Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary

Quaternary Geochronology xxx (2017) 1e15

Contents lists available at ScienceDirect

Quaternary Geochronology journal homepage: www.elsevier.com/locate/quageo

Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary Brittany L. Grimm a, *, Howard J. Spero a, Juliana M. Harding b, Thomas P. Guilderson c, d a

Department of Earth and Planetary Sciences, University of California, Davis, CA 95616, USA Department of Marine Science, Coastal Carolina University, Conway, SC 29528, USA c Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory, Livermore, CA 94550, USA d Department of Ocean Sciences, University of California, Santa Cruz, CA 95064, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 August 2016 Received in revised form 20 February 2017 Accepted 22 March 2017 Available online xxx

This study utilizes a combined stable isotope and 14C dating approach to determine the radiocarbon reservoir age correction, DR, for the James River, Virginia estuary from 17th century Crassostrea virginica shells of known collection dates. DR, which can vary spatially and temporally, is a locality-specific adjustment applied to the global ocean reservoir, R, to further account for the offset between the atmospheric and marine 14C calibration curves. To assess the temporal variability in DR, continuous d18O sampling along the oyster shell hinge provides a seasonal record throughout the oyster's life. This is then used to identify sampling locations for 14C measurements based on calcite precipitated during the Summer (>19
C) and Fall through Spring (F-Sp, <15
C) months. The resulting seasonal DR values range from 151 ± 46 to þ109 ± 55 14C years (260 years) due to changes in the contribution and age of dissolved inorganic carbon (DIC) from marine and freshwater sources in the James River estuary. The F-Sp samples display a larger DR range than the Summer samples, as do the shells precipitated during drought conditions (1606e1612) when compared to shells from the remainder of the 17th century. The largest intrashell DR variability, 195 14C years, is similarly found in a drought shell and is attributed to variability caused by the extreme regional 1606e1612 drought. Early land use changes related to European development and farming practices also altered the age of DIC in the James River estuary. We estimate that the soil inorganic carbon (SIC) contributing to freshwater DIC ranged from 0 to ~1800 years old and reflected both the drought and land use changes that occurred during the 17th century. Using only the Summer samples, which represent the majority of shell calcite, we obtain a mean DR ¼ 32 ± 11 14C years (1s) for 17th century James River estuary DR at the very onset of European colonization. Employing a seasonally resolved sampling method will provide the greatest constraint on 14C measurements in an estuarine environment where multiple carbon sources can fluctuate on seasonal timescales and as a result of large scale environmental change. © 2017 Elsevier B.V. All rights reserved.

Keywords: Radiocarbon Reservoir age (DR) Sclerochronology Crassostrea virginica Dissolved inorganic carbon (DIC) Estuary

1. Introduction Radiocarbon (14C) is one of the most important dating tools used by geologists and archaeologists to reconstruct the timing of changes in Earth and human history over the past 50,000 years. 14C can be used in the traditional sense of assigning a date to a specific event, but it is also used as a tracer for processes such as ocean

* Corresponding author. E-mail addresses: [email protected] (B.L. Grimm), hjspero@ucdavis. edu (H.J. Spero), [email protected] (J.M. Harding), [email protected] (T.P. Guilderson).

circulation, changes in air-sea equilibration, and upwelling (Druffel et al., 2004; Andrus et al., 2005; Ferguson et al., 2013), as well as carbon flow through terrestrial ecosystems (Gaudinski et al., 2000; Trumbore, 2000; Raymond et al., 2004). As a result, extensive research has been devoted to refining the 14C technique, including intercalibrating radiocarbon with absolute geochronometers such as U/Th (Bard et al., 1990; Fairbanks et al., 2005; Reimer et al., 2013), constraining carbon source variations in 14C from different environments (Hogg et al., 1998; Ascough et al., 2010; Rick et al., 2012), and selection of the most appropriate datable material (Ingram and Southon, 1996; Hogg et al., 1998; Kennett et al., 2002; Ascough et al., 2005b).

http://dx.doi.org/10.1016/j.quageo.2017.03.002 1871-1014/© 2017 Elsevier B.V. All rights reserved.

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

2

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

Interpretations of 14C data from identifiable terrestrial fossils are generally straightforward because photosynthesis incorporates 14 CO2 directly from the atmosphere (Anderson et al., 1947). Because the residence time of atmospheric CO2 is ~4e10 years (Craig, 1957; Levin and Hesshaimer, 2000), the initial 14C content of terrestrial material is closely linked to the atmospheric 14C/12C ratio. Dating organic carbon from plants and animals in aquatic environments is more complicated due to a mixing of carbon pools with different ages. Lakes, estuaries, and oceans each contain dissolved inorganic carbon (DIC) with 14C content that is often lower than the atmosphere. The difference between the atmospheric 14C/12C ratio and the co-occurring 14C/12C content of a carbon pool is known as the radiocarbon reservoir effect (Stuiver and Polach, 1977; Stuiver and Braziunas, 1993). The magnitude and sign of the 14C reservoir effect varies depending on environment. Freshwater lake samples often appear older than their true calendar age because the DIC pool contains a fraction of carbonate-derived ‘dead’ carbon (carbon that lacks 14C because its age exceeds ~50 kyr) from groundwater that interacts with surrounding bedrock (Deevey et al., 1954; Broecker and Walton, 1959). There may also be slow atmosphere-water mixing rates resulting from stratification or ice cover (Abbott and Stafford, 1996). Oceanic environments are also depleted in 14C with respect to the atmosphere because the deep oceanic carbon reservoirs require millennia to circulate before interacting with the atmosphere again (Broecker and Peng, 1982). The oceanic 14C surface reservoir effect is referred to as the global ocean reservoir, R, and is considered to have an average age of ~400 years (Stuiver and Braziunas, 1993). Processes such as changes in ocean circulation rate, upwelling, or mixing of water masses can cause regional offsets from 400 years (Goodfriend and Flessa, 1997; Kennett et al., 1997; Ferguson et al., 2013), thereby necessitating the use of locality-specific reservoir age corrections, referred to as DR (Stuiver and Braziunas, 1993), to compute radiocarbon ages. The magnitude of DR varies spatially and temporally and must be established to compute a more accurate age of marine sample material (see review by Ascough et al., 2005a). In estuaries, river and seawater DIC mix to produce a final inorganic carbon pool that is a function of the DIC concentration and radiocarbon age of each component (Ingram and Southon, 1996; Ulm, 2002; Culleton et al., 2006; Russell et al., 2010; Lougheed et al., 2016). Sources of 14C in freshwater rivers, ranging in age from modern to radiocarbon-dead, include atmospheric 14C, inorganic carbon derived from remineralized soil organic carbon, and carbon from the interaction of water with bedrock (Hope et al., 1994; Raymond et al., 1997, 2000, 2004; Hossler and Bauer, 2012, 2013). These latter two comprise the soil inorganic carbon (SIC) component of the freshwater DIC pool. Within an estuary, the relative proportions of marine and freshwater-derived 14C will change depending on the amount of freshwater input and marine incursion, which can be calculated if salinity is known. Attempts have been made to deconvolve the marine and estuarine inorganic carbon system to determine region specific DR. These include combining 14C data from paired terrestrial and aquatic material (Kennett et al., 1997; Ingram, 1998; Ascough et al., 2005b), quantifying 14C and U/Th ages from the same samples (Bard et al., 1990; Yu et al., 2010), or dating marine organisms such as mollusks from archaeological sites of known calendar age (Colman et al., 2002; Culleton et al., 2006; Rick et al., 2012; Rick and Henkes, 2014). Application of paired terrestrial and aquatic materials allow for the direct determination of DR if contemporaneity can be assumed. Unfortunately, the coeval nature of such samples is often difficult to confidently confirm in paired marine and terrestrial deposits. The combined use of U/Th dating and 14C measurements to establish DR introduces its own set of assumptions and

uncertainties (Edwards et al., 1987). Historically dated archaeological sites allow the direct comparison of estuarine 14C measurements to specific time periods, thereby reducing the uncertainties when computing DR for a defined watershed. Accuracy in determining the radiocarbon reservoir ages applied to estuarine mollusks (e.g., Eastern oyster, Crassostrea virginica) is critical for archaeological studies of early North American sites that predate the historical record. Of particular importance in defining DR are issues related to monthly, seasonal, and interannual changes in the estuarine carbon system which can vary with decadal changes in regional precipitation patterns and/or watershed landuse modifications (Ingram and Southon, 1996; Brush, 2001; Raymond and Bauer, 2001b). A suite of archaeological sites available from Virginia's James River estuary are uniquely suited to quantify variations in DR from 17th century estuarine waters because the burial date of their C. virginica oyster shells can be established from the historical record (Fig. 1, Table 1). These sites include three Jamestown Fort era wells (1607e1624, Kelso and Straube, 2004; Hudgins et al., 2008; Schmidt and Straube, 2012) and trash pits from Nansemond Pallizado (1636e1646; Luccketti, 2010; Pecoraro, 2015) and Bacon's Castle (1676; Luccketti, 1990; Dean, 2005). The sealed wells and trash pits were used because the archaeology supports material deposition during unique time periods of week(s) to about three months (Luccketti, 1990; Kelso and Straube, 2004; Hudgins et al., 2008; Schmidt and Straube, 2012; Pecoraro, 2015). These sites are representative of the James River geographic range of oyster habitat (Haven and Fritz, 1985; Harding et al., 2010). Previous radiocarbon studies of mollusks from freshwater and marine environments recommend homogenizing shell CaCO3 for 14 C analyses to smooth short term (tidal, days, or weeks; e.g. physiological processes) and sub-annual (months; e.g. seasonal upwelling or productivity) variability that may be present (Hogg et al., 1998; Culleton et al., 2006). However, this variability can impart significant uncertainty to an age determination because 14C content in a system may differ seasonally (Druffel et al., 2004;

Fig. 1. Map of (a) Chesapeake Bay and (b) James River estuary showing the location of archaeological oyster and water collection localities used in this study (modified from Harding et al., 2010). Shaded areas of the river indicate the fresh-salt water transition zone today (downriver shading) and during the drought that spanned 1606e1612 (upriver shading). Deep Water Shoal (site E) is the modern limit for oyster survival based on their minimum salinity tolerance of 5e7 psu (Shumway, 1996; Mann et al., 2009).

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

3

Table 1 Archaeological sites containing oyster shells used in this study and their associated temporal period. Period

Shell ID

Site locality

Archaeological site ID

Coordinates

Estimated years preserved

Contact/Drought (C): 1606-1612

C-1 C-2 C-3 C-4 E-1 E-2 M-1 L-1 L-2

Jamestown 1609 Well

Structure 185 JR 2718 W layera

37
120 N 76
460 W

1603e10

Jamestown 1611 Well Jamestown 1617 Well

Structure 177 JR 2158 Z layera,b Structure 170 JR 1101 F layera

37
120 N 76
460 W 37
120 N 76
460 W

1609e11 1621e24

Nansemond Pallizado trash pit Bacon's Castle trash pit

44SK-192-5c 44SY117-274c

36
520 N 76
300 W 37
060 N 76
430 W

1634e40 Sep.eDec. 1676

Contact/Drought (C): 1606-1612 Early development (E): 1612-1624 Mid-century (M): 1635-1650 Late century (L): 1676 a b c

Samples provided by W. Kelso and B. Straube, Jamestown Rediscovery Project, Preservation Virginia. Oxygen isotope data from shell C-4 were published in Harding et al. (2010) as JTZ-1.2. Samples provided by N. Luccketti, James River Institute for Archaeology.

