Experimental validation of environmental controls on the δ13C of Arctica islandica (ocean quahog) shell carbonate

Available online at www.sciencedirect.com

Geochimica et Cosmochimica Acta 84 (2012) 395–409 www.elsevier.com/locate/gca

Experimental validation of environmental controls on the d13C of Arctica islandica (ocean quahog) shell carbonate Erin C. Beirne a,⇑, Alan D. Wanamaker Jr. a, Scott C. Feindel b a

Department of Geological & Atmospheric Sciences, Iowa State University, Ames, IA 50011, USA b Mook’s Sea Farm, Walpole, ME 04573, USA

Received 9 May 2011; accepted in revised form 19 January 2012; available online 28 January 2012

Abstract The marine bivalve species, Arctica islandica, was reared under experimental conditions for 29 weeks in the Gulf of Maine in order to determine the relationship between the carbon isotope composition of shell carbonate (d13CS) and ambient seawater dissolved inorganic carbon (d13CDIC), as well as to approximate the metabolic contribution (CM) to shell material. Three experimental environments were compared: two flow-through tanks (one at ambient seawater conditions, one with a supplemental food source) and an in situ cage. Each environment contained 50 juveniles and 30 adults. Both juvenile (2– 3 years) and adult (19–64 years) specimens displayed average percent CM of less than or equal to 10% when using three different proxies of respired carbon: digestive gland, adductor muscle and sediment. Hence, the primary control on d13CS values is ambient DIC. The relationship between d13CDIC and d13CS for 114 individuals used in the study was: d13 CDIC ¼ d13 CS  1:0& ð0:3&Þ No ontogenetic effect on d13CS was observed, and growth rates did not generally impact d13CS values. Based on the results of this study, shell material derived from the long-lived ocean quahog (A. islandica) constitutes a viable proxy for paleo-DIC from the extratropical Atlantic Ocean. Ó 2012 Elsevier Ltd. All rights reserved.

1. INTRODUCTION Given the exponential growth of atmospheric carbon dioxide (CO2) concentrations in recent decades (IPCC, 2007) and the importance of the world’s oceans in carbon sequestration (Sabine et al., 2004), proxies sensitive to oceanic carbon dynamics from the geologic record are of particular interest. Specifically, the lack of marine-based proxy archives that have been calibrated with ambient ocean conditions and that reflect aspects of the carbon cycle, limits our knowledge of past carbon cycling dynamics.

⇑ Corresponding author. Current address: Department of Earth

and Planetary Sciences, Harvard University, Cambridge, MA 02138, USA. E-mail address: [email protected] (E.C. Beirne). 0016-7037/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.gca.2012.01.021

Approximately half of the CO2 released into the atmosphere through fossil fuel combustion and cement-manufacturing is sequestered by the world’s oceans (Sabine et al., 2004). Recent declines in the carbon isotope ratio (13C/12C) of atmospheric carbon dioxide from the combustion of 13C-depleted fossil fuels provides a means of documenting rates of oceanic carbon sequestration (e.g. Swart et al., 2010), as well as reconstructing the partial pressure of CO2 (Druffel and Benavides, 1986). This trend toward lower (more negative) d13C values in seawater dissolved inorganic carbon (DIC), known as the 13C-Suess Effect (after Druffel and Benavides, 1986; Keeling et al., 1979; Suess, 1953), is caused by the addition of isotopically “light” CO2 from fossil fuels to the atmospheric reservoir. The evolution in atmospheric d13CO2 values during the last millennium has been shown through proxy reconstructions from Antarctic ice core records (Friedli et al., 1986; Francey

396

E.C. Beirne et al. / Geochimica et Cosmochimica Acta 84 (2012) 395–409

et al., 1999). In the marine environment, this trend toward 13 C- depletion is seen through increasingly negative d13C values of DIC (d13CDIC) in marine surface waters (Gruber et al., 1999, 2002; Quay et al., 2003, 2007) and in the d13C values of the accretionary hard parts of calcium carbonate secreting marine organisms, such as foraminiferas, corals, sclerosponges, coralline algae and mollusks (e.g. Nozaki et al., 1978; Druffel and Benavides, 1986; Bo¨hm et al., 1996, 2002; Butler et al., 2009; Cage and Austin, 2010; Swart et al., 2010; Scho¨ne et al., 2011; Williams et al., 2011). Mollusk shell carbonate, especially from bivalves, is one of the few available archives of extratropical marine carbon cycling (Weidman and Jones, 1993; Butler et al., 2009; Poulain et al., 2010; Scho¨ne et al., 2011); however, the nonequilibrium incorporation of carbon isotopes into shell material, resulting from kinetic or metabolic effects (i.e. “vital” effects, see Section 4.1 in discussion), complicates the comparison between the d13C of shell carbonate (d13CS) and d13CDIC of ambient seawater. In the case of marine mollusks, values of d13CS are largely controlled by three factors: d13CDIC of ambient seawater, d13C of respired carbon (d13CR) and the proportion of metabolic carbon (CM) incorporated into the shell material (McConnaughey, 1989; McConnaughey et al., 1997; McConnaughey and Gillikin, 2008). Therefore, in order to reconstruct d13CDIC from d13CS, one must know (or estimate) values of d13CR and CM for the bivalve species of interest. McConnaughey et al. (1997) calculate a theoretical value of 10% metabolic carbon (or CM = 0.1) for aquatic mollusks based on the “respiratory gas exchange model.” However, recent experimental and empirical studies have documented several deviations from this model. First, CM appears to vary from species to species (e.g. Gillikin et al., 2006, 2007; Lorrain et al., 2004), even in the marine environment. Second, depending upon the study species, CM may vary over the lifetime of an individual (Klein et al., 1996; Lorrain et al., 2004; Gillikin et al., 2005; Butler et al., 2011). Additionally, there is some evidence that productivity may have a secondary influence on CM (e.g., Lartaud et al., 2010). The viability of d13CS as an archive for paleo-DIC hinges upon the relatively small (ideally less than 10%) and constant percent contribution of metabolic carbon to shell carbonate throughout ontogeny. Poulain et al. (2010) provide the first rigorous, speciesspecific experimental calibration between d13CS and d13CDIC in molluskan carbonate, using the Manila Clam (Ruditapes philippinarum). By systematically changing both the isotopic composition of the food and water, the authors conclude that d13Cphytoplankton is the best proxy for d13CR, that the study species maintained a relatively constant %CM (12%) over the experimental period, and that nearly 80% of the variability in d13CS for this species was explained by the ambient d13CDIC conditions. Such calibrations allow for confident use of a given species in reconstructions of paleoenvironmental conditions, and greater accuracy in the determination of paleo-d13CDIC. Several recent studies have used the aragonitic shells of Arctica islandica (L.), the ocean quahog, in carbon isotope studies of anthropogenic CO2 input to the global oceans.

Scho¨ne et al. (2011) show that annual shell carbonate samples from A. islandica, dating back to AD 1753, display d13C trends similar to atmospheric records reconstructed from ice cores and instrumental records (Friedli et al., 1986; Francey et al., 1999; Keeling et al., 2005) and sclerosponge records from the tropics (Bo¨hm et al., 1996, 2002). Similarly, Butler et al. (2009) hypothesize that a trend in recent centuries toward more negative d13CS values from A. islandica shells (dead- and live-collected) from the Irish Sea provides evidence of the 13C-Suess Effect in the temperate oceans. Despite these applications, the species has not been tested to determine the fidelity of shell d13C as a proxy for environmental d13CDIC. Therefore, d13C reconstructions utilizing A. islandica carbonate can neither decisively rule out a variable contribution of metabolic CO2 to shell carbonate at different stages of ontogeny, nor can they confidently calculate the original d13CDIC values from the isotopic composition of the shell material for A. islandica, because a transfer function between d13CS and d13CDIC has not yet been developed. A. islandica represents an excellent bioarchive for a range of paleoenvironmental applications, which are explored at length by Scho¨ne et al. (2005) and Wanamaker et al. (2008a,b, 2011a). The characteristics of A. islandica that make it an ideal proxy for paleo-DIC reconstructions are its annually deposited growth increments (Jones, 1980), its extremely long life-span (Scho¨ne et al., 2005; Wanamaker et al., 2008b), its demonstrated utility in developing master shell chronologies (Witbaard et al., 1997, 2003; Marchitto et al., 2000; Scho¨ne et al., 2003; Scourse et al., 2006; Butler et al., 2009, 2010; Stott et al., 2010) and its large geographical distribution in the temperate North Atlantic (Dahlgren et al., 2000). The existence of several absolutely-dated shell chronologies, as well as the potential to establish a network of shell-based chronologies in the North Atlantic (see Wanamaker et al. (2011b)) make A. islandica an excellent proxy candidate for investigating extratropical carbon cycling of the shelf seas. A transfer function relating d13CS to ambient d13CDIC may be immediately applied to existing stable carbon isotope data sets from A. islandica to explore spatiotemporal variability in d13CDIC. The study presented in this paper was designed to quantify the percent of metabolic contribution to the shell material of A. islandica juveniles and adults grown in laboratory and in situ environments, over a 29-week experimental period. Carbon isotope values of 175 carbonate samples from 114 individuals are reported. Analysis of d13CS from the experimental period is compared with ambient d13CDIC, ontogenetic age, individual growth rate and d13C of various soft tissues in order to calculate average percent metabolic contribution to shell carbonate. Additionally, we evaluate the range of percent metabolic contributions obtained for individuals and groups based upon the d13C from each of five individual tissues, sediment and average particulate organic carbon (POC) values. By isotopically characterizing and quantifying the metabolic contribution to shell carbonate, we present the first species-specific equation (transfer function) for A. islandica to relate d13CS values to oceanic d13CDIC.

