Timing of shell ring formation and patterns of shell growth in the sea scallop Placopecten magellanicus based on stable oxygen isotopes.
Article Type: Report
Subject: Scallops (Physiological aspects)
Shells (Structure)
Oxygen (Isotopes)
Oxygen (Research)
Authors: Chute, Antonie S.
Wainright, Sam C.
Hart, Deborah R.
Pub Date: 08/01/2012
Publication: Name: Journal of Shellfish Research Publisher: National Shellfisheries Association, Inc. Audience: Academic Format: Magazine/Journal Subject: Biological sciences; Zoology and wildlife conservation Copyright: COPYRIGHT 2012 National Shellfisheries Association, Inc. ISSN: 0730-8000
Issue: Date: August, 2012 Source Volume: 31 Source Issue: 3
Topic: Event Code: 310 Science & research
Product: Product Code: 0913070 Scallops NAICS Code: 114112 Shellfish Fishing SIC Code: 0913 Shellfish
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 303011387
Full Text: ABSTRACT The ratio of [sup.18]O to [sup.16]O in the shell material of bivalves depends on the ambient water temperature at the time the shell material was deposited. By analyzing samples of shell material taken sequentially from the umbo to the shell margin, we obtained the oxygen isotopic records from the shells of 14 sea scallops (Placopecten magellanicus) and compared the isotope data with the visible rings on the upper valve. Using generalized additive models, we show that ring formation is related significantly to water temperature, and that rings were typically laid down near the annual temperature maximum. Shell ring formation was generally annual, although 2 of the mid-Atlantic scallops appeared to have laid down 2 rings in 1 y. Some of the scallops appear to form new shell material and increase in shell height over the entire year for the first few years of life, and in later years reduce or halt accretion at the shell margin during the coldest temperatures. The isotopic records obtained from near the umbo of the shells suggest that all but one of the scallops were spawned in the fall.

KEY WORDS: Placopecten magellanicus, oxygen isotopes, annual rings, growth, scallop

INTRODUCTION

Visible bands of a different color or texture (rings) are often formed on the shells of bivalves when they undergo periods of modified shell deposition (Stevenson & Dickie 1954, Merrill et al. 1966, Haag & Commens-Carson 2008). Ring formation may coincide with the reallocation of resources to the gonads during spawning, or as the result of changes in metabolic processes because of environmental extremes. These hypotheses are not mutually exclusive because different mechanisms may be operating in different geographical areas, and possibly both may operate simultaneously or at different times during the life of the organism. Water temperatures, which can be associated with environmental stress, physiological processes, and spawning, may be particularly important in the formation and timing of rings (Naidu 1970, Jones & Quitmeyer 1996). False annual rings, or checks, may be formed as a result of physical trauma, such as having the shell or mantle damaged by contact with fishing gear (Merrill et al. 1966, Caddy 1989).

Results of previous studies support the hypothesis that rings on the upper valve of most sea scallops Placopecten magellanicus (Gmelin 1791) are formed once a year. The methods used to determine annual periodicity include visual examination of shells collected monthly in the Bay of Fundy (Stevenson & Dickie 1954), comparison of growth estimates from tagged scallops with those from shell rings (Posgay 1963, Merrill et al. 1966), and comparison of growth observed in size composition data from annual surveys to that inferred by shell rings (Hart & Chute 2009b). The assumption that the rings are true annuli allows for the estimation of growth curves based on the annual growth increment between rings or shell size at age analysis (e.g., Verrill 1897, Posgay 1979, Serchuk et al. 1979, MacDonald & Thompson 1985, Thouzeau et al. 1991, Hart & Chute 2009a). Recently, growth increments between the rings on sea scallop shells, assumed to be a 1-y increase in shell height, were used to estimate annual population growth for size-structured stock assessment models and to predict potential productivity of sea scallops (Northeast Fisheries Science Center (NEFSC) 2010)). Growth increment data are less prone to error than age estimates from counting rings and are particularly useful for sea scallops because the first 1 or 2 rings are sometimes obscure (Claereboudt & Himmelman 1996, Hart & Chute 2009b). However, both methods require the assumption of annual ring formation.

Another method that can be used to verify the annual formation of rings is the seasonal variation in the ratio of oxygen stable isotopes in the calcite of the shell itself. The ratio of the heavy-to-light oxygen isotopes in the shell material is determined predominantly by thermodynamic equilibrium between the [sup.18]O and [sup.16]O isotopes (which differ in their rotational energies) in calcite and seawater at the time of shell deposition, as described in Urey (1948) and Zeebe and WolfGladrow (2001). As a result, calcite that has been deposited at equilibrium with seawater has a higher [sup.18]O/[sup.16]O ratio (or [delta][sup.18]O when expressed relative to a standard) than that of seawater. Furthermore, the difference between [delta][sup.18][O.sub.calcite] and [delta][sup.18][O.sub.seawater] increases as temperature decreases. This temperature dependency was first described in an empirical equation (for use as a paleothermometer of minerals) by McCrae (1950). Modifications of McCrae's equation for use as a temperature recorder of biogenic minerals followed and are reviewed by Bemis et al. (1998).

Because the amount of [sup.18]O incorporated into the shell is temperature dependent, [delta][sup.18]O values from sequential calcite samples taken along the axis of growth of the shell produce a record of relative water temperature over the life of the scallop during the time it was depositing shell material. This information can be used to estimate the approximate time of year when the visible rings are formed as well as to infer the periods of time when the scallop is depositing shell material actively. Starting in the 1950s, stable oxygen isotope ratios from both fossil and recent mollusc shells have been used to recreate temperature time series (e.g., Urey et al. 1951, Epstein et al. 1953). Samples of shell material from long-lived animals, such as the ocean quahog Arctica islandica that regularly lives for more than 200 y, can provide historical ocean water temperature data (Weidman et al. 1994). Shell isotope records have been used to verify annual growth increments in several bivalve (Jones et al. 1983, Dare & Deith 1990, Brey & Mackensen 1997, Ivany et al. 2003, Lomovasky et al. 2007, Haag & Commens-Carson 2008), and univalve (Gurney et al. 2005, Naylor et al. 2007) mollusc species.