Andrus et al., 2005; Ferguson et al., 2013) and selecting shell samples for 14C analyses is often a random process. In this study, we present oxygen and carbon isotope (d18O and d13C) and radiocarbon data from the hinges of C. virginica shells collected from 17th century James River estuary archaeological sites. The sites, including the 1607 Jamestown Colony, have reliable collection dates of ±5 years. Samples for 14C analyses were chosen based on seasonal d18O profiles, thereby allowing us to compute seasonal James River estuary variations in DR. These data are then used to model the contribution of the different inorganic carbon sources to the estuary to better constrain SIC ages from the surrounding forested watershed prior to Western European land-use changes. 2. 17th century James River and C. virginica geochemistry C. virginica shells are abundant artifacts in most East Coast midAtlantic archaeological sites (Galtsoff, 1964; Stenzel, 1971). The 17th century oysters used in this study are divided into four temporal periods defined as contact/drought (1606e1612), early development (1612e1624), mid-century (1635e1650), and late century (1676, Table 1). These four periods each encompass large changes within the James River watershed beginning with the initial European contact and concurrent six-year drought spanning 1606e1612 (Stahle et al., 1998; Cronin et al., 2000; Brush, 2001; Kelso, 2006). The native Algonquins populated the area prior to the colonists’ appearance (Rountree and Turner, 2002), but introduction of tobacco in 1612 led to rapid land clearing and development below the fall line in the James River watershed (Hatch, 1957; Miller, 1986; Wennersten, 2001). This land use change intensified through the mid- to late century as plantations and settlements were established along the length of the James River and in other nearby estuaries (Hatch, 1957; Wennersten, 2001). The oysters, which require salinities of at least 5e7 psu (practical salinity unit; Shumway, 1996), were likely collected from the mesohaline (9e23 psu) region of the estuary extending from Jamestown Island through Middle Ground (Fig. 1). Because the hinge contains a complete life history record of an oyster's post-larval benthic life (Rhoads and Lutz, 1980; Jones, 1983), researchers have reconstructed ambient environmental conditions through analyses of hinge oxygen (d18O) and carbon (d13C) isotope profiles (e.g., Kirby et al., 1998; Surge et al., 2001; Harding et al., 2010). C. virginica d18Ocalcite is primarily influenced by temperature and d18Owater, with the latter parameter covarying with salinity (Epstein et al., 1953; Keith et al., 1964). During calcification, the oxygen isotope ratio (18O/16O) of the CaCO3 varies in a predictable manner via a temperature-dependent fractionation between the mineral and water. This fractionation, in which a temperature increase causes d18Ocalcite to decrease with a slope

of ~ 0.22‰
C1, is the foundation for d18Ocalcite thermometry that is used to reconstruct past environmental temperatures from calcium carbonate shells (Epstein et al., 1953). The James River estuary is a partially mixed coastal plain estuary (Pritchard, 1952) in which d18Owater, the second controlling factor of d18Ocalcite, is a function of mixing between marine and freshwater sources. Whereas near shore Atlantic Ocean seawater d18O is relatively constant with a value ~ 1‰ (VSMOW; Schmidt et al., 1999), the d18O of James River freshwater (S ¼ 0 psu) is ~ 7‰ (Harding et al., 2010). Hence, estuarine d18Owater covaries with salinity with a slope of ~0.25‰ psu1 (Harding et al., 2010). The interplay between seasonal changes in salinity and temperature leads to cyclical variations in shell d18O. Salinity varies seasonally with river discharge as influenced by watershed precipitation. Typically, higher rainfall from March through May leads to lower salinities with increasing salinities observed from June through January corresponding to the seasonal dry period (Nichols, 1972; Brooks and Fang, 1983; Mann et al., 2009). Temperature also varies seasonally with a low in January, increase at the end of February, peak in August, and a decrease thereafter (Mann et al., 2009). While not perfectly matched, the combined swings in salinity and temperature lead to an amplification of the d18O signal as both factors contribute to either an increase or decrease in shell d18O. This is observed in the oyster shell as a d18O decrease during spring followed by a summer d18O minimum plateau that then leads to a d18O increase and maximum during the fall and winter (Harding et al., 2010). Shell d13C in C. virginica is a function of the d13C of the DIC in the river as well as the contribution of respired CO2 to the shell (Surge et al., 2001; McConnaughey and Gillikin, 2008). James River freshwater d13CDIC is a function of the proportional contributions of atmospheric CO2 and SIC that flows into the river with ground water seepage (Raymond et al., 2000, 2004; Hossler and Bauer, 2012, 2013). Importantly, in the tidally-influenced mesohaline region of the James River estuary, d13CDIC will reflect mixing between 13 C-enriched marine d13CDIC and 13C-depleted freshwater d13CDIC sources in proportion to their isotopic ratio and DIC concentration (Spiker, 1980; Ingram et al., 1996). This relationship between DIC sources suggests that d13CDIC should also covary with salinity (Bratton et al., 2003; Cronin et al., 2005). The use of shell d13C as a salinity proxy is complicated by the fact that bivalves incorporate metabolic carbon into their shell during calcification (Tanaka et al., 1986; McConnaughey and Gillikin, 2008). Metabolic carbon can contribute highly depleted carbon isotope values to the shell (d13C ¼ ~ 28‰) for rivers in areas dominated by C3 vegetation, potentially obscuring the d13CDIC signal (Mook and Tan, 1991). McConnaughey et al. (1997) found that shells contain 10% metabolic CO2, although Gillikin et al. (2007) reported 37% metabolic carbon incorporation in clam

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

4

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

shells. Surge et al. (2001) observed that Floridian C. virginica precipitate their shell near isotopic equilibrium with estuarine d13CDIC, except during cold months when the shell is offset from d13CDIC by þ1‰. An ontogenetic trend in increasing metabolic carbon contribution to the shell has also been observed in scallops and clams (Lorrain et al., 2004; Gillikin et al., 2007). This potential physiological complication must be considered when interpreting salinity vs. d13CDIC relationships derived from C. virginica, particularly at different locations where temperature and oyster physiology may vary during the same calendar months.

through a series of dry ice traps in a vacuum extraction line and collected in 6 mm Pyrex tubes at liquid nitrogen temperatures. After pumping away non-condensable gases, the tubes were flame sealed. CO2 was then analyzed for d13C using a Fisons Optima IRMS and are reported relative to VPDB. Analytical precision for d13CDIC is 0.08‰ (±1s) based on 20 repeat analyses of an in-house DIC standard. 3.3. Salinity and temperature calculations using shell d18O

3. Materials and methods

We use the salinity (S) vs. d18Owater relationship for the James River, VA described by Harding et al. (2010):

3.1. Stable isotope sampling


. 18 S ðWreck ShoalÞ ¼ d Owater þ 7:08 0:25

C. virginica shells were selected from five sealed archaeological features for analysis (Table 1). Shells with an intact left valve with a straight hinge (the ligamental attachment area) were chosen as per Harding et al. (2010) as they preserve the entire ontogenetic history of the oyster (Carriker, 1996; Kirby et al., 1998). The hinge was separated from the rest of the valve just below the resilifer (ligament attachment), embedded in epoxy, and sectioned parallel to the maximum growth axis. The resulting thick section was then polished and microsampled for geochemical analysis using an automated Merchantek/New Wave automated microdrilling system. Shell samples were drilled along a hinge transect following the direction of shell growth (perpendicular to growth signatures) with a spatial resolution of ~250 mm. Sample drill holes were 50e100 mm deep with distance between each sample hole ranging from 40 to 100 mm. The hinge exposed for sampling contains dense, foliated calcite and, in most cases, chalky layers composed of porous calcite (Galtsoff, 1964; Carriker, 1996; Surge et al., 2001). While Surge et al. (2001) observed little geochemical difference between the two types of calcite, we only sampled the foliated calcite regions of the shell and avoided the chalky calcite layers as recommended by Kirby et al. (1998). Drilled samples yielded 15e60 mg of calcite powder. Prior to stable isotope analyses, sample powder was roasted in vacuo for 30 min at 450
C to remove water. d18O and d13C values were obtained using a Fisons Optima isotope ratio mass spectrometer (IRMS) housed in the Stable Isotope Laboratory (SIL) in the Department of Earth & Planetary Sciences at the University of California, Davis. Calcite samples were reacted at 90
C in a common acid bath autocarbonate device containing 8 mL of 105% orthophosphoric acid. Oxygen and carbon isotope data for calcite are reported relative to the Vienna Pee Dee Belemnite (VPDB) standard. All isotope data are presented in per mil (‰) units using delta notation where d ¼ [(Rsample/Rstandard)e1] x 1000. Analytical precision for calcite is 0.05 and 0.06‰ (±1s) for d13C and d18O, respectively, based on 140 analyses of an in-house Carrara marble standard interspersed among analytical sessions. 3.2. James River d13CDIC Water samples from Wreck Shoal, Deep Water Shoal, and Fort Monroe in the James River were collected between February and September of 2010 and analyzed for their d13CDIC (Fig. 1, Table S1). At the time of collection the water was filtered, poisoned with HgCl2 to eliminate metabolic activity, and sealed in glass vials with no air space. All samples were collected within 1 m of the substrate and encompass a salinity gradient of 7.1e25.1 psu (±0.1 psu) based on measurements made with a YSI 85 handheld conductivity probe at the time of collection. For d13CDIC analyses, 5 mL of water was injected into a Pyrex reaction vessel containing ~1 mL of 105% H3PO4 in vacuo. The resulting CO2 was cryogenically purified

(1)

to calculate salinity from d18Owater (see Eq. (3) below). Eq. (1) was derived from water samples collected at Wreck Shoal, downriver of Jamestown Island, during 2006 and 2007. Water data (d18Owater) are presented relative to the Vienna Standard Mean Ocean Water (VSMOW). Whereas water temperatures at Wreck Shoal range between <4
C and 30
C (Mann et al., 2009), C. virginica shell growth in the James River is limited to water temperatures between 8 and 25
C (Mann and Evans, 2004; Harding et al., 2010). Therefore, we make the assumption here that river temperatures outside the C. virginica growth range will produce a geochemical hiatus in late summer (JulyeAugust, >25
C) and winter (mid-December through midMarch, <8
C). We follow the method of Harding et al. (2010) and assume the absolute d18O minima and maxima recorded during every season in the shell reflect shell calcite that was precipitated at these temperature limits. Mann et al. (2009) showed that temperatures in the James River between 1993 and 2006 surpassed these limits every year. Constraining the water temperature during oxygen isotope extremes allows us to invert the T vs. d18O relationship of Epstein et al. (1953):