Environmental controls on d13C shell of Arctica islandica

2. METHODS AND MATERIALS 2.1. Collection trip A commercial quahog fishing vessel, F.V. Three of A Kind, was employed on November 21, 2009 to collect live A. islandica via a bespoke dredge in the Gulf of Maine. Approximately 380 adults and 355 juveniles were collected from a depth of 82 m at 44°260 9.82900 N, 67°26.00 18.04500 W off Jonesport, Maine (Fig. 1). Animals were transported to the Darling Marine Center (University of Maine, Orono) in Walpole, Maine and placed in flow-through seawater tanks until the beginning of the experimental period. 2.2. Measurement and marking One hundred and twenty adults and 200 juveniles were randomly chosen for four experimental growth environments. The animals were measured three times along the maximum growth axis using digital calipers (±0.03 mm) and then marked with a number and a line along the ventral margin using Mark-texÒ markers, in order to visually indicate the beginning of the experimental period. A subset of animals were caliper measured in March and May to determine seasonal growth, and all animals were caliper measured at the end of the experimental period in August 2010. The biomarker, calcein, was chosen to visually indicate the beginning of experimental growth (Riascos et al., 2007). To the best of our knowledge, A. islandica have not been successfully marked with calcein, therefore, the concentration and length of exposure to calcein were based

Fig. 1. Generalized map of the Gulf of Maine. Animals were collected off Jonesport, Maine and transported, live, to the Darling Marine Center in Walpole, Maine. Black circles mark the locations of the initial sample collection trip (collection site) and the location of the in situ cage (Darling Marine Center).

397

on best estimates from studies published using other marine mollusks (Moran and Marko, 2005; Riascos et al., 2007; Poulain et al., 2010). Adults were submerged in a 100 mg/ L calcein solution for six days at approximately 9 °C in January 2010. Juveniles were placed in a solution of the same concentration for a period of 48 h. Juveniles in the laboratory growth environments were calcein marked in January, March and May; juveniles from the in situ environment were marked in January and May. Age determinations of a representative subset of animals were made from acetate peel replicas of polished and etched shells at the end of the experiment. The method used for shell preparation was described in detail by Ropes (1984), Scourse et al. (2006) and Butler et al. (2009). Briefly, thick sections of shell were cut and embedded in epoxy and resin. Epoxy blocks were cut using an IsometÒ slow-speed saw and then polished on a BuehlerÒ variable speed grinding and polishing wheel (grits P400, P800, P1200, and nylon pad 0.3 lm alumina paste). Polished shells were then etched with 0.1 M HCl for approximately 180 s. Acetate peels were made and mounted on glass slides. Slides were viewed using a Nikon SMZ1500 transmitted light microscope under 3 – 8x magnification. Annual growth increments were counted in the hinge region and recorded using Buehler OmnimetÒ software. We estimate up to a 1% error in our age estimates based on band counting (i.e., the animals were not crossdated; Black et al., 2008). 2.3. Laboratory and in situ growth environments (and monitoring parameters) The Darling Marine Center (DMC) is located adjacent to the tidally-influenced Damariscotta River estuary (Fig. 1). In the DMC flow-through lab, two raceway tanks (239  58.5  28 cm, 425 L capacity) were set-up to receive seawater at a rate of approximately 12–15 L/min pumped from a depth of approximately 10 m from the Damariscotta River. Each raceway tank held 30 adults (each in an individual, numbered flowerpot) and 50 juveniles in sediment collected from the tidal flats of the Damariscotta River during the experimental period, January 9, 2010–August 3, 2010. Tank A was held at ambient seawater conditions, and temperature was monitored hourly using a TidbitÒ temperature logger (±0.2 °C). Tank A outflow was directly discharged to the Damariscotta River. Tank B received an additional 100,000 cells/mL/day of Shellfish Diet 1800 (Reed Mariculture) through a drip-feeding mechanism and also contained a TidbitÒ temperature logger. Tank B outflow entered a barrel containing a HydrolabÒ MiniSonde, which logged temperature (±0.1 °C), salinity (±0.2 ppt), and pH (±0.2 units), hourly over the course of the experiment. A time series of each salinity (a), temperature (b) and pH (c) measurements is shown in Fig. 2. Two cages (91.5  46  30.5 cm) were deployed in the Damariscotta River estuary on January 18, 2010. Each cage contained 30 adults and 50 juveniles in plastic bins filled with sediment. Cage C was located at 43°550 49.500 N, 69°350 0.1800 W, at a depth of approximately 30 m. However, this cage was accidentally hauled away by a fishing boat two weeks into the experimental period. Although the cage

398

E.C. Beirne et al. / Geochimica et Cosmochimica Acta 84 (2012) 395–409

Fig. 2. Time-series of salinity (a; ±0.2 ppt), temperature (b; ±0.1 °C), and pH (c; ±0.2 units) from the Hydrolab MiniSonde at the outflow of Tank B over the 29-week experimental period. Hourly measurements were binned weekly. The temperature series includes four individual records from three tidbit loggers (±0.2 °C) and the Hydrolab MiniSonde. pH values have been drift corrected and normalized to spot checks taken with an Metrohm handheld pH meter (±0.003 units). The time-series of d13CDIC (d) includes data from Tank A (blue circles) and Tank B (green triangles). The red square is the average value of two surface water samples taken at the site of animal collection in Jonesport, Maine on November 23, 2009.

and most animals were recovered months later, the data were unusable. Cage D was located at 43°550 54.3600 N, 69°340 46.1400 W, at a depth of 10 m, which is the same depth as the source water for the flow-through tanks. It should be noted that Cage D and Tank A environments are assumed to be identical with respect to water conditions due to the similarity of the depth of Cage D and the depth from which water was supplied to Tank A (10 m). A TidbitÒ temperature logger recorded the temperature at this site hourly. Cage D animals were harvested on August 10, 2010. Water samples were collected roughly every two weeks during the experimental period from Tank A and Tank B for analysis of d18O and d13CDIC. In this paper, we will discuss only the results of d13CDIC water samples (Fig. 2d). DIC vials were prepared according to the method described by Taipale and Sonninen (2009). Briefly, 100 lL of 85% phosphoric acid (H3PO4) was injected into 12 mL Labco ExetainerÒ vials, which were then capped, filled with a pure helium headspace, covered with ParafilmÒ, and shipped to the experimental site in Maine. A seawater sample was then filtered through a 0.45 lm filter (Cameoe 30GN Nylon prefilter). One milli liter of filtered water was then injected into an ExetainerÒ using a gas-tight syringe. Vials were covered with ParafilmÒ and sent back to Iowa State University and stored at 4 °C until isotopic analysis.