The 2 previous studies of P. magellanicus growth using stable oxygen isotopes analyzed 2 shells apiece. Krantz et al. (1984) analyzed shells collected off Virginia (37[degrees]15' N) and suggested that 2 visible rings were formed annually. Tan et al. (1988) analyzed scallops from Browns Bank off Nova Scotia (42[degrees]50' N) and suggested the formation of a single ring per year during the temperature minimum. Other studies of sea scallops from Nova Scotia have also indicated 1 annual ring is formed during the temperature minimum (Stevenson & Dickie 1954, Roddick et al. 1999). In this article, we examine oxygen isotopic records from 14 P. magellanicus shells collected from 8 locations off the eastern U.S. coast to determine the timing of ring formation and to investigate patterns of growth and spawning.

MATERIALS AND METHODS

The 14 sea scallops used for this study were collected live during the annual NEFSC summer scallop survey or, in 2 cases, from commercial scallop vessels. Scallops from 8 different sites throughout the sea scallop range in U.S. waters were selected for analysis (Table 1, Fig. 1). The collection sites were distributed latitudinally between the sampling locations of Krantz et al. (1984) off the coast of Virginia, 37[degrees]15' N), and Tan et al. (1988) Brown's Bank, Nova Scotia, 42[degrees]50' N). Many of the scallops were collected in different locations and years, which allows for the examination of the isotopic record under a variety of conditions. The collection sites were all in rotational fishing areas that have been closed and reopened periodically to scallop fishing, except for the Nantucket Lightship site, which has been closed to fishing since 1994. Five collection sites were in the Mid-Atlantic Bight (Hudson Canyon, Elephant Trunk, and Delmarva rotational areas), 1 was on Nantucket Shoals (Nantucket Lightship closed area) and 2 were on Georges Bank (closed areas I and II, Fig. 1) for a total of 8 collection sites. Four scallops each were analyzed from 2 sites (Nantucket Lightship and Elephant Trunk) to evaluate within-site variability and replicability, whereas a single scallop was analyzed at the other sites. The collection sites ranged in depth from 53 98 m, and had mostly sandy bottom with some gravel in the Georges Bank/Nantucket Shoals sites and some silt in the deeper mid-Atlantic sites. The scallops analyzed ranged in shell height (SH; body size measured from the umbo to the shell margin) from 101-137 mm, with a mean of 123 mm (Table 1).

Shells with the least infestation by boring organisms were selected for isotope analysis from among all shells collected at a site. The visual location of each ring on the shell was marked, and the distance to each ring from the umbo was measured in millimeters using a caliper. The shell was then scrubbed in warm water and allowed to dry completely. A dentist's drill was used to sample the top layer of calcite to a depth of approximately 0.5 mm, removing powdered calcite from lines ~1.5 cm in length perpendicular to the long axis of the shell and parallel to the shell margin at the time of deposition (Fig. 2). Samples were taken from the outside of the left (upper) valve, starting at the umbo and moving toward the shell margin. For the first few samples at the umbo, where the shell was very narrow, it was impossible to produce enough shell material to analyze by drilling a line, so a rounder area was excavated (these samples were so close to the umbo they are not visible in Fig. 2). The number of samples per shell ranged from 64-88, depending on the size of the shell.

For the isotopic analysis, subsamples of approximately 0.2-0.6 mg calcite powder were placed in 4-mL gas-tight screw-cap glass vials in a helium atmosphere. Phosphoric acid (0.5 mL; specific gravity, ~1.92) was added to the vials and allowed to react with the carbonate sample in a sonicator bath for 20 min, releasing C[O.sub.2] gas into the headspace. The headspace was sampled at room temperature with a Gilson gas autosampler through a rubber septum in the vial cap, and was swept with a stream of helium through a magnesium perchlorate water trap, concentrated chromatographically (Europa ANCA GSL), then passed to a Europa 20-20 stable isotope ratio mass spectrometer. The C[O.sub.2] from each sample was compared with 4 pulses of C[O.sub.2] from a reference tank calibrated previously versus a standard reference material (Pee Dee Belemnite, PDB; NBS-19). Scallops 9 and 10 were analyzed on a Thermo Advantage Delta V mass spectrometer with a Gas Bench II carbonate analysis system. Differences in protocol were as follows: A total of 0.2 mg was reacted with acid in 10-mL vials at 70[degrees]C rather than room temperature, without sonication, and the 4 reference peaks were compared with 4 sample peaks rather than 1 peak. Results from the 2 mass spectrometers, based on scallops from the same region, were comparable in every respect. The raw data from each sample consisted of the [sup.18]O/[sup.16]O of the sample C[O.sub.2] relative to that of the reference C[O.sub.2] pulses. The reference gas-corrected isotope ratios were then corrected for machine drift by analyzing NBS-19 or a working standard (chalk) after every 6th sample, and applying a linear correction factor to the samples in the order they were analyzed. Results are expressed in delta notation, relative to PDB:

[delta][sup.18]O = [R.sub.sa] - [R.sub.st]/[R.sub.st] x 1,000,

where [R.sub.sa] and [R.sub.st] are the [sup.18]O/[sup.16]O ratios of the sample and standard (PDB), respectively. The [delta][sup.18]O data for each shell were plotted as a function of distance from the umbo and were smoothed using a Henderson 7-term moving average appropriate for periodic data (Henderson 1916) as well as a weighted 3-point moving average smoother (with the middle point weighted twice that of the points on each side).

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

To test the relationship quantitatively between ring formation and temperature, we developed generalized additive models (GAMs). For an isotope sample at SH H, where H is larger than the SH of the first visible ring and less than the SH of the last visible ring, define h by h = (H - [H.sub.L])/([H.sub.U] - [H.sub.L]), where [H.sub.L] and [H.sub.U] are the SHs of the first visible ring larger and smaller than H, respectively. Thus, h is between 0 and 1 and indicates the relative position between the 2 adjacent visible rings. The GAMs estimate smoothed (by splines) functions s of h that predict the values of [delta][sup.18]O , i.e. [delta][sup.18]O = s(h). If the visible rings were laid down twice a year or at random times, there would be no clear annual cycle observed. We used as a null hypothesis that there was no relationship between the relative position h and temperature compared with the alternative hypothesis that [delta][sup.18]O is a smoothed function of h. Statistical significance was determined by a chi-square likelihood ratio test (Crawley 2007). Akaike's information criterion (AIC) and Akaike weights were used to evaluate further the relative strength of evidence for the alternative models (Burnham & Anderson 2002). The scallop isotope data for each region (Georges Bank including Nantucket Shoals, scallops 1-6; Mid-Atlantic Bight, scallops 7-14) were modeled separately.

