Comparative sclerochronology of modern and mid-Pliocene (c. 3.5 Ma) Aequipecten opercularis (Mollusca, Bivalvia): an insight into past and future climate change in the north-east Atlantic region

Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

Contents lists available at ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p a l a e o

Comparative sclerochronology of modern and mid-Pliocene (c. 3.5 Ma) Aequipecten opercularis (Mollusca, Bivalvia): an insight into past and future climate change in the north-east Atlantic region Andrew L.A. Johnson a,⁎, Jonathan A. Hickson a, Annemarie Bird a, Bernd R. Schöne b, Peter S. Balson c, Timothy H.E. Heaton d, Mark Williams c,e a

Geographical, Earth and Environmental Sciences, School of Science, University of Derby, Derby DE22 1GB, UK Earth System Science Research Centre, Department of Applied and Analytical Paleontology (INCREMENTS), Institute of Geosciences, University of Mainz, 55128 Mainz, Germany c British Geological Survey, Keyworth, Nottingham NG12 5GG, UK d NERC Isotope Geosciences Laboratory, British Geological Survey, Keyworth, Nottingham NG12 5GG, UK e Department of Geology, University of Leicester, Leicester LE1 7RH, UK b

a r t i c l e

i n f o

Article history: Received 25 June 2009 Received in revised form 19 September 2009 Accepted 22 September 2009 Available online 3 October 2009 Keywords: Sclerochronology Climate Pliocene North Sea Aequipecten Arctica

a b s t r a c t Records of environment contained within the accretionarily deposited tissues of fossil organisms afford a means of detailed reconstruction of past climates and hence of rigorous testing of numerical climate models. We identify the environmental factors controlling oxygen and carbon stable-isotopic composition, and microgrowth-increment size, in the shell of modern examples of the Queen Scallop, Aequipecten opercularis. This understanding is then applied in interpretation of data from mid-Pliocene A. opercularis from eastern England. On the basis of oxygen-isotope evidence we conclude that winter minimum seafloor temperature was similar to present values (typically 6–7 °C) in the adjacent southern North Sea and that summer maximum seafloor temperature was a few degrees lower than present values (typically 16–17 °C). This contrasts with evidence from other proxies that winter and summer temperatures were higher than present. The pattern of seasonal variation in microgrowth-increment size suggests the existence of intense thermal stratification in summer. We therefore conclude that summer surface temperatures were much higher (maxima well over 20 °C) than those recorded isotopically on the seafloor and that the annual range of surface temperature (probably over 14 °C) was greater than now at the times in the mid-Pliocene when the investigated A. opercularis were alive. Taken in conjunction with other proxy evidence of warmer winters as well as summers, the data point to substantial fluctuation (up to 10 °C) in winter minimum temperatures during the mid-Pliocene in the northeast Atlantic region. This fluctuation may be attributable to variation in the strength of the Gulf Stream/North Atlantic Drift. Since the Pliocene has been widely used as a test-bed for numerical models of a greenhouse Earth, the results have implications for prediction of future climate in the north-east Atlantic region under the influence of anthropogenic global warming. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Sclerochronology (Buddemeier and Maragos, 1974) is the study of time-series data contained within the accretionarily deposited mineral tissues of plants (e.g. the skeletal materials of certain multicellular algae) and animals (e.g. the shells of many invertebrates). As in the longer-established sister-field of dendrochronology (involving records in organic tissue: wood), the size of increments is an important datum. However, to a much greater extent than in dendrochronology, geochemical time-series (e.g. of stable-isotopic composition or traceelement concentration) are used in sclerochronology, a reflection of the relative immunity of mineral materials from post-mortem chem⁎ Corresponding author. Tel.: +44 1332 591721; fax: +44 1332 597747. E-mail address: [email protected] (A.L.A. Johnson). 0031-0182/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.palaeo.2009.09.022