 T ¼ 16:5  4:30 d18 Ocalc  d18 Owater  0:20
2
þ 0:14* d18 Ocalc  d18 Owater  0:20

(2)

and solve for d18Owater (Eq. (3)) at the minimum and maximum d18Ocalcite values by assuming growth temperatures of 25
C and 8
C, respectively:




d18 Owater ¼ d18 Ocalcite þ 0:20  4:30  ð18:49  0:56  ð16:5  TÞÞ0:5

. ð0:28Þ

(3)

Here, we utilize the constant 0.20 to adjust for the differences in VPDB and VSMOW standards as discussed in Bemis et al. (1998). Inserting the computed d18Owater value into Eq. (1) allows us to solve for salinity during the warmest (25
C) and coldest (8
C) periods represented in the shells. 3.4. Radiocarbon Radiocarbon sampling regions were chosen from selected areas of C. virginica hinges based on seasonal d18O profiles. When possible, each oyster shell hinge was sampled to obtain three 14C measurements. Each radiocarbon sample corresponded to an area previously drilled for stable isotopes and included 3 to 10 consecutive stable isotope samples, containing either a Summer (Sum) d18O minimum or a Fall-Spring (F-Sp) d18O maximum. Prior to sample collection, the top 10 mm of each sample was removed with

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

5

a cleaning pass of the drill bit to eliminate potential contamination on the calcite surface. A 14C-free Carrara marble sample was also drilled to establish a background correction for sample processing and drilling prior to graphitization. Note the term “F-Sp” denotes 14 C samples that precipitated immediately before and after calcification stopped at the minimum temperature of 8
C, while “Sum” indicates a 14C sample taken from margin regions where the d18O decrease from winter cessation levels out. Drilled calcite powder (1e2 mg) was converted to CO2 using a vacuum extraction line in the SIL at UC Davis. Samples were reacted in a common acid bath containing 105% H3PO4 at 90
C and the resulting CO2 was cryogenically purified to trap H2O and eliminate non-condensable gases. CO2 was subsequently collected and sealed in 6 mm Pyrex tubes and later converted to graphite in the presence of H2 and a Fe catalyst at ~570
C (Vogel et al., 1987) at the Lawrence Livermore National Laboratory Center for Accelerator Mass Spectrometry (CAMS). Graphite targets were analyzed by AMS and were 13 C-corrected using d13C values generated on a small amount of each sample at UC Davis. Sample preparation backgrounds were determined from blanks and the 14C-free Carrara marble calcite standards analyzed with the samples. All data are reported as fraction modern (fM), D (‰), and conventional 14C age (years before present (yr BP); Stuiver and Polach, 1977). Data presented here include d13C and background corrections and have been agecorrected to the estimated year of harvest as per Stuiver and Polach (1977). DR and its associated uncertainty (1s) were determined using equations from Jones et al. (2007) and Stuiver et al. (1986), respectively:

oysters analyzed. The remaining 7 oysters display broad d18O minimum plateaus indicating elevated calcification rates during summer and early fall. However, summer d18O values are not constant across these plateaus, suggesting habitat temperature and/or salinity varied during the summer (Fig. 2). An abrupt increase in d18O values occurs with the presumed reduction in river temperature (and/or increase in salinity) during the fall and winter (Fig. 2), while the opposite occurs in the spring as river temperatures rise and/or salinity decreases during the seasonal wet period. Because James River C. virginica calcification ceases below 8
C (Mann and Evans, 2004), we observe an abrupt transition between fall/winter and spring d18O values. Shell d13C profiles show a similar seasonal structure. In almost all oysters, a decrease in d13Ccalcite occurs when d18Ocalcite decreases during the onset of warming in spring (e.g., C-1 and C-3, Fig. 2). However, in some oysters (e.g., C-2 and C-4), d13C values increase prior to the fall d18O increase. The most positive d13C values in each shell generally occur with the d18O maxima that record oyster calcification immediately prior to growth cessation at 8
C.

DRðtÞ¼ Mm ðtÞ  Mcalib ðtÞ

d13 CDIC ¼ 0:31*S  8:25

DRðtÞs ¼

(4)

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi

sMm ðtÞ2 þ sMcalib ðtÞ2

(5)

where Mm is the conventional 14C age of an oyster of known calendar age (t) and Mcalib is the marine model age for that time (t) from the Marine13 calibration curve (Reimer et al., 2013). 4. Results 4.1. Oyster d18O and d13C A total of nine C. virginica shells from archaeological features of known deposition history were analyzed for shell d18O and d13C (Table 2). These shells span the initial European colonization of the James River watershed. Each shell displays seasonal cycles and records several years of James River environmental conditions. Assuming sequential d18O maxima and minima record an annual temperature cycle, 7 out of 9 shells contain 3e5 years of growth (Table 2, Fig. 2). Oysters C-1 and M1 contain 7 years of growth and appear to have deposited shell at a much lower rate than the other

4.2. James River d13CDIC and salinity We generate d13CDIC v. S relationships for the James River estuary using water data from three sites (Fig. 3a, Table S1). Wreck Shoal and Deep Water Shoal are upriver from the confluence of the James River and Chesapeake Bay (Fig. 1) and yield data that indicate they are recording the same mixing line:




 R2 ¼ 0:92; n ¼ 20

(6)

Data from Fort Monroe at the intersection of the James River and the Chesapeake Bay yield a different d13CDIC vs. S relationship suggesting these data are also influenced by DIC from Chesapeake Bay and/or lower James River tributaries. We therefore omit Fort Monroe data from further discussions of James River salinity. The yintercept in Eq. (6) was subsequently modified by þ1.26‰ to account for the addition of 13C-depleted CO2 due to fossil fuel burning, commonly known as the Suess Effect (Fig. 3b; Bohm et al., 2002). This yields the relationship:

d13 CDIC ¼ 0:31*S  6:99

(7)

that can be used to estimate salinity for the James River estuary prior to ~1850 and the initiation of the Industrial Revolution. It is important to note that while water samples were not collected over the course of the entire year, the salinity range represented in the samples encompasses that seen throughout the year at the Wreck Shoal and Deep Water Shoal locations (Mann et al., 2009). This indicates that the full range of d13CDIC is represented, as well. Salinity, computed from d18Owater, is derived from seasonal

Table 2 Oyster shell stable isotope data summary. Shell

Estimated oyster settlement year

Oyster age (yrs)a

d18O min Sum (‰, VPDB)

d18O max F-Sp (‰, VPDB)

d13C min (‰, VPDB)

d13C max (‰, VPDB)

C-1 C-2 C-3 C-4 E-1 E-2 M-1 L-1 L-2

Fall 1603 Summer 1606 Spring 1607 Spring 1609 Spring 1621 Spring 1621 Summer 1634 Summer 1672 Summer 1673

7 5 4 3 4 4 7 5 4

7.20 6.31 6.91 5.14 5.86 5.57 5.50 5.26 5.28

1.15 1.26 2.56 0.36 0.69 1.36 0.81 0.04 0.17

5.92 4.72 5.23 3.08 4.30 7.57 3.74 3.44 2.60

0.77 2.08 1.83 0.06 0.34 0.09 0.05 0.29 0.50

a

Age estimate based on the number of summers (d18O minima) recorded in the shell, per Harding et al. (2010).

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

6

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

Fig. 2. Oxygen (open diamond) and carbon (closed triangle) isotope and DR values (closed diamond) for the James River archaeological oyster shells spanning the 17th century. Grey bars indicate the location of 14C sampling with respect to the d18O seasonal profile. Note that the scale on the x-axis differs between graphs based on the hinge length of each oyster. d18O data from C-4 were originally published in Harding et al. (2010). DR error bars are ±1s.

d18Ocalcite extremes using Eqs. (1) and (3), and the assumption that

the calcite precipitation cessation temperature of 25
C or 8
C characterizes summer and winter isotopic extremes, respectively. The d13Ccalcite data corresponding to these d18Ocalcite values were then plotted against the salinity derived from d18Ocalcite computed d18Owater (Fig. 3b, Table S2). d13Ccalcite shows a positive relationship with salinity in the James River estuary and the linear regression through these d13Ccalcite v. d18O-derived salinity data:

d13 Ccalcite ¼ 0:28*S  6:40




R2 ¼ 0:35; n ¼ 73



(8)

is indistinguishable from the Suess-corrected d13CDIC v. S relationship (Eq. (7)) generated from modern water samples (t-test, alpha ¼ 0.05, P > 0.05). These results suggest that C. virginica d13Ccalcite is controlled primarily by the d13CDIC of the water in which

the oyster precipitated its shell and can therefore be used to directly reconstruct James River estuarine salinity without having to invoke assumptions about temperature extremes corresponding to calcification cessation temperatures. Because Eqs. (7) and (8) are statistically identical, we use the Suess-corrected d13CDIC v. S relationship (7) when calculating James River salinity because it was generated directly from river water samples. We have determined that “summer” CaCO3 values in all shells record temperatures >19
C by using the cyclical seasonal patterns, the d13CDIC vs. S relationship described with Eq. (7), and temperatures calculated using Eqs. (1) and (2). Based on average recorded water temperatures in the James River (Mann et al., 2009), these “summer” data correspond to approximately mid-May through mid-October. With few exceptions, F-Sp shell calcite is precipitated in water temperatures spanning 8e15
C, corresponding to the