We also used GAMs to test the idea that at least some scallops grow continuously while young, but then reduce or stop growing during the coldest temperatures as they get older.

For this purpose, we constructed a more complex GAM:

[delta][sup.18]O = s(h) + [a + b [(h - 1/2).sup.2]] x Age group (1)

where Age group is 0 for isotope samples collected at SHs less than the year 3 visual ring, and 1 for samples collected at SHs larger than this ring. If there is substantially less or no growth during the coldest months at older ages, then this model should have a lower AIC and have a higher Akaike weight than those from the model [delta][sup.18]O = s(h), which does not take into account changes in [delta][sup.18]O with age. The GAMs were fit using the statistical program R (R Development Core Team 2005) with the mgcv package (Wood 2006); the degree of smoothness was chosen automatically during the model-fitting process using a cross-validation procedure.

A 2-step procedure was used to estimate bottom water temperatures quantitatively at the time of deposition of each shell sample. First, [delta][sup.18][O.sub.seawater] was estimated using equations that relate [delta][sup.18][O.sub.seawater] to salinity. Fairbanks (1982) introduced equations specific to inshore Mid-Atlantic Bight water and Gulf of Maine water. The [delta][sup.18]O of seawater depends on its origin and whether it has been influenced primarily by rain or snow (in the form of glacial meltwater), because they are formed by processes that fractionate oxygen isotopes differently (Epstein & Mayeda 1953). We found that when we applied both equations to our data, the maximum difference between the 2 estimates of mean water temperature at deposition for any shell was 1.07[degrees]C, with an average of 0.66[degrees]C, and differences were both positive and negative. The conditions at our collection sites are probably influenced by several water types, but at the depths the shells were collected, much of the water likely originated in the Gulf of Maine. Thus, we used the GOM equation (Fairbanks 1982):

[delta][sup.18][O.sub.seawater] = 0.421S - 14.66, (2)

where [delta][sup.18][O.sub.seawater] is the per mil difference relative to SMOW and S is salinity in parts per thousand. The resulting [delta][sup.18][O.sub.seawater] was adjusted downward by 0.22[per thousand] so it could be expressed relative to PDB (Craig 1961). Then, the estimated [delta][sup.18][O.sub.seawater] relative to PDB and the observed [delta][sup.18][O.sub.shell] calcite were used to predict ambient water temperature at the time of calcite deposition using the equation from Epstein et al. (1953):

[T.sub.water] = 16.5 - 4.3([delta][sup.18][O.sub.shell calcite] - [delta][sup.18][O.sub.seawater])

+ 0.14[([delta][sup.18][O.sub.shell calcite] - [delta][sup.18][O.sub.seawater]).sup.2], (3)

where [T.sub.water] is measured in degrees Celsius and [delta][sup.18][O.sub.shell] calcite and [delta][sup.18][O.sub.seawater] are both relative to PDB. The estimated ambient water temperature during the time the scallop was accreting shell material can be used to approximate the time of year the scallop was actively growing.

[FIGURE 3 OMITTED]

The salinity values used to estimate [delta][sup.18][O.sub.seawater] in Eq (2) are mean annual values from the Marine Resources Monitoring, Assessment and Prediction (MARMAP) research cruises (Sherman 1980) that collected oceanographic data at fixed stations distributed over the continental shelf from Cape Hatteras to the Gulf of Maine during 1977 to 1987. Data from all MARMAP stations within 30 km and within 25% of the depth of the sites where the shells that were collected were used to calculate an annual mean salinity for the site. The same MARMAP data were used to construct mean annual temperature cycles for each of the sites. A mean bottom temperature was generated for every day of the year (Fig. 3) by fitting the MARMAP data using the methods of Mountain et al. (2004).

[FIGURE 4 OMITTED]

Predicted temperatures were then adjusted by the difference between the means of the MARMAP bottom temperature data for each site and the means of the predicted temperatures over approximately the first 2 annual cycles of growth for each scallop. For example, if the mean observed annual temperature from the MARMAP data was 2[degrees]C less than the mean predicted temperature from the shell samples, the predicted temperature values were reduced by 2[degrees]C. The adjustment procedure accounts for differences among locations that would be reflected ideally in Eq (3).

RESULTS

Isotopic Analysis

Annual cycles of [delta][sup.18]O are especially clear for the first ~100 mm of shell, corresponding to the first 3-4 y when growth is relatively fast and the annual increase in SH is greatest (Fig. 4). The sampling resolution is highest during the first few years as well because 25 or more samples can be taken in the space between the rings. The annual cycles shorten with time and represent a smaller fraction of the SH because scallop shells typically grow proportionately less in height each year after about 90 mm. The relative amplitude of the annual [delta][sup.18]O cycles appears to decline with age in some scallops (especially noticeable in shells 2, 4, 5, and 9; Fig. 4).

Typically, [delta][sup.18]O peaked around 2[per thousand] at the coldest time of the year and fell to about 0.25[per thousand] at the warmest. The range of all values was 2.5 to -1.5[per thousand]. These values are comparable with those from other isotopic analyses of P. magellanicus shells from the northwest Atlantic (Krantz et al. 1984, Tan et al. 1988). In the current study, shells collected from the same site in the same year tend to show a similar pattern of peaks, but their [delta][sup.18]O values are offset from one another (compare scallops 3 and 4 with scallop 5 from the Nantucket Lightship closed area; Fig. 4).

Comparison of Isotopic Analysis with Visible Rings

The location of the visible rings relative to the isotopic records of each scallop suggests that most rings are laid down once during the annual temperature cycle and are true annuli (Fig. 4). However, the isotopic record suggests that 2 of the Elephant Trunk scallops (shells 11 and 12) may have deposited 2 rings during 1 annual cycle (as inferred from the isotopes) in the last year of their lives. Scallop 10, also from the same site, had an extra visual ring at 95 mm that does not match the isotope annual cycle.