ical alteration. The time-series data facilitate both measurement of elapsed time and identification of environmental changes during the ontogeny/astogeny of the individual/colony concerned. Dependent on the taxon involved and the techniques used, both the chronological precision and the range of environmental parameters documented may be much greater than from dendrochronological data (e.g. Richardson, 2001; Schöne et al., 2002; Schöne and Surge, 2005; Hallmann et al. 2009). There have been numerous studies of marine environmental conditions involving time-series of growth-increment size and stableisotope ratios (18O/16O, 13C/12C) from bivalve molluscs (e.g. Richardson, 2001; Schöne and Surge, 2005). Here we review and substantially supplement such data for modern examples of the Queen Scallop, Aequipecten opercularis (Fig. 1), discussing what environmental variables may be reflected therein. We then undertake a similar exercise

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

165

Fig. 1. Right valve of Pliocene Aequipecten opercularis (UD 52795) with enlargement showing microgrowth increments of varying size. Scale bar = 10 mm. Growth-increment and isotopic data from this specimen are shown in Fig. 8 and Tables 3 and 4.

for representatives from the mid-Pliocene Coralline Crag Formation of Suffolk, eastern England, a deposit which accumulated under greenhouse conditions shortly before the onset of major northern-hemisphere glaciation. On the basis of the understanding obtained from study of modern A. opercularis, and of a test of the Coralline Crag results employing the bivalve Arctica islandica, we interpret aspects of the local mid-Pliocene marine environment and subsequently relate these findings to regional and global circumstances. We conclude that in the greenhouse world of the mid-Pliocene there was substantial variation in oceanic heat supply to the north-east Atlantic region. This may give an indication of the future behaviour of the Gulf Stream/ North Atlantic Drift in response to global warming. 2. Methods 2.1. Stable isotopes Hickson et al. (1999, 2000) and Johnson et al. (2000) describe methods used in earlier stable-isotope work on A. opercularis. They extracted samples at c. 1 mm intervals through shell ontogeny, omitting the first c. 10 mm of growth (representative of a few months) because of the difficulty of obtaining sufficiently large, temporally well-resolved samples. For the first 1–2 years of (rapid) growth this sampling protocol achieves at least a monthly temporal resolution, sufficient to characterise seasonal fluctuations in isotopic composition. The supplementary data reported here (seven further profiles from Pliocene shells, one from a modern shell and a single nearmargin value from another modern shell) were obtained following essentially the same procedures but with analysis conducted partly at the NERC Isotope Geosciences Laboratory (NIGL), Keyworth, UK (VG Isocarb + Optima system), and partly at the Institute of Geosciences (IGF), University of Frankfurt, Germany (Finnigan MAT 253 continuous-flow mass spectrometer equipped with a GasBench II). The relevant laboratory is indicated in the text. 18O/16O and 13C/12C ratios are reported as δ18Oshell and δ13Cshell values versus VPDB by compar-

ison with within-run laboratory standards calibrated against NBS-19 and NBS-18, with a precision (1 SD) typically b0.1‰ for both δ18O and δ13C. In conformity with earlier published work on A. opercularis calcite, estimated temperatures from δ18O values were calculated using the equation of O'Neil et al. (1969), with 0.26‰ subtracted from the water value (δ18Owater; measured versus VSMOW) to compare with VPDB (Coplen et al., 1983). The Ar. islandica specimen from the Coralline Crag Formation used to test the results from A. opercularis was sectioned and sampled across the third annual increment by means of a series of immediately adjacent 300 µm-diameter boreholes equidistant from the outer surface. Being representative of the early, relatively rapid phase of growth (Schöne et al. 2005), the third annual increment offered the prospect of high temporal resolution in sampling. Estimated temperatures for the aragonite mineralogy were calculated using the equation of Grossmann and Ku (1986), modified so that water values relate to VPDB rather than VSMOW (Schöne et al., 2005). 2.2. Growth increments Although we did record the positions of growth-breaks (‘growthrings’) marking the boundaries of major increments, our study of A. opercularis focused on microgrowth increments on the outer shell surface (Fig. 1). In contrast to previous work employing a microscope fitted with an eyepiece graticule (Johnson et al., 2000), increments were measured on digital images using Panopea© (2004, Peinl and Schöne) software, the shells having been coated before photography with a sublimate of NH4Cl to improve the clarity of increments. A. opercularis valves are of low convexity so measurement of 2D-images of the surface introduces only a trivial error. Some shells (those analysed at NIGL) were isotopically sampled before growth-increment analysis. Of these, some could no longer be measured, while others had to be measured anterior or posterior of the sampling grooves and a correction factor (the ratio of shell height measured along the midline to that measured along the relevant anterior or

