Determining provenance of marine metal pollution in French bivalves using Cd, Zn and Pb isotopes

Available online at www.sciencedirect.com

ScienceDirect Geochimica et Cosmochimica Acta 121 (2013) 155–167 www.elsevier.com/locate/gca

Determining provenance of marine metal pollution in French bivalves using Cd, Zn and Pb isotopes Alyssa E. Shiel a,⇑, Dominique Weis a, Daniel Cossa b, Kristin J. Orians a a

Pacific Centre for Isotopic and Geochemical Research, Department of Earth and Ocean Sciences, University of British Columbia (UBC), 6339 Stores Road, Vancouver, BC V6T 1Z4, Canada b IFREMER, Center for the Mediterranean Sea, BP 330, F-83507 La Seyne sur Mer, France Received 18 September 2012; accepted in revised form 4 July 2013; available online 23 July 2013

Abstract Cadmium, Zn and Pb isotopic compositions (MC-ICP-MS) and elemental concentrations (HR-ICP-MS) have been used to distinguish between natural and anthropogenic sources of these metals in bivalves collected from the coastlines of France (English Channel, Atlantic and Mediterranean coasts). The Cd isotopic signatures (d114Cd = 1.08& to 0.52&) exhibited by bivalves from the coastlines of France, excluding those from NE France, are within the range of those exhibited by bivalves from the USA East coast (d114Cd = 1.20& to 0.54&). This indicates the high prevalence of industry, as well as the low natural contributions of Cd from North Atlantic waters in both regions. Thus, the significance of anthropogenic Cd sources is similar. These significant anthropogenic contributions are identified for bivalves with a large range in tissue Cd concentrations. Importantly, French bivalves from the Gironde estuary and Marennes-Ole´ron basin (regions of historic and modern importance for oyster farming, respectively) exhibited the highest Cd levels of the study. Their Cd isotopic signatures indicate historical smelting emissions remain the primary Cd source despite the cessation of local smelting activities in 1986 and subsequent remedial efforts. No significant variability is observed in the d66Zn values of the French bivalves (0.53&), with the exception of the much heavier compositions exhibited by oysters from the polluted Gironde estuary (1.19–1.27&). Lead isotopes do not fractionate during processing like Cd and Zn. They can, therefore, be used to identify emissions from industrial processes and the consumption of unleaded gasoline and diesel fuel as metal sources to French bivalves. Cadmium and Zn isotopes are successfully used here as tracers of anthropogenic processing emissions and are combined with Pb isotope “fingerprinting” techniques to identify metal sources. Ó 2013 Elsevier Ltd. All rights reserved.

1. INTRODUCTION Oysters and mussels are sessile organisms and bioaccumulators. Their tissues have metal concentrations that are much higher than seawater and that are representative of time-integrated bioavailable metal levels. Bivalve species, such as oysters and mussels, have been used to quantitatively ⇑ Corresponding author. Present address: Department of Geology, University of Illinois at Urbana-Champaign, 208 Natural History Building, 1301 W. Green St., Urbana, IL 61801, USA. Tel.: +1 217 333 2695. E-mail address: [email protected] (A.E. Shiel).

0016-7037/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.gca.2013.07.005

monitor spatial and temporal trends in the metal levels of coastal waters for almost 40 years following Goldberg’s proposal (Goldberg, 1975). France has been monitoring bivalve metal levels as a part of the National Mussel Watch Program since 1979 (see ROCCH-RNO website). Some limitations are associated with the use of only elemental concentrations for monitoring metal levels, e.g., it is difficult to differentiate between anthropogenic and natural metal sources or to determine if relatively low metal concentrations represent the natural baseline level. In addition, for some metals, tissue concentrations vary depending on the species. New geochemical tools, such as Cd and Zn isotopic signatures, may provide strong evidence as to metal sources.

156

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167

Cadmium and Zn are among a growing number of nontraditional stable isotope systems under investigation for environmental applications (e.g., Weiss, 2008). These tools have been demonstrated as especially powerful in combination with radiogenic isotope tracers of source, e.g., Pb (Cloquet et al., 2005; Shiel et al., 2010, 2012). While Pb isotopes have a long history of use in environmental investigations, a significant amount of interest has developed around the potential for the use of both Cd and Zn isotopic systematics because of their environmental importance. For Cd and Zn, studies have demonstrated that large isotope effects are associated with industrial and anthropogenic processes (e.g., smelting and refining, waste incineration and fuel combustion; e.g., Cloquet et al., 2005, 2006; Mattielli et al., 2009; Shiel et al., 2010). Pollution tracing studies have utilized the Cd and Zn isotopic compositions of a variety of environmental archives (e.g., soils, sediments, rainwater, particulate matter and lichens; Cloquet et al., 2008; Rehka¨mper et al., 2011). For both Cd and Zn large isotope effects are also associated with biological uptake. Both laboratory and seawater depth profiles have identified significant fractionation during uptake by marine phytoplankton (e.g., Lacan et al., 2006; Bermin et al., 2006; John et al., 2007; Ripperger et al., 2007). For Zn, diffusion and adsorption lead to small isotope fractionation (Cloquet et al., 2008). While for Cd, insignificant fractionation accompanies inorganic processes (Rehka¨mper et al., 2011). The significant fractionation observed for Cd and Zn isotopes is in contrast to the Pb isotopic composition of samples, which reflects the source. As a result, Pb isotopes are used to identify the source in a technique called fingerprinting, whereas Cd and Zn isotopes are used to gain process information. In a study of North America, Shiel et al. (2012) focused on determining the origin of Cd, Zn and Pb in oysters harvested from sites along the coastlines of British Columbia (B.C., Canada), Hawaii and the eastern USA. Lead isotopes were used to “fingerprint” important anthropogenic sources, narrowing the identified potential anthropogenic sources of Cd and Zn at marine sites. Significant variation observed among the Cd isotopic compositions of these oysters was suggested to result from mixing of natural (or background) Cd with anthropogenic inputs. The relative size of the anthropogenic inputs was found to be significant for many sites along the eastern USA. Shiel et al. (2012) also used Cd isotopes to reveal that despite relatively high Cd concentrations in B.C. oysters, the Cd source is primarily natural, most likely the upwelling of Cd rich intermediate waters along the B.C. coast. In addition, a trend was observed between Cd and Pb isotopic compositions for B.C. oysters suggesting that the small anthropogenic contributions were largely from a common source. However, they observed no significant variability among the Zn isotopic compositions of B.C. oysters, which fell in the range reported for North Pacific seawater and plankton tows (John, 2007). Shiel et al. (2012) were able to resolve ambiguities encountered when considering only one isotopic system by using Cd, Pb and to a lesser extent Zn isotope systems in combination as tracers of these elements from a common source.

We apply these new tools to trace the source of Cd, Zn and Pb at sites along the French coastline, selecting collection sites impacted to various degrees by anthropogenic inputs, and giving special attention to identified “hotspots” for metal contamination (Claisse, 1989). Of particular interest are the high Cd levels of the historic and modern oyster regions (Gironde estuary and the Marennes-Ole´ron basin, respectively) on the Atlantic coast of France. High metal concentrations in the Gironde estuary are attributed to the local history of metal contamination from industries (primarily Zn ore smelting and coal mining) along the Riou Mort, a small tributary of the Lot River, which feeds into the Gironde estuary via the Garonne River (Jouanneau et al., 1990; Grousset et al., 1999; Audry et al., 2004). Smelting activities occurred near Viviez–Decazeville, France from 1842 to 1987 (Audry et al., 2004). The smelting facility closed after an accidental release of effluent, rich in Cd and Zn among other heavy metals, associated with the tailings pile in 1986. In fact, high Cd levels were revealed in oysters from sites within the Gironde estuary from 1979, reaching a high in 1988, with site averages between 10.2 and 129 lg g1 dry weight (Claisse, 1989). In addition, all oysters from sites within the Gironde estuary had concentrations well above the 5.0 lg g1 dry weight (1.0 lg g1 wet weight) maximum Cd level for bivalves imposed by the Commission of the European Communities (EC/881/2006 which replaced EC No. 466/2001). These high Cd levels led to the prohibition of production, selling and consumption of oysters from the Gironde estuary (Audry et al., 2004). In addition to leading to a complete shutdown of shellfish aquaculture in the Gironde estuary, elevated Cd levels found in oysters grown in the Marennes-Ole´ron basin (30 km north of the Gironde estuary), the largest oyster farming area in Europe, are also suggested to result from smelting activities in the region (Parra et al., 1999; Boutier et al., 2000). In this study, we use Cd, Zn and Pb isotopes as tracers of sources of these metals in bivalves collected from sites along the French coastline (English Channel, Atlantic and Mediterranean coasts). Isotope systematics are used to assess the relative impacts of anthropogenic sources on coastal sites, in particular on resident bivalve populations. Cadmium, Zn and Pb isotopic tracers are used in combination to help resolve ambiguities encountered with the use of only one of these isotope systems. Well-documented pollution in some areas, especially the Gironde estuary, provides strong support for the use of these isotope systems to trace anthropogenic sources in marine environments. In addition, isotopes are used to assess the influence of pollution from the Gironde estuary system on the adjacent Marennes-Ole´ron basin (located 30 km to the north), the largest oyster farming area in Europe. 2. SAMPLES AND METHODS 2.1. Samples and collection Bivalve samples from the coastlines of France (Fig. 1) were selected to gain an appreciation for the variability of the Cd, Zn and Pb isotopic compositions recorded in their

