 ± 29(Gut content) 6.3 4.2 ± 0.5 727 ± 28 4.4 128 ± 6 769 ± 37 5.1 79 ± 4 1220 ± 43

n.m.: not measured.

Table 4Analysis of covariance of the effects of fish tissues/organs, fish species and 210Pbconcentration (covariate) on 210Po concentration. The degrees of freedom (df), Sumof squares (SS), and P values are shown. Significant P values (P < 0.10) are bolded.The explained variation (R2

adj) was 0.491.

Source of variation df SS P

210Pb 1 5,738 0.618Tissue 6 1,563,478 <0.001Species 2 119,350 0.092Tissue � 210Pb 6 219,081 0.187Species � 210Pb 2 10,689 0.789Tissue � Species 11 76,932 0.974Tissue � Species � 210Pb 11 168,379 0.737Residuals 21 468,958

Table 5Ratios of 210Pb and 210Po activities in fish tissue/organ from C. carpio, C. labrosus andC. auratus for each sample.

Site 1 Site 2 Site 3 Blank

210Po/210Pb

Cyprinus carpioMuscle 1.5 ± 0.2 5.9 ± 0.9 6.0 ± 0.6 2.3 ± 0.8Spine 0.4 ± 0.1 n.m. 0.6 ± 0.1 0.4 ± 0.2Gonads 6.9 ± 0.6 56 ± 8 22 ± 2 2.5 ± 1.0Gills 1.4 ± 0.2 4.2 ± 0.4 3.6 ± 0.3 1.8 ± 0.4Kidney 10.9 ± 1.0 61 ± 4 43 ± 5 10 ± 4Hepatopancreas 5.5 ± 0.5 44 ± 5 10.0 ± 1.4 7 ± 2Gut 51 ± 6 99 ± 6 82 ± 7 37 ± 13(Gut content) 22.6 ± 1.6 81 ± 5 16.6 ± 1.1 4.6 ± 0.7Chelon labrosusMuscle 4.2 ± 0.4 5.9 ± 0.9 5.0 ± 0.8 9 ± 6Spine 0.7 ± 0.1 0.3 ± 0.1 0.9 ± 0.1 0.5 ± 0.4Gonads 4.6 ± 0.6 9.3 ± 1.6 25 ± 2 0.6 ± 0.5Gills 1.1 ± 0.1 1.3 ± 0.1 0.8 ± 0.1 2.2 ± 0.5Kidney 16.2 ± 1.6 4.2 ± 0.4 2.7 ± 0.2 16 ± 3Hepatopancreas 21 ± 2 7.8 ± 0.5 4.8 ± 0.3 13 ± 4Gut n.m. 41 ± 3 16.9 ± 1.1 17 ± 4(Gut content) 51 ± 4 3.6 ± 0.2 27.5 ± 1.8 4.4 ± 0.6Carassius auratusMuscle 35 ± 6 11.5 ± 1.5 82 ± 18Spine 0.9 ± 0.1 0.3 ± 0.1 1.1 ± 0.1Gonads 115 ± 11 12.0 ± 1.9 98 ± 11Gills 7.3 ± 0.5 5.0 ± 0.4 3.8 ± 0.3Kidney 59 ± 6 22 ± 3 37 ± 7Hepatopancreas 53 ± 9 11.9 ± 1.9 69 ± 21Gut 121 ± 17 17 ± 2 63 ± 6(Gut content) 173 ± 22 6.0 ± 0.4 15.4 ± 1.0

n.m.: not measured.