166

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

posterior trajectory) applied to render measurements equivalent to those taken along the midline (also the axis to which isotope data were referred). For the Ar. islandica specimen we simply recorded the positions of the characteristic autumn growth-breaks bounding the annual increments (cf. Schöne et al., 2005). The individual was 34 years old at death.

to the local environment. The lower mean value of δ13C in 20th century specimens (−0.04‰) compared to that from pre-industrial Holocene specimens (+0.79‰) can reasonably be ascribed to the ‘Suess Effect’: the addition of 12C-rich carbon to the atmosphere and oceans by human activities, such as the burning of fossil fuels (Hickson et al., 2000; Johnson et al., 2000). 3.2. Variation in microgrowth-increment size

3. Stable-isotope and microgrowth-increment sclerochronology of modern Aequipecten opercularis We take ‘modern’ here to include subfossil Holocene material, whose environment was certainly very similar to that of other investigated shells live-collected from the area concerned (southern North Sea). The brief review of stable-isotope data is principally to illustrate the temperature dependence of δ18O in A. opercularis and hence the potential not only for determining temperature per se but also for marking off seasons and thus years during ontogeny. Existing δ18O information is then used to provide an independent temporal framework for discussion of some of the new microgrowth-increment data, and new δ18O information is used to test an explanation proposed for microgrowth-increment variation in wild individuals. 3.1. Variation in δ18O and δ13C Serial ontogenetic sampling of A. opercularis cultured in monitored, semi-natural conditions in the southern North Sea, and of wild forms from the same area for whose lifespans instrumental records were available, revealed cycles in δ18Oshell which are clearly related to seasonal temperature variation (Hickson et al. 1999). Observed δ18Oshell minima were close to those predicted for the summer temperature maximum (at the relevant value for δ18Owater), showing that absolute summer temperatures can be read from shell data, provided δ18Owater is known or can be reliably inferred. Maxima of δ18Oshell in the wild individuals were somewhat lower than predicted for the winter temperatures (and δ18Owater) experienced. However, the shells concerned had prominent winter ‘growth-rings’, indicating a slow-down and, perhaps, cessation of growth. Investigated subfossil Holocene shells (965–2535 years BP) from the southern North Sea generally lacked such growth interruptions and exhibited δ18Oshell maxima close to what would be expected for the coldest period of the winter (Hickson et al., 2000; Johnson et al., 2000). It is clear therefore that A. opercularis can provide an essentially faithful and full record of winter as well as summer temperature. This is only obtainable from early ontogeny because of a general slow-down in growth thereafter, leading to time-averaging of samples and typically a trend towards lower annual range in δ18Oshell (Hickson et al., 2000; see also Goodwin et al. 2003). In order to obtain a realistic estimate of seasonal temperature range it is obviously preferable to select shells whose early growth lacked interruptions, whether in reaction to seasonal temperature changes or to traumatic events such as storms and attempted predation. Ontogenetic variation in δ13Cshell of southern North Sea A. opercularis (cultured and wild, including subfossil) is less systematic than variation in δ18O and more difficult to explain (Hickson et al., 1999, 2000; Johnson et al., 2000). Some shells exhibited a positive correlation between δ13C and δ18O, a pattern observed in A. opercularis from the English Channel by Heilmayer et al. (2004) and attributed by them to seasonal (hence temperature- and δ18Oshell-related) fluctuations in planktonic productivity. This is implausible because higher planktonic productivity in spring and summer should lead to a concentration of 13C in the environmental ‘pool’ and hence a negative correlation between δ13Cshell and δ18Oshell. The observed co-variation cannot be explained as a direct temperature effect because temperature has little or no impact on carbon-isotope fractionation (Romanek et al., 2002). However, it could well be a consequence of temperature-controlled respiration on the seafloor, which would affect the rate of return of 12C-rich organic carbon