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

7

model age at the time of collection (Table 4, Reimer et al., 2013), to calculate DR using Eqs. (4) and (5). DR values range from 151 ± 46 to þ109 ± 55 14C years (±1s, Table 4 and Fig. 4). Samples from the contact/drought period display the largest range in DR values of 260 14C years (Fig. 4). Two samples corresponding to 1605 (predrought) and 1607 (early drought) yield elevated DR values of þ91 ± 38 and þ109 ± 55 14C years, respectively, while a third sample from 1608 (early drought) displays the lowest DR value of 146 ± 83 14C years. The remainder of post-drought shells from the watershed show DR values that are indistinguishable from or lower than DR ¼ 0 (with one exception: M1), displaying a reduced DR range of 122 ± 64 to þ46 ± 42 14C years for a total DR range of 168 14C years. Contact/drought period oysters also have the largest intrashell DR variance observed (195 14C years, C-2) compared to oysters from later in the 17th century where DR variance is  160 14 C years. The intrashell DR values in every contact/drought oyster with more than one 14C sample are all distinct, while all postdrought oysters have at least two intrashell DR values that are indistinguishable within 1s. When separated by season, the F-Sp DR values have a larger DR range of 255 14C years compared to the Sum DR range of 197 14C years. We find that hinge calcite precipitated during summer, as defined by T > 19
C, and during F-Sp, as defined by T < 15
C, comprises ~54 ± 11% and 21 ± 8% of the sampled hinge transects, respectively. The remainder of the hinges precipitated when temperatures varied between 15 and 19
C. To calculate the most accurate DR that can be applied to James River 14C measurements, we use only the Sum DR values and exclude the more variable F-Sp values to get a pooled mean Sum DR ¼ 32 ± 11 14C years. While Sum DR values are inherently less variable than those from F-Sp, the hypothesis that individual Sum DR values are similar to the pooled mean Sum DR value is rejected (Chi square: T0 ¼ 29.61, c20.05 ¼ 23.68, df ¼ 14, Long and Rippeteau, 1974; Ward and Wilson, 1978). This indicates high natural interannual variability during summer conditions, possibly due to salinity fluctuations experienced by the oyster. However, the pooled Sum DR is the most accurate descriptor available, as the smaller range of DR values within the season reflects less variability in 14C concentration, resulting in a more precise characterization of DR. 4.4. Freshwater carbon mixing model Fig. 3. (a) Relationship between d13CDIC and salinity for three sites in the James River (Fig. 1): Fort Monroe (closed circle), Deep Water Shoal (closed diamond), and Wreck Shoal (open diamond). The linear regression for the Wreck Shoal and Deep Water Shoal data is used in this study as the modern d13CDIC v. S relationship for the James River (Eq. (6)). (b) The relationship between oyster shell d13C data and salinity computed from shell d18O seasonal minima and maxima data, assuming calcification limits of 25 and 8
C, respectively (Eqs. (1)e(3)). The dashed line is the linear regression through all seasonal calcite data. The solid black line is the modern d13CDIC vs. S relationship corrected by 1.26‰ to account for the change in atmospheric CO2 during the past ~150 years due to fossil fuel burning (Eq. (7), Bohm et al., 2002).

months of about mid-November to mid-April, excluding the coldest periods (Mann et al., 2009). The few exceptions in which temperatures higher than 15
C are encountered in F-Sp samples could result from shallow water temperatures exceeding 15
C for short periods of time.

4.3.

14

C data and James River DR

Radiocarbon ages for 17th century James River shells are presented in Table 3 and range from 825 ± 50 to 515 ± 40 years BP. These conventional 14C ages are used, along with the Marine13

To deconvolve the age of DIC sources contributing to the 14C content of the James River estuary, we developed a simple mixing model based on the following equation: 14

CRiver ½DICRiver  ¼ ð% MWÞ½DICMW ð14 CMW Þ þ ð1  %MWÞ½DICFW ð14 CFW Þ

(9)

where the fraction modern 14C measured in the shell (14Criver, ±1s) is a function of the percentage of freshwater (FW) and marine water (MW) in which the shell calcified, and their respective DIC concentrations and 14C content (Table 5, Fig. 5). The percentage of marine water is determined using the calculated average salinities experienced during calcite precipitation (Eq. (7)) and the assumption that marine salinity is z 36 psu. Almost all salinities represent mesohaline conditions (~9e23 psu), with only one oyster, E2, displaying salinities below 9 psu. The James River freshwater carbonate system, or [DICFW], used in our model was computed using paired pH and total alkalinity (TAlk) data (n ¼ 6) from 1999 (Chesapeake Bay Program, 2015; http://www.chesapeakebay.net, station TF5.0J) and average [Ca], [Mg], [Na], [K], [Cl], and [SiO2] data from the USGS spanning 1985e1995 (2015; http://nwis.waterdata.usgs.gov/va/nwis/qw,

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

8

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

Table 3 Radiocarbon data from James River oyster shells during initial contact/drought conditions (C) through early development (E), mid-century (M), and late century (L). CAMS #

Shell

Seasona

Cal. year of sample collectiona

Nb

d13C (‰, VPDB)

Fraction modern (fM)

±1s

D (‰)

±1s

14

±1s

163019 159367 159366 159365 159369 159368 159370 163022 163023 163025 163024 163026 163028 163027 163031 163029 163030 163035 163036 163037 163032 163034 163033

C-1 C-2 C-2 C-2 C-3 C-3 C-4 C-4 E-1 E-1 E-1 E-2 E-2 E-2 M-1 M-1 M-1 L-1 L-1 L-1 L-2 L-2 L-2

F-Sp 2 Sum 2 F-Sp 2 Sum 3 F-Sp 1 Sum 2 Sum 2 Sum 3 Sum 1 F-Sp 1 Sum 2 Sum 1 F-Sp 1 Sum 2 F-Sp 2 Sum 3 Sum 4 Sum 1 Sum 2 F-Sp 2 Sum 1 F-Sp 1 Sum 3

1605 1607 1608 1608 1608 1608 1610 1611 1621 1622 1622 1621 1622 1622 1636 1636 1637 1672 1673 1674 1673 1674 1675

4 7 5 5 5 9 N/A N/A 6 3 10 10 4 9 5 7 8 8 5 4 9 4 7

2.02 3.54 2.69 3.83 2.84 3.72 1.73 1.26 2.34 1.09 1.86 5.08 1.66 2.83 1.29 2.29 2.87 2.91 2.25 0.09 2.42 0.60 2.29

0.9049 0.9147 0.9025 0.9247 0.9320 0.9148 0.9323 0.9217 0.9117 0.9291 0.9108 0.9186 0.9276 0.9257 0.9230 0.9241 0.9116 0.9376 0.9266 0.9314 0.9316 0.9217 0.9203

0.0030 0.0030 0.0053 0.0037 0.0085 0.0036 0.0041 0.0035 0.0033 0.0061 0.0035 0.0035 0.0038 0.0033 0.0040 0.0031 0.0039 0.0046 0.0038 0.0040 0.0059 0.0035 0.0037

56.5 46.6 59.4 36.2 28.6 46.6 28.6 39.7 51.3 33.3 52.3 44.1 34.9 36.8 41.3 40.1 53.2 30.3 41.8 37.0 36.7 47.0 48.6

3.0 3.0 5.3 3.7 8.5 3.6 4.1 3.5 3.3 6.1 3.5 3.5 3.8 3.3 4.0 3.1 3.9 4.6 3.8 4.0 5.9 3.5 3.7

805 715 825 630 570 715 565 655 745 590 750 680 605 620 645 635 745 515 610 570 570 655 665

30 30 50 35 80 35 40 35 30 60 35 35 35 30 35 30 35 40 35 35 60 35 35

C agec (yr BP)

a Seasons are numbered from the start of the oyster's life. Sum 1 is the first full summer recorded in the d18Ocalcite record after the oyster's settlement. Calendar year of sample collection was subsequently determined assuming each d18O minimum represents a summer season and counting forward from the estimated settlement year (Table 2). b Each 14C sample averages hinge material corresponding to n stable isotope samples (Tables S3e11). c Conventional 14C age (Stuiver and Polach, 1977).

Table 4 James River DR values spanning initial contact through the late 17th century. Shell

Season

Cal. year of sample collection

C-1 F-Sp 2 1605 C-2 F-Sp 2 1608 C-3 F-Sp 1 1608 E-1 F-Sp 1 1622 E-2 F-Sp 1 1622 M-1 F-Sp 2 1636 L-1 F-Sp 2 1674 L-2 F-Sp 1 1674 F-Sp DR range ¼ 255 years C-2 Sum 2 1607 C-2 Sum 3 1608 C-3 Sum 2 1608 C-4 Sum 2 1610 C-4 Sum 3 1611 E-1 Sum 1 1621 E-1 Sum 2 1622 E-2 Sum 1 1621 E-2 Sum 2 1622 M-1 Sum 3 1636 M-1 Sum 4 1637 L-1 Sum 1 1672 L-1 Sum 2 1673 L-2 Sum 1 1673 L-2 Sum 3 1675 Sum DR range ¼ 197 years Pooled mean Sum samples (n ¼ 15) a

14

±1s

Marine13 model age (yr BP)

±1s

DRa (14C yr)

±1s

805 825 570 590 605 645 570 655

30 50 80 60 35 35 35 35

715 716 716 712 712 699 628 628

23 23 23 23 23 23 23 23

90 109 146 122 107 54 58 27

38 55 83 64 42 42 42 42

715 630 715 565 655 745 750 680 620 635 745 515 610 570 665

30 35 35 40 35 30 35 35 30 30 35 40 35 60 35

715 716 716 716 716 712 712 712 712 699 699 640 628 628 628

23 23 23 23 23 23 23 23 23 23 23 23 23 23 23

0 86 1 151 61 33 38 32 92 64 46 125 18 58 37

38 42 42 46 42 38 42 42 38 38 42 46 42 64 42

¡32

11

C age (yr BP)

DR calculations from Eqs. (4) and (5) and the Marine13 calibration curve of Reimer et al. (2013).

station 02035000). These data were obtained from monitoring stations in Cartersville, VA, where salinity ¼ 0 and were converted to [DIC] using Web-PHREEQC (https://www.ndsu.edu/webphreeq/ ), a web version of the USGS geochemical modeling program available for download at http://wwwbrr.cr.usgs.gov/projects/ GWC_coupled/phreeqc/. Input parameters were 6.38e9.06 for the pH range, 25
C for temperature, and 1 g/cc for density. Calculated

[DICFW] is 1139 mmol/kg, while [DIC] of the marine component is 2050 mmol/kg (Wang et al., 2013). The corresponding fM of marine water is derived from the Marine13 calibration curve (Reimer et al., 2013). Rearranging Eq. (9) to solve for the fraction modern 14C of the freshwater produces:

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

Fig. 4. Sum (closed circles) and F-Sp (open squares) DR values (±1s) calculated using Eqs. (4) and (5) for each sample. The year(s) of 14C collection describes oyster growth periods based on archaeological data (Table 3). DR ¼ 0 (horizontal black line) indicates the global ocean reservoir, R, of ~400 years (Stuiver and Braziunas, 1993).