J. Garcia-Orellana et al. / Journal of Environmental Radioactivity 155-156 (2016) 46e5450

also tested using the 210Po/210Pb activity ratio (Table 5). Ratiosranged from 11 to 120, 3 to 90, 5 to 97, 1 to 83 and 4 to 120 for gut,kidney, hepatopancreas, muscle and gonads, respectively, showinga preferential 210Po accumulation with respect to 210Pb. Asdescribed in previous works conducted in the marine environment,210Po is incorporated to the organism by the absorption withingested food (e.g. Carvalho, 2011). It is associated with metal-lothioneins and ferratin (Durand et al., 1999) and has been linked toS uptake in the fish digestion (Cherry and Shannon, 1974). On theother hand, for spines, the 210Po/210Pb activity ratio, which rangedfrom 0.3 to 1.1 with a median of 0.6, is due to the higher affinity of210Pb and Ca to bone tissue. High concentration of 210Pb in spinesallows for a significant production of 210Po and could justify thecorrelation observed for both radionuclides in this tissue (Fig. 3). Inthe case of gills, 210Po/210Pb activity ratio ranged from 0.7 to 7.3with a median of 3.6 indicating some preferential accumulation of210Pb, as in spines. The relatively higher concentration of 210Po(Fig. 3) in gills could be related either to some incorporation of 210Pothat was dissolved inmarsh water or to an excretion of 210Pb (Olsonet al., 1998).

Considering that both radionuclides are incorporated to the fishbodymainly through ingestion, concentrations of 210Po and 210Pb ingut content of each individual sample can be used to normalize theaccumulation of both radionuclides for each tissue/organ. Values

Fig. 3. Correlation of 210Po and 210Pb content in spines and gills.

Fig. 4. Mean 210Pb and 210Po accumulation in fish tissues of C. carpio, C. labrosus andC. auratus normalised by the gut content (food).

J. Garcia-Orellana et al. / Journal of Environmental Radioactivity 155-156 (2016) 46e54 51

lower or higher than those found in gut content would indicate alow or high accumulation due to ingestion. Fig. 4 shows how theingestion causes an accumulation preferably of 210Pb in spines andgills, and 210Po in gut, kidney and hepatopancreas, as it was alreadyshown by the 210Po/210Pb activity ratios. Yet, the accumulationfactor is much higher for 210Pb than 210Po, indicating that whilst210Pb is constantly accumulated in fish tissues/organ and it slowlydecays (T1/2 ¼ 22.3 y), 210Po (T1/2 ¼ 138 d) may accumulate but itdecays much faster than 210Pb.

3.3. Variation with sampling site and size

According to Rodellas et al. (2012), the area close to ST1 wascharacterized by low salinities (~1.5) and relatively low 226Ra and222Rn concentrations (80 ± 5 Bq m�3 and (42 ± 3)$103 Bq m�3,respectively), as a consequence of fresh groundwater inputs fromthe shallow systems, which do not represent a relevant source ofnatural radionuclides. The area surrounding ST2 was affected byinputs of high-226Ra, high-222Rn, and brackish groundwater inputs,whichmix with freshwater from the ST1. The salinity of ST2 was 6.7with 226Ra and 222Rn concentrations of 1450 ± 80 Bq m�3 and(495 ± 17)$103 Bq m�3, respectively. Finally, ST3 is located at theoutlet of the wetland, where water from the marshland dischargeto the sea, and thus represents an integration of water fromdifferent sources. This latter station was characterized by having asalinity of 5.6 with 226Ra and 222Rn concentrations of680 ± 30 Bq m�3 and (295 ± 15)$103 Bq m�3, respectively.

Although there was a significant difference of 226Ra and 222Rnbetween sites, dissolved 210Pb and 210Po concentrations along themarshland were relatively homogenous. Results of 210Pb and 210Poin marsh water are reported in Garcia-Orellana et al. (2013). Valuesof 210Pb in dissolved fraction ranged from 13 to 23 Bq m�3. Lowervalues were observed for dissolved 210Po, ranging from 2.5 to4.6 Bq m�3. In the particulate fraction, 210Pb ranged from 2.5 to7.6 Bq m�3, and 210Po ranged from 0.8 to 3.8 Bq m�3. Both partic-ulate and dissolved fractions showed higher 210Pb activitiescompared to 210Po. 210Pb and 210Po in dissolved fractions werehigher in sampling sites close to the areas where SGD inflowsthrough marsh sediments developing a spring pool (e.g., betweenST4 and ST6). Indeed, most of the 210Pb (from 70% to 90%) and the210Po (from 60% to 85%) were in the dissolved fraction for all thestations except for ST3 (marsh outlet). Regarding the particulatefraction, which can be particularly relevant for the fish food chain(e.g. benthic diatoms, epiphytic algae, small invertebrates, detritus…), concentrations of 210Pb and 210Po associated with particleswere significantly higher at ST3.