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167 W 4°

W 2°

0°

E 4°

E 6°

157

E 8°

Dunkerque

N 50.0°

English

hannel

C

Oye Plage, Dunkerque-Calais Ambleteuse-Boulogne Noyelles-Godault

Le Havre Rouen Cap de la Hève, Seine estuary

Paris

Aber Benoît

FRANCE Pointe de Chemoulin, Loire estuary

Boyardville, La Mouclière, Les Palles, Marennes-Oléron basin

N 46.0°

Atlantic Ocean

Lyon

La Fosse, Gironde estuary

Bordeaux N 44.0°

Decazeville

Etang du Prévost

Anse de Carteau

Narbonne

Etang de Bages

N 42.0°

200 km

Mediterranean Sea

Fig. 1. Map of France showing the locations of sampling sites. The sites along the French Mediterranean coast are unique in that they are coastal lagoons (Etang de Bages and Etang du Pre´vost) and a semi-enclosed basin (Anse de Carteau). Note the Marennes-Ole´ron basin and Anse de Carteau are sites of major oyster and mussel production, respectively. Facilities of significance in the discussion are shown here. Black star: smelting and refining of Pb and Zn near Noyelles-Godault and of Zn near Decazeville (both now closed); white star: iron and steel mill near Dunkerque; grey star: Cd pigment plant near Narbonne (now closed).

bivalve tissues. Oyster and mussel tissue samples (dried powders) from France (Fig. 1) have been provided by the national mussel watch project handled by the Re´seau National d’Observation de la qualite´ du milieu marin (RNO) of the Institut Francßais de Recherche pourl’Exploitation de la Mer (IFREMER). Information about the IFREMER mussel watch project, including details about the sampling sites and sample collection is provided by Claisse (1989). The majority of oyster and mussel samples from France (Fig. 1) were collected in 2004 and 2005. However, archived samples from 1984 (Boyardville, Marennes Ole´ron basin) and 1987 (La Fosse, Gironde estuary) are also included to capture the relatively immediate impact of an industrial spill in 1986 upstream of the Gironde estuary. As no one species exists at all sites, our study uses three species of oysters (Crassostrea gigas) and mussels (Mytilus edulis, M. galloprovincialis). The species of each sample, as well as the collection year are indicated in Tables 1–3. Differences in bioaccumulation of Zn are expected between oysters and mussels; Zn is concentrated more than 10-fold in the oyster C. virginica relative to the mussel M. edulis (O’Connor, 1993). However, differences are not expected in the bioaccumulation of Cd and Pb between oysters and mussels (O’Connor, 1993), with the exception of specimens from Cd contaminated environments (Claisse, 1989).

2.2. Sample preparation Detailed descriptions of the sample preparation, which led to the dried oyster tissue powders provided for use in this study, are provided by Claisse (1989). In brief, the bivalves were cleaned of epibiota with seawater at the site, depurated for 24 h to eliminate feces and pseudofeces, shucked and then frozen for transport. After returning to the lab, frozen samples were thawed, ground, homogenized and freeze-dried. Samples from each site were pooled as >10 oysters or >50 mussels. Details regarding the sample preparation for isotopic analysis, inclusive of sample digestion and anion exchange chromatographic procedures for Cd, Zn and Pb, are provided by Shiel et al. (2009, 2012). 2.3. Data presentation The majority of studies report variations in Cd and Zn isotopic compositions using a relative notation, typically as either delta (d) or epsilon (e) values, where the values are reported in parts per thousand (i.e., per mil, &) or parts per ten thousand (i.e., per myriad), respectively. Cadmium and Zn isotopic ratios are expressed in this study using the standard delta (d) per mil (&) notation as follows:

158

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167

Table 1 Cadmium concentrations (lg g1 dry weight) and isotopic compositions of bivalve tissues. Sample collection sitea, year France English Channel Oye plage, Dunkerque-Calais Ambleuteuse-Boulogne Cap de la He`ve, Seine estuary Aber Benoıˆt, North Brittanyd Aber Benoıˆt, North Brittany dup.d Atlantic Ocean Pointe de Chemoulin, Loire estuary Boyardville, Marennes-Ole´ron basinf Boyardville, Marennes-Ole´ron basinf La Mouclie`re, Marennes-Ole´ron basin Les Palles, Marennes-Ole´ron basin La Fosse, Gironde estuaryd La Fosse, Gironde estuary dup.d,e La Fosse, Gironde estuaryd Mediterranean Sea Etang de Bages, Roussillond Etang de Bages, Roussillon dup.d,e Etang du Pre´vost, Thaud a b c d e f

Bivalve species

[Cd]

d111Cd/110Cdb

d112Cd/110Cdb

d113Cd/110Cdb

d114Cd/110Cdb

nc

2004 2004 2004 2005 2005

M. edulis M. edulis M. edulis C. gigas C. gigas

0.48 0.57 2.2 1.4 1.4

0.09 ± 0.13 0.15 ± 0.07 0.22 ± 0.08 0.15 ± 0.11 0.14 ± 0.10

0.10 ± 0.16 0.28 ± 0.16 0.44 ± 0.11 0.29 ± 0.17 0.25 ± 0.19

0.21 ± 0.25 0.34 ± 0.31 0.68 ± 0.16 0.49 ± 0.06 0.42 ± 0.18

0.20 ± 0.22 0.49 ± 0.36 0.88 ± 0.23 0.63 ± 0.14 0.52 ± 0.31

3 3 3 2 3

2004

M. edulis

1.8

0.15 ± 0.05

0.31 ± 0.06

0.47 ± 0.21

0.62 ± 0.20

3

1984

C. gigas

0.13 ± 0.07

0.33 ± 0.21

0.51 ± 0.24

0.63 ± 0.35

3

1984

M. edulis

1.9

0.20 ± 0.10

0.36 ± 0.14

0.62 ± 0.23

0.72 ± 0.21

3

2004

M. edulis

1.1

0.22 ± 0.06

0.45 ± 0.08

0.70 ± 0.04

0.92 ± 0.07

3

2004

C. gigas

2.8

0.28 ± 0.05

0.55 ± 0.05

0.81 ± 0.12

1.08 ± 0.09

3

1987 1987 2005

C. gigas C. gigas C. gigas

0.27 ± 0.06 0.27 ± 0.04 0.26 ± 0.04

0.51 ± 0.02 0.52 ± 0.13 0.52 ± 0.05

0.74 ± 0.13 0.81 ± 0.31 0.79 ± 0.12

0.99 ± 0.12 1.06 ± 0.30 1.03 ± 0.17

4 3 5

2005

M. galloprovincialis M. galloprovincialis M. galloprovincialis

5.7

0.14 ± 0.08

0.25 ± 0.05

0.40 ± 0.04

0.51 ± 0.11

5

5.7

0.13 ± 0.13

0.26 ± 0.14

0.41 ± 0.07

0.51 ± 0.20

3

0.37

0.09

0.17

0.19

0.27

1

2005 2005

12

129 129 29

Sample collection sites labelled to be consistent with those assigned by the RNO (IFREMER, France). Ratios are reported permil (&) as the mean ± 2 standard deviation (SD). n Refers to the number of replicate isotopic measurements. [Cd] provided by the RNO (IFREMER). dup. refers to a full procedual duplicate, inclusive of the analytical separation and isotopic analysis. Note [Cd] of the Boyardville oyster sample is 6.3 [Cd] of the Boyardville mussel.