The plots of the ring locations over the [delta][sup.18]O data (Fig. 4) and the plots of [delta][sup.18]O as a function of the relative position of the sample between visible rings (Fig. 5) show a clear pattern of lower [delta][sup.18]O (and hence higher temperatures) near the rings, indicating that most of the rings are laid down near the temperature maximum, which on the northwest Atlantic shelf occurs during the late summer or autumn. The simple GAMs [delta][sup.18]O = s(h) were highly statistically significant (Fig. 5; P < 0.001 in both the Georges Bank and mid-Atlantic regions), indicating that mean temperatures were significantly warmer during ring formation than during the other portions of the year. They reduced AIC considerably compared with the null model (reduction of 45 AIC units for the Georges Bank scallops and 133 for the mid-Atlantic scallops), and the relative probabilities (Akaike weights) that the null model is true is less than 0.001 in both regions, again demonstrating strong evidence that mean temperatures are warmer during ring formation.

Predicted Temperature at the Time of Shell Deposition

Plots of the estimated bottom water temperatures as a function of the distance from the umbo were made for each shell (Fig. 6). Lines showing the mean annual maximum and minimum temperatures from the MARMAP data from each site were drawn on the graph for reference. The resulting graph can be used to estimate the portion of the annual temperature cycle during which the scallop was actively accreting shell material.

The patterns of annual predicted temperature cycles mirror annual cycles of [delta][sup.18]O. The water temperatures predicted from the [delta][sup.18]O data from the first 2 or 3 y of a scallop's life (SH, up to ~90 mm) generally reflected the range of the observed annual bottom water temperature at the location where the scallop was collected. The reduction in amplitude after approximately 100 mm in SH that can be seen in the [delta][sup.18]O record of shells 2, 4, 5, and 9 is also evident in the estimated temperature plots (Fig. 6). The plots of these shells suggest accretion of shell material occurs over the full range of the annual temperature cycle for the first 2 or 3 y, then slows or ceases during portions of the year.

[FIGURE 5 OMITTED]

The GAMs that allowed for changed patterns of growth (split model) for when the scallops were older than 3 y reduced the AIC substantially (by 7 AIC units) compared with the base model [delta][sup.18]O = s(h) for the Georges Bank scallops but increased AIC (by 4) for the mid Atlantic (Fig. 5C and D). Akaike weights give a relative 97% probability that the split model is true on Georges Bank, compared with 3% with the nonsplit model. By contrast, the split model has only a relative chance of 13% of being true in the mid Atlantic. This confirms the idea that reduced growth during the coldest months often occurs on Georges Bank, but does not appear to occur as a general pattern in the mid Atlantic.

Using the daily mean temperature estimates derived from the MARMAP data, it was possible to match observed temperatures with the water temperature predicted from the shell calcite samples. From this, it can be determined when shell growth was halted. For example, 2 of the 4 scallops (shells 4 and 5) collected from the Nantucket Lightship area seemed to show a pattern of temperature range truncation. Beyond 100 mm from the umbo, no samples from these shells gave a predicted water temperature less than 7.7[degrees]C. For that location, the MARMAP data show water temperatures are less than 7.7[degrees]C from about ordinal days 22-153, roughly mid January until the end of May. Scallop 2 from closed area I showed no growth below 4.9[degrees]C after an SH of 100 ram, which translates roughly to mid March to mid May. Scallop 10, from the Elephant Trunk closed area, showed no growth below 10.3[degrees]C after an SH of 63mm, corresponding to mid January to September. This scallop showed an interesting pattern in that the first 2 y and the last 3 y of shell growth appeared to occur during only slightly overlapping temperature ranges. The Elephant Trunk area has an unusual annual temperature cycle with a cooling period in the summer (Fig. 3); this scallop may have slowed or ceased growth at different times of year depending on its age.

[FIGURE 6 OMITTED]

DISCUSSION

Our results support the viability of using stable oxygen isotope analysis to study growth patterns in P. magellanicus. Growth and ring formation occur in response to a number of environmental and physiological stimuli. Populations that inhabit an extensive geographical area are necessarily exposed to different conditions depending on their location within that area, such as temperature extremes, food limitation, or seasonal spawning that serve as those stimuli. The results of the current study, combined with the 2 previous stable isotope studies of sea scallops (Krantz et al. 1984, Tan et al. 1988), demonstrate some of the different patterns of growth and ring formation that can be found in the population of P. magellanicus. A clinal effect in the timing and cause of external ring formation resulting from latitudinal differences has been noted or theorized in several bivalve species (Tan et al. 1988, Dare & Deith 1990, Jones & Quitmeyer 1996).

Our results indicate that most of the visual rings appear to have been laid down at or near the water temperature maximum. By contrast, there is evidence that sea scallops from the Bay of Fundy and Brown's Bank deposit rings during the temperature minimum (Stevenson & Dickie 1954, Tan et al. 1988). It is possible that environmental stress may be greater during the temperature minimum for the scallops collected at these more northern sites, leading to the observed difference in the timing of ring formation. However, Brown's Bank shares a similar temperature regime with much of the Georges Bank area, so there may be other distinctive oceanographic or biological factors present on Brown's Bank that influence sea scallop ring formation.

The data from this study suggest the possibility that some scallops from the southern portion of the Mid-Atlantic Bight (scallops 11 and 12 from this study) may, occasionally, form 2 visible rings per year. Assuming that this is correct, we can offer a hypothesis for these occurrences based on the unusual annual temperature cycle in the Elephant Trunk region where these scallops were collected. The area seems to experience a warming period around June, then cools again before reaching the temperature maximum in the fall (Fig. 3). The [delta][sup.18]O plots for shells 11 (SH, ~112 mm) and 12 (SH, ~85 mm) show evidence for small warming periods and ring formation during the spring or early summer, with second rings deposited nearer the temperature maximum that occurs during early November. Scallops from this site may lay down an extra ring in response to this warming period that may indicate a spring or early summer spawning event. Sea scallops progressively put more energy into spawning with age (Langton et al. 1987). It is possible that these scallops produced 2 rings during that year because the greater energy put into semiannual spawning altered their growth pattern. The observations of twice-annual ring formation in Krantz et al. (1984) may have been the result of a similar mechanism.