Differences on 210Pb and 210Po concentrations in fish amongdifferent sites are comparatively smaller than differences amongfish tissues/organs or among fish species (Table 2), likely becausethe three fish species can move throughout the marshland. How-ever, some slightly differences can be observed. C. labrosus con-centrations at ST1 were lower than in the other two sites andC. auratus concentrations at ST3 showed higher concentrationscompared to those obtained for other species in other sites.

Another factor that might influence the different concentrationof radionuclides in individual fish samples may be age/size. Fishsize and weight are strongly correlated (Table 1), althoughC. auratus individuals presented similar dimensions. Fish lengthwas also less important than fish tissue/organ or species to explainthe concentrations of radionuclides in Peníscola (Table 2). Severalstudies showed that some radionuclides have a relationship be-tween their concentration in fish tissue/organ and fish weight (e.g.Smith et al., 2002). Although there is not a clear trend for 210Po,likely because the age of the collected fish (>2 y) is longer than thehalf-life of this radionuclide (138 d), there is a slight relationship

J. Garcia-Orellana et al. / Journal of Environmental Radioactivity 155-156 (2016) 46e5452

between the fish weight and the concentration of 210Pb in thosetissues/organs samples where a greater accumulation of 210Pb wasobserved (i.e. gills and spine) (Fig. 5).

3.4. Bioaccumulation factors (BAF)

The bioaccumulation factor (BAF), also named concentrationratio (CR), is defined as the ratio between the radionuclide con-centration (fresh weight) in biota and its concentration in water,and it is commonly used as a reference indicator for the radionu-clide's accumulation in aquatic environments (IAEA, 2010).Considering the concentration of 210Pb and 210Po in water reportedin Garcia-Orellana et al. (2013), BAFs for 210Pb and 210Po estimatesin this work appear to vary depending on the specific tissue/organconsidered. BAFs for 210Pb in C. auratus, C. carpio and C. labrosusranged from 6.4$101 to 4.0$103 L kg�1, 45 and 6.7$103 L kg�1, and 72and 1.6$104 L kg�1, respectively. Lowest 210Pb-BAFs were found onmuscle and gonads, while the highest values were obtained inspine for C. carpio and C. labrosus and in gill for C. auratus.

On the other hand, BAFs for 210Po in C. auratus, C. carpio andC. labrosus ranged from 3.2$103 to 1.1$105 L kg�1, 8.5$102 and8.7$104 L kg�1, and 1.3$103 and 2.3$105 L kg�1, respectively. 210Pb-BAFs have the same trend in three different species, with lowestvalues in muscle, gills and gonads, and highest values in gut, kidneyand hepatopancreas.

3.5. Comparison with marine and freshwater studies

Results show that all analyzed species have higher 210Pb and210Po concentrations in different internal tissues/organ comparedto existing data in continental waters (Lambrecht et al., 1992;Cherry et al., 1994; Hameed et al., 1997; Carvalho et al., 2007;NKS, 2009). In a work conducted in lakes of Finland (NKS, 2009),210Pb and 210Po concentrations in fish were considerably lowercompared to the present study. 210Pb and 210Po concentrationsranged from 0.014 ± 0.003 to 0.13 ± 0.02 Bq kg�1 (f.w.) and from0.08 ± 0.02 to 1.86 ± 0.35 Bq kg�1 (f.w.) for edible parts, respec-tively, compared to concentrations of 210Pb and 210Po inmuscle thatranged from 1.1 ± 0.1 to 21.3 ± 0.8 Bq kg�1 (f.w.) for 210Pb and from4.4 ± 0.6 to 77 ± 2 Bq kg�1 (f.w.) for 210Po. Hameed et al. (1997) alsomeasured the 210Pb concentration in fish bones and musclecollected in Kaveri river system (India). Bone and muscle 210Pbconcentrations ranged from 0.21 to 0.32 Bq kg�1 (f.w.) and 0.74 and

Fig. 5. 210Pb concentration in gills and spines of C. carpio, C. labrosus and C. auratus as afunction of the fish weight.