j

d Cd ¼ dk Zn ¼

! ðj Cd=110 CdÞsample  1  1000 ðj Cd=110 CdÞstandard ! ðk Zn=64 ZnÞsample  1  1000 ðk Zn=64 ZnÞstandard

where j are the Cd isotopes in the ratio numerator (111, 112, 113 or 114) and k are the Zn isotopes in the ratio numerator (66, 67 or 68). Isotopic compositions for all the ratios mentioned above are reported. Cadmium isotopic compositions are expressed relative to the PCIGR-1 Cd reference material (Shiel et al., 2009), which has an isotopic composition that is within error of that of NIST SRM 3108 Cd (d114Cd = 0.06&; calculated from Abouchami et al., 2012; Shiel et al., 2012). Zinc isotopic compositions are expressed relative to the “Lyon-JMC” Zn reference material (Mare´chal et al., 1999), which has a slightly lighter Zn isotopic composition than the PCIGR-1 Zn reference material (d66Zn = 0.12 ± 0.05&, n = 5; Shiel et al., 2009) used as the in-house bracketing standard. Delta values for Cd (dry plasma mode) and Zn (wet plasma mode) are reported as combined external normalization—SSB

corrected. For Cd and Zn the external normalization was made using Ag or Cu, respectively (Shiel et al., 2009). Zinc delta values for dry plasma mode are reported as SSB corrected. For Pb, the 206Pb/204Pb, 207Pb/204Pb, 208Pb/204Pb, 206 Pb/207Pb and 208Pb/206Pb ratios are reported. 2.4. Analytical methods Experimental work was carried out in metal-free Class 1000 clean laboratories at the Pacific Centre for Isotopic and Geochemical Research (PCIGR), University of British Columbia (UBC). Elemental analysis was carried out on an ELEMENT2 (Thermo Finnigan, Germany) high-resolution inductively coupled plasma mass spectrometer (HR-ICPMS). The trace element analysis method and instrument set-up are described by Shiel et al. (2009). Isotopic analysis for Cd and Pb was performed on a Nu Plasma (Nu 021; Nu Instruments Ltd., UK) multi-collector inductively coupled plasma mass spectrometer (MC-ICP-MS). Isotopic analysis for Zn was performed in both dry and wet plasma modes on a Nu Plasma (Nu 021; Nu Instruments Ltd., UK) MC-ICP-MS and a Nu Plasma II (Nu 214; Nu Instruments

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167

159

Table 2 Zinc concentrations (lg g1 dry weight) and isotopic compositions of bivalve tissues. Sample collection sitea, year

Bivalve species

[Zn]

d66Zn/64Znb

d67Zn/64Znb

d68Zn/64Znb

nc

Franced English Channel Aber Benoıˆt, North Brittanye Aber Benoıˆt, North Brittanyf

2005 2005

C. gigas C. gigas

1320 1320

0.54 ± 0.03 0.51 ± 0.04

0.82 ± 0.08 0.78 ± 0.07

1.04 ± 0.06 1.02 ± 0.07

3 3

Atlantic Ocean La Fosse, Gironde La Fosse, Gironde La Fosse, Gironde La Fosse, Gironde

1987 1987 2005 2005

C. C. C. C.

gigas gigas gigas gigas

8350 8350 3570 3570

1.27 ± 0.04 1.27 ± 0.10 1.19 ± 0.02 1.15 ± 0.04

1.95 ± 0.20 1.91 ± 0.12 1.78 ± 0.07 1.72 ± 0.08

2.50 ± 0.09 2.51 ± 0.12 2.33 ± 0.06 2.29 ± 0.08

3 3 3 3

2005 2005 2005 2005

M. M. M. M.

galloprovincialis galloprovincialis galloprovincialis galloprovincialis

0.52 ± 0.12 0.55 ± 0.06 0.53 ± 0.10 0.58 ± 0.07

0.79 ± 0.15 0.80 ± 0.10 0.81 ± 0.14 0.87 ± 0.16

1.03 ± 0.24 1.09 ± 0.07 1.01 ± 0.16 1.15 ± 0.13

2 4 3 3

estuarye estuaryf estuarye estuaryf

Mediterranean Sea Etang de Bages, Roussillone Etang de Bages, Roussillonf Etang du Pre´vost, Thaue Etang du Pre´vost, Thauf a b c d e f

90.0 90.0 116 116

Sample collection sites are labelled to be consistent with those assigned by the RNO (IFREMER, France). Ratios are reported permil (&) as the mean ± 2 standard deviation (SD). n refers to the number of replicate isotopic measurements. [Zn] provided by the RNO (IFREMER). Samples measured using Cu external normalization combined with SSB in wet plasma. These values are discussed in the text. Samples measured using SSB in dry plasma.

Ltd., UK) MC-ICP-MS, respectively. The isotopic analysis methods for Cd, Zn and Pb in dry plasma mode are described by Shiel et al. (2010, 2012). The collector set up and analysis routine for Zn were the same for both runs in dry and wet plasma modes. Zinc isotope measurements in wet plasma mode comprised of three blocks of 15  10 s integrations with a 60 s ESA deflected baseline before each block. The majority of bivalve samples were analyzed for both Cd and Pb isotopic compositions. Zinc isotopic analyses were performed only on a subset of samples chosen to exemplify total variability and extremes in environmental site conditions (especially industrial presence). The linear data arrays for both Cd and Zn ratios are consistent with mass-dependent fractionation. 3. RESULTS Cadmium and Zn concentrations anddelta values for bivalve samples are reported in Tables 1 and 2, respectively. Cadmium concentrations for samples from France (this study) are compared to those of bivalve samples from the USA and Canada (Shiel et al., 2012) in Fig. 2. Variations in bivalve Cd tissue concentration and isotopic composition are shown in Fig. 3. A plot of d114Cd vs. Cd concentration is provided in Electronic Annex EA-1. Lead concentrations and isotopic ratios are reported in Table 3 and shown in Fig. 4. Bivalve tissue d114Cd and d66Zn values are compared to those reported in the literature in Figs. 5 and 6, respectively. 3.1. Cd isotopes Oysters from the English Channel and the Atlantic coasts of France have Cd concentrations ranging from 1.4 (Aber Benoıˆt, English Channel) to 29 lg g1 dry weight

(Gironde estuary, Atlantic Ocean), for samples collected in 2004–2005. Cadmium concentrations are as high as 129 lg g1 dry weight (Gironde estuary) when including samples collected in the mid-1980s (Table 1). French mussels have Cd concentrations ranging from 0.37 (Etang du Pre´vost, Mediterranean Sea) to 5.7 lg g1 dry weight (Etang de Bages, Mediterranean Sea) (Table 1). There is a significant correlation (r2 = 0.7) between d114Cd and 1/[Cd] for the bivalves from the coasts of the Atlantic and English Channel (as the concentration increases, d114Cd decreases; Electronic Annex EA-1). For French bivalves, d114Cd values vary from 0.88& to 0.20& for sites on the English Channel, from 1.08& to 0.62& for sites on the Atlantic coast and from 0.52& to 0.29& for sites on the Mediterranean coast (Fig. 3). The heaviest d114Cd value (0.20&) is exhibited by mussels from Oye Plage (English Channel) in northeast France and the lightest (1.08& to 0.99&) by French oysters from the Gironde estuary and Marennes-Ole´ron basin (i.e., Les Palles) (Table 1; Fig. 3). For bivalves collected from Boyardville (Atlantic coast) in 1984, the Cd concentration of the oyster sample is 6.3 that of the mussel sample, while an insignificant difference is observed between their d114Cd values (Fig. 3). 3.2. Zn isotopes Oysters from the English Channel and the Atlantic coasts of France have Zn concentrations ranging from 1320 (Aber Benoıˆt, English Channel) to 3570 lg g1 dry weight (Gironde estuary) for samples collected in 2005, and up to 8350 lg g1 dry weight (Gironde estuary) for the sample collected in 1987 (Table 2). Mussels from the French Mediterranean coast have Zn concentrations ranging from 90 (Etang de Bages) to 116 lg g1 dry weight (Etang du Pre´vost) (Table 2).