3.2.1. Cultured shells While there have been a number of studies of variation in the annual growth increment within modern A. opercularis (e.g. Taylor and Venn, 1978; Richardson et al., 1982; Heilmayer et al., 2004), the only well-controlled experimental investigation of microgrowth-increment variation published hitherto is by Broom and Mason (1978). Their data (1978, tables 1 and 2) indicate a correlation between microgrowth-increment size and growth-rate, with increments being deposited approximately daily at times of high growth-rate. Gruffydd (1981) and Owen et al. (2002) noted a similar correlation between increment size and growth rate in the Great Scallop, Pecten maximus. Broom and Mason's experiment involved specimens about six months in age, cultured for a further year (February 1975 to February 1976) in two settings: an outdoor tank with a continuous supply of unfiltered seawater and a cage suspended at a depth of 2 m in Langstone Harbour, Hampshire, UK. Specimens in the former setting showed a spring increase in microgrowth-increment size to an early summer maximum, followed by a minor autumn peak (Fig. 2A); those in the latter setting also showed early summer and autumn peaks, the autumn peak being 63% higher in this case (Fig. 2B). No individuals

Fig. 2. Microgrowth-increment size (blue), temperature (orange) and chlorophyll ‘a’ concentration (green) from two experimental cultures of A. opercularis conducted in southern England (created from data in Broom and Mason, 1978). No individuals grew after 19 November 1975 in either culture, hence the dashed line representing a decline in increment height to zero by this date, and no growth thereafter. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

grew between mid-November and mid-February in the second year of life, hence microgrowth-increment size can be taken as zero over this interval (dashed lines in Fig. 2). Broom and Mason regarded food availability in the form of standing phytoplankton crop as the dominant single control on growth (hence microgrowth-increment size). Notwithstanding some mismatches (Fig. 2) between the patterns of phytoplankton abundance (measured as chlorophyll a) and increment size, perhaps explained by the additional and at times dominant influence of temperature (Broom and Mason, 1978; Johnson et al., 2000), growth was clearly affected by planktonic food supply in these experimental settings. Johnson et al. (2000) measured microgrowth-increment size in A. opercularis which had been reared in artificial culture from very early in ontogeny: for about a further 1.5 years in Scotland and then for four months (October 1994 to February 1995) in suspended nets in the southern North Sea. They recognised (2000, fig. 5a–d) a seasonal pattern of increment size variation similar to that identified by Broom and Mason (1978) for the same interval of ontogeny (first to second winter). We remeasured (using Panopea©) increment size over the period of North-Sea culture in six second-year (1+) shells and identified an overall decrease in four of them, in accordance with the results of Johnson et al. (2000, fig. 5a–d). Seven first-year (0+) shells (not previously investigated) all showed an overall decrease in increment size over the period of North-Sea culture (Fig. 3). Oxygenisotope samples extracted 1 mm from the left and right ventral margins of the specimen (UD 53357) which had grown the most in culture (19 mm) signify a temperature of 8.3 °C (Fig. 3; analysis at NIGL). By comparison with data from an adjacent temperature-logger (Hickson, 1997, fig. 5.2.3), this suggests formation of the shell material in late December and hence that the individual continued to grow into January. It is clear, therefore, that in suspended culture first-year (0+) shells may grow deep into the winter but form microgrowth increments of diminishing size. Since chlorophyll a concentration changed little over the course of the North-Sea experiment (Hickson, 1997, fig. 5.2.5), it might be inferred that temperature was an important control on increment size. However, multiple upward excursions in increment size (Fig. 3) within the context of what appears to have been a fairly uniform temperature reduction from 12° to 6 °C over the course of the experiment (Hickson, 1997, fig. 5.2.3), argue against this. It is more likely that these upward excursions reflect autumn blooms of particular, highly nutritive phytoplankton species and that the overall