2.4. Sampling methods 2.4.1. Carbonate sampling New growth in all sampled animals was determined by caliper measurements taken at the beginning and end of the study and verified in cross-section by calcein marks. Calcein marks of a subset of animals were viewed in cross-section and photographed at the end of the experimental period at the Bates College Imaging Center (Lewiston, Maine). A Nikon Modular Focus microscope with an epi-fluoresence illuminator was used to view the calcein marks. Images of the calcein marks were taken using a Nikon DS-Ri1 camera attached to the microscope (Fig. 3). Growth in both juveniles and adults was, in most cases, sufficient to allow for hand sampling of carbonate using a DremelÒ variable speed drill. Adults were sampled by first removing the periostracum using a razor blade, then measuring the initial maximum shell height of the individual with digital calipers and using a diamond drilling tip (Dremel #7144) to grind the outer shell margin along the maximum growth axis back to the initial shell height. Juveniles were also hand sampled using the DremelÒ drill. However, juveniles from Tanks A and B were calcein marked on three occasions (January 7, 2010; March 24, 2010; May 24, 2010), and juveniles from Cage D were marked on two occasions (January 7, 2010; May 24, 2010). These marking periods were evident on the outside

Environmental controls on d13C shell of Arctica islandica

399

Fig. 3. Images of adult (a, b) and juvenile (c, d) specimens in cross-section under white (b, d) and fluorescent (a, c) light. Bright lines in (b) and (d) are consistent with calcein marking events. Adults were marked once at the beginning of the experiment, and juveniles were marked in January, March and May. Numbers 1, 2 and 3 (c, d, e) correspond to growth periods, January–March (1), March–May (2) and May–August (3). Photo (f) highlights the position of external checks (also visible in (e)) at the end of each growth period (white lines), as well as the areas hand sampled for carbonate analysis (black rectangles). This image was produced at the Bates College Imaging Center in Lewiston, Maine.

of the shell due to the presence of a slightly lighter and raised band on the outer shell surface, corresponding to each calcein marking event (Fig. 3c–f). Periostracum was gently removed using a diamond drilling tip. Carbonate from each of the growth periods was milled out of five transects from the beginning to the end of the periods, March– May and May–August. Carbonate powder from all five transects was homogenized in one vial per growth period, prior to analysis. Due to smaller overall growth during the winter, a swath of carbonate was milled corresponding to the entire January–March period (Fig. 3f). These samples were taken using a carbide drilling tip (Brasseler #H52.11.003). Carbonate powder was stored in sterile plastic vials until isotopic analysis. 2.4.2. Tissue sampling Five soft tissues (digestive gland, foot, gills, mantle and adductor muscles) were separated and analyzed from one adult in November 2009, six adults and 18 juveniles in December 2009 and 15 adults and 15 juveniles in August 2010. In the case of juveniles, tissues were grouped by environment due to the small volume of each individual tissue (e.g., all Tank A digestive glands were put into one container for analysis). Tissues were separated, rinsed with deionized water, frozen in a commercial freezer and then freeze-dried for 24 h (after Gillikin et al. (2007)). Tissues were subsequently homogenized and stored in glass vials until isotopic analysis. 2.4.3. POC sampling Samples of particulate organic carbon (POC) were collected in January, March, May and November of 2010 according to the procedure described by Lorrain et al. (2003). Briefly, 500 mL of seawater (taken from Tank A) was filtered through precombusted glass fiber filters (Whatman GF/F), which were then dried overnight, wrapped in foil and shipped to Iowa State University (ISU). Once at ISU, filters were exposed to concentrated hydrochloric acid fumes, left in a hood to drive off residual acid and then dried overnight. Eight to ten 4 mm punchouts were then taken from each filter and stacked into silver capsules, which were placed in a desiccator to await isotopic analysis.

2.4.4. Sediment sampling Sediment from the top 0.5 cm of each of 10 flowerpots in Tank A was sampled on November 10, 2010. The sediment in the pots was not changed during the course of the experiment and remained in the tanks after the animals were harvested; therefore, the top layer of sediment was taken to approximate the average POC that had been collecting from January to November 2010. Sediment was placed in 15 mL sterile, plastic centrifuge tubes, covered with ParafilmÒ, and sent on ice from Maine to Iowa. Once at ISU, the sediment samples were stored at 6 °C for 5 days. A small amount of sediment was smeared onto individually labeled pieces of weigh paper. Each sample was freeze-dried overnight, then homogenized with a methanol-rinsed mortar and pestle (after Bouillon et al. (2004) and Gillikin et al. (2007)). 2.5. Isotopic analyses All stable isotope analyses were performed on a ThermoFinnigan Delta Plus XL isotope ratio mass spectrometer (IRMS), in the Department of Geological and Atmospheric Sciences at ISU. DIC and carbonate analyses were carried out using a GasBench, with an automated sampler (CombiPal), coupled to the IRMS. Standards used on the GasBench were NBS-18, NBS-19 and LSVEC with a longterm, combined uncertainty of 0.08& for d13C. At least one standard was run for every five samples analyzed on the GasBench. DIC samples, stored at 4 °C, were brought to room temperature and run using the GasBench “long method,” consisting of four injections of reference CO2, followed by ten injections of sample gas, followed by two injections of reference CO2. Powdered carbonate samples weighing 300–500 lg were each placed in an ExetainerÒ, capped and flushed with pure helium gas. 100 ll of 95% phosphoric acid was then injected into the vial, and the carbonate sample was allowed to react with the acid in the heating block of the GasBench at 34 °C for 18–24 h. Carbonate analyses were also conducted using the GasBench “long method.” All carbon isotope values are reported relative to the international standard Vienna PeeDee Belemnite (VPDB).

400

E.C. Beirne et al. / Geochimica et Cosmochimica Acta 84 (2012) 395–409

Homogenized tissue samples (0.75–1.0 mg) were weighed into pressed tin capsules, folded and placed in a desiccator until analysis. Algal paste samples were dried in an oven overnight at 50 °C, freeze-dried, homogenized and weighed (0.75 mg) into tin capsules, folded and placed in a desiccator until analysis. Approximately 3 mg of homogenized sediment was weighed out into silver cups, four drops of 5% HCl were added, then the cups were put in an oven overnight at 40 °C (after Bouillon et al. (2004) and Gillikin et al. (2007)). Cups were then folded and placed in a desiccator until analysis. Analyses of soft tissues, algal paste, sediment and POC filters were conducted on a Costech elemental analyzer (EA) interfaced to the IRMS at ISU. Samples in either pressed tin or silver capsules were combusted at 1050 °C and analyzed for d13C and d15N. Each analysis consisted of three injections of reference N2 gas, followed by analysis of one sample nitrogen peak, followed by a mass jump, analysis of one sample CO2 peak, and finally three reference CO2 injections. At least one standard was analyzed for every five samples. Standards run on the EA include internal Acetanilide and Sucrose yielding a long-term precision of 0.06& for d13C and 0.10& for d15N.

64 years. All adults that were analyzed for d13CS values were also aged. Both adults and juveniles displayed measurable growth over the experimental period in all three environments. Total experimental growth of adults measured in August averaged 2.18 mm (±0.77; n = 25), 1.83 mm (±0.77; n = 24), and 1.15 mm (±0.79; n = 25) for Tank A, Tank B and Cage D, respectively. Assuming a constant rate of growth over the entire experimental period, adults grew 0.08 mm/week (±0.03), 0.06 mm/week (±0.03), and 0.04 mm/week (±0.03) in Tank A, Tank B and Cage D, respectively (Fig. 4a). All of the measurements reported in this paper were made using digital calipers. Juveniles grew a total of 12.60 mm (±1.05; n = 15), 12.19 mm (±1.14; n = 15), 9.66 mm (±1.51; n = 15) in Tank

2.6. Calculations 2.6.1. Metabolic contribution and offset from isotopic equilibrium Calculations of percent metabolic contribution (%CM) were made using the following equation (modified from McConnaughey et al. (1997)): %CM ¼ ððd13 CS  ðarb þ d13 CHCO3 ÞÞ=ðd13 CResp  d13 CHCO3 ÞÞ  100

ð1Þ

where d13 CS;HCO3 ;Resp represent the carbon isotope values of shell carbonate, bicarbonate and respired carbon, respectively, and ear-b is the enrichment factor between aragonite and bicarbonate (2.7& ± 0.6 in Romanek et al. (1992)). The offset from observed isotopic equilibrium was calculated by subtracting d13 CHCO3 from d13CS. Calculations of d13 CHCO3 from measured d13CDIC were made using equations provided by Zeebe and Wolf-Gladrow (2005) and additional coefficients from Stumm and Morgan (1996) (a table of d13 CHCO3 values can be found in Electronic Annex EA-1) in order to make a direct comparison with the equilibrium-based model of Romanek et al. (1992). Offset from the Romanek et al. (1992) model was reported as the difference between the offset from observed equilibrium with bicarbonate and 2.7& (i.e. ear-b). 3. RESULTS 3.1. Specimen age and growth rate A representative sampling of 15 juveniles (five from each environment) was aged to characterize the age class of all juveniles in the experiment. All juveniles examined were between 2 and 3 years old at the beginning of the experimental period. Adults ranged in age from 19 to

Fig. 4. Comparison of d13C of shell versus growth rate in adults (a) and juveniles (b), and versus biological age (c).