14

CFW ¼½14 CRiver *½DICRiver   ð%MWÞ*½DICMW  *ð14 CMW Þ=½ð1  %MWÞ*½DICFW 

(10)

with the results presented in Table 5 as fM. Fig. 6 presents possible ages for the 14C freshwater component, known to be a mixture between atmospheric CO2 equilibration and soil inorganic carbon (SIC), divided into the four temporal periods of contact/drought, early development, mid-century, and late century. Base years were chosen for each period according to the closest 5 year interval that encompassed most, if not all, of the estimated years of sample collection. The fM of atmospheric 14C associated with each base year (1610, 1620, 1640, and 1675 for each period, respectively) serves as the 14C content for the atmospheric component of the model (Fig. 5, Reimer et al., 2013). Mixing lines of different SIC ages are then used to constrain the potential percentage and age of the SIC that is contributing carbon to the James River. Although there is an infinite range of combinations between these two variables, if we assume that 50% of the freshwater DIC pool is derived from atmosphere exchange, as proposed by Hossler and Bauer (2012), then SIC comprises the remainder of the pool (Fig. 6). With this assumption, the age of the DIC derived from remineralized soil carbon plus weathered bedrock in the contact/ drought and early development periods ranges from zero years (relative to the contemporary atmosphere at the time) to ~1800 years. SIC age during the mid- and late century periods displays a similar age between ~250 and ~1500 years. These ages are constrained using Eq. (10) to solve for the minimum and maximum 14 CFW (fM) from the set of shells within each period (Table 5). 5. Discussion The seasonally-constrained 17th century James River DR values that we obtain in this study span a range of 260 years (151 ± 46 to þ109 ± 55 14C years) and can be attributed to the mixing of inorganic carbon from marine and freshwater end-member sources. Converting shell d13C data to salinity using Eq. (7) indicates that

9

the shell calcite we sampled for 14C analysis precipitated in water that was between 11 and 64% marine (Table 5) with the remaining inorganic carbon sourced from James River freshwater. Inorganic carbon in James River freshwater could be derived from three possible sources. The first is atmospheric CO2, which contributes ~50% (±10%) of the DIC in rivers through exchange across the air:water interface (Hossler and Bauer, 2012). The other two possible sources, weathering of carbonate bearing minerals and remineralization of soil carbon, comprise the total SIC pool. Potential sources of weatherable CaCO3 in the James River watershed include the fossiliferous Pliocene carbonates of the Yorktown formation found in the Coastal Plain (Hobbs, 2004, 2009), as well as carbonate rich formations of the Blue Ridge and Valley and Ridge Provinces from which the James River headwaters originate (Southworth et al., 2008). Unlike atmospheric CO2, carbon derived from carbonate mineral dissolution would be devoid of 14C because radioactive decay would have effectively eliminated radiocarbon after ~50 kyr. The 14C content of the remaining freshwater DIC is derived from the remineralization of soil organic matter with a poorly constrained age. Wang et al. (1998) divided soil organic matter into three carbon pools, each with different ages. Respired CO2 from soil organic carbon contributes to the freshwater DIC pool via pore water flow and subsequent drainage into the river (Raymond et al., 2004). The active pool of rapidly decomposing plant litter has a 14C content close or equal to that of the atmosphere. However, the passive pool of stabilized organic matter remains in soil over timescales of several thousand years or more and therefore contributes a more negative 14C signature to the soil DIC mixture. The third, intermediate, pool is less well defined and has a 14C content that is in between the active and passive pools. 14C age of organics also increases with depth in the soil, as the active pool resides at the surface and the older refractory carbon is found at depth (Gaudinski et al., 2000; Trumbore, 2000). Potential ages and the percent contribution of the SIC derived from a combination of remineralized soil carbon and weathered carbonate are estimated in our model and reveal that SIC is sourced from both the active and intermediate soil organic pools described by Wang et al. (1998). This is not surprising as several studies have demonstrated the export of modern to millennial aged soil organic carbon into river systems (Raymond and Bauer, 2001b; Raymond et al., 2004, and references therein). Our maximum SIC age estimate of ~1800 years suggests that the final passive pool contributes little to no inorganic carbon to the James River. Nevertheless, it is likely that SIC of different ages mix to produce the observed 14C age recorded in the shells (Trumbore, 2000; Raymond et al., 2004) and any variability in our base assumption that 50% of the SIC is derived from the atmosphere will affect the reconstructed SIC ages. We note that our model does not account for variable 14C production in the atmosphere over time (Reimer et al., 2013). For example, 14C was produced at a higher rate in the atmosphere during the century preceding the year 1610 (Fig. 5). Our model does not correct for this difference in production, instead only using the 14 C content obtained from the 14C curve for the year in question in the calculations. Two shells, one within the contact/drought period and another from the early development era, yielded results that indicated the age of the SIC was greater than or equal to that of atmospheric 14C and only “fit” the model when the shell 14C measurements were expanded to their 1s error limit. Nevertheless, our model provides the first quantitative SIC age estimate for the James River estuary during the 17th century historical period. Two broad-scale environmental perturbations are revealed by the SIC model. First, land use change after the arrival of western Europeans significantly altered the terrestrial ecosystem. Prior to the colonists' arrival in 1607 the native Algonquin population

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

10

Shell

14

C-1 C-2 C-2 C-2 C-3 C-3 C-4 C-4 E-1 E-1 E-1 E-2 E-2 E-2 M-1 M-1 M-1 L-1 L-1 L-1 L-2 L-2 L-2

F-Sp Sum F-Sp Sum F-Sp Sum Sum Sum Sum F-Sp Sum Sum F-Sp Sum F-Sp Sum Sum Sum Sum F-Sp Sum F-Sp Sum

a

C sample 2 2 2 3 1 2 2 3 1 1 2 1 1 2 2 3 4 1 2 2 1 1 3

Average d13Ccalcite (‰, VPDB)

Salinity Eq. (7) (psu)

Marine %a

Shell

1.76 3.47 2.85 2.64 3.15 4.11 1.52 1.63 3.57 0.67 2.09 5.00 1.81 5.73 1.29 1.95 2.52 2.78 2.24 0.15 2.35 0.11 2.02

16.9 11.4 13.4 14.0 12.4 9.2 17.7 17.3 11.0 20.4 15.8 6.4 16.7 4.1 18.4 16.3 14.4 13.6 15.3 23.0 15.0 22.2 16.0

47 32 37 39 34 26 49 48 31 57 44 18 46 11 51 45 40 38 43 64 42 62 45

0.9049 0.9147 0.9025 0.9247 0.9320 0.9148 0.9323 0.9217 0.9117 0.9291 0.9108 0.9186 0.9276 0.9257 0.9230 0.9241 0.9116 0.9376 0.9266 0.9314 0.9316 0.9217 0.9203

14

C (GM)b

±1s

[DIC] marinec (mmol/kg)

[DIC] freshc (mmol/kg)

[DIC] Total (mmol/kg)

14

0.0030 0.0030 0.0053 0.0037 0.0085 0.0036 0.0041 0.0035 0.0033 0.0061 0.0035 0.0035 0.0038 0.0033 0.0040 0.0031 0.0039 0.0046 0.0038 0.0040 0.0059 0.0035 0.0037

951 639 752 788 696 518 993 973 620 1146 889 362 940 228 1033 914 812 764 861 1295 842 1248 902

604 780 716 696 747 848 581 592 790 495 639 936 610 1011 558 625 682 709 655 410 666 437 632

1555 1419 1468 1484 1444 1365 1573 1565 1410 1640 1528 1297 1550 1239 1591 1539 1494 1473 1516 1706 1507 1685 1534

0.8895 0.9147 0.8897 0.9360 0.9481 0.9148 0.9563 0.9332 0.9090 0.9593 0.9047 0.9199 0.9467 0.9281 0.9335 0.9340 0.9048 0.9514 0.9290 0.9522 0.9402 0.9128 0.9139

C Fresh Eq. (10) (GM)

Assume marine salinity (S) ¼ 36 psu and freshwater S ¼ 0. The marine and atmospheric 14C content (fM) for the four temporal periods (contact ¼ 1610, early develop. ¼ 1620, mid-century ¼ 1640, late century ¼ 1675) were obtained from the Marine13 and IntCal13 calibration curves, respectively (Fig. 5, Reimer et al., 2013). c Contribution of [DIC] based on the percentage of each source in the river; [DIC] of 100% marine water ¼ 2025 mmol/kg (Wang et al., 2013) and 100% freshwater [DIC] ¼ 1139 mmol/kg (http://nwis.waterdata.usgs.gov; http:// www.chesapeakebay.net). b

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

Table 5 Calculations for the freshwater carbon mixing model (Eq. (10)).

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

11

Fig. 5. IntCal13 (black diamonds) and Marine13 (black circles) 14C calibration curves (Reimer et al., 2013). Time periods examined in this study are highlighted with solid vertical lines.

employed slash and burn techniques to clear fields, maintaining root networks (Rountree and Turner, 2002). These dense root networks retained freshwater and maintained soil horizons within the terrestrial environment, as did the extensive system of marshes and beaver dams prevalent in the riparian zone prior to colonization (Miller, 1986). The English hunted beavers for their pelts leading to localized extinction in the area by 1700 and removal of the dam network (Miller, 1986; Wennersten, 2001), while John Rolfe's introduction of tobacco as a cash crop in 1612 immediately resulted in clear cutting of forests throughout the watershed below the fall line. By 1629, yearly tobacco shipments from the Virginia Colony were up to 1.5 million pounds and continuing to grow with English settlements extending along the James River shoreline from the fall line to the confluence with the Chesapeake Bay (Hatch, 1957). These activities disturbed soil horizons at depth, thereby remobilizing older organic carbon with lower 14C content that contributed to the older age estimates described by our model during the mid- and late century periods (Fig. 6). Agricultural practices and other causes of soil disturbance are known to export older soil profiles (Trumbore, 2000; Raymond and Bauer, 2001a; Raymond et al., 2004; Ewing et al., 2006). This point is highlighted with an age of SIC not less than 250 years, assuming a 50% contribution of SIC to the freshwater DIC component. This is likely a direct result of the disturbance of deeper soil horizons and the destruction of marshes that retain older soil C (Raymond et al., 2004). The introduction of a small amount of ancient carbon from sedimentary bedrock weathering could also be considered to explain the >1000 year old C found in all temporal periods. Both the contact/drought and early development periods have soil carbon with possible ages of zero relative to the contemporary atmosphere, which may reflect the original watershed structure before European agricultural and land-use disruptions. The second environmental disturbance is found within the contact/drought period, which displays the largest range of

Fig. 6. Model of James River watershed soil inorganic carbon (SIC) age in the freshwater DIC pool based on a three component mixing model (Eqs. (9) and (10)). Dashed lines reflect mixing between atmospheric CO2 and different SIC ages for the contact/ drought (1610), early development (1620), mid-century (1640), and late century (1675) periods (solid circle convergence point, atmospheric 14C content for base year). We assume 50% of freshwater DIC is due to atmospheric CO2 through equilibration (Hossler and Bauer, 2012) with the remaining DIC due to dissolved SIC. The dark horizontal line indicates the 50% SIC estimate. The light shaded boxes form an envelope within the vertical domain indicating ±10% uncertainty (Hossler and Bauer, 2012) and the horizontal domain indicating the 14CFW data constraint (Table 5).