160

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167

Table 3 Lead concentrations (lg g1 dry weight) and isotopic compositions of bivalve tissues. Sample collection sitea, year France English Channel Oye plage, Dunkerque-Calais Oye plage, Dunkerque-Calais rep.d AmbleuteuseBoulogne Cap de la He`ve, Seine estuary Cap de la He`ve, Seine estuary rep.d Aber Benoıˆt, North Brittanye Aber Benoıˆt, North Brittany dup.e,f Atlantic Ocean Pointe de Chemoulin, Loire estuary Boyardville, Marennes-Ole´ron basin Boyardville, Marennes-Ole´ron basin La Mouclie`re, Marennes-Ole´ron basin Les Palles, Marennes-Ole´ron basin La Fosse, Gironde estuarye La Fosse, Gironde estuary rep.d,e La Fosse, Gironde estuary dup.e,f La Fosse, Gironde estuary La Fosse, Gironde estuarye Mediterranean Sea Etang de Bages, Roussillone Etang du Pre´vost, Thaue Etang du Pre´vost, Thau dup.e,f Anse de Carteau, Golfe de Fos

Bivalve species

[Pb]

206

Pb/204Pb 2SEb,c 207Pb/204Pb 2SEb,c 208Pb/204Pb 2SEb,c 206Pb/207Pb 2SEb,c 208Pb/206Pb 2SEb,c

2004 M. edulis

1.8 18.2015

10

15.6226

9

38.176

2

1.16508

1

2.09742

4

2004 M. edulis

1.8 18.2009

9

15.6220

7

38.175

2

1.16509

2

2.09743

4

2004 M. edulis

1.4 18.2225

11

15.6262

11

38.206

3

1.16615

2

2.09662

4

2004 M. edulis

5.7 18.1819

6

15.6271

7

38.194

2

1.16355

1

2.10053

3

2004 M. edulis

5.7 18.1837

6

15.6277

6

38.197

2

1.16358

1

2.10046

3

2005 C. gigas

0.90 18.3787

9

15.6390

7

38.351

2

1.17518

1

2.08670

3

2005 C. gigas

0.90 18.3798

12

15.6378

12

38.350

3

1.17534

2

2.08657

4

2004 M. edulis

1.5 18.1476

8

15.6343

8

38.150

2

1.16076

1

2.10223

3

1984 C. gigas

2.0 18.4302

6

15.6536

5

38.828

1

1.17739

2

2.10673

3

1984 M. edulis

1.8 18.3860

6

15.6567

6

38.782

1

1.17433

1

2.10931

4

2004 M. edulis

1.6 18.4534

7

15.6636

6

38.742

2

1.17810

1

2.09943

3

2004 C. gigas

1.2 18.4639

6

15.6571

6

38.729

2

1.17927

1

2.09758

3

1987 C. gigas

3.4 18.3920

9

15.6556

9

38.478

3

1.17479

1

2.09209

4

1987 C. gigas

3.4 18.3917

10

15.6547

9

38.477

2

1.17485

2

2.09202

3

1987 C. gigas

3.4 18.3917

8

15.6536

7

38.474

2

1.17488

2

2.09190

4

2004 C. gigas

2.6 18.4855

7

15.6636

6

38.596

2

1.18019

1

2.08789

3

2005 C. gigas

3.1 18.5035

12

15.6623

10

38.678

3

1.18134

2

2.09047

4

2005 M. 0.40 18.4924 galloprovincialis 2005 M. 1.7 18.3733 galloprovincialis 2005 M. 1.7 18.3716 galloprovincialis 2004 M. 1.1 18.4451 galloprovincialis

10

15.6630

9

38.488

2

1.18064

1

2.08130

4

6

15.6523

5

38.432

2

1.17385

1

2.09169

4

12

15.6522

11

38.429

3

1.17374

1

2.09175

3

8

15.6580

6

38.446

2

1.17801

1

2.08435

3

a

Sample collection sites labelled to be consistent with those assigned by the RNO (IFREMER, France). Ratios are reported as the mean ± 2 standard error (SE). Reported error values are the ten-thousandth (206Pb/204Pb, thousandth (208Pb/204Pb) or hundred-thousandth (206Pb/207Pb, 208Pb/206Pb) decimal digit. c All data have been normalized to the NIST SRM 981 triple spike Pb ratios of Galer and Abouchami, 1998. d rep. refers to a replicate analysis of the Pb sample solution. e [Pb] provided by the RNO (IFREMER). f dup. refers to a full procedual duplicate, inclusive of the analytical separation and isotopic analysis. b

207

Pb/204Pb),

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167

14

Frequency

12 10 8

English Channel-France

<2 <4 <6 <8 <10 France-mussels France-oysters USA East Coast mussels USA East Coast oysters Hawaii (USA)-oysters B.C. (Canada)-oysters

6 4

[Cd] (µg g-1 dry weight)

8 7 6 5 4 3 2 1 0

16

0.5

Oye Plage

0.6

Ambleteuse

2.2

Seine estuary

1.4

Aber Benoît

Atlantic Ocean-France

1.8

Loire estuary

12 (1984)

1.9 1.1

Marennes-Oléron

(2004)

2.8 129

(1987)

29

(2005)

2 0

161

Gironde estuary

Mediterranean Sea-France <5

<10 <15 <20 <25 <30

[Cd] (μg g-1 dry weight)

5.7

Etang de Bages

0.4

Etang du Prévost

–1.50 –1.00

–0.50 0.00 δ114Cd (‰)

0.50

1.00

Fig. 2. Histogram of the Cd concentrations of oyster and mussel soft tissue samples collected between 2002 and 2006 from France (this study), the USA and Canada (Shiel et al., 2012). Oyster (solid color) and mussel (spots) samples are differentiated, as significant differences in Cd accumulation exist between the two species in polluted environments (Claisse, 1989). Unlike the bivalve samples from the USA, the French and the B.C. samples do not include gut contents. Due to the high relative concentration of gut conents, whole oysters (inclusive of the gut) will have somewhat higher Cd concentrations than those shown here; e.g., the Cd concentrations for B.C. whole oysters are reported to range from 3.78 to 22.1 lg g1 dry weight for oyster farming areas (Bendell and Feng, 2009).

Fig. 3. Plot of variations in the Cd isotopic composition of bivalve samples from France. The grey ellipses denote 2 standard deviation on the mean Cd value, except when this value is less than the longterm reproducibility calculated for the in-house secondary standards (±0.14&), and then the latter is used. Cadmium concentrations in bivalve tissues are given on the left in black and grey for oysters and mussels, respectively. For the Marennes-Ole´ron, samples collected in 1984 are from Boyardville and in 2004 are from La Mouclie`re (top) and Les Palles (bottom).

The Zn isotopic compositions of the French bivalves range from d66Zn = 0.52–1.27& (Fig. 6). The lightest d66Zn values are exhibited by the bivalve samples from the English Channel and Mediterranean Sea (0.53&) (Fig. 6). The heaviest d66Zn values are exhibited by the oysters from the Gironde estuary on the Atlantic coast (1.19& and 1.27&, for oysters collected both in 1987 and 2005, respectively) (Fig. 6). The oysters from the Gironde estuary (Atlantic coast) have by far the highest d66Zn values of the whole study.

1.16355 to 1.17534 for sites on the English Channel, from 1.16076 to 1.18134 for sites on the Atlantic coast and from 1.17374 to 1.18064 for sites on the Mediterranean coast (Fig. 4). A shift is observed between the Pb isotopic compositions of bivalves sampled in the mid-1980s and 2004–2005 from the Gironde estuary and Marennes-Ole´ron basin (Fig. 4). For bivalves collected from Boyardville (Marennes-Ole´ron basin) in 1984, the Pb concentration and isotopic composition of oysters (Fig. 4) is similar but not identical to that of mussels (1.97 lg g1 dry weight, 206 Pb/207Pb = 1.17739 and 1.76 lg g1 dry weight, 206 Pb/207Pb = 1.17433; respectively); any difference may be attributable to differences in their ages, tissue turnover rates, sampling season or locations within the site.