2.03 Bq kg�1 (f.w.), respectively, which are values significantlylower than those obtained in this work.

Several studies reported the 210Pb and 210Po concentration levelsin fish in rivers or lakes affected by U mining industry (i.e. Clulowet al., 1998; �Strok and Smodi�s, 2011; Skipperud et al., 2013).Clulow et al. (1998) analyzed fish tissues (bone, muscle and gutcontents) of lake trout (Salvelinus namaycush) and whitefish (Cor-egonus clupeaformis and Prosopium cylindraceum) from four lakes ina watershed affected by U mining and milling operations at ElliotLake, Ontario (Canada), and showed that in lakes, trout bone 210Pband 210Po levels were higher than in muscle. Another interestingexample was reported by �Strok and Smodi�s (2011), who compared210Pb and 210Po in river fishes from two rivers in Slovenia, one ofthem affected by an ancient uranium mine at �Zirovski vrh. Resultsin Squalius cephalus, Salmo trutta, and Barbus barbus collected inPoljanska Sora (�Zabja vas), one of the rivers affected by the uraniummine, showed higher concentrations compared to the blank riverBrebov�s�cica (Gorenja Dobrava), although these concentrationswere lower than those measured in this work. On the other hand,Skipperud et al. (2013) reported 210Pb and 210Pb concentrations indifferent fish organs from three different fish species in TobosharPit Lake, which is located in a uranium mining area in Tajikistan.210Pb concentrations in C. auratus ranged from 25 to 330 and22e190 Bq kg�1 (f.w.) in liver and bone, respectively. 210Po con-centrations in liver, bone and muscle ranged from 590 to 9400,270e1400 and 130e1300 Bq kg�1 (f.w.), respectively. These mea-surements were in the same range with the concentrations ob-tained for the Peníscola marsh.

There are also many works that reported 210Pb and 210Po con-centrations of biota in seawater. Themore extensiveworkwas doneby Carvalho (2011), which presented a detailed study of the dis-tribution of 210Po and 210Pb concentrations in marine organismsand its transfer into the marine food chain. Comparisons betweenvalues reported by Carvalho (2011) and values reported in thisstudy are shown in Table 6. Most of the 210Pb and 210Po concen-trations reported here are higher than those reported by Carvalho(2011), although some of them, such as the case of the Sardinapilchardus, accumulated extremely high levels of radionuclidesfrom themarine environment. Solea solea and Sparus sp. are speciesthat were investigated in other studies, and results showed thatlow activities of 210Po were found in gills, skins, bones and muscle,and the highest activities were observed in the liver and intestine(Connan et al., 2007) of these species. However, results varied dueto the species’ lifestyle (Connan et al., 2007). Results obtained byConnan et al. (2007) agreewith the results obtained in this study, as210Po concentration in fish tissues/organs samples matched theresulting pattern in their study.

4. Conclusions

Samples of fish tissues/organs from C. carpio, C. labrosus and C.auratus in Peníscola marsh, a wetland nourished by submarinegroundwater discharge (SGD), showed significantly high concen-trations of 210Pb and 210Po compared to blanks (up to 55 and 66times, respectively). Ingestion was determined to be the majorincorporation route into fish for 210Pb and 210Po, as high levels of210Pb and 210Po were found in gut content, as well as in organsinvolved in digestion and metabolism (i.e. gut, kidney and hepa-topancreas). There is no apparent relationship between 210Poaccumulation and fish size or weight within the same species,although there seems to be some correlation between high 210Pblevels and size or weight in gills and spines. Generally, the highestlevels of 210Pb and 210Po concentration in fish tissues/organs werefound in C. labrosus rather than in C. auratus and C. carpio. Resultsconfirmed that the level to which a radionuclide is accumulated in

Table 6Comparison between 210Pb concentrations (in Bq kg�1 f.w.) reported by authors in a previous study (Carvalho, 2011) and 210Pb concentrations (in Bq kg�1 f.w.) reported by thispresent study.