Environmental controls on d13C shell of Arctica islandica

A, Tank B and Cage D, respectively. Growth measurements of Tank A and B juveniles made in March, May and August allowed for the calculation of average seasonal growth rates. Tank A juveniles grew at an average rate of 0.23 mm/ week (±0.03) during the winter period (January–March), 0.51 mm/week (±0.03) during the spring bloom (March– May), and 0.56 mm/week (±0.10) during the summer period (May–August). Tank B juveniles grew at average rates of 0.20 mm/week (±0.04), 0.46 mm/week (±0.05), and 0.57 mm/week (±0.10) during the winter, spring and summer, respectively. Cage D juveniles were measured in May and August, yielding growth rate estimates of 0.30 mm/ week (±0.04) and 0.37 mm/week (±0.13) during the periods of January–May and May–August, respectively (Fig. 4b). All estimates of seasonal growth rates assume a constant growth rate over each period. Total growth and growth rates of juveniles are based on only the subset of animals (n = 45) harvested in August and used for carbonate analyses. 3.2. Carbon isotope results 3.2.1. DIC and shell carbonate The average d13CDIC value of water collected from Tank A between January and August was 0.5& (±0.1; n = 13). Tank B samples also averaged 0.5& (±0.1; n = 11). Fig. 2d is a time-series of d13CDIC measurements in both environments over the experimental period, during which time the carbon isotopic value of DIC was relatively stable. Two surface water samples taken at the collection site in Jonseport, Maine at the time of animal collection (November 2009) averaged 0.5& (±0.4; n = 2). 3.2.1.1. Adults. Shell carbonate samples from 53 adults were analyzed. Four to five adults from each environment were harvested in May, and therefore integrated only the January–May period. The remaining 39 animals were harvested in August and represent the entire January through August

401

growth period. Adults harvested in May had average d13CS values of 1.7& (±0.3; n = 5) from Tank A, 2.0& (±0.2; n = 5) from Tank B, and 1.7& (±0.3; n = 4) from Cage D. Adults harvested in August had average values of 1.3& (±0.3; n = 14) in Tank A, 1.4& (±0.2; n = 15) Tank B and 1.5& (±0.2; n = 10) in Cage D. The average of all adult measurements was 1.5& ± 0.3. 3.2.1.2. Juveniles. Stable carbon isotope values were obtained for carbonate samples from 45 individual juveniles (15 from each environment) harvested in August, and 20 additional individuals harvested from Tank A (10 in March, 10 in May). Individuals harvested in August were sampled from each of the seasonal growth periods where evident on the external shell surface (Fig. 3). One hundred twenty-two juvenile carbonate values are included in the following results. During the January–March period, juveniles in Tank A averaged 1.3& (±0.2; n = 24) and Tank B averaged 1.2& (±0.2; n = 13). The period integrating March through May averaged 1.8& (±0.1; n = 21) in Tank A and 1.7& (±0.2; n = 14) in Tank B. Cage D juveniles averaged 1.6& (±0.2; n = 14) from January to May. From May to August, carbon isotope values averaged 1.4& (±0.1; n = 13), 1.3& (±0.2; n = 12) and 1.4& (±0.2; n = 11) in Tank A, Tank B and Cage D, respectively. Fig. 5 depicts the range of values for each environment over the entire 29-week experimental period. The average of all juvenile measurements was 1.5& ± 0.3. 3.3. Environmental organic carbon and soft tissue 3.3.1. Soft tissue analyses Five types of soft tissue and three separate harvests (November, December and August) were included in this study. November samples represent animals harvested at the collection site; those harvested in December are from animals harvested directly prior to the experimental period; and, August tissues are representative of animals at the end

Fig. 5. Carbon isotope values of shell carbonate from juveniles (left) and adults (right) in each growth environment. Crosses represent the median of each population, lower and upper bounds of the boxes display the first and third quartile and lines indicate the maximum and minimum values within each group.

402

E.C. Beirne et al. / Geochimica et Cosmochimica Acta 84 (2012) 395–409

Fig. 6. d13C values of individual tissues from juvenile and adult specimens. Sediment and particulate organic carbon (POC) values are also reported (see Fig. 5 for explanation of box and whisker symbols).

of the growth in experimental environments. Results were grouped by age (adults or juveniles) and tissue type, but are integrated over the three time periods and three environments (Fig. 6). Although individual tissues were separated and analyzed from 33 juveniles, tissues from several individuals were combined by tissue type to obtain a large enough sample for analysis. As a result, averages represent nine values of each tissue type. Adult digestive gland values averaged 21.4& (±1.5; n = 23), and juveniles averaged 21.4& (±1.3; n = 9). The average of adult foot values was 18.4& (±0.4; n = 21), and juvenile foot values were 19.7& (±1.0; n = 9). Gills of adults averaged 19.6& (±0.7; n = 21) and juveniles averaged 20.0& (±1.2; n = 9). Mantle tissue from adults had average values of 18.7& (±0.6; n = 22) and juvenile mantle tissue was 19.6& (±1.0; n = 9). Muscle values averaged 17.9& (±0.4; n = 23) for adults and 18.9& (±0.7; n = 9) for juveniles. In both juveniles and adults, digestive gland values were the most depleted and muscle values were the most enriched in 13C. For this reason, subsequent calculations of metabolic carbon will focus on these two soft tissues. 3.3.2. POC samples and shellfish diet 1800 Filter samples collected from Tank A in May and November 2010 had sufficient sample to yield isotopic results. Filters collected in January and March had insufficient sample for d13C determinations. The one filter collected in May had a carbon isotope value of 26.4&, while an average of four replicate filters taken in November 2010 (post-experiment) was 24.4& (±0.4). The shellfish paste added to Tank B had an average stable carbon isotope composition of 30.4& (±0.3; n = 2). 3.3.3. Sediment samples Ten sediment samples from the upper 0.5 cm of ten separate flowerpots in the Tank A flow- through system yielded an average value of 20.8& (±0.3). This value is ta-

ken to represent the particulate carbon reaching the animals over the course of the 29-week experiment. In the absence of regularly spaced POC filter samples spanning the experimental period, we will use the average d13Csediment in lieu of d13CPOC when calculating metabolic contributions to shell carbonate. No monitoring was performed to determine the accumulation rate of POC on the sediment surface; therefore, it should be noted that these values are only rough approximations average POC during the period of January–November 2010. 3.4. Percent metabolic contribution and offset from isotopic equilibrium Values of d13CR used in final calculations were average carbon isotope values of the digestive gland, muscle and sediment. d13 CHCO3 values used in Eq. (1) were calculated from average d13CDIC measurements made over the time period represented by the individual carbonate sample and were specific to environment. Values of d13CDIC assigned to Cage D were the same as those in Tank A because water sampled from the flowing seawater lab was pumped from the same approximate depth as Cage D (10 m). Average %CM was 8.7% (±1.4; Fig. 7) using the digestive gland, 10.0% (±1.6) using the muscle, 9.0% (±1.4) using the average sediment value, integrated over all environments, all growth periods and both age classes (n = 175). The average difference between d13CS and d13CDIC was + 1.0& (±0.3; n = 175), and between d13CS and d13 CHCO3 was + 0.8& (±0.3). In contrast, equilibrium fractionation between aragonite and bicarbonate, as defined by abiotic mineral synthesis, is 2.7& (Rubinson and Clayton, 1969; Romanek et al., 1992). Romanek et al. (1992) noted an enrichment of 2.7& ± 0.6 between bicarbonate and aragonite during inorganic precipitation. Presuming an enrichment of 2.7& for shell carbonate from ambient bicarbonate, our data averaged 1.9& (±0.3) lower in d13C

Environmental controls on d13C shell of Arctica islandica

403

Fig. 7. Percent metabolic contribution (%CM) to shell material was calculated using three proxies for respired carbon – digestive gland, muscle and sediment – from juveniles and adults. The dashed line represents the 10% metabolic contribution hypothesized for aquatic mollusks using the respiratory gas exchange model (McConnaughey et al., 1997).