It is possible that scallop 10, collected in 2009, did not grow during the summer of 2006, when anomalously slow growth was observed in scallops from that area (Hart & Chute 2009b). If this is the case, the isotope record would be missing the high-temperature peak during that year, so that what appears to be 1 annual temperature cycle is actually 2 y, meaning that the visual ring record is correct and the scallop is 8 y old (assuming that the first visible ring occurred at year 2, as is often the case). The site where this scallop was collected has been sampled annually since 2003, when a very high density of juvenile scallops, likely all 2-y-olds, was recorded at this site (Hart & Chute 2009b). No evidence of new recruitment was observed in the following years, so it is highly probable that scallop 10 (as well as scallops 11-13 from the same site) was spawned in 2001. If scallop 10 deposited little new shell material during 2006 (SH, ~94-100 mm), then the annual temperature cycle in the isotope record would be dampened to the point where there was no visible peak and the scallop would appear to be younger that it actually is.

Our observations of growth cessation during some times of the year, as estimated here using predicted temperature at the time of shell deposition, agrees with previous studies showing that growth in SH occurs at different rates over the annual temperature cycle. Harris and Stokesbury (2006) analyzed the growth of tagged scallops from 2 sites in the Great South

Channel at about 41[degrees] N latitude. They found faster growth for scallops at a site where scallops were tagged in May than a site where scallops were tagged in September. Because many of the scallops were at large less than a year (or 1 y plus a fraction of a year), the May-tagged scallops were more likely to reflect a period of faster summer growth whereas the September-tagged scallops were more likely to reflect a time of slower winter growth.

This study suggests that at least some sea scallops may accrete shell material continuously throughout the entire year during their first few years of life, and then only part of the year when older. This pattern, during which shell accretion stops or slows during certain times of the year, leaving the [delta][sup.18]O values recorded by the shell in a reduced range of temperatures, is often found in bivalves (Stevenson & Dickie 1954, Jones et al. 1983, Harrington 1989, Dare & Deith 1990, Owen et al. 2002, Goodwin et al. 2003, Heilmayer et al. 2003, Ivany et al. 2003, Nakashima et al. 2004, Chauvaud et al. 2012). Postsettlement sea scallops are subject to high predation rates from numerous predators (Hart & Chute 2004, Hart 2006); rapid uninterrupted growth allows scallops to reach a partial size refuge from these predators (Jones et al. 1983). With age, scallops also put an increasing amount of their energy into reproduction (Thompson & MacDonald 2006), and thus may show a seasonal reduction in shell growth. For the individuals that appeared to undergo a change to seasonal growth with age, such as scallops 2, 4, 5, and 9, the change did not appear until between 80 mm and 100 mm in SH. The largest shell we analyzed for this study was 137 mm. Future investigations analyzing older, larger scallops, concentrating on achieving higher sampling resolution toward the shell margin, could elucidate patterns of intermittent growth.

The GAMs suggest that older scallops from Georges Bank were more likely to cease shell deposition during the coldest time of year than those collected from the Mid-Atlantic Bight. Primary production is higher in the Mid-Atlantic Bight than Georges Bank during the fall and winter (Yoder et al. 2002, Schofield et al. 2008). This pattern may affect the timing of both spawning and shell growth. As the water temperature decreases in the fall, the mid-Atlantic scallops may have a better supply of food than scallops on Georges Bank, may be more likely to continue growth through the winter, and may be able to prepare for spring spawning. Consistent with this idea, Brust et al. (2001) found that scallops collected off Virginia at 38[degrees] N latitude grew at the highest rates during November to April. This may be why the spring spawn is often dominant in the mid Atlantic, whereas scallops spawn primarily in the fall on Georges Bank (Posgay & Norman 1958, DuPaul et al. 1989, Schmitzer et al. 1991, Dibacco et al. 1995).

The [delta][sup.18]O plots may also be able to provide an indication of when a scallop was spawned. For most scallops sampled in this study, the [delta][sup.18]O data indicate increasing ambient water temperature during early growth, with a temperature maximum occurring when the scallop is between 10 mm and 20 mm in SH. Growth of early postsettlement sea scallops has been estimated at between 0.047-0.199 mm/day (Wildish & Saulnier 1992), 0.032-0.057 mm/day (Parsons et al. 1993), and 0.008-0.028 mm/day (Milke et al. 2004). At an average daily growth rate of 0.04 mm/day, a scallop will have grown to about 13 mm 1 y after spawning (allowing for a 35-day planktonic period), so that it is likely these scallops reached their first birthday at or near the temperature maximum. Thus, the temperature patterns observed for the first year of growth for most of the scallops is consistent with a fall spawn date. Scallop 14, the most southerly scallop of our samples, appears to be an exception to this pattern, with apparently cooling temperatures as the scallop grew between 7 mm and 20 mm in SH. A scallop that was spawned in May would reach 7 mm approximately 6 mo later in November at roughly the temperature maximum (assuming a slightly faster growth rate of 0.05 mm/day, consistent with the higher temperatures at this site). This scallop, therefore, may have been spawned in the spring.

Although the growth rings on many sea scallops are not difficult to locate visually, there is a degree of subjectivity involved with identifying rings on shells that introduces some uncertainty into inferences of scallop growth based on shell ring analysis. For example, there are often subrings between the major rings, and the rings on pure white scallops (such as scallop 6) and the outermost rings on older scallops appear as changes in texture of the shell rather than color. It is possible another age reader might have interpreted at least some of our rings differently. However, we have used other methods to verify that the growth inferred by our reader (A. S. C.) matches well with observed scallop growth (Hart & Chute 2009b).

The MARMAP ocean water temperature data we used to estimate annual temperature ranges for our sampling sites were collected from 1977 to 1987, and the mean daily temperatures are the mean for that time period. The use of these data represents a source of uncertainty, but the MARMAP data set was the best available series of year-round oceanographic measurements for our collection sites, because sampling occurred during every month of the year for 11 y. The average values probably represent a compressed version of the actual annual temperature range in any given year at a specific location. There is also evidence of a recent warming trend in the years since these data were collected (NEFSC Ecosystem Assessment Program 2009).

Our study was intended to examine the utility of oxygen stable isotopes for verification of annuli in scallops living over a range of latitudes, depths, and years. Our results confirm the utility of oxygen stable isotope ratios as a tool for annulus verification and elucidation of growth patterns throughout a range of environmental conditions, and encourage follow-up studies to test for specific patterns in the timing of ring deposition, growth, and spawning.