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

12

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

possible SIC age (within 1s), from zero to ~1800 years. We hypothesize that this large range is related to environmental changes during the severe drought of 1606e1612 (Stahle et al., 1998). Drought and deforestation in the James River watershed after 1606 changed the hydrologic cycle. Originally, mature forests retained water through a combination of canopy, leaf litter, and root dynamics (e.g., Jackson et al., 2005; Piao et al., 2007). During the 160612 drought, concurrent reduced leaf litter and soil moisture likely allowed rainwater from the rare precipitation events to soak deeper into the soil where older carbon resides, thereby mobilizing this older carbon and adding it to the freshwater DIC pool. Precipitation runoff intensified after the onset of tobacco farming and related deforestation (Wennersten, 2001; Rountree and Turner, 2002), evidenced by an increase in sedimentation rates in the Chesapeake Bay (Miller, 1986; Cooper and Brush, 1991, 1993). Conversely, the presence of modern-aged soil C during the contact/drought period is easily explained by root respiration and the degradation of surface detritus that is less than a decade and often less than a few years old (Trumbore, 2000; Taylor et al., 2015). The large range of possible SIC age highlights what we believe to be natural seasonal variation in the James River system, which is likely enhanced during times of environmental stress. Our seasonal 14C sampling of shells across several chronological periods in the James River reveals seasonal variability and broad environmental change. Colman et al. (2002) suggested a DR ¼ 0 14C years, or the standard ocean reservoir correction, for the Chesapeake Bay while more recently Rick et al. (2012) produced a DR ¼ 129 ± 22 14C years for the Chesapeake Bay western shore. This value was derived from three Potomac River C. virginica shells and one mainstem Chesapeake Bay shell, from just north of the James River mouth, spanning the late 19th and early 20th century (Rick et al., 2012). Our season-specific average DR of 32 ± 11 14C years is considerably different than both the calculated western shore DR value and the DR value of 82 ± 46 14C years derived from the mainstem Chesapeake Bay shell in Rick et al. (2012). These reservoir age differences from locations along the Chesapeake Bay western shore are significant. Of the four western shore shells used to produce the mean DR proposed by Rick et al. (2012), only one was collected from near the James River, but in a zone that is dominated by non-James River waters (e.g., Chesapeake Bay mainstem above the James River mouth), while the other three were from the Potomac River. The bedrock geology and environmental conditions of the James and Potomac watersheds are quite different (Hobbs, 2004; Southworth et al., 2008), suggesting the proportion of weathered carbonate and soil carbon entering the river could be different. Furthermore, the lower Chesapeake Bay shell presented in the Rick et al. (2012) study was collected in 1916, more than two centuries after our oldest shells were collected. As noted earlier, significant land use changes occurred during and after the 17th century, which likely altered the amount and signature of the SIC 14C content exported to the estuary (Brush, 2001; Wennersten, 2001). The large positive DR of 129 14C years from Rick et al. (2012), obtained from more recent shells than we analyze here, agrees with studies that show an increase in agriculture and development results in the export of more aged soil profiles (Raymond and Bauer, 2001a; Raymond et al., 2004). Stable isotope profiles from mollusks have been previously combined with 14C dating in coastal environments to identify periods of seasonal upwelling (Andrus et al., 2005; Culleton et al., 2006; Jones et al., 2007; Ferguson et al., 2013). However, Culleton et al. (2006) recommended that 14C measurements on mollusks should be conducted on shell material that spans multiple growth increments in an effort to smooth short-term variability. Recently, Rick and Henkes (2014) investigated intrashell variability in DR from a historical 1898 oyster shell from the mouth of the Potomac

River. The authors sampled the shell at a variety of intervals spanning several growth bands and discovered a DR range of 100 14 C years within the shell. Following the recommendation of Culleton et al. (2006), Rick and Henkes (2014) suggested that researchers should sample large areas of a shell in an effort to homogenize 14C variability in order to date estuarine oysters. However, it is now clear that analyzing bulk shell without a thorough understanding of seasonal variability and the proportion of calcite precipitated during each season may generate error around the 14C age that is considerably larger than the analytical uncertainty of the measurement. The current study demonstrates that a linked stable isotope and radiocarbon approach is effective in an estuarine environment that has changing proportions of fresh and marine water on a seasonal timescale. Shell carbon isotope data can be used to calculate the salinity in which the shell calcified, thereby enabling researchers to better quantify the multiple variables in the system that affect DR values. Whereas salinity is often calculated from the d18O profile within a shell, these calculations (Eqs. (1)e(3)) rely on an assumption of calcification temperature. In the James River estuary, we suggest that C. virginica shell d13C is primarily controlled by river d13CDIC so salinity can be computed by an independent geochemical proxy and d18O can be used to directly determine temperature with reduced uncertainty (Fig. 3). We note that some studies have argued that d13Ccalcite is not an appropriate proxy for d13CDIC due to the complication of metabolic carbon input (Gillikin et al., 2006; Chauvaud et al., 2011), while others have stated the exact opposite (Poulain et al., 2010; Beirne et al., 2012; Marchais et al., 2015). Data from the James River estuary agree with the latter conclusion, as the relationship between d13Ccalcite vs. computed salinity is statistically indistinguishable from that of d13CDIC vs. S (Fig. 3b), indicating that the dominant control on shell d13C is the d13C value of the DIC. Surge et al. (2001) also arrived at this conclusion with C. virginica from Florida, although the authors noted a small offset in shell d13C during the winter months (17e26
C, JanuaryeFebruary 1998e99). As James River shells cease growth below 8
C, a temperature minima which the Florida oysters examined by Surge et al. (2001) never experienced, we do not expect to see a similar offset at the coldest temperatures recorded by the James River oysters. We note, however, that a small amount of metabolic carbon, or an ontogenetic change in the amount of metabolic carbon incorporated into the shell, may contribute to the scatter observed in this relationship. Our data show that seasonal sampling for 14C quantifies the considerable DR range that exists in an estuarine system on an annual basis. During the 17th century, we record a total range of 260 14C years across the full dataset and an intrashell DR range of 195 14C years within a single oyster (C-2). DR values from Sum samples are more consistent among the time periods sampled and display a smaller DR range than those from F-Sp samples (197 14C years vs. 255 14C years). Random sampling of a shell for 14C analysis without accounting for seasonality could introduce a substantial amount of error during time periods with large instrashell variability such as the contact/drought period of the early 17th century. While the post-drought shells display less extreme DR ranges (largest instrashell DR range <160 14C years), the range is still considerable. Using a combined stable isotope-radiocarbon approach, it is possible to elucidate seasonal cycles and 14C variability to obtain the most comprehensive environmental data for a sample. Nevertheless, we recognize that the primary objective for analyzing shell 14C is age control. Because a larger portion of the shell hinge, 54 ± 11%, is precipitated in summer months compared to only 21 ± 8% in F-Sp, it is likely that random sampling of shell calcite will yield data that are strongly weighted towards summer DR values. Accordingly, for dating applications prior to or at the

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

onset of European influence in the James River estuary, we recommend a DR of 32 ± 11 14C years. 6. Conclusion To our knowledge, the C. virginica radiocarbon ages we present here are the first such data to constrain 17th century North American estuarine reservoir ages. We calculate a mean value of DR ¼ 32 ± 11 14C years (n ¼ 15 Sum samples) that can be applied to future studies from the James River estuary, while noting the entire 23 sample dataset yields a 260 14C year range in DR values from 151 ± 46 to þ109 ± 55 14C years. The seasonal sampling technique we employ here allows us to identify variability in DR, an important consideration for oysters that grow in a dynamic estuarine environment as the various components that contribute 14C to the estuary fluctuate on this time scale. Estuaries and coastal areas are often plentiful in material that can be 14C dated to reconstruct human and geologic history, so we recommend utilizing a seasonally resolved sampling method to provide the greatest constraint on DR interpretations. Our mixing model revealed broad environmental changes due to an extensive drought in 1606e1612 and land use change that persisted and increased throughout the ~70 year dataset. Contact/ drought and early development SIC ranged from modern (0 years) to ~1800 years old, reflecting both the original watershed structure prior to the agricultural disruption and the possibility for rare precipitation during times of environmental stress to interact with older carbon within the soil. SIC ages during the mid- to late century were no younger than 250 years, likely due to increased agricultural activities that disturbed soil horizons and released older 14C into the river. The model further highlights natural variation present within the estuary, and therefore the need to consider all sources of carbon that contribute to a 14C measurement and how each may change on varying timescales. Acknowledgements We thank B. Straube, W. Kelso, and Preservation Virginia for providing us access to Crassostrea virginica shells from the Jamestown archaeological excavation and N. Lucketti, James River Institute for Archaeology, for providing the Nansemond Pallizado and Bacon's Castle shells. We also thank R. Zeebe for assistance with James River carbon system calculations, P. Zermeno for help pre~ ez for paring the 14C samples for graphitization, and I. Montan comments on an early version of the manuscript and microdrill access. Research funds were provided to BLG by the UC Davis Dept. of Earth and Planetary Science Durrell Fund for graduate student research. A portion of this work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.quageo.2017.03.002. References Abbott, M.B., Stafford, T.W., 1996. Radiocarbon geochemistry of modern and ancient Arctic lake systems, Baffin Island, Canada. Quat. Res. 45, 300e311. Anderson, E.C., Libby, W.F., Weinhouse, S., Reid, A.F., Kirshenbaum, A.D., Grosse, A.V., 1947. Natural radiocarbon from cosmic radiation. Phys. Rev. 72, 931e936. Andrus, C.F.T., Hodgins, G.W.L., Sandweiss, D.H., Crowe, D.E., 2005. Molluscan ~ o-related upwelling variation in Peru. In: radiocarbon as a proxy for El Nin Mora, G., Surge, D. (Eds.), Isotopic and Elemental Tracers of Cenozoic Climate Change. Geological Society of America, pp. 13e20. Special Paper 395. Ascough, P., Cook, G., Dugmore, A., 2005a. Methodological approaches to