3.3. Pb isotopes Bivalve samples from the coasts of France have Pb concentrations ranging from 0.40 (Etang de Bages, Mediterranean coast) to 3.1 lg g1 dry weight (Gironde estuary, Atlantic coast) for samples collected in 2004 and 2005, and up to 34 lg g1 dry weight (Gironde estuary) for samples collected in the mid-1980s (Table 3). Lead isotopic signatures of the same bivalve samples range from 206 Pb/207Pb = 1.16076 to 1.18134. These values are intermediate between those of French aerosols and pre-industrial sediments (Fig. 4). The values for mussels from the Seine and Loire estuaries are consistent with those published by Couture et al. (2010). The lowest 206Pb/207Pb value is exhibited by Loire estuary (Atlantic coast) mussels (206Pb/207Pb = 1.16076) and the highest by Gironde estuary (Atlantic coast) oysters (206Pb/207Pb = 1.18134; 2005 sample) (Fig. 4). The 206Pb/207Pb values vary from

4. DISCUSSION 4.1. Seawater dissolved Cd and diet as Cd sources for bivalves Field studies have demonstrated that Cd accumulation in oyster tissues is primarily related to direct uptake of dissolved Cd in seawater rather than to tropic transfer (Lekhi et al., 2008; Strady et al., 2011a). This has been confirmed in controlled laboratory experiments, which have further found that the trophic transfer of Cd in oysters is dependent on algae species (Ettajani et al., 2001; Strady et al., 2011c). Lacan et al. (2006) demonstrated that phytoplankton preferentially take up light Cd isotopes. Thus, the preferential uptake of light Cd isotopes by phytoplankton leaves the surface seawater enriched in the heavier Cd

162

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167 2.18 2.16

Br (1 ok .0 en 4, Pb Hil 2.2 or l/Mt 2) e .I sa

French gasoline Aerosols (W. Europe)

208Pb/206Pb

2.14

Auto exhaust-France Highway runoff-France Aerosols-France Industrial (smelter) Industrial (MSWC)

Seine River

2.12

Garonne River and tributaries

2.10 2.08

Loire River

2.06 2.04 1.08

Pre-industrial sediments

1.10

1.12

1.14 206

Bivalve samples: Oye Plage Ambleteuse Seine estuary Aber Benoît

Gironde

1.16

1.18

1.20

1.22

Pb/207Pb

Marennes-Oléron

1987

Boyardville, 1984

2004/5

La Mouclière/Les Palles, 2004

Mediterranean Etang de Bages Etang du Prévost Anse de Carteau

Loire estuary

Fig. 4. Plot of 208Pb/206Pb vs. 206Pb/207Pb for the French bivalves, compared with those of pre-industrial sediments from the Atlantic Ocean (Sun, 1980), Seine, Loire and Gironde Rivers (ElbazPoulichet et al., 1986); suspended particulate matter and water from the Garonne River and its tributaries (Elbaz-Poulichet et al., 1986; Grousset et al., 1999), Seine and Loire Rivers (ElbazPoulichet et al., 1986); aerosols from western Europe and France (1994–1998; Bollho¨fer and Rosman, 2001); auto exhaust (1987) and highway runoff (1992/3) (France; Monna et al., 1995); industrial emissions from a municipal solid waste combustor (MSWC; Cloquet et al., 2005) and smelter (Cloquet et al., 2006) in France and Broken Hill and Mt. Isa Pb ores (Sangster et al., 2000, and references within). There is an insignificant difference between the Pb isotope ratios of the Garonne River and its tributaries collected in the early 1980s and the late 1990s (Elbaz-Poulichet et al., 1986; Grousset et al., 1999) likely due to the dominance of leaded gasoline emissions as a Pb source. For this study, the error (2SE) is smaller than the symbol size.

isotopes (Ripperger et al., 2007). The oyster tissue Cd isotopic composition is thought to primarily reflect that of the seawater with a smaller contribution from their diet.

4.2. Cd isotope systematics The French bivalves exhibit a wide range in Cd concentration, from the lowest, 0.37 lg g1 dry weight, to the highest of the study, 29 lg g1 dry weight (for 2004–2005 samples) (Table 1). With the exception of samples from highly polluted areas (Fig. 2), these samples have relatively low Cd contents compared to oysters from the North Pacific (Shiel et al., 2012) (Fig. 2) likely owing to the comparatively low Cd concentrations in North Atlantic seawater. In addition, bivalves from the French coasts (Fig. 1) exhibit large overall variation in d114Cd values (Fig. 3), from relatively heavy isotopic compositions, such as observed in bivalves from Oye Plage (English Channel) and Etang du Pre´vost (Mediterranean Sea), to relatively light isotopic compositions, such as those observed in bivalves from sites within the polluted Gironde estuary and the nearby Marennes-Ole´ron basin (Atlantic coast). While the lightest Cd isotopic compositions of this study are comparable to those reported for bivalve samples from the Atlantic coast of the

USA (Shiel et al., 2012), the observed total variation for French oysters is larger (Fig. 5). This range in Cd isotopic composition, in addition to the observed range in Cd concentration, highlights the combination of low natural contributions in the North Atlantic and the significance of anthropogenic sources. Therefore, the range in d114Cd is largely attributed to variability in the relative contributions of anthropogenic inputs at the French sites. For bivalve samples collected from the English Channel, increasing Cd concentrations correspond to increasingly light Cd isotopic compositions (Electronic Annex EA-1). This trend is suggested to reflect mixing of natural Cd and variable contributions of anthropogenic Cd. The mussel sample from Oye Plage (English Channel) has the lowest Cd concentration and exhibits the heaviest Cd isotopic composition of the French bivalves (d114Cd = 0.20&). While the heaviest Cd isotopic composition of this study, it is still significantly lighter than seawater (Ripperger et al., 2007) (Fig. 5). In contrast, mussels from the Seine estuary have the highest Cd concentration and exhibit the lightest Cd isotopic composition of the bivalves from the English Channel (Fig. 3). The Seine River and estuary receive wastewater and industrial effluents from neighboring urban areas including Paris, Rouen and Le Havre (Fig. 1). Cadmium contamination of the Seine estuary is largely attributed to the disposal (banned in 1992) of calcium sulfate, a Cd-rich waste byproduct produced by phosphoric acid plants located near Rouen and Le Havre (Fig. 1), into the Seine River and directly into the estuary (Nakhle´ et al., 2007). This industrial waste is thought to be responsible for the observed elevation in Cd concentrations and the relatively light d114Cd observed in the Seine estuary mussels. The Cd isotopic compositions (d114Cd = 1.08 to 0.62&) of the bivalves from the Atlantic coast of France reflect the relatively large industrial presence near the collection sites (Fig. 3). The Cd content of oysters from the Gironde estuary decreases markedly between 1987 and 2005 (from 129.1 to 28.7 lg g1 dry weight), corresponding to the shutdown of the Zn smelter near Viviez–Decazeville (Table 1) in 1986, after a major pollution event, i.e., large release of effluent associated with the tailings pile. Despite the significant decrease in the Cd concentrations over this time, the Cd isotopic signatures (d114Cd = 1&) of the oyster samples (Fig. 5) are within error, and comparable to that of smelting dust (from the Metaleurop Pb smelter and refinery in Noyelles-Godault, Northeast France, closed in 2003; Cloquet et al., 2005). This suggests the dominant source of Cd pollution in the Gironde estuary is historical Cd emissions from the metallurgical industry, despite both cessation of industrial activities and subsequent remediation efforts which have decreased Cd and Zn levels in river and estuary waters and sediments and in resident organisms. The present-day source of these metals is likely remobilization of sediments and associated metals by, e.g., flooding, riverbed dredging and other anthropogenic activities (Audry et al., 2004). The source of high Cd levels in oysters from the Marennes-Ole´ron basin, the largest oyster farming area in Europe, has been suggested to be the Gironde estuary,

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167

163

Bivalves:

Bivalves: English Channel-France bivalves Atlantic coast-France bivalves Mediterranean Sea-mussels USA East Coast bivalves1 Pacific coast-Canada oysters1

English Channel-France oysters Atlantic coast-Gironde estuary, France-oysters Mediterranean Sea-mussels Pacific coast-British Columbia, Canada-oysters1 Marine samples: N Atlantic Ocean seawater2

Marine samples: N Atlantic Ocean Mediterranean seawater3

NE Pacific Ocean seawater Stn. P43

seawater2

N Pacific Ocean seawater2

N Pacific Ocean seawater3

Plankton tows, worldwide2

N Pacific Ocean seawater2

Mineral/Industrial samples: Galena (PbS)4

1.6‰, 3.8‰2

Sphalerite (ZnS)4

Mineral/Industrial samples: Sphalerite (ZnS)4 Zn ore

Pb-Zn metallurgical plant (N France):

concentrates5

Smelter ZnO

Zn and Pb enriched ores5

fume5

Zn and Pb refining emissions (roasting, blast furnace)5

Smelter dust6 Waste incineration

Slag6

Main chimney emissions5

dust6

Environmental samples5

Environmental samples: Polluted soils, Pb-Zn refinery (France)7

–1.50

–1.00

–0.50

0.00

0.50

1.00

Zn ore treatment plant (Riou-Mort-Lot River system, SW France): Tailings6 Percolating water6 Polluted stream sediments6