Tissue 210Pb 210Po

Bq kg�1 (f.w.) Bq kg�1 (f.w.)

Carvalho (2011) Present study Carvalho (2011) Present study

Muscle 0.15e2.1 1.1e21 0.5e66 4.4e77Spine 0.7e31 4.0e362 5.9e197 8.6e706Gonads 0.2e22 2.9e55 4.5e275 20e183Hepatopancreas 0.3e134 4.4e33 5.4e2140 56e477Gut 0.3e100 2.5e27 10e28,000 187e666

J. Garcia-Orellana et al. / Journal of Environmental Radioactivity 155-156 (2016) 46e54 53

an organism depends on a wide range of factors, such as itschemical characteristics and speciation in water or sediment, bio-logical processes, including rates of uptake from water or diet,excretion, and metabolic transformation. Results from this studysuggest that the high accumulation of 210Po and 210Pb in wetlandfish is associated with SGD inputs supplying brackish waterenriched in 226Ra and 222Rn. This work represents one of the firstevidences of direct ecological effects derived from SGD. Measuredconcentrations of 210Pb and 210Po in fish were even higher thanthose concentrations measured in rivers or lakes affected by Umining industry, indicating that semi-enclosed coastal areas withsignificant inputs of SGD should be studied because of its potentialrisk for biota.

Acknowledgements

The authors would like to thank the assistance in the field and inthe laboratory work provided by our colleagues at the Laboratori deRadioactivitat Ambiental (Universitat Aut�onoma de Barcelona).This project has been funded partially by the Spanish Governmentproject EDASMAR (ref. CGL2006-09274/HID), the Consejo deSeguridad Nuclear project (ref. 2686-SRA), and the Generalitat deCatalunya (2014 SGR 484 and 1356). V.R. and G.C.C. acknowledgefinancial support through PhD fellowships from the Spanish Gov-ernment (AP2008-03044) and the University of Girona (BR2010/10), respectively. V.R. also acknowledges financial support from theEuropean Union's FP7 (Marie Curie Actions PCOFUND-GA-2013-609102), through the PRESTIGE programme coordinated byCampus France. Support for the research of PM was receivedthrough the prize ICREA Academia, funded by the Generalitat deCatalunya. This research is a contribution to the ANR @RActionchair (ANR-14-ACHN-0007-01) and Labex OT-Med (ANR-11-LABEX-0061) funded by the “Investissements d’Avenir” programthrough the A*MIDEX project (ANR-11-IDEX-0001-02) of theFrench National Research Agency (ANR). We also want to thank thecollaboration of Maribel Forner (Hotel Marina), Camping Eden andthe Peníscola Municipality. We alsowant to thank Lluís Benejam forhis support during fish sampling.

References

Barto�n, K., 2015. MuMIn: Multi-Model Inference. R Package Version 1.15.1. https://CRAN.R-project.org/package¼MuMIn.

Brown, J.E., Gjelsvik, R., Roos, P., Kålås, J.A., Outola, I., Holm, E., 2011. Levels andtransfer of 210Po and 210Pb in nordic terrestrial ecosystems. J. Environ. Radioact.102 (5), 430e437.

Burnham, K.P., Anderson, D.R., 2002. Model Selection and Multimodel Inference: aPractical Information-Theoretic Approach, second ed. Springer-Verlag, NewYork.

Carvalho, F.P., Fowler, S.W., 1994. A double-tracer technique to determine therelative importance of water and food as sources of polonium-210 to marineprawns and fish. Mar. Ecol. Prog. Ser. 103, 251e264.

Carvalho, F.P., Oliveira, J.M., Lopes, I., Batista, A., 2007. Radionuclides from pasturanium mining in rivers of Portugal. J. Environ. Radioact. 98, 298e314.