13

determining the marine radiocarbon reservoir effect. Prog. Phys. Geogr. 29, 532e547. Ascough, P.L., Cook, G.T., Church, M.J., Dunbar, E., Einarsson, A., McGovern, T.H., Dugmore, A.J., Perdikaris, S., Hastie, H., Frioriksson, A., Gestsdottir, H., 2010. Temporal and spatial variations in freshwater 14C reservoir effects: lake Mývatn, Northern Iceland. Radiocarbon 52, 1098e1112. Ascough, P.L., Cook, G.T., Dugmore, A.J., Scott, E.M., Freeman, S., 2005b. Influence of mollusk species on marine DR determinations. Radiocarbon 47, 433e440. Bard, E., Hamelin, B., Fairbanks, R.G., Zindler, A., 1990. Calibration of the 14C timescale over the past 30,000 years using mass spectrometric U-Th ages from Barbados corals. Nature 345, 405e410. Beirne, E.C., Wanamaker Jr., A.D., Feindel, S.C., 2012. Experimental validation of environmental controls on the d13C of Arctica islandica (ocean quahog) shell carbonate. Geochim. Cosmochim. Acta 84, 395e409. Bemis, B.E., Spero, H.J., Bijma, J., Lea, D.W., 1998. Reevaluation of the oxygen isotopic composition of planktonic foraminifera: experimental results and revised paleotemperature equations. Paleoceanography 13, 150e160. Bohm, F., Haase-Schramm, A., Eisenhauer, A., Dullo, W.C., Joachimski, M.M., Lehnert, H., Reitner, J., 2002. Evidence for preindustrial variations in the marine surface water carbonate system from coralline sponges. Geochem. Geophys. Geosyst 3, 1e13. Bratton, J.F., Colman, S.M., Seal, R.R., 2003. Eutrophication and carbon sources in Chesapeake Bay over the last 2700 yr: human impacts in context. Geochim. Cosmochim. Acta 67, 3385e3402. Broecker, W.S., Peng, T.-H., 1982. Tracers in the Sea. Lamont-Doherty Geological Observatory, Columbia University. Broecker, W.S., Walton, A., 1959. The geochemistry of C14 in freshwater systems. Geochim. Cosmochim. Acta 16, 15e38. Brooks, T., Fang, C., 1983. James River Slack Water Data Report: Temperature, Salinity, and Dissolved Oxygen, 1971-1980. VA. Inst. of Mar. Sci. Data Report No. 12. Brush, G.S., 2001. Natural and anthropogenic changes in Chesapeake Bay during the last 1000 years. Hum. Ecol. Risk Assess. 7, 1283e1296. Carriker, M.R., 1996. The shell and ligament. In: Kennedy, V.S., Newell, R.I.E., Eble, A.F. (Eds.), The Eastern Oyster: Crassostrea Virginica. Maryland Sea Grant, College Park, pp. 75e168. bault, J., Clavier, J., Lorrain, A., Strand, Ø., 2011. What's hiding Chauvaud, L., The behind ontogenetic d13C variations in mollusk Shells? New insights from the great scallop (Pecten maximus). Estuaries Coasts 34, 211e220. Chesapeake Bay Program, 2015. Water Quality Database (1984-present). http:// www.chesapeakebay.net/data/downloads/cbp_water_quality_database_1984_ present. Colman, S.M., Baucom, P.C., Bratton, J.F., Cronin, T.M., McGeehin, J.P., Willard, D., Zimmerman, A.R., Vogt, P.R., 2002. Radiocarbon dating, chronologic framework, and changes in accumulation rates of Holocene estuarine sediments from Chesapeake Bay. Quat. Res. 57, 58e70. Cooper, S.R., Brush, G.S., 1991. Long-term history of Chesapeake Bay anoxia. Science 254, 992e996. Cooper, S.R., Brush, G.S., 1993. A 2,500-year history of anoxia and eutrophication in Chesapeake Bay. Estuaries 16, 617e626. Craig, H., 1957. The natural distribution of radiocarbon and the exchange time of carbon dioxide between atmosphere and sea. Tellus 9, 1e17. Cronin, T., Willard, D., Karlsen, A., Ishman, S., Verardo, S., McGeehin, J., Kerhin, R., Holmes, C., Colman, S., Zimmerman, A., 2000. Climatic variability in the eastern United States over the past millennium from Chesapeake Bay sediments. Geology 28, 3e6. Cronin, T.M., Thunell, R., Dwyer, G.S., Saenger, C., Mann, M.E., Vann, C., Seal, R.R., 2005. Multiproxy evidence of Holocene climate variability from estuarine sediments, eastern North America. Paleoceanography 20, PA4006. Culleton, B.J., Kennett, D.J., Ingram, B.L., Erlandson, J.M., Southon, J.R., 2006. Intrashell radiocarbon variability in marine mollusks. Radiocarbon 48, 387e400. Dean, C.E., 2005. APVA Preservation Virginia: Bacon's Castle. The Creative Company, Lawrenceburg. Deevey, E.S., Gross, M.S., Hutchinson, G.E., Kraybill, H.L., 1954. The natural C14 contents of materials from hard water lakes. Proc. Natl. Acad. Sci. 40, 285e288. Druffel, E.R.M., Griffin, S., Hwang, J., Komada, T., Beaupre, S.R., DruffelRodriguez, K.C., Santos, G.M., Southon, J., 2004. Variability of monthly radiocarbon during the 1760s in corals from the Galapagos Islands. Radiocarbon 46, 627e631. Edwards, R.L., Chen, J.H., Wasserburg, G.J., 1987. 238U-234U-230Th-232Th Systematics and the precise measurement of time over the past 500,000 years. Earth Planet. Sci. Lett. 81, 175e192. Epstein, S., Buchsbaum, R., Lowenstam, H.A., Urey, H.C., 1953. Revised carbonatewater isotopic temperature scale. Geol. Soc. Am. Bull. 64, 1315e1325. Ewing, S.A., Sanderman, J., Baisden, W.T., Wang, Y., Amundson, R., 2006. Role of large-scale soil structure in organic carbon turnover: evidence from California grassland soils. J. Geophys. Res. Biogeosciences 111, G03012. Fairbanks, R.G., Mortlock, R.A., Chiu, T.C., Cao, L., Kaplan, A., Guilderson, T.P., Fairbanks, T.W., Bloom, A.L., Grootes, P.M., Nadeau, M.J., 2005. Radiocarbon calibration curve spanning 0 to 50,000 years BP based on paired 230Th/234U/ 238U and 14C dates on pristine corals. Quat. Sci. Rev. 24, 1781e1796. Ferguson, J.E., Johnson, K.R., Santos, G., Meyer, L., Tripati, A., 2013. Investigating d13C and D14C within Mytilus californianus shells as proxies of upwelling intensity. Geochem. Geophys. Geosystems 14, 1856e1865. Galtsoff, P.S., 1964. The American oyster: Crassostrea virginica Gmelin. U. S. Fish

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

14

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15

Wildl. Serv. Fish. Bull. 64, 397e456. Gaudinski, J.B., Trumbore, S.E., Davidson, E.A., Zheng, S.H., 2000. Soil carbon cycling in a temperate forest: radiocarbon-based estimates of residence times, sequestration rates and partitioning of fluxes. Biogeochemistry 51, 33e69. Gillikin, D.P., Lorrain, A., Bouillon, S., Willenz, P., Dehairs, F., 2006. Stable carbon isotopic composition of Mytilus edulis shells: relation to metabolism, salinity, d13C(DIC) and phytoplankton. Org. Geochem. 37, 1371e1382. Gillikin, D.P., Lorrain, A., Meng, L., Dehairs, F., 2007. A large metabolic carbon contribution to the d13C record in marine aragonitic bivalve shells. Geochim. Cosmochim. Acta 71, 2936e2946. Goodfriend, G.A., Flessa, K.W., 1997. Radiocarbon reservoir ages in the Gulf of California: roles of upwelling and flow from the Colorado River. Radiocarbon 39, 139e148. Harding, J.M., Spero, H.J., Mann, R., Herbert, G.S., Sliko, J.L., 2010. Reconstructing early 17th century estuarine drought conditions from Jamestown oysters. Proc. Natl. Acad. Sci. 107, 10549e10554. Hatch, C.E., 1957. The First Seventeen Years: Virginia, 1607e1624. University of Virginia Press, Charlottesville. Haven, D., Fritz, L., 1985. Setting of the American oyster Crassostrea virginica in the James River, Virginia, USA: temporal and spatial distribution. Mar. Biol. 86, 271e282. Hobbs, C., 2009. York river geology. J. Coast. Res. 57, 10e16. Hobbs, C.H., 2004. Geological history of Chesapeake Bay, USA. Quat. Sci. Rev. 23, 641e661. Hogg, A.G., Higham, T.F.G., Dahm, J., 1998. 14C dating of modern marine and estuarine shellfish. Radiocarbon 40, 975e984. Hope, D., Billett, M.F., Cresser, M.S., 1994. A review of the export of carbon in river water: fluxes and processes. Environ. Pollut. 84, 301e324. Hossler, K., Bauer, J.E., 2012. Estimation of riverine carbon and organic matter source contributions using time-based isotope mixing models. J. Geophys. Res. Biogeosciences 117, G03035. Hossler, K., Bauer, J.E., 2013. Amounts, isotopic character and ages of organic and inorganic carbon exported from rivers to ocean margins: 1. Estimates of terrestrial losses and inputs to the Middle Atlantic Bight. Glob. Biogeochem. Cycles 27, 331e346. Hudgins, C., Schmidt, D., Straube, B., 2008. Summary of James Fort-period burial ground. In: Kelso, W.M., Straube, B. (Eds.), 2000-2006 Interim Report on the APVA Excavations at Jamestown, Virginia. Association for the Preservation of Virginia's Antiquities, Richmond, pp. 55e69, 71-76. Ingram, B.L., 1998. Differences in radiocarbon age between shell and charcoal from a Holocene shellmound in northern California. Quat. Res. 49, 102e110. Ingram, B.L., Ingle, J.C., Conrad, M.E., 1996. Stable isotope record of late Holocene salinity and river discharge in San Francisco Bay, California. Earth Planet. Sci. Lett. 141, 237e247. Ingram, B.L., Southon, J.R., 1996. Reservoir ages in eastern Pacific coastal and estuarine waters. Radiocarbon 38, 573e582. Jackson, R.B., Jobb agy, E.G., Avissar, R., Roy, S.B., Barrett, D.J., Cook, C.W., Farley, K.A., Le Maitre, D.C., McCarl, B.A., Murray, B.C., 2005. Trading water for carbon with biological carbon sequestration. Science 310, 1944e1947. Jones, D.S., 1983. Sclerochronology: reading the record of the Molluscan shell. Am. Sci. 71, 384e391. Jones, K.B., Hodgins, G.W.L., Dettman, D.L., Andrus, C.F.T., Nelson, A., EtayoCadavid, M.F., 2007. Seasonal variations in Peruvian marine reservoir age from pre-bomb Argopecten purpuratus shell carbonate. Radiocarbon 49, 877e888. Keith, M., Anderson, G., Eichler, R., 1964. Carbon and oxygen isotopic composition of mollusk shells from marine and fresh-water environments. Geochim. Cosmochim. Acta 28, 1757e1786. Kelso, W.M., 2006. Jamestown: the Buried Truth. University of Virginia Press, Charlottesville. Kelso, W.M., Straube, B., 2004. Jamestwon Rediscovery 1994-2004. Association for the Preservation of Virginia's Antiquities, Richmond, pp. 131e135. Kennett, D.J., Ingram, B.L., Erlandson, J.M., Walker, P., 1997. Evidence for temporal fluctuations in marine radiocarbon reservoir ages in the Santa Barbara Channel, southern California. J. Archaeol. Sci. 24, 1051e1059. Kennett, D.J., Ingram, B.L., Southon, J.R., Wise, K., 2002. Differences in 14C age between stratigraphically associated charcoal and marine shell from the Archaic Period site of Kilometer 4, southern Peru: old wood or old water? Radiocarbon 44, 53e58. Kirby, M.X., Soniat, T.M., Spero, H.J., 1998. Stable isotope sclerochronology of Pleistocene and recent oyster shells (Crassostrea virginica). Palaios 13, 560e569. Levin, I., Hesshaimer, V., 2000. Radiocarbon - a unique tracer of global carbon cycle dynamics. Radiocarbon 42, 69e80. Long, A., Rippeteau, B., 1974. Testing contemporaneity and averaging radiocarbon dates. Am. Antiq. 39, 205e215. Lorrain, A., Paulet, Y.M., Chauvaud, L., Dunbar, R., Mucciarone, D., Fontugne, M., 2004. d13C variation in scallop shells: increasing metabolic carbon contribution with body size? Geochim. Cosmochim. Acta 68, 3509e3519. Lougheed, B.C., Lubbe, H., Davies, G.R., 2016. 87Sr/86Sr as a quantitative geochemical proxy for 14C reservoir age in dynamic, brackish waters: Assessing applicability and quantifying uncertainties. Geophys. Res. Lett. 43, 735e742. Luccketti, N.M., 1990. Archaeological excavations at Bacon's Castle, Surry county, Virginia. In: Kelso, W.M., Most, R. (Eds.), Earth Patterns: Essays in Landscape Archaeology. University Press of Virginia, Charlottesville, pp. 21e42. Luccketti, N.M., 2010. Nansemond Pallizado and Virginia palisade Fortifications. In: Klingelhofer, E. (Ed.), First Forts: Essays on the Archaeology of Proto-colonial