δ114Cd (‰) Fig. 5. Cdisotopic composition variations of French oysters and mussels plotted with those of other bivalves, seawater, plankton, geological and anthropogenic materials. The grey ellipses indicate error as reported by referenced authors; for this study the grey fields denote 2 standard deviation on the mean Cd value for replicate analyses of each sample except when this value is less than the long-term reproducibility calculated for the in-house secondary standards (±0.14&), and then the latter is used. In several cases the error is smaller than the symbol and so no grey field is shown. Data sources: 1Shiel et al. (2012), 2Ripperger et al. (2007), 3Lacan et al. (2006), 4Schmitt et al. (2009), 5Shiel et al. (2010), 6Cloquet et al. (2005), 7Cloquet et al. (2006).

located 30 km to the south (Parra et al., 1999; Boutier et al., 2000). Although the Cd concentrations of bivalves from sites in the Marennes-Ole´ron basin (La Mouclie`re and Les Palles: 1.1 and 2.8 lg g1 dry weight, respectively) are much lower than those of the oysters from the Gironde estuary (28.7 lg g1 dry weight), their Cd isotopic compositions are within error (Fig. 3). These light Cd isotopic compositions reflect relatively large contributions of anthropogenically derived light Cd, mixing with relatively heavy natural Cd. The similarity between the Cd isotopic composition of the oysters from the Marennes-Ole´ron basin and those from the Gironde estuary suggests a common source, despite large differences among Cd concentrations. This supports the conclusion of Strady et al. (2011a,b), whose work found the Gironde estuary plume travels north along the coast to the Marennes-Ole´ron basin leading to temporal dissolved and particulate Cd concentrations in the Marennes-Ole´ron basin that are closely related to Gironde estuary water discharges. Therefore, smelting

Polluted soils6 Sediment cores6: Unpolluted

–1.00

–0.50

Polluted

0.00

0.50

δ66Zn

1.00

1.50

2.00

(‰)

Fig. 6. Zn isotopic composition variations of French oysters and mussels plotted with those of other bivalves, seawater, plankton, geological and anthropogenic materials. The grey ellipses indicate error as reported by referenced authors; for this study the grey fields denote 2 standard deviation on the mean Zn value for replicate analyses of each sample except when this value is less than the long-term reproducibility calculated for the in-house secondary standards (±0.06&), and then the latter is used. In several cases the error is smaller than the symbol and no grey field is shown. Data sources: 1Shiel et al., 2012; 2John, 2007; 3Bermin et al., 2006; 4 Sonke et al., 2008; 5Mattielli et al., 2009; 6Sivry et al., 2008.

operations in the Riou-Mort impact, not only, the Gironde estuary, but also the Marennes-Ole´ron basin. Both oyster and mussel samples collected in 1984 at Boyardville (Marennes-Ole´ron basin) were included in this study. Cadmium metal concentrations in bivalves for this collection year are the highest reported for this site (Claisse, 1989), reflecting the strength of regional industrial Cd emissions at the time. Consistent with previous findings that significant differences exist between the accumulation of Cd in oysters versus mussels in contaminated environments (Claisse, 1989), the Cd concentration of the Boyardville oyster is 6.3 that of the mussel. Despite this difference in Cd concentration, the d114Cd values of the oyster and mussel samples are within error (Fig. 3), suggesting that

164

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167

differences in Cd uptake among bivalve species do not affect the Cd isotopic composition of the accumulated Cd, i.e., that there is no Cd isotope fractionation. The French Mediterranean sites are coastal lagoons and will therefore have the tendency to accumulate pollutants over time. The significantly higher Cd concentration of the mussels from Etang de Bages (5.7 lg g1 dry weight), as compared to those from Etang du Pre´vost (0.4 lg g1 dry weight), is accompanied by a somewhat lighter Cd isotopic composition (d114Cd = 0.51 as compared to -0.27&, respectively). The Cd contamination at Etang de Bages is attributed largely to Cd emissions associated with Cd pigment plant activities in the nearby city of Narbonne (Fig. 1), which ceased in the early 1990s (Claisse, 1989). In contrast, the Cd isotopic composition of mussels collected from Etang du Pre´vost is consistent with the lightest d114Cd values reported for Mediterranean seawater (Lacan et al., 2006) (Fig. 5). 4.3. Zn isotopesystematics French bivalves from the English Channel and the Mediterranean coast have Zn isotopic compositions (Fig. 6) within error of each other and those of seawater from the English Channel and Atlantic Ocean (Bermin et al., 2006; John, 2007). Their compositions are similar to those of oysters from the North Pacific (average d66Zn = 0.42&) (Shiel et al., 2012). However, oysters from the polluted Gironde estuary have distinctly heavier Zn isotopic compositions. These relatively heavy Zn isotopic compositions are consistent with those of water percolated through the tailings pile (d66Zn = 1.18–1.40&) at the former Viviez–Decazeville smelter and polluted stream sediments collected downstream of the smelter (0.91&) (note that sediments collected upstream have a Zn isotopic composition close to those of other French bivalves, 0.32&) (Sivry et al., 2008) (Fig. 6). Despite the significant difference between the Zn concentrations of oyster samples collected from the Gironde estuary in 1987 and 2005 (8350 to 3570 lg g1 dry weight, respectively), the Zn isotopic signatures (d66Zn = 1.27 and 1.19&, respectively) are not significantly different. This suggests that, similar to Cd, the dominant source of Zn in the Gironde estuary remains historical releases from the smelting facility and mining near Viviez–Decazeville (Fig. 6). 4.4. Pb isotope systematics The Pb contents of bivalves (Table 3) from France (0.40–5.7 lg g1 dry weight) are higher than those of the B.C. oysters, exhibiting similar variability to those of the USA East coast bivalves (Shiel et al., 2012). All French bivalve (Fig. 4) samples have Pb isotopic compositions consistent with industrial, as opposed to automotive Pb sources (Deboudt et al., 1999) or preindustrial North Atlantic sediments (Sun, 1980). Lead deposited in the environment from the use of leaded gasoline (banned in France in 2000) is characterized by a low radiogenic Pb isotopic signature (e.g., auto exhaust collected in 1987 before the introduction of unleaded gasoline in France in 1989; Monna et al., 1995; Fig. 4). This is because the tetra-ethyl lead

(TEL) additive was produced primarily from Pb ores from the Broken Hill, (Australia), Mt. Isa (Australia) and Sullivan (B.C.) mines (Monna et al., 1995), which all exhibit unradiogenic Pb isotopic signatures. The decrease in the Pb concentration of aerosols in Europe, associated with the progressive phase-out of leaded gasoline, is linked with a systematic change in 206Pb/207Pb, i.e., increasing since 1979 (Grousset et al., 1994). Bivalve samples from English Channel sites (Oye Plage, Ambleteuse and the Seine estuary) and the Loire estuary (Atlantic coast) are characterized by relatively low 206 Pb/207Pb values, when compared to other French bivalve samples. Although the Metaleurop Pb smelter in NoyellesGodault (Northeast France) closed in 2003 (Franssens et al., 2004), Pb emissions (29,000 kg Pb in 2002; DRIRE Nord Pas-de-Calais, 2003) from the smelter are expected to represent a significant contribution of Pb to the bivalves collected in 2004 at proximal sites. A nearby iron and steel mill (Dunkerque, Northeast France) emitted a significant amount of Pb (about half the quantity of the Pb smelter, 15,319 kg) in 2002 (DRIRE Nord Pas-de-Calais, 2003). A mixture of Pb emissions from these two facilities may be largely responsible for the Pb isotopic signatures exhibited by the bivalves from Northeast France. The Pb isotopic signature of the oyster sample from Aber Benoıˆt (English Channel) is closer to that of the radiogenic pre-industrial sediments (Sun, 1980) endmember than to those of the other bivalves sampled from the English Channel (Fig. 4), indicating a larger contribution of natural Pb at Aber Benoıˆt. The oysters from the Loire estuary are characterized by the most unradiogenic Pb isotopic signature of the study (Fig. 4). Interestingly, these oysters are located downstream from a now-closed alkyl-Pb plant (closed in 1996), a historically significant source of very unradiogenic Pb (e.g., Broken Hill ore) to the Loire estuary. Resuspension of this Pb in the river or estuary is suggested to account for the relatively low 206Pb/207Pb observed in Loire oysters (Couture et al., 2010). The Pb isotopic compositions of the bivalves from the Gironde estuary and Marennes-Ole´ron basin (Fig. 4) plot with values reported for the Garonne River and its tributaries (SPM and sediments) in the late-1990s (Grousset et al., 1999) and the mid-1980s (Elbaz-Poulichet et al., 1986). Lead in bivalves from the Gironde estuary and the Marennes-Ole´ron basin (Atlantic coast) is suggested to derive largely from local industries, primarily regional mining and smelting operations (closed in 1987; Audry et al., 2004). A small shift in the Pb isotopic composition toward the radiogenic natural endmember (pre-industrial river sediments; Elbaz-Poulichet et al., 1986) is observed between the mid-1980s and 2004–2005 for bivalves from both the Gironde estuary and the Marennes-Ole´ron basin (Fig. 4). However, insignificant change is observed in the Pb concentrations of the bivalve tissues over the same time (Table 3). This shift is consistent with increasing contributions of Pb from a natural endmember and likely results from the closure of regional mining and smelting operations. This is in contrast to the insignificant change observed in the Cd and Zn isotopic compositions of bivalve tissues over that time (see Sections 4.1 and 4.2), which indicate the