Carvalho, F.P., 2011. Polonium (210Po) and lead (210Pb) in marine organisms and their

transfer in marine food chains. J. Environ. Radioact. 102, 462e472.Cherry, R.D., Shannon, L.V., 1974. The alpha radioactivity of marine organisms. At.

Energy Rev. 12, 3e45.Cherry, R.D., Heyraud, M., 1982. Evidence of high natural radiation doses in certain

mid-water oceanic organisms. Science 218.Cherry, R.D., Heyraud, M., Rindfuss, R., 1994. Polonium-210 in teleost fish and in

marine mammals: interfamily differences and possible association betweenpolonium-210 and red muscle content. J. Environ. Radioact. 24, 273e291.

Clulow, F.V., Dav, N.K., Lim, T.P., Avadhanula, R., 1998. Radionuclides (lead-210,polonium-210, thorium-230, and thorium-232) and thorium and uranium inwater, sediments, and fish from lakes near the city of Elliot Lake, Ontario,Canada. Environ. Pollut. 99, 199e213.

Connan, O., Germain, P., Solier, L., Gouret, G., 2007. Variations of 210Po and 210Pb invarious marine organisms from Western english channel: contribution of 210Poto the radiation dose. J. Environ. Radioact. 97 (2e3), 168e188.

Durand, J.P., Carvalho, F.P., Goudard, F., Pieri, J., Fowler, S.W., Cotret, O., 1999. 210Pobinding to metallothioneins and fer- ritin in the liver of teleost marine fish. Mar.Ecol. Prog. Ser. 177, 189e196.

Fisher, N.S., Beaugelin-Seiler, K., Hinton, T.G., Baumann, Z., Madigan, D.J., Garnier-Laplace, J., 2013. Evaluation of radiation doses and associated risk from theFukushima nuclear accident to marine biota and human consumers of seafood.Proc. Natl. Acad. Sci. U. S. A. 110 (26), 10670e10675.

Fleishman, D.G., 1973. Accumulation of artificial radionuclides in freshwater fish. In:Klechkovskii, V.M. (Ed.), Radioecology. John Wiley and Sons, Inc, New York,pp. 347e370.

Fowler, S.W., 2011. 210Po in the marine environment with emphasis on its behaviourwithin the biosphere. J. Environ. Radioact. 102, 448e461.

García-Berthou, E., 2001. Size- and depth-dependent variation in habitat and diet ofthe common carp (Cyprinus carpio). Aquat. Sci. 63, 466e476.

García-Berthou, E., Moreno-Amich, R., 1993. Multivariate analysis of covariance inmorphometric studies of the reproductive cycle. Can. J. Fish. Aquatic Sci. 50,1394e1399.

Garcia-Orellana, J., Rodellas, V., Casacuberta, N., Lopez-Castillo, E., Vilarrasa, M.,Moreno, V., Garcia-Solsona, E., Masqu�e, P., 2013. Submarine groundwaterdischarge: natural radioactivity accumulation in a wetland ecosystem. Mar.Chem. 156, 61e72.

Hameed, P.S., Shaheed, K., Somasundaram, S.S.N., 1997. Bioaccumulation of 210Pb inthe Kaveri river ecosystem, India. J. Environ. Radioact. 37, 17e27.

IAEA, 2010. Handbook of Parameter Values for the Prediction of RadionuclideTransfer in Terrestrial and Freshwater Environments. Technical Reports SeriesNo. 472. IAEA, Vienna.

ICRP (International Commission on Radiation Protection), 1979. Limits of intakes ofradionuclides by workers. Part 1. In: ICRP Publication 30. Pergammon Press,Oxford, p. 116.

Kottelat, M., Freyhof, J., 2007. Handbook of European Freshwater Fishes. Publica-tions Kottelat, Cornol, Switzerland, p. 646.

Koulikov, A.O., Ryabov, I.N., 1992. Specific cesium activity in freshwater fish and thesize effect. Sci. Total Environ. 112, 125e142.

Lambrecht, A., Foulquier, L., Gariner-Laplace, J., 1992. Natural radioactivity in theaquatic component of the main French rivers. Radiat. Prot. Dosim. 45, 253e256.