ACKNOWLEDGMENTS

We thank Maureen Taylor for calculating the annual temperature and salinity cycles for our collection sites, and David Rudders and the crew and scientists aboard R/V Albatross IV, R/V Hugh R. Sharp, and F/V Celtic for collecting the scallops used for this study. Stephen Artabane and Courtney Higgins assisted in the mass spectrometry lab. This work benefited from discussions with Olaf Heilmayer, Stephen Howe, Linda Ivany, Cynthia Jones, Burton Shank, and Chris Weidman. Special thanks to Larry Jacobson for his advice and input, to Jay Burnett, and to an anonymous reviewer for providing valuable comments and suggestions. We are also grateful to Dr. Neil Ringler for his donation of a dentist's drill.

LITERATURE CITED

Bemis, B. E., H. J. Spero, J. Bijma & D. W. Lea. 1998. Reevaluation of the oxygen isotopic composition of planktonic foraminifera: experimental results and revised paleotemperature equations. Paleoceanography 13:150-160.

Brey, T. & A. Mackensen. 1997. Stable isotopes prove shell growth bands in the Antarctic bivalve Laternula elliptica to be formed annually. Polar Biol. 17:465-468.

Brust, J. C., W. D. DuPaul & J. E. Kirkley. 2001. The effects of a regulatory gear restriction on the recruiting year class in the sea scallop, Placopecten magellanicus (Gmelin, 1791), fishery. J. Shellfish Res. 20:1035-1041.

Burnham, K. P. & D. R. Anderson. 2002. Model selection and multimodel inference: a practical information-theoretic approach. New York: Springer-Verlag. 488 pp.

Caddy, J. F. 1989. A perspective on the population dynamics and assessment of scallop fisheries, with special reference to sea scallop, Placopecten magellanicus (Gmelin). In: J. F. Caddy, editor. Marine invertebrate fisheries: their assessment and management. New York: Wiley-Interscience. pp. 559-589.

Chauvaud, L., Y. Patry, A. Jolivet, E. Cam, C. Le Goff, O. Strand, G. Charrier, J. Thebault, P. Lazure, K. Gotthard & J. Clavier. 2012. Variation in size and growth of the great scallop Pecten maximus along a latitudinal gradient. PLoS ONE 7:e37717.

Claereboudt, M. R. & J. H. Himmelman. 1996. Recruitment, growth and production of giant scallops (Placopecten magellanicus) along an environmental gradient in Baie des Chaleurs, eastern Canada. Mar. Biol. 124:661-670.

Craig, H. 1961. Standard for reporting concentrations of deuterium and oxygen-18 in natural waters. Science 33(3467):1833-1834.

Crawley, M. J. 2007. The R book. New York: Wiley. 950 pp.

Dare, P. J. & M. R. Deith. 1990. Age determination of scallops, Pecten maximus (Linnaeus, 1758), using stable oxygen isotope analysis, with some implications for fisheries management in British waters. In: S. E. Shumway & P. A. Sandifer, editors. An international compendium of scallop biology and culture. Baton Rouge, LA: World Aquaculture Society. pp. 118-133.

Dibacco, C., G. Robert & J. Grant. 1995. Reproductive cycle of the sea scallop, Placopecten magellanicus (Gmelin, 1791), on northeastern Georges Bank. J. Shellfish Res. 14:59-69.

DuPaul, W. M., J. E. Kirkley & A. C. Schmitzer. 1989. Evidence of a semiannual reproductive cycle for the sea scallop, Placopecten magellanicus (Gmelin, 1791), in the mid-Atlantic region. J. Shellfish Res. 8:173-178.

Epstein, S., R. Buchsbaum, H. A. Lowenstam & H. C. Urey. 1953. Revised carbonate-water isotopic temperature scale. Bull. Geol. Soc. Am 64:1315-1326.

Epstein, S. & T. Mayeda. 1953. Variation of [O.sup.18] content of waters from natural sources. Geochim. Cosmochim. Acta 4:213-224.

Fairbanks, R. G. 1982. The origin of continental shelf and slope water in the New York Bight and Gulf of Maine: evidence from [H.sub.2.sup.18]O/[H.sub.2.sup.16]O ratio measurements. J. Geophys. Res. C Oceans 87:5796-5808.

Goodwin, D. H., B. R. Schone & D. L. Dettman. 2003. Resolution and fidelity of oxygen isotopes as paleotemperature proxies in bivalve mollusk shells: models and observations. Palaios 18:110-125.

Gurney, L. J., C. Mundy & M. C. Porteus. 2005. Determining age and growth of abalone using stable oxygen isotopes: a tool for fisheries management. Fish. Res. 72:353-360.

Haag, W. R. & A. M. Commens-Carson. 2008. Testing the assumption of annual shell ring deposition in freshwater mussels. Can. J. Fish. Aquat. Sci. 65:493-508.

Harrington, R. J. 1989. Aspects of growth deceleration in bivalves: clues to understanding the seasonal [sup.18]O and [sup.13]C record: a comment on Krantz et al. (1987). Palaeogeogr. Palaeoclimatol. Palaeoecol. 70:

399-407. Harris, B. P. & K. D. E. Stokesbury. 2006. Shell growth of sea scallops (Placopecten magellanieus) in the southern and northern Great South Channel, USA. ICES J. Mar. Sci. 63:811-821.

Hart, D. R. 2006. Effects of sea stars and crabs on sea scallop Placopecten magellanicus recruitment in the Mid-Atlantic Bight (USA). Mar. Ecol. Prog. Ser. 306:209-221.

Hart, D. R. & A. S. Chute. 2004. Essential fish habitat source document: sea scallop, Placopecten magellanicus, life history and habitat characteristics, 2nd edition. NOAA technical memorandum NMFS NE-189. U.S. Department of Commerce, Woods Hole, Massachusetts. 21 pp.

Hart, D. R. & A. S. Chute. 2009a. Estimating von Bertalanffy growth parameters from growth increment data using a linear mixed-effects model with an application to the sea scallop Placopecten magellanicus. ICES J. Mar. Sci. 66:2165-2175.