Fortifications. Brill, Leiden, pp. 85e104. Mann, R., Evans, D.A., 2004. Site selection for oyster habitat rehabilitation in the Virginia portion of the Chesapeake Bay: a commentary. J. Shellfish Res. 23, 41e49. Mann, R., Southworth, M., Harding, J.M., Wesson, J.A., 2009. Population studies of the native eastern oyster, Crassostrea virginica, (Gmelin, 1791) in the James River, Virginia, USA. J. Shellfish Res. 28, 193e220. bault, J., Jean, F., Marchais, V., Richard, J., Jolivet, A., Flye-Sainte-Marie, J., The Richard, P., Paulet, Y.-M., Clavier, J., Chauvaud, L., 2015. Coupling experimental and field-based approaches to decipher carbon sources in the shell of the great scallop, Pecten maximus (L.). Geochim. Cosmochim. Acta 168, 58e69. McConnaughey, T.A., Burdett, J., Whelan, J.F., Paull, C.K., 1997. Carbon isotopes in biological carbonates: respiration and photosynthesis. Geochim. Cosmochim. Acta 61, 611e622. McConnaughey, T.A., Gillikin, D.P., 2008. Carbon isotopes in mollusk shell carbonates. Geo-Mar. Lett. 28, 287e299. Miller, H.M., 1986. Transforming a “splendid and delightsome land”: colonists and ecological change in the Chesapeake 1607-1820. J. Wash. Acad. Sci. 76, 173e187. Mook, W., Tan, F., 1991. Stable carbon isotopes in rivers and estuaries. In: Degens, E.T., Kempe, S., Richey, J.E. (Eds.), Biogeochemistry of Major World Rivers. John Wiley, Hoboken, pp. 245e264. Nichols, M.M., 1972. Sediments of the James River estuary, Virginia. Geol. Soc. Am. Memoirs 133, 169e212. Pecoraro, L.J., 2015. “Mr. Gookin Out of Ireland, Wholly upon His Owne Adventure”: an Archaeological Study of Intercolonial and Transatlantic Connections in the Seventeenth Century. Dissertation. Boston University. , N., Labat, D., Zaehle, S., 2007. Piao, S., Friedlingstein, P., Ciais, P., de Noblet-Ducoudre Changes in climate and land use have a larger direct impact than rising CO2 on global river runoff trends. Proc. Natl. Acad. Sci. 104, 15242e15247. Poulain, C., Lorrain, A., Mas, R., Gillikin, D.P., Dehairs, F., Robert, R., 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, 75e82. Pritchard, D.W., 1952. Salinity distribution and circulation in the Chesapeake Bay estuarine system. J. Mar. Res. 11, 106e123. Raymond, P.A., Bauer, J.E., 2001a. Riverine export of aged terrestrial organic matter to the North Atlantic Ocean. Nature 409, 497e500. Raymond, P.A., Bauer, J.E., 2001b. Use of 14C and 13C natural abundances for evaluating riverine, estuarine, and coastal DOC and POC sources and cycling: a review and synthesis. Org. Geochem. 32, 469e485. Raymond, P.A., Bauer, J.E., Caraco, N.F., Cole, J.J., Longworth, B., Petsch, S.T., 2004. Controls on the variability of organic matter and dissolved inorganic carbon ages in northeast US rivers. Mar. Chem. 92, 353e366. Raymond, P.A., Bauer, J.E., Cole, J.J., 2000. Atmospheric CO2 evasion, dissolved inorganic carbon production, and net heterotrophy in the York River estuary. Limnol. Oceanogr. 45, 1707e1717. Raymond, P.A., Caraco, N.F., Cole, J.J., 1997. Carbon dioxide concentration and atmospheric flux in the Hudson River. Estuaries 20, 381e390. Reimer, P.J., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk Ramsey, C., Buck, C.E., Cheng, H., Edwards, R.L., Friedrich, M., 2013. IntCal13 and Marine13 radiocarbon age calibration curves 0-50,000 years cal BP. Radiocarbon 55, 1869e1887. Rhoads, D.C., Lutz, R.A., 1980. Skeletal Growth of Aquatic Organisms: Biological Records of Environmental Change. Plenum Press, New York. Rick, T., Henkes, G.A., 2014. Radiocarbon variability in Crassostrea virginica shells from the Chesapeake Bay. Radiocarbon 56, 305e311. Rick, T.C., Henkes, G.A., Lowery, D.L., Colman, S.M., Culleton, B.J., 2012. Marine radiocarbon reservoir corrections (DR) for Chesapeake Bay and the Middle Atlantic Coast of north America. Quat. Res. 77, 205e210. Rountree, H.C., Turner, E.R., 2002. Before and after Jamestown: Virginia's Powhatans and Their Predecessors. University Press of Florida, Gainesville. Russell, N., Cook, G.T., Ascough, P.L., Dugmore, A.J., 2010. Spatial variation in the marine radiocarbon reservoir effect throughout the Scottish post-Roman to late Medieval period: north Sea values (500-1350 BP). Radiocarbon 52, 1166e1181. Schmidt, D., Straube, B., 2012. James Fort structures. In: Kelso, W.M., Straube, B., Schmidt, D. (Eds.), 2007-2010 Interim Report on the Preservation Virginia Excavations at Jamestown, Virginia. Preservation Virginia, Richmond, pp. 29e41. Schmidt, G.A., Bigg, G.R., Rohling, E.J., 1999. Global Seawater Oxygen-18 Database v1.21. http://data.giss.nasa.gov/o18data/. Shumway, S.E., 1996. Natural environmental factors. In: Kennedy, V.S., Newell, R.I.E., Eble, A.F. (Eds.), The Eastern Oyster: Crassostrea Virginica. Maryland Sea Grant. College Park, pp. 467e513. Southworth, S., Brezinski, D.K., Orndorff, R.C., Repetski, J.E., Denenny, D.M., 2008. Geology of the Chesapeake and Ohio Canal National Historical Park and Potomac River Corridor, District of Columbia, Maryland, West Virginia. U.S. Geological Survey Professional Paper 1691. Spiker, E.C., 1980. The behavior of 14C and 13C in estuarine water: effects of in situ CO2 production and atmospheric exchange. Radiocarbon 22, 647e654. Stahle, D.W., Cleaveland, M.K., Blanton, D.B., Therrell, M.D., Gay, D.A., 1998. The lost Colony and Jamestown droughts. Science 280, 564e567. Stenzel, H.B., 1971. Oysters. In: Moore, R.C. (Ed.), Treatise on Invertebrate Paleontology. Part N. Mollusca 6. Bivalvia. Geological Society of America and the University of Kansas, Boulder, pp. 953e1184. Stuiver, M., Braziunas, T.F., 1993. Modeling atmospheric 14C influences and 14C ages of marine samples to 10,000 BC. Radiocarbon 35, 137e189.

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002

B.L. Grimm et al. / Quaternary Geochronology xxx (2017) 1e15 Stuiver, M., Pearson, G.W., Braziunas, T., 1986. Radiocarbon age calibration of marine samples back to 9000 Cal Yr BP. Radiocarbon 28, 980e1021. Stuiver, M., Polach, H.A., 1977. Reporting of 14C data - discussion. Radiocarbon 19, 355e363. Surge, D., Lohmann, K.C., Dettman, D.L., 2001. Controls on isotopic chemistry of the American oyster, Crassostrea virginica: implications for growth patterns. Paleogeogr. Paleoclimatol. Paleoecol. 172, 283e296. Tanaka, N., Monaghan, M.C., Rye, D.M., 1986. Contribution of metabolic carbon to mollusk and barnacle shell carbonate. Nature 320, 520e523. Taylor, A.J., Lai, C.-T., Hopkins, F.M., Wharton, S., Bible, K., Xu, X., Phillips, C., Bush, S., Ehleringer, J.R., 2015. Radiocarbon-based partitioning of soil respiration in an old-growth coniferous forest. Ecosystems 18, 459e470. Trumbore, S., 2000. Age of soil organic matter and soil respiration: radiocarbon constraints on belowground C dynamics. Ecol. Appl. 10, 399e411. Ulm, S., 2002. Marine and estuarine reservoir effects in central Queensland, Australia: determination of DR values. Geoarchaeology 17, 319e348. USGS, 2015. USGS Water Quality Data for Virginia. http://nwis.waterdata.usgs.gov/ va/nwis/qw. Vogel, J.S., Southon, J.R., Nelson, D.E., 1987. Catalyst and binder effects in the use of

15

filamentous graphite for AMS. Nucl. Instrum. Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 29, 50e56. Wang, Y., Huntington, T.G., Osher, L.J., Wassenaar L.I., Trumbore, S.E., Amundson, R.G., Harden, J.W., McKnight, D.M., Schiff, S.L., Aiken, G.R., Lyons, W.B., Aravena, R.O., Baron, J.S., 1998. Carbon cycling in terrestrial environments. In: Kendall, C., McDonnell, J.J. (Eds.), Isotope Tracers in Catchment Hydrology. Elsevier, Amsterdam, pp. 577e602. Wang, Z.A., Wanninkhof, R., Cai, W.-J., Byrne, R.H., Hu, X., Peng, T.-H., Huang, W.-J., 2013. The marine inorganic carbon system along the Gulf of Mexico and Atlantic coasts of the United States: insights from a transregional coastal carbon study. Limnol. Oceanogr. 58, 325e342. Ward, G.K., Wilson, S.R., 1978. Procedures for comparing and combining radiocarbon age determinations: a critique. Archaeometry 20, 19e31. Wennersten, J.R., 2001. The Chesapeake: an Environmental Biography. Maryland Historical Society, Baltimore. Yu, K.F., Hua, Q.A., Zhao, J.X., Hodge, E., Fink, D., Barbetti, M., 2010. Holocene marine 14 C reservoir age variability: evidence from 230Th-dated corals in the South China Sea. Paleoceanography 25, PA3205.

Please cite this article in press as: Grimm, B.L., et al., Seasonal radiocarbon reservoir ages for the 17th century James River, Virginia estuary, Quaternary Geochronology (2017), http://dx.doi.org/10.1016/j.quageo.2017.03.002