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167

relative importance of the dominant Cd and Zn sources have not changed. This striking difference between the reaction rates to environmental changes of Pb versus Cd and Zn isotopic signatures is a strong argument for the combined use of these isotope systems to trace the sources, fate and behavior of these metals in the environment. Mussels collected from French Mediterranean sites (coastal lagoons/semi-enclosed basins) show increasing Pb concentrations as the Pb isotopic compositions become less radiogenic (Table 3; Fig. 4). This trend indicates increasing contributions from anthropogenic Pb sources and decreasing contributions from natural Pb sources (e.g., pre-industrial sediments; Sun, 1980). The Pb isotopic compositions of mussels from Etang du Pre´vost fall within the range reported for recent shells from the Mediterranean lagoon, whereas the mussels from Etang de Bages and Anse de Carteau have more radiogenic signatures similar to those of ancient shells from the Mediterranean lagoon (Labonne et al., 1998). The higher Pb concentration and less radiogenic Pb signature of the Etang du Pre´vost mussels likely results from the accumulation of metal pollution associated with urban waste effluents discharging in the lagoon until 2006 (LER, 2008). 5. CONCLUSIONS Our investigation of Cd, Zn and Pb isotopic signatures in oysters and mussels from France resulted in the following conclusions:

165

(3) The Pb isotopic compositions of French bivalves are consistent with industrial, as opposed to automotive Pb sources. Shifts toward natural values are observed in the Pb isotopic compositions of Gironde estuary bivalves between the mid-1980s and 2004–2005, which are likely related to the closure of the smelter near Viviez–Decazeville in 1987. (4) The results of this study demonstrate the effective use of Cd isotopes, and to lesser extent Zn isotopes, to trace industrial emissions of these metals in the environment. While Cd and Zn isotopes are used as indicators of anthropogenic processes that fractionate these elements, Pb isotopes are used to trace (fingerprint) the source.

ACKNOWLEDGEMENTS We thank Didier Claisse at IFREMER for providing the bivalve samples. We are grateful to Jane Barling for her assistance with the Nu Plasma MC-ICP-MS and to Vivian Lai for her assistance with the ELEMENT2 HR-ICP-MS. We are greatly appreciative of the constructive reviews by three anonymous reviewers and to Claudine Stirling for editorial handling. This study was funded by NSERC Discovery grants to Dominique Weis and Kristin J. Orians. Alyssa E. Shiel also gratefully acknowledges support from NSF ADVANCE Grants 0620101 and 0620087.

APPENDIX A. SUPPLEMENTARY DATA (1) Significant variation in d114Cd values, as well as Cd concentrations, exists among bivalves collected from the coasts of France, from relatively pristine (relatively heavy d114Cd) to heavily polluted (relatively light d114Cd) isotopic signatures. The d114Cd values of Gironde estuary oysters are characteristically light, reflecting historical smelting emissions inputs into the watershed—similarly light signatures in bivalves from the Marennes-Ole´ron basin reveal a common source. For the smelting polluted sites in the Gironde estuary and the Marennes-Ole´ron basin, decreases in the Cd concentrations between the mid-1980s and 2004–2005 are not accompanied by a shift toward natural isotopic compositions, suggesting the dominant Cd source remains emissions from historical smelting and mining activities. (2) The Zn isotopic compositions of the majority of French bivalves are similar to those of North Pacific oysters and fall within a narrow range of those reported for seawater and plankton tows. Oysters from the smelting/mining polluted Gironde estuary exhibit exceptionally heavy Zn isotopic compositions. The heavy composition reflects the primary Zn source, effluent associated with the tailings pile, and is expected to in part directly result from the 1986 accidental release of effluent from the tailings pile. Differences between the origin of smelter associated Cd and Zn to the Gironde estuary likely reflect optimization of the production technique for Zn, rather than Cd, and the lower boiling point of Cd compared to Zn (i.e., light Cd isotopes will preferentially escape as emissions during smelting processes).

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.gca.2013.07.005. REFERENCES Abouchami W., Galer S. J. G., Horner T. J., Rehka¨mper M., Wombacher F., Xue Z., Lambelet M., Gault-Ringold M., Stirling C., Scho¨nba¨chler M., Shiel A. E., Weis D. and Holdship P. F. (2012) A common reference material for cadmium isotope studies—NIST SRM 3108 Cd. Geostand. Geoanal. Res.. http:// dx.doi.org/10.1111/j.1751-908X.2012.00175.x. Audry S., Blanc G. and Schafer J. (2004) Cadmium transport in the Lot-Garonne River system (France) – temporal variability and a model for flux estimation. Sci. Total Environ. 319, 197–213. Bendell L. I. and Feng C. (2009) Spatial and temporal variations in cadmium concentrations and burdens in the Pacific oyster (Crassostrea gigas) sampled from the Pacific north-west. Mar. Pollut. Bull. 58, 1137–1143. Bermin J., Vance D., Archer C. and Statham P. J. (2006) The determination of the isotopic composition of Cu and Zn in seawater. Chem. Geol. 226, 280–297. Bollho¨fer A. and Rosman K. J. R. (2001) Isotopic source signatures for atmospheric lead: the Northern Hemisphere. Geochim. Cosmochim. Acta 65, 1727–1740. Boutier B., Chiffoleau J.-F., Gonzalez J.-L., Lazure P., Auger D. and Truquet I. (2000) Influence of the Gironde estuary outputs on cadmium concentrations in the coastal waters: consequences on the Marennes-Oleron bay (France). Oceanol. Acta 23, 745–757. Claisse D. (1989) Chemical contamination of French coasts – the results of a 10 years mussel watch. Mar. Pollut. Bull. 20, 523– 528.