Hart, D. R. & A. S. Chute. 2009b. Verification of Atlantic sea scallop (Placopecten magellanicus) shell growth rings by tracking cohorts in fishery closed areas. Can. J. Fish. Aquat. Sci. 66:751-758.

Heilmayer, O., T. Brey, M. Chiantore, R. Cattaneo-Vietti & W. E. Arntz. 2003. Age and productivity of the Antarctic scallop, Adamussium colbecki, in Terra Nova Bay (Ross Sea, Antarctica). J. Exp. Mar. Biol. Ecol. 288:239-256.

Henderson, R. 1916. Note on graduation by adjusted average. Trans. Am. Soc. Actuaries 17:43-48.

Ivany, L. C., B. H. Wilkinson & D. S. Jones. 2003. Using stable isotopic data to resolve rate and duration of growth throughout ontogeny: an example from the surf clam, Spisula solidissima. Palaios 18:126-137.

Jones, D. S. & I. R. Quitmeyer. 1996. Marking time with bivalve shells: oxygen isotopes and season of annual increment formation. Palaios 11:340-346.

Jones, D. S., D. F. Williams & M. A. Arthur. 1983. Growth history and ecology of the Atlantic surf clam, Spisula solidissima (Dillwyn), as revealed by stable isotopes and annual shell increments. J. Exp. Mar. Biol. Ecol. 73:225-242.

Krantz, D. E., D. S. Jones & D. F. Williams. 1984. Growth rates of the sea scallop, Placopecten magellanicus, determined from the [sup.18]O/[sup.16]O record in shell calcite. Biol. Bull. Mar. Biol. Lab. Woods Hole 167:186-199.

Langton, R. W., W. E. Robinson & D. Schick. 1987. Fecundity and reproductive effort of sea scallops Placopecten magellanicus from the Gulf of Maine. Mar. Ecol. Prog. Ser. 37:19-25.

Lomovasky, B. J., T. Brey, A. Baldoni, M. Lasta, A. Mackensen, S. Campodonico & O. Iribarne. 2007. Annual shell growth increment formation in the deepwater Patagonian scallop Zygoehlamys patagonica. J. Shellfish Res. 26:1055-1063.

MacDonald, B. A. & R. J. Thompson. 1985. Influence of temperature and food availability on the ecological energetics of the giant scallop Placopecten magellanicus: I. Growth rates of shell and somatic tissue. Mar. Ecol. Prog. Ser. 25:279-294.

McCrae, J. M. 1950. On the isotopic chemistry of carbonates and a paleotemperature scale. J. Chem. Phys. 18:849-857.

Merrill, A. S., J. A. Posgay & F. E. Nichy. 1966. Annual marks on shell and ligament of sea scallop (Placopecten magellanicus). Fish Bull. 65:299-311.

Milke, L. M., V. M. Bricelj & C. C. Parish. 2004. Growth of post-larval sea scallops, Placopecten magellanicus, on microalgal diets, with emphasis on the nutritional role of lipids and fatty acids. Aquaculture 234:293-317.

Mountain, D. G., M. H. Taylor & C. Bascunan. 2004. Revised procedures for calculating regional average water properties for Northeast Fisheries Science Center cruises. U.S. Department of Commerce, Northeast Fisheries Science Center ref. doc. 04-08. U.S. Department of Commerce, Woods Hole, Massachusetts. 62 pp.

Naidu, K. S. 1970. Reproduction and breeding cycle of the giant scallop Placopecten magellanicus (Gmelin) in Port au Port Bay, Newfoundland. Can. J. Zool. 48:1003-1012.

Nakashima, R., A. Suzuki & T. Watanabe. 2004. Life history of the Pliocene scallop Fortipecten, based on oxygen and carbon isotope profiles. Palaeogeogr. Palaeoclimatol. Palaeoecol. 211:299-307.

Naylor, J. R., B. M. Manighetti, H. L. Neil & S. W. Kim. 2007. Validated estimation of growth and age in the New Zealand abalone Haliotis iris using stable oxygen isotopes. Mar. Freshw. Res. 58:354-362.

Northeast Fisheries Science Center (NEFSC). 2010. 50th Northeast Regional Stock Assessment Workshop (50th SAW) Assessment Report. U.S. Department of Commerce, Northeast Fisheries Science Center ref doc. 10-17. U.S. Department of Commerce, Woods Hole, Massachusetts. 844 pp.

NEFSC Ecosystem Assessment Program. 2009. Ecosystem assessment report for the Northeast U.S. Continental Shelf large marine ecosystem. U.S. Department of Commerce, Northeast Fisheries Science Center ref doc. 09-11. U.S. Department of Commerce, Woods Hole, Massachusetts. 34 pp.

Owen, R., H. Kennedy & C. Richardson. 2002. Isotopic partitioning between shell calcite and seawater: effect of shell growth rate. Geochim. Cosmochim. Acta 66:1727-1737.

Parsons, G. J., S. M. Robinson, J. C. Roff & M. J. Dadswell. 1993. Daily growth rates as indicated by valve ridges in postlarval giant scallop (Placopecten magellanicus) (Bivalvia: Pectinidae). Can. J. Fish. Aquat. Sci. 50:456-464.

Posgay, J. A. 1963. Tagging as a technique in population studies of the sea scallop. ICNAF Spec. Publ. 4:268-271.

Posgay, J. A. 1979. Population assessment of the Georges Bank sea scallop stocks. Rapp. P.-V. Reun. CIEM 175:109-113.

Posgay, J. A. & K. D. Norman. 1958. An observation on the spawning of the sea scallop, Placopecten magellanicus (Gmelin), on Georges Bank. Limnol. Oceanogr. 3:478.

R Development Core Team. 2005. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. 409 pp.

Roddick, D., E. Kenchington, J. Grant & S. Smith. 1999. Temporal variation in sea scallop (Placopecten magellanicus) adductor muscle RNA/DNA ratios in relation to gonosomatic cycles, off Digby, Nova Scotia. J. Shellfish Res. 18:405-413.

Schmitzer, A. C., W. D. Dupaul & J. E. Kirkley. 1991. Gametogenic cycle of sea scallops Placopecten magellanicus Gmelin 1791 in the Mid-Atlantic Bight. J. Shellfish Res. 10:221-228.