Fig. 8. The shell carbonate data from this study are plotted against average bicarbonate values from each growth period. Individual values, as well as averages from adults and juveniles are plotted relative to the experimentally determined relationship between inorganic aragonite and bicarbonate (Romanek et al., 1992). Data from Owen et al. (2008) are presented for comparison of biologically precipitated carbonate; however, in the case of the Owen et al. (2008) data, shell values are plotted against total DIC.

relative to equilibrium fractionation (Fig. 8). A table of individual isotopic and growth related data for the animals included in this study can be found in Electronic Annex EA-2. 4. DISCUSSION Minimum and maximum calculations of percent metabolic contributions averaged 8.7 and 10.0%, based on the most negative and positive carbon isotope values used to approximate respired carbon. These averages represent

175 carbonate samples from 114 individuals, including a wide range of ontogenetic ages grown in laboratory and in situ environments, under ambient and food-supplemented conditions. This range is in good agreement with several other studies that have calculated percent metabolic contribution to shell material of aquatic mollusks (McConnaughey et al., 1997; Kennedy et al., 2001; Lorrain et al., 2004; Gillikin et al., 2006). Additionally, the averages are very near the hypothesized 10% contribution for aquatic mollusks given by the respiratory gas exchange model (McConnaughey et al., 1997). Alternatively, many studies

404

E.C. Beirne et al. / Geochimica et Cosmochimica Acta 84 (2012) 395–409

of marine mollusks have documented significantly higher %CM (e.g. up to 38%, Gillikin et al., 2007), as well as variable %CM over the lifetime or shell region (i.e. lateral vs. ventral margins) of an individual (e.g. Klein et al., 1996; Lorrain et al., 2004; Gillikin et al., 2005; Butler et al., 2011). With regard to ontogenetic effects, researchers have noted both 13C enrichment (Klein et al., 1996; Gillikin et al., 2009) and depletion (Krantz et al., 1987; Kennedy et al., 2001; Elliot et al., 2003; Lorrain et al., 2004; Gillikin et al., 2007) with increasing biological age; however, no strong age-related trends in d13CS were observed in the individual A. islandica included in this study (Fig. 4c). 4.1. Kinetic and metabolic isotope effects Carbon isotope disequilibrium due to vital effects may be the result of either kinetic or metabolic isotope effects, or some combination thereof (summarized in Shanahan et al. (2005)). Under the kinetic model, fractionation associated with hydration and hydroxylation at the calcification site discriminates against the heavy isotope of carbon, leaving the calcifying fluids (extrapallial fluid, or EPF) depleted in 13C relative to the source DIC, which is a combination of environmental and metabolic sources (McConnaughey, 1989). The membrane separating the calcification fluid from the surrounding water is permeable to CO2, but impermeable to bicarbonate, HCO 3 (see Shanahan et al. (2005)). However, direct exchange with ambient fluids around the periostracum allows for the addition of HCO 3 from the surrounding water, which is deprotonated to produce carbonate ions for calcification (Vander Putten et al., 2000; McConnaughey and Gillikin, 2008). Additionally, the different forms of carbon in solution (i.e., CO2, HCO 3, CO2 ) have different isotopic signatures, which can compli3 cate computation of the metabolic contribution to shell carbonate depending upon which species is most abundant (Zhang et al., 1995). Over the range of pH and temperatures of ambient seawater observed during the experimental period (Fig. 2b and c), HCO 3 accounted for 92–95% of the dissolved inorganic species of carbon; therefore, d13CDIC will largely reflect d13 CHCO3 . Values of d13 CHCO3 were approximately 0.2& more positive than measured values of d13CDIC given average measurements of temperature and pH for each growth environment and growth period (see Electronic Annex EA-1). Several studies have noted that kinetic isotope effects appear to minimally affect marine mollusks (e.g. McConnaughey, 1989; Klein et al., 1996; Shanahan et al., 2005; Wanamaker et al., 2006, 2007; Gillikin et al., 2007). This is attributed to the activity of carbonic anhydrase (CA), an enzyme that catalyzes the reaction of bicarbonate to CO2, and thereby facilitates diffusion across the membrane separating internal fluids from EPF (Paneth and O’Leary, 1985). Kinetic isotope effects should influence oxygen and carbon simultaneously and are therefore evidenced by a strong, positive correlation between d18O and d13C. In the present study, there is a very weak correlation (r2 = 0.003) between these two parameters (see Electronic Annex EA3) indicating that kinetic isotope effects are not responsible

for the observed offset from equilibrium precipitation of calcium carbonate. Metabolic isotope effects are more commonly invoked to explain positive and negative departures of d13CS from equilibrium with ambient seawater. Metabolic isotope effects are the result of biological processes, such as respiration and photosynthesis, which alter the isotopic composition of CO2 and HCO 3 originating from the food source (Klein et al., 1996). For example, Klein et al. (1996) suggest that shell carbonate enrichment with varying position around the shell is a function of mantle metabolic activity, where higher rates of “metabolic pumping” are consistent with higher %CM and more negative d13CS values. The authors observe that mantle metabolic activity is inversely proportional to growth rate, and is lower at the ventral, rather than the lateral margin of marine bivalve Mytilus trossulus. Lorrain et al. (2004) also invoke metabolic isotope effects to explain a decrease in d13CS throughout ontogeny within several individual scallops (Pecten maximus). They hypothesize that early in ontogeny the ratio of respired to precipitated carbon is low, therefore the shell values will substantially reflect environmental or “external” DIC. Later in ontogeny, respired to precipitated carbon ratios are high. Therefore, d13CS becomes more negative as the proportion of 13C depleted metabolic carbon (“internal” DIC) incorporated into shell material increases. The respiratory gas exchange model proposed by McConnaughey et al. (1997) was introduced to explain the large observed difference in the contribution of metabolic isotope effects to air-breathing versus aquatic animals. As concerns aquatic mollusks, the authors acknowledge several assumptions inherent to the estimated value resulting from the respiratory gas exchange model they propose. These assumptions relate to environmental and physiological differences among the various species, including the role of oxygen demand, direct exchange between calcification fluids and ambient seawater (see McConnaughey and Gillikin (2008)), variable ratios of CO2/O2 (Gillikin et al., 2006), and ontogenetic effects. Nevertheless, the results of the present study indicate that the 10% metabolic contribution to shell carbonate predicted by the respiratory gas exchange model is a near approximation of the observed behavior of A. islandica. 4.2. Growth rate effects It is well established that A. islandica shell growth is more rapid during the early stages of shell formation and slows considerably after maturity (Jones, 1980). In the present study, growth rates are approximately an order of magnitude higher in juveniles than in adults. In the winter months (January–March), when temperatures did not exceed 6 °C, juveniles still grew an average of 0.21 mm/week. However, growth rate is not a primary control on the d13CS in either juveniles or adults (Fig. 4a and b). Fig. 4b illustrates that growth rates in juveniles are highest in the May–August growth period, yet d13CS is very similar to the January– March period. During the March–May period, d13CS correlates positively with increasing growth rate (r2 = 0.23;

Environmental controls on d13C shell of Arctica islandica

p < 0.012). This time interval roughly corresponds to the spring bloom in the Damariscotta River estuary (Townsend and Spinrad, 1986), during which there are higher than average rates of primary productivity. When rates of photosynthesis increase, isotopically light carbon is preferentially taken up by phytoplankton and thus, removed from the ambient water, leaving the water more enriched in 13C (Gruber et al., 1999). If more rapid calcification corresponds to a greater proportion of environmental DIC in shell carbonate, as suggested by Lorrain et al. (2004), the carbonate deposited during the spring bloom could conceivably reflect more positive d13C values relative to other times of year. However, only a slight increase in d13CDIC was observed in water samples collected at the end of March (Fig. 2d). This also coincides with a period of decreased salinity (Fig. 2a), which would act to lower d13CDIC, perhaps muting the effect of productivity on d13CDIC; however, over the limited range of observed salinity and DIC values, no systematic covariance between salinity and d13CDIC was observed during the experimental period. If calcification occurs preferentially during the hours of active photosynthesis, a single sample taken to represent a two-week d13CDIC window may be insufficient to capture a trend that has been incorporated in the shell carbonate. We suggest that productivity may be a second order control on d13CS for juvenile A. islandica during the spring bloom; however, more work is needed to establish the diurnal range of d13CDIC values during the spring bloom, collect information from independent proxies of productivity (e.g. chlorophyll or dissolved oxygen concentration), as well as to estimate the daily periods of calcification for A. islandica. Due to the preliminary evidence that productivity may modestly influence carbon isotopic composition of juvenile shell material during the spring bloom, caution must be exercised when attempting to reconstruct seasonal variation in d13CDIC using this species. Averaged over the eightmonth experimental period, there seems to be little effect from productivity on d13CS in juveniles or adults. A recent study conducted on A. islandica (Butler et al., 2011) suggests that discernable carbon isotopic trends in the biologically younger portions (0–40 years) of the shell are evidence that the species shows age-related vital effects on CM that may limit the use of ontogenetically younger portions of the shell. Notably, there appears to be no significant relationship between age and isotopic value after the first 40 years in the data provided by Butler et al. (2011). However, d13CDIC was unknown for the specimens analyzed, therefore true environmental variability could not be eliminated as a possible explanation for the observed variability in d13CS. Although the animals included in our study represent only 2–64 years biological age, the carbon isotope composition of shell carbonate from juveniles and adults averaged over the entire 29-week experimental period was identical, within analytical uncertainty, in spite of an order of magnitude difference in growth rate. 4.3. Ontogenetic effects The identical average d13CS values for adults and juveniles in this study (1.5& ± 0.3) averaged across growth