167

trend to increment-size reduction reflects a decline in frequency of such blooms and their ultimate cessation with the onset of winter. Phytoplankton dynamics in the southern North Sea are consistent with this explanation (several taxa have autumn blooms; Reid et al., 1990) and the disproportionate autumn increases in increment size relative to chlorophyll a concentration shown by Broom and Mason's data (Fig. 2) add credence to the notion that the type of food available, as well as its quantity, has an influence on increment size. Whatever the precise cause, it can be concluded that in the circumstances of the experimental cultures described above, increment size variation shows a broadly seasonal pattern, with smaller (or no) microgrowth increments in the winter months. 3.2.2. Wild (subfossil Holocene) shells We had at our disposal three live-collected indigenous shells from the southern North Sea which had been isotopically analysed by Hickson et al. (1999). Microgrowth increments were, however, difficult to measure on these, partly due to effacement in isotopic sampling and partly due to poor definition originally. In order to characterise the increment patterns of wild shells from the southern North Sea we therefore investigated four late Holocene specimens (965–2535 years BP) for which isotopic data, and hence a chronological framework in terms of seasonal temperature change, were available (Hickson et al., 2000). Summary increment statistics are given in Table 1, and increment and isotope profiles in Fig. 4. While some shells (Fig. 4C, D) show slight seasonal variation in increment size, the dominant pattern, sustained through the year, is of low-amplitude, high-frequency fluctuation with a period in the order of one month, judging from the number of cycles between successive summers or winters (Fig. 4). This sustained, short-period fluctuation can have nothing to do with temperature or the availability/type of planktonic food, the factors inferred to control increment size in the experimental cultures discussed above. Factors varying on this short timescale, and likely (cf. Johnson et al., 2007) to influence the growth of scallops in their natural seafloor habitat, are the availability of benthic food (living or dead organic material suspended by wave or current action) and turbidity (likewise increased by wave or current action but to the detriment of growth). Given what seems to be an approximately monthly timescale for this pattern, and the pronounced tidality of the southern North Sea, variation in tidal current strength seems a likely controlling influence, as inferred by Clark (2005) for fortnightly fluctuations in microgrowthincrement size of Pliocene scallops from the Gulf of California. The longer-period pattern in the southern North Sea could reflect the occurrence there of more extreme spring tides, and hence stronger tidal currents, in every other spring/neap cycle (i.e. every lunar month; Proudman Oceanographic Laboratory 2009). Whether or not this is the determining factor, the above data indicate that microgrowthincrement size in modern, natural populations of A. opercularis in the southern North Sea is not significantly influenced by the factors which control increment size in experimental cultures, as identified above. The dominant influence of planktonic food availability and type in these unnatural settings very probably reflects substantial removal Table 1 Maximum, minimum, range and mean of microgrowth-increment size for late Holocene A. opercularis from the Southern North Sea Basin. MEAN (bold figures) = average for column. Details of specimens in Fig. 4. BGS = British Geological Survey, Keyworth.

Fig. 3. Variation in microgrowth-increment size (five-point moving averages) of seven 0+ (first-year) A. opercularis specimens (UD 53357–53363) collected as spat close to Ardtoe (western Scotland), transported by car to Barry (south Wales), maintained in a tank on-board ship for about one week, and then deployed in nets in the southern North Sea from 7 October 1994 to c. 30 January 1995. Star marks position of oxygen isotope samples (right and left valves of UD 53357; N4/B1 of Hickson, 1997, Appendix A) indicating a temperature of 8.3 °C (δ18Owater: + 0.16‰; equation of O'Neil et al., 1969); this is representative of late December conditions (Hickson, 1997, fig. 5.2.3). Strong upward excursions in increment size within the first half of post-deployment growth (‘autumn’) probably reflect increased food availability from phytoplankton blooms.