NKS, 2009. Nordic Nuclear Safety Research, Po-210 and Other Radionuclides inTerrestrial and Freshwater Environments. A Deliverable report for the NKS-Bactivity October 2008 GAPRAD.

Olsson, P.E., Kling, P., Hogstrand, C., 1998. Mechanisms of heavy metal accumulationand toxicity in fish. In: Langston, W., Bebianno, M. (Eds.), Metal Metabolism inAquatic Environments. Chapman & Hall, London, pp. 321e350.

R Core Team, 2015. R: a Language and Environment for Statistical Computing. RFoundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.

Rigaud, S., Puigcorb�e, V., C�amara-Mor, P., Casacuberta, N., Roca-Martí, M., Garcia-Orellana, J., Benitez-Nelson, C.R., Masqu�e, P., Church, T., 2013. A methodsassessment and recommendations for improving calculations and reducinguncertainties in the determination of 210Po and 210Pb activities in seawater.Limnol. Oceanogr. Methods 11, 561e571.

Rodellas, V., Garcia-Orellana, J., Rodellas, R., Garcia-Solsona, E., Masqu�e, P.,Domínguez, J.A., Ballesteros, B., Mejías, M., 2012. Quantifying groundwaterdischarge from different sources into a Mediterranean wetland by using 222Rn

J. Garcia-Orellana et al. / Journal of Environmental Radioactivity 155-156 (2016) 46e5454

and Ra isotopes. J. Hydrology 466e467, 11e22.Sax�en, R., Koskelainen, U., 2000. Distribution of 137Cs and 90Sr in various tissues and

organs of freshwater fish in Finnish lakes. Boreal Environ. Res. 7, 105e112.Shaheed, K., Somasundaram, S.S.N., Shahul Hameed, P., Iyengar, M.A.R., 1997.

A study of polonium-210 distribution aspects in the riverine ecosystem ofKaveri, Tiruchirappalli, India. Environ. Pollut. 95 (3), 371e377.

Simkiss, K., Taylor, M.G., 1995. Transport of metals across membranes. In: Tessier, A.,Turner, D.R. (Eds.), Metal Speciation and Bioavailability in Aquatic Systems.Wiley, Chichester, pp. 1e44.

Skipperud, L., Jørgensen, a G., Heier, L.S., Salbu, B., Rosseland, B.O., 2013. Po-210 andPb-210 in water and fish from Taboshar uranium mining Pit Lake, Tajikistan.J. Environ. Radioact. 123, 82e89.

Smith, J.T., Kudelsky, A.V., Ryabov, I.N., Daire, S.E., Boyer, L., Blust, R.J.,Fernandez, J.A., Hadderingh, R.H., Voitsekhovitch, O.V., 2002. Uptake andelimination of radiocaesium in fish and the “size effect”. J. Environ. Radioact. 62

(2), 145e164.Sokal, R.R., Rohlf, F.J., 2012. Biometry: the Principles and Practice of Statistics in

Biological Research, fourth ed. W. H. Freeman and Co., New York, p. 937.�Strok, M., Smodi�s, B., 2011. Chemosphere levels of 210Po and 210Pb in fish and

molluscs in Slovenia and the related dose assessment to the population. Che-mosphere 82, 970e976.

Stewart, G.M., Fowler, S.W., Fisher, N.S., 2008. The Bioaccumulation of U- and Th-series Radionuclides in Marine Organisms. In: Radioactivity in the Environ-ment, 13. Elsevier.

Williams, R.J.P., 1981. Physico-chemical aspects of inorganic element transferthrough membranes. Philos. Trans. R. Soc. Lond. B 294, 57e74.

Zarroca, M., Linares, R., Rodellas, V., Garcia-Orellana, J., Roqu�e, C., Bach, J., Masqu�e, P.,2014. Delineating coastal groundwater discharge processes in a wetland area bymeans of electrical resistivity imaging, 224Ra and 222Rn. Hydrol. Process. 28,2382e2395.