405

environment and growth period, imply that little to no ontogenetic effect exists for the specimens analyzed. Additionally, when d13CS values were compared to biological age (Fig. 4c), a very weak, but significant (r2 = 0.04, p = 0.03) trend towards depletion of 13C with increasing age was observed. This comparison is somewhat limited by the uneven distribution of biological ages of the animals studied. Our findings are supported by those of Strahl and Abele (2010) who noted low, but stable cell turnover and metabolic rates for A. islandica (ranging 7–148 years old), which points to insignificant physiological and metabolic differences between juveniles and adults of this species. Therefore, any difference in d13CS between juvenile and adult A. islandica must be explained by variable rates of calcification (see discussion of metabolic isotope effects in Section 4.1). However, many of the studies that show a strong, unidirectional trend in d13CS over the course of ontogeny have been conducted on short-lived bivalves (Klein et al., 1996; Lorrain et al., 2004; Gillikin et al., 2007, 2009), unlike A. islandica. 4.4. Values of respired carbon A variety of proxies for respired carbon have been suggested for marine mollusks. Gillikin et al. (2006) consider POC versus phytoplankton as the primary source of metabolic carbon to Mytilus edulis, concluding that d13Cphytoplankton is more accurately reflected in the d13C of the soft tissues and is thus, a better proxy for d13CR. The authors calculate d13Cphytoplankton by subtracting the average fractionation between phytoplankton and DIC (20&) from measured values of d13CDIC (after Fry (2002)). When the same transformation is applied to the average d13CDIC obtained during our 29-week experiment, the resulting estimation of d13Cphytoplankton is 19.5&. This value falls between our maximum and minimum approximations of d13CR and therefore would predict values of %CM within our calculated range. However, we are not aware of any selectivity estimates or gut content analyses specific to A. islandica that would justify the assumption that their diet is primarily composed of phytoplankton as opposed to POC, as in the case of Mytilus edulis (Gillikin et al., 2006). Future experimental work using A. islandica should include monthly d13CPOC sampling and selectivity determinations. Poulain et al. (2010) use digestive gland, mantle, muscle, phytoplankton and total soft tissue to calculate CM in an experiment where the food source is changed rapidly to a significantly depleted value (58&). When each proxy was evaluated, d13Cphytoplankton predicted the most consistent value of CM, and muscle tissues took the longest period of time to respond. In the present study, relatively little difference was noted between Tank A and Tank B carbon isotope composition of shell carbonate (Fig. 5) and tissues. Had there been a significant contribution of the shellfish diet (d13Cshellfish paste = 30.4&) to the tissues of the animals in Tank B, we would have likely seen more negative d13C values of tissues and potentially shell carbonate. Therefore, we conclude that the shellfish paste did not constitute a large part of the diet of the animals in Tank B

406

E.C. Beirne et al. / Geochimica et Cosmochimica Acta 84 (2012) 395–409

(either due to too small a ration or selection against the paste) and ignore it in the forthcoming calculations of %CM. d13CR represents assimilated carbon, therefore, as noted by Poulain et al. (2010), when integrating shell material over a long time period (e.g. annual averages), a longer turnover time tissue, such as muscle, may be the most appropriate proxy for respired carbon. If sampling shell carbonate on a finer scale (e.g. seasonal or diurnal resolution), carbon assimilated into shorter turnover time tissues (such as the gills or digestive gland; Strahl and Abele, 2010), or periodic POC samples may give the most accurate estimation of CM. In the present study, we were interested in bracketing the highest and lowest metabolic contribution by using the most enriched and depleted sources of assimilated carbon, the muscle and digestive gland, respectively. 4.5. Offset from predicted equilibrium We find that A. islandica shell aragonite is depleted in C by 1.9 ± 0.3& relative to anticipated equilibrium between calcium carbonate and ambient bicarbonate. Owen et al. (2008) found that for the larval stage of the scallop Placopecten magellanicus cultured at various DIC conditions, shell aragonite d13C values were lower than predicted equilibrium values by 1.8 ± 0.2&, in this case presuming that d13CDIC and d13 CHCO3 were roughly equivalent (Romanek et al., 1992; Fig. 8). The comparison between A. islandica and larval scallop carbonate data is strictly to illustrate the offset from predicted equilibrium observed in a range of experimentally cultured biogenic carbonates and is not meant to imply a similarity between the study organisms. Based on our findings, the equations relating d13CS to 13 d CDIC and d13 CHCO3 for A. islandica are:

reconstructions. By applying the experimentally determined offset between d13CS and d13CDIC for this species, existing carbon isotope data from previously published studies may be used to reconstruct annual and sub-annual d13CDIC time-series in a variety of extratropical marine environments. These reconstructions may constitute a significant expansion of the spatiotemporal network of paleo-DIC data. Based upon the findings of this study, caution should be exercised when attempting to reconstruct seasonal variation in d13CDIC using this species, due to the preliminary evidence that productivity may exert a secondary influence on the carbon isotopic composition of the shell during the spring bloom. Additionally, future work is needed to confirm the veracity of these conclusions in deep water environments where food is less abundant and growth rates are slower, in a variety of geographic regions, and across variable DIC and salinity conditions. ACKNOWLEDGEMENTS

13

d13 CDIC ¼ d13 CS  1:0&ð0:3Þ; and; 13

13

d CHCO3 ¼ d CS  0:8& ð0:3Þ

ð2Þ ð3Þ

This estimation is limited by the relatively uniform d13CDIC under which the experimental animals were cultured, as well as the artificially shallow and high-flow culture conditions. Nevertheless, we find a surprisingly consistent offset from 13C equilibrium despite the variety of age classes, growth environments, and growth periods sampled. 5. CONCLUSIONS This study demonstrates that A. islandica deposits its shell material with a consistent and measurable offset from carbon isotope equilibrium under the experimental conditions described above. The results indicate that the average percent metabolic contribution to A. islandica shell carbonate is roughly 10% based upon the three proxies used for this calculation. When considering analytical uncertainty, the 29-week averages of d13CS are equal between juveniles and adults, suggesting that vital effects are minimal for this species. The offset from predicted equilibrium, based upon careful environmental monitoring, remained relatively consistent across the ontogenetic ages, growth environments and seasons examined in this study. We therefore conclude that A. islandica constitutes a viable proxy for paleo-DIC

We are very grateful for the assistance of R. VanDenBroeke, M. Hosford, J. Feenstra, S. Griffin, E. Lower and A. Lane in the preparation and processing of tissue and shell samples. M. Mathison was essential in maintaining the quality performance of the IRMS and all laboratory equipment. We thank Rob Russell (Maine Department of Marine Resources) and the crew of the F.V. Three of a Kind for providing the opportunity and the means to collect study animals. The experimental design of this project was greatly improved by the input of B. Beal, D. Gillikin and A. Lorrain. We appreciate the help of T. Miller at the Darling Marine Center and A. Bradley at Harvard University. We are grateful for the use of imaging facilities at Bates College in Lewiston, Maine and the assistance of W. Locke and M. Duvall. Funding for this project was in part provided by Iowa State University to A.D.W., and by the National Science Foundation to A.D.W. (NSF OCE-1003438). We also thank E. Grossman, C. Lazareth and an anonymous reviewer for their insightful comments on this manuscript.

APPENDIX A. SUPPLEMENTARY DATA Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.gca.2012.01. 021. REFERENCES Black B. A., Gillespie D. C., MacLellan S. E. and Hand C. M. (2008) Establishing highly accurate production-age data using the tree-ring technique of crossdating: a case study for Pacific geoduck (Panopea abrupta). Can. J. Fish. Aquat. Sci. 65, 2572– 2578. Bo¨hm F., Joachimski M. M., Lehnert H., Morgenroth G., Kretschmer W., Vacelet J. and Dullo W. C. (1996) Carbon isotope records from extant Caribbean and South Pacific sponges: evolution of d13C in surface water DIC. Earth Planet. Sci. Lett. 139, 291–303. Bo¨hm F., Haase-Schramm A., Eisenhauer A., Dullo W. C., Joachimski M. M., Lehnert H. and Reitner J. (2002) Evidence for preindustrial variations in the marine surface water carbonate system from coralline sponges. Geochem. Geophys. Geosyst. 3(3). doi:10.1029/2001GC000264.