Schofield, O., R. Chant, B. Cahill, R. Castelao, D. Gong, A. Kahl, J. Kohut, M. Montes-Hugo, R. Ramadurai, P. Ramey & S. Glenn. 2008. The decadal view of the Mid-Atlantic Bight from the COOLroom: is our coastal system changing? Oceanography (Wash. D.C.) 21:108-117.

Serchuk, F. M., P. W. Wood, J. A. Posgay & B. E. Brown. 1979. Assessment and status of sea scallop (Placopecten magellanicus) populations off the northeast coast of the United States. Proc. Natl. Shellfish. Assoc. 69:161-191.

Sherman, K. 1980. MARMAP: a fisheries ecosystem study in the northwest Atlantic: fluctuations in the ichthyoplankton--zooplankton component and their potential for impact on the system. In: F. P. Diemer, F. J. Vernberg, D. Z. Mierkes & W. Belle, editors. Advanced concepts in ocean measurement for marine biology. Columbia, SC: Baruch Institute of Marine Biology and Coastal Research, University of South Carolina Press. pp. 9-37.

Stevenson, J. A. & L. M. Dickie. 1954. Annual growth rings and rate of growth of the giant scallop, Placopecten magellanicus (Gmelin) in the Digby area of the Bay of Fundy. J. Fish. Res. Bd. Canada 11:660-671.

Tan, F. C., D. Cai & D. L. Roddick. 1988. Oxygen isotope studies on sea scallops, Placopecten magellanicus, from Browns Bank, Nova Scotia. Can. J. Fish. Aquat. Sci. 45:1378-1386.

Thompson, R. J. & B. A. MacDonald. 2006. Physiological integrations and energy partitioning. In S. E. Shumway & G. J. Parsons, editors. Scallops: biology, ecology, and aquaculture, 2nd edition. Amsterdam: Elsevier. pp. 493-516.

Thouzeau, G., G. Robert & S. J. Smith. 1991. Spatial variability in distribution and growth of juvenile and adult sea scallops Placopecten magellanicus (Gmelin) on eastern Georges Bank (Northwest Atlantic). Mar. Ecol. Prog. Ser. 74:205-218.

Urey, H. C. 1948. Oxygen isotopes in nature and in the laboratory. Science 108:489-496.

Urey, H. C., H. A. Lowenstam, S. Epstein & C. R. McKinney. 1951. Measurement of paleotemperatures and temperatures of the upper cretaceous of England, Denmark, and the southeastern United States. Bull. Geol. Soc. Am 62:399-416.

Verrill, A. E. 1897. A study of the family Pectinidae, with a revision of the genera and subgenera. Trans. Conn. Acad. Arts Sci 10: 41-95.

Weidman, C. R., G. A. Jones & K. C. Lohmann. 1994. The long-lived mollusk Arctica islandica: a new paleoceanographic tool for the reconstruction of bottom temperatures for the continental shelves of the northern North Atlantic Ocean. J. Geophys. Res. C Oceans 99):18305-18314.

Wildish, D. J. & A. M. Saulnier. 1992. The effect of velocity and flow direction on the growth of juvenile and adult giant scallops. J. Exp. Mar. Biol. Ecol. 155:133-143.

Wood, S. N. 2006. Generalized additive models: an introduction with R. Boca Raton, FL: Chapman & Hall/CRC. 416 pp.

Yoder, J. A., S. E. Schollaert & J. E. O'Reilly. 2002. Climatological phytoplankton chlorophyll and sea surface temperature patterns in continental shelf and slope waters off the northeast U.S. Coast. Limnol. Oceanogr. 47:672-682.

Zeebe, R. E. & D. Wolf-Gladrow. 2001. C[O.sub.2] in seawater: equilibrium, kinetics, isotopes. New York: Elsevier. 346 pp.

ANTONIE S. CHUTE, (1) * SAM C. WAINRIGHT (2) AND DEBORAH R. HART (1)

(1) Northeast Fisheries Science Center, 166 Water Street, Woods Hole, MA 02543; (2) Department of Science, U.S. Coast Guard Academy, 27 Mohegan Avenue, New London, CT 06320-8101

* Corresponding author. E-mail: toni.chute@noaa.gov

DOI: 10.2983/035.031.0308
TABLE 1.
Dates and locations of stations where scallops were collected.

Scallop   Shell Height
No.           (mm)       Date Collected          Source

1             125             2005        Commercial vessel
2             137           8/3/2005      NEFSC scallop survey
3             121          8/14/2003      NEFSC scallop survey
4             135          8/14/2003      NEFSC scallop survey
5             132          8/14/2003      NEFSC scallop survey
6             135          8/11/2005      NEFSC scallop survey
7             116          7/14/2005      NEFSC scallop survey
8             119          7/20/2005      NEFSC scallop survey
9             lot           6/9/2009      NEFSC scallop survey
10            111          5/13/2009      NEFSC scallop survey
11            127          7/15/2005      NEFSC scallop survey
12            111          7/15/2005      NEFSC scallop survey
13            114            6/2006       Commercial vessel
14            136          7/17/2006      NEFSC scallop survey

Scallop   Decimal     Decimal      Area
No.       Latitude   Longitude   Collected   Depth (m)

1          41.07      -66.89       CAII         72
2          41.19      -68.91        CAI         98
3          40.57      -69.72        NLS         63
4          40.57      -69.72        NLS         63
5          40.57      -69.72        NLS         63
6          40.57      -69.70        NLS         62
7          39.10      -73.09       HCCA         72
8          39.02      -73.54       HCCA         53
9          38.76      -73.24       HCCA         80
10         38.56      -73.79        ET          60
11         38.55      -73.81        ET          58
12         38.55      -73.81        ET          58
13         38.56      -73.80        ET
14         37.53      -74.61        DMV         60

* Data unavailable.

CAI, closed area I (Georges Bank); CAII, closed area II (Great
South Channel); DMV, Delmarva closed area; ET, Elephant Trunk
closed area; HCCA, Hudson Canyon closed area; NEFSC, Northeast
Fisheries Science Center; NLS, Nantucket Lightship closed area.
See Figure 1 for locations of closed areas.
Gale Copyright: Copyright 2012 Gale, Cengage Learning. All rights reserved.