166

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167

Cloquet C., Carignan J., Lehmann M. F. and Vanhaecke F. (2008) Variations in the isotopic composition of zinc in the natural environment and the use of zinc isotopes in biogeosciences: a review. Anal. Bioanal. Chem. 390, 451–463. Cloquet C., Carignan J., Libourel G., Sterckeman T. and Perdrix E. (2006) Tracing source pollution in soils using cadmium and lead isotopes. Environ. Sci. Technol. 40, 2525–2530. Cloquet C., Rouxel O., Carignan J. and Libourel G. (2005) Natural cadmium isotopic variations in eight geological reference materials (NIST SRM 2711, BCR 176, GSS-1, GXR-1, GXR2, GSD-12, Nod-P-1, Nod-A-1) and anthropogenic samples, measured by MC-ICP-MS. J. Geostand. Geoanal. Res. 29, 95–106. Couture R.-M., Chiffoleau J.-F., Auger D., Claisse D., Gobeil C. and Cossa D. (2010) Seasonal and decadal variations in lead sources to Eastern North Atlantic mussels. Environ. Sci. Technol. 44, 1211–1216. DRIRE Nord Pas-de-Calais. (2003) L’Industrie au Regard de l’Environnement, e´dition 2003. Available at: . Deboudt K., Flament P., Weis D., Mennessier J. P. and Maquinghen P. (1999) Assessment of pollution aerosols sources above the straits of Dover using lead isotope geochemistry. Sci. Total Environ. 236, 57–74. Elbaz-Poulichet F., Holliger P., Martin J. M. and Petit D. (1986) Stable lead isotopes ratios in major French rivers and estuaries. Sci. Total Environ. 54, 61–76. Ettajani H., Berthet B., Amiard J. C. and Chevolot L. (2001) Determination of cadmium partitioning in microalgae and oysters: contribution to the assessment of trophic transfer. Env. Contam. Toxicol. 40, 209–221. Franssens M., Flament P., Deboubt K., Weis D. and Perdrix E. (2004) Evidencing lead deposition at the urban scale using “short-lived” isotopic signatures of the source term (Pb–Zn refinery). Atmos. Environ. 38, 5157–5168. Galer S. J. G. and Abouchami W. (1998) Practical application of lead triple spiking for correction of instrumental mass discrimination. Mineral. Mag. 62A, 491–492. Goldberg E. D. (1975) The mussel watch: a first step in global marine monitoring. Mar. Pollut. Bull. 6, 111–112. Grousset F. E., Jouanneau J. M., Castaing P., Lavaux G. and Latouche C. (1999) A 70 year record of contamination from industrial activity along the Garonne River and its tributaries (SW France). Estuar. Coast. Shelf Sci. 48, 401–414. Grousset F. E., Quetel C. R., Thomas B., Buatmenard P., Donard O. F. X. and Bucher A. (1994) Transient Pb isotopic signatures in the Western-European atmosphere. Environ. Sci. Technol. 28, 1605–1608. John S.G. (2007) The Marine Biogeochemistry of Zinc Isotopes. Ph. D. thesis, Massachusetts Institute of Technology. John S. G., Geis R. W., Saito M. A. and Boyle E. A. (2007) Zinc isotope fractionation during high-affnity and low-affinity zinc transport by the marine diatom Thalassiosira oceanica. Limnol. Oceanogr. 52, 2710–2714. Jouanneau J.-M., Boutier B., Chiffoleau J.-F., Latouche C. and Philipps I. (1990) Cadmium in the Gironde fluvio-estuarine system: behavior and flow. Sci. Total Environ. 97(98), 465–479. LER (Laboratoire Environnement Ressources) (2008) Re´seau de Suivi Lagunaire du Languedoc-Roussilon: Bilan des re´sultats 2007. Rapport Ifremer N° RST/LER/LR/08.010. Available at: . Labonne M., Ben Othman. D. and Luck J. M. (1998) Recent and past anthropogenic impact on a Mediterranean lagoon: lead isotope constraints from mussel shells. Appl. Geochem. 13, 885–892.

Lacan F., Francois R., Ji Y. and Sherrell R. M. (2006) Cadmium isotopic composition in the ocean. Geochim. Cosmochim. Acta 70, 5104–5118. Lekhi P., Cassis D., Pearce C. M., Ebell N., Maldonado M. T. and Orians K. (2008) Role of dissolved and particulate cadmium in the accumulation of cadmium in cultured oysters (Crassostrea gigas). Sci. Total Environ. 393, 309–325. Mare´chal C. N., Te´louk P. and Albare`de F. (1999) Precise analysis of copper and zinc isotopic compositions by plasma-source mass spectrometry. Chem. Geol. 156, 251–273. Mattielli N., Petit J. C. J., Deboudt K., Flament P., Perdrix E., Taillez A., Rimetz-Planchon J. and Weis D. (2009) Zn isotope study of atmospheric emissions and dry depositions within a 5 km radius of a Pb–Zn refinery. Atmos. Environ. 43, 1265– 1272. Monna F., Benothman D. and Luck J. M. (1995) Pb isotopes and Pb, Zn and Cd concentrations in the rivers feeding a coastal pond (Thau, Southern France) – constraints on the origin(s) and flux(es) of metals. Sci. Total Environ. 166, 19–34. Nakhle´ K., Cossa D., Claisse D., Beliaeff B. and Simon S. (2007) Cadmium and mercury in Seine estuary flounders and mussels: the results of two decades of monitoring. ICES J. Mar. Sci. 64, 929–938. O’Connor T. P. (1993) The NOAA national status and trends mussel watch program: national monitoring of chemical contamination in coastal United States. In Environmental Statistics, Assessment and Forecasting (eds. C. R. Cothern and N. P. Ross). Lewis Publications, Boca Raton, FL, USA. Parra M., Castaing P., Jouanneau J. M., Grousset F. and Latouche C. (1999) Nd–Sr isotopic composition of present-day sediments from the Gironde Estuary, its draining basins and the West Gironde mud patch (SW France). Cont. Shelf Res. 19, 135–150. Rehka¨mper M., Wombacher F., Horner T. J. and Xue Z. (2011) Natural and anthropogenic Cd isotope variations. In Handbook of Environmental Isotope Geochemistry (ed. M. Baskaran). Springer, Berlin, Germany. Ripperger S., Rehka¨mper M., Porcelli D. and Halliday A. N. (2007) Cadmium isotope fractionation in seawater – a signature of biological activity. Earth Planet. Sci. Lett. 261, 670–684. ROCCH-RNO. . Sangster D. F., Outridge P. M. and Davis W. J. (2000) Stable lead isotope characteristics of lead ore deposits of environmental significance. Environ. Rev. 8, 115–147. Schmitt A. D., Galer S. J. G. and Abouchami W. (2009) Massdependent cadmium isotopic variations in nature with emphasis on the marine environment. Earth Planet. Sci. Lett. 277, 262–272. Shiel A. E., Barling J., Orians K. J. and Weis D. (2009) Matrix effects on the multi-collector inductively coupled plasma mass spectrometric analysis of high-precision cadmium and zinc isotope ratios. Anal. Chim. Acta 633, 29–37. Shiel A. E., Weis D. and Orians K. J. (2010) Evaluation of zinc, cadmium and lead isotope fractionation during smelting and refining. Sci. Total Environ. 408, 2357–2368. Shiel A. E., Weis D. and Orians K. J. (2012) Tracing cadmium, zinc and lead sources in bivalves from the coasts of Western Canada and the USA using isotopes. Geochim. Cosmochim. Acta 76, 175–190. Sivry Y., Riotte J., Sonke J. E., Audry S., Scha¨fer J., Viers J., Blanc G., Freydier R. and Dupre´ B. (2008) Zn isotopes as tracers of anthropogenic pollution from Zn-ore smelters the Riou MortLot River system. Chem. Geol. 255, 295–304. Sonke J. E., Sivry Y., Viers J., Freydier R., Dejonghe L., Andre´ L., Aggarwal J. K., Fontan F. and Dupre´ B. (2008) Historical variations in the isotopic composition of atmospheric zinc deposition from a zinc smelter. Chem. Geol. 252, 145–157.

A.E. Shiel et al. / Geochimica et Cosmochimica Acta 121 (2013) 155–167 Strady E., Blanc G., Baudrimont M., Scha¨fer J., Robert S. and Lafon V. (2011a) Roles of regional hydrodynamic and tropic contamination in cadmium bioaccumulation by Pacific oysters in the Marennes-Ole´ron Bay (France). Chemosphere 84, 80–90. Strady E., Kervella S., Blanc G., Robert S., Stanisie`re J. Y., Coynel A. and Scha¨fer J. (2011b) Spatial and temporal variations in trace metal concentrations in surface sediments of the Marennes Ole´ron Bay. Relation to hydrodynamic forcing. Cont. Shelf Res. 31, 997–1007. Strady E., Scha¨fer J., Baudrimont M. and Blanc G. (2011c) Tracing cadmium contamination kinetics and pathways in oysters

167

(Crassostrea gigas) by multiple stable Cd isotope spike experiments. Ecotoxicol. Environ. Safety 74, 600–606. Sun S. S. (1980) Lead isotopic study of young volcanic-rocks from mid-ocean ridges, ocean islands and island arcs. Philos. Trans. R. Soc. Lond. A 297, 409–445. Weiss D. J. (2008) Application of nontraditional stable-isotope systems to the study of sources and fate of metals in the environment. Environ. Sci. Technol. 42, 655–664. Associate editor: Claudine Stirling