Environmental controls on d13C shell of Arctica islandica Bouillon S., Moens T. and Dehairs F. (2004) Carbon sources supporting benthic mineralization in mangrove and adjacent seagrass sediments (Gazi Bay, Kenya). Biogeosciences 1, 71–78. Butler P. G., Scourse J. D., Richardson C. A., Wanamaker, Jr., A. D., Bryant C. L. and Bennell J. D. (2009) Continuous marine radiocarbon reservoir calibration and the 13C Suess effect in the Irish Sea: results from the first multi-centennial shell-based marine master chronology. Earth Planet. Sci. Lett. 279, 230–241. Butler P. G., Richardson C. A., Scourse J. D., Wanamaker, Jr., A. D., Shammon T. M. and Bennell J. D. (2010) Marine climate in the Irish Sea: analysis of a 489-year marine master chronology derived from growth increments in the shell of the clam Arctica islandica. Quatern. Sci. Rev. 29, 1614–1632. Butler P. G., Wanamaker, Jr., A. D., Scourse J. D., Richardson C. A. and Reynolds D. R. (2011) The stability of shell d13C with respect to biological age in mature specimens of the long-lived bivalve Arctica islandica. Palaeogeogr. Palaeocl. 302. doi:10.1016/j.palaeo.2010.03.038. Cage A. G. and Austin W. E. N. (2010) Marine climate variability during the last millennium: the Loch Sunart record, Scotland, UK. Quatern. Sci. Rev. 29, 1633–1647. Dahlgren T. G., Weinberg J. R. and Halanych K. M. (2000) Phylogeography of the ocean quahog (Arctica islandica): influences of paleoclimate on genetic diversity and species range. Mar. Biol. 137, 487–495. Druffel E. R. M. and Benavides L. M. (1986) Input of excess CO2 to the surface ocean based on 13C/12C ratios in a banded Jamaican sclerosponge. Nature 321, 58–61. Elliot M., deMenocal P. B., Linsley B. K. and Howe S. S. (2003) Environmental controls on the stable isotopic composition of Mercenaria mercenaria: potential application to paleoenvironmental studies. Geochem. Geophys. Geosyst. 4. doi:10.1029/ 2002GC000425. Francey R. J., Allison C. E., Etheridge D. M., Trudinger C. M., Enting I. G., Leuenberger M., Langenfelds R. L., Michel E. and Steele L. P. (1999) A 1000-year high precision record of d13C in atmospheric CO2. Tellus 51B, 170–193. Friedli H., Lotscher H., Oeschger H., Siegenthaler U. and Stauffer B. (1986) Ice Core record of 13C/12C ratio of atmospheric CO2 in the past two centuries. Nature 324, 237–238. Fry B. (2002) Conservative mixing of stable isotopes across estuarine salinity gradients: a conceptual framework for monitoring watershed influences on downstream fisheries production. Estuaries 25, 264–271. Gillikin D. P., Lorrain A., Navez J., Taylor J. W., Keppens E., Baeyens W. and Dehairs F. (2005) Strong biological controls on Sr/Ca ratios in aragonitic marine bivalve shells. Geochem. Geophys. Geosyst. 6, Q05009. doi:10.1029/2004GC000874. Gillikin D. P., Lorrain A., Bouillon S., Willenz P. and Dehairs F. (2006) Stable carbon isotopic composition of Mytilus edulis shells: relation to metabolism, salinity, d13CDIC and phytoplankton. Org. Geochem. 37, 1371–1382. Gillikin D. P., Lorrain A., Meng L. and Dehairs F. (2007) A large metabolic carbon contribution to the d13C record in marine aragonitic bivalve shells. Geochim. Cosmochim. Acta 71, 2936– 2946. Gillikin D. P., Hutchinson K. A. and Kumai Y. (2009) Ontogenic increase of metabolic carbon in freshwater mussel shells (Pyganodon cataracta). J. Geophys. Res. Biogeo. 114, G01007, doi:10.1029/2008JG000829. Gruber N., Keeling C. D., Bacastow R. B., Guenther P. R., Lueker T. J., Wahlen M., Meijer H. A. J., Mook W. G. and Stocker T. F. (1999) Spatiotemporal patterns of carbon-13 in the global surface oceans and the oceanic Suess effect. Global Biogeochem. Cycles 13, 307–335.

407

Gruber N., Keeling C. D. and Bates N. R. (2002) Interannual variability in the North Atlantic Ocean carbon sink. Science 298, 2374–2378. IPCC (2007) Fourth Assessment Report (AR4), climate change 2007: the physical science basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge, United Kingdom and New York, NY, USA. 966pp. Jones D. S. (1980) Annual cycle of shell growth increment formation in two continental shelf bivalves and its paleoecologic significance. Paleobiology 6, 331–340. Keeling C. D., Mook W. G. and Tans P. P. (1979) Recent trends in the 13C/12C ratio of atmospheric carbon dioxide. Nature 277, 121–123. Keeling C. D., Piper S. C., Bacastow R. B., Wahlen M., Whorf T. P., Heimann M. and Meijer H. A. (2005) Atmospheric CO2 and 13 CO2 exchange with the terrestrial biosphere and oceans from 1978 to 2000: observations and carbon cycle implications. In A History of CO2 and Its Effects on Plants, Animals, and Ecosystems (eds. J. R. Ehleringer and M. D. Dearing). Springer, New York, pp. 83–113. Kennedy H., Richardson C. A., Duarte C. M. and Kennedy D. P. (2001) Oxygen and carbon stable isotopic profiles of the fan mussel, Pinna nobilis, and reconstruction of sea surface temperatures in the Mediterranean. Mar. Biol. 139, 1115–1124. Klein R. T., Lohmann K. C. and Thayer C. W. (1996) Sr/Ca and 13 C/12C ratios in skeletal calcite of Mytilus trossulus: covariation with metabolic rate, salinity, and carbon isotopic composition of seawater. Geochim. Cosmochim. Acta 60, 4207–4221. Krantz D. E., Williams D. F. and Jones D. S. (1987) Ecological and paleoenvironmental information using stable isotopic profiles from living and fossil molluscs. Palaeogeogr. Palaeocl. 58, 249– 266. Lartaud F., Emmanuel L., de Rafelis M., Pouvreau S. and Renard M. (2010) Influence of food supply on the d13C signature of mollusc shells: implications for palaeoenvironmental reconstitutions. Geo-Mar. Lett. 30, 23–34. Lorrain A., Savoye N., Chauvaud L., Paulet Y. M. and Naulet N. (2003) Decarbonation and preservation method for the analysis of organic C and N contents and stable isotope ratios of lowcarbonated suspended particulate material. Anal. Chim. Acta 491, 125–133. Lorrain A., Paulet Y.-M., Chauvaud L., Dunbar R., Mucciarone D. and Fontugne M. (2004) D13C variation in scallop shells: increasing metabolic carbon contribution with body size?. Geochim. Cosmochim. Acta 68 3509–3519. Marchitto T. M., Jones G. A., Goodfriend G. A. and Weidman C. R. (2000) Precise temporal correlation of Holocene mollusk shells using sclerochronology. Quatern. Res. 53, 236–246. McConnaughey T. (1989) 13C and 18O isotopic disequilibrium in biological carbonates: I. Patterns. Geochim. Cosmochim. Acta 53, 151–162. McConnaughey T. A. and Gillikin D. P. (2008) Carbon isotopes in mollusk shell carbonates. Geo-Mar. Lett. 28, 287–299. McConnaughey T. A., Burdett J., Whelan J. F. and Paull C. K. (1997) Carbon isotopes in biological carbonates: respiration and photosynthesis. Geochim. Cosmochim. Acta 61, 611–622. Moran A. L. and Marko P. B. (2005) A simple technique for physical marking of larvae of marine bivalves. J. Shellfish Res. 24, 567–571. Nozaki Y., Rye D. M., Turekian K. K. and Dodge R. E. (1978) 200-year record of 13C and 14C variations in a Bermuda coral. Geophys. Res. Lett. 5, 825–828. Owen E. F., Wanamaker, Jr., A. D., Feindel S. C., Scho¨ne B. R. and Rawson P. D. (2008) Stable carbon and oxygen isotope

408

E.C. Beirne et al. / Geochimica et Cosmochimica Acta 84 (2012) 395–409

fractionation in bivalve (Placopecten magellanicus) larval aragonite. Geochim. Cosmochim. Acta 72, 4687–4698. Paneth P. and O’Leary M. H. (1985) Carbon isotope effect on dehydration of bicarbonate ion catalyzed by carbonic-anhydrase. Biochemistry-US 24, 5143–5147. Poulain C., Lorrain A., Mas R., Gillikin D. P., Dehairs F., Robert R. and Paulet Y. M. (2010) Experimental shift of diet and DIC stable carbon isotopes: influence on shell d13C values in the Manila clam Ruditapes philippinarum. Chem. Geol. 272, 75–82. Quay P., Sonnerup R., Westby T., Stutsman J. and McNichol A. (2003) Changes in the 13C/12C of dissolved inorganic carbon in the ocean as a tracer of anthropogenic CO2 uptake. Global Biogeochem. Cycles. 17, 1045–1061. Quay P., Sonnerup R., Stutsman J., Maurer J., Kortzinger A., Padin X. A. and Robinson C. (2007) Anthropogenic CO2 accumulation rates in the North Atlantic Ocean from changes in the 13C/12C of dissolved inorganic carbon. Global Biogeochem. Cycles. 21, GB1009, doi:10.1029/2006GB002761. Riascos J., Guzman N., Laudien J., Heilmayer O. and Oliva M. (2007) Suitability of three stains to mark shells of Concholepas concholepas (Gastropoda) and Mesodesma donacium (Bivalvia). J. Shellfish Res. 26, 43–49. Romanek C. S., Grossman E. L. and Morse J. W. (1992) Carbon isotopic fractionation in synthetic aragonite and calcite: effects of temperature and precipitation rate. Geochim. Cosmochim. Acta 56, 419–430. Ropes J. W. (1984) Procedures for preparing acetate peels and evidence validating the annual periodicity of growth lines formed in the shells of ocean quahogs, Arctica islandica. Mar. Fish. Rev. 46, 27–35. Rubinson M. and Clayton R. N. (1969) Carbon-13 fractionation between aragonite and calcite. Geochim. Cosmochim. Acta 33, 997–1002. Sabine C. L., Feely R. A., Gruber N., Key R. M., Lee K. and Bullister J. L., et al. (2004) The oceanic sink for anthropogenic CO2. Science 305, 367–371. Scho¨ne B. R., Oschmann W., Rossler J., Castro A. D. F., Houk S. D., Kroncke I., Dreyer W., Janssen R., Rumohr H. and Dunca E. (2003) North Atlantic Oscillation dynamics recorded in shells of a long-lived bivalve mollusk. Geology 31, 1037–1040. Scho¨ne B. R., Fiebig J., Pfeiffer M., Gleb R., Hickson J., Johnson A., Dreyer W. and Oschmann W. (2005) Climate records from a bivalve Methuselah (Arctica islandica, Mollusca; Iceland). Palaeogeogr. Palaeocl. 228, 130–148. Scho¨ne B. R., Wanamaker, Jr., A. D., Fiebig J., The´bault J. and Kreutz K. (2011) Annually resolved d13Cshell chronologies of long-lived bivalve mollusks (Arctica islandica) reveal oceanic carbon dynamics in the temperate North Atlantic during recent centuries. Palaeogeogr. Palaeocl. 302. doi:10.1016/ j.palaeo.2010.02.002. Scourse J. D., Richardson C., Forsythe G., Harris I., Heinemeier J., Fraser N. M., Briffa K. and Jones P. (2006) First crossmatched floating chronology from the marine fossil record: data from growth lines of the long-lived bivalve mollusc Arctica islandica. Holocene 16, 967–974. Shanahan T. M., Pigati J. S., Dettman D. L. and Quade J. (2005) Isotopic variability in the aragonite shells of freshwater gastropods living in springs with nearly constant temperature and isotopic composition. Geochim. Cosmochim. Acta 69, 3946– 3966. Stumm W. and Morgan J. J. (1996) Aquatic Chemistry – Chemical Equilibria and Rates in Natural Waters. John Wiley & Sons, Inc., New York. Stott K. J., Austin W. E. N., Sayer M. D. J., Weidman C. R., Cage A. G. and Wilson R. J. S. (2010) The potential of Arctica

islandica growth records to reconstruct coastal climate in north west Scotland, UK. Quatern. Sci. Rev. 29, 1602–1613. Strahl J. and Abele D. (2010) Cell turnover in tissues of the longlived ocean quahog Arctica islandica and the short-lived scallop Aequipecten opercularis. Mar. Biol. 157, 1283–1292. Suess H. E. (1953) Natural radiocarbon and the rate of exchange of carbon dioxide between the atmosphere and the sea. In Nuclear Processes in Geologic Settings (ed. W. Aldrich). University of Chicago Press, Chicago, pp. 52–56. Swart P. K., Greer L., Rosenheim B. E., Moses C. S., Waite A. J., Winter A., Dodge R. E. and Helmle K. (2010) The 13C Suess effect in scleractinian corals mirror changes in the anthropogenic CO2 inventory of the surface oceans. Geophys. Res. Lett. 37, L05604. doi:10.1029/2009GL041397. Taipale S. J. and Sonninen E. (2009) The influence of preservation method and time on the d13C value of dissolved inorganic carbon in water samples. Rapid Commun. Mass. Sp. 23, 2507– 2510. Townsend D. W. and Spinrad R. W. (1986) Early Spring phytoplankton blooms in the Gulf of Maine. Cont. Shelf Res. 6, 515–529. Vander Putten E., Dehairs F., Keppens E. and Baeyens W. (2000) High resolution distribution of trace elements in the calcite shell layer of modern Mytilus edulis: environmental and biological controls. Geochim. Cosmochim. Acta 64, 997–1011. Wanamaker, Jr., A. D., Kreutz K. J., Borns, Jr., H. W. and Introne D. S. (2006) An aquaculture-based method for calibrated bivalve isotope paleothermometry. Geochem. Geophys. Geosys. 7, Q09011, 1–13. doi:10.1029/2005GC001189. Wanamaker, Jr., A. D., Kreutz K. J., Borns H. W., Introne D. S., Feindel S., Funder S., Rawson P. D. and Barber B. J. (2007) Experimental determination of salinity, temperature, growth, and metabolic effects on shell isotope chemistry of Mytilus edulis collected from Maine and Greenland. Paleoceanography 22, PA2217, doi:10.1029/2006PA001352. Wanamaker, Jr., A. D., Kreutz K. J., Scho¨ne B. R., Pettigrew N. R., Borns, Jr., H. W., Introne D. S., Belknap D., Maasch K. A. and Feindel S. (2008a) Coupled North Atlantic Slope Water forcing on Gulf of Maine temperatures over the past millennium. Clim. Dynam. 31, 183–191. Wanamaker, Jr., A. D., Heinemeier J., Scourse J. D., Richardson C. A., Butler P. G., Eiriksson J. and Knudsen K. L. (2008b) Very long-lived mollusks confirm 17th century AD tephrabased radiocarbon reservoir ages for north icelandic shelf waters. Radiocarbon 50, 399–412. Wanamaker, Jr., A. D., Kreutz K. J., Scho¨ne B. R. and Introne D. S. (2011a) Gulf of Maine shells reveal changes in seawater temperature seasonality during the Medieval Climate Anomaly and the Little Ice Age. Palaeogeogr. Palaeocl. 302, 43–51. Wanamaker, Jr., A. D., Hetzinger S. and Halfar J. (2011b) Reconstructing mid- to high-latitude marine climate and ocean variability using bivalves, coralline algae, and marine sediment cores from the northern hemisphere. Palaeogeogr. Palaeocl. 302, 1–9. Weidman C. R. and Jones G. A. (1993) A shell-derived time history of bomb 14C on Georges Bank and its Labrador Sea implications. J. Geophys. Res. Oceans 98, 14577–14588. Williams B., Halfar J., Steneck R. S., Wortmann U. G., Hetzinger S., Adey W., Lebednik P. and Joachimski M. (2011) Twentieth century d13C variability in surface water dissolved inorganic carbon recorded by coralline algae in the northern North Pacific Ocean and the Bering Sea. Biogeosciences Discuss. 7, 5801–5828. Witbaard R., Duineveld G. C. A. and DeWilde P. A. W. J. (1997) A long-term growth record derived from Arctica islandica

Environmental controls on d13C shell of Arctica islandica (Mollusca, Bivalvia) from the Fladen Ground (northern North Sea). J. Mar. Biol. Assoc. UK 77, 801–816. Witbaard R., Jansma E. and Klaassen U. S. (2003) Copepods link quahog growth to climate. J. Sea Res. 50, 77–83. Zeebe R. E. and Wolf-Gladrow D. (2005) CO2 in Seawater: Equilibrium, Kinetics, Isotopes, 3rd ed. Elsevier, Boston.

409

Zhang J., Quay P. D. and Wilbur D. O. (1995) Carbon-isotope fractionation during gas–water exchange and dissolution of CO2. Geochim. Cosmochim. Acta 59, 107–114. Associate editor: Juske Horita