BGS number

Zt 9952 Zt 9953 Zt 9955 Zt 9957 MEAN

Microgrowth-increment height (mm) Maximum (fivepoint averages)

Minimum (fivepoint averages)

Range (fivepoint averages)

Mean (raw data for midline)

0.643 0.322 0.401 0.375 0.435

0.420 0.187 0.194 0.199 0.250

0.223 0.135 0.207 0.176 0.185

0.501 0.251 0.276 0.279 0.327

168

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

Fig. 4. Microgrowth-increment (blue), and oxygen (red) and carbon (black dash) stable-isotope data for late Holocene A. opercularis from the Southern North Sea Basin. (A)–(D): BGS Zt 9952 (1430 years BP), Zt 9953 (965 years BP), Zt 9955 (2535 years BP) and Zt 9957 (1845 years BP), respectively. Raw increment data shown by dashed line, five-point moving averages by continuous bold line; positions of growth breaks shown by triangles (large where prominent, small where faint). Ages and isotope data from Hickson et al. (2000); anomalous isotope data from the sample at 30 mm height in Zt 9952 (A) have been excluded because they almost certainly reflect some sampling or instrumental aberration (Hickson et al., 2000, p. 6). Note that the isotope axes are reversed such that more depleted δ18O values, representative of higher temperatures, plot towards the top. All the shells show at least one complete cycle of seasonal variation in δ18O/temperature. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

from (suspended cages/nets) or the complete absence of (tank) a sedimentary substratum, hence reducing or excluding the influence of benthic food supply and turbidity. 3.2.3. Test of explanation for microgrowth-increment variation in wild A. opercularis If the above explanation for modern increment patterns in the southern North Sea is true, it should be the case that the range of variation in increment size is comparably small in other settings where the sea-floor is frequently disturbed by current or wave action. We tested this proposition by an investigation of increment-size variation in 27 modern, wild A. opercularis from a variety of tidal, depth and climatic settings on the European and African margins of the North Atlantic Ocean (including northern North Sea) and Mediterranean Sea. The range of ontogenetic variation in increment size (Table 2; fivepoint averages) is rarely more than 0.2 mm in Atlantic forms but nearly always more than 0.2 mm in the Mediterranean Sea, with an average of over 0.3 mm amongst specimens from the Gulf of Tunis. With the marginal exception of Madeira (borderline meso-microtidal), all the Atlantic settings have intermediate to large tidal ranges, hence tidal currents must operate. Wave action must also disturb the seafloor in the shallower settings. The Mediterranean settings (microtidal) have no significant tidal currents and in the fairly deep (50 m) context of the Gulf of Tunis the seafloor must be relatively immune from wave action during the summer because of the development of a steep thermocline (see temperature/depth information in Table 2) and hence pronounced stratification of the water-column. The proposition is therefore supported: increment-size variation is relatively small where the sea-floor is frequently disturbed by current or wave action.

The large increment-size range of Mediterranean A. opercularis, and particularly of Gulf-of-Tunis forms, cannot relate to temperature variation: Gulf-of-Tunis shells (average increment-range: 0.306 mm) experienced a seasonal temperature range (2 °C) much less than shells with a smaller increment-size range (e.g. those from Falmouth: average increment-range 0.153 mm; temperature range 6 °C). We tested whether the large increment-size range of Gulf-of-Tunis forms reflects seasonal variation in phytoplankton productivity by deriving an indication of the seasonal timing of shell deposition from the δ18O profile of one specimen (Fig. 5; analysis at IGF). Within this temporal context it is clear that the maximum and minimum values in the corresponding increment-size profile relate, respectively, to autumn and winter. Zarrad et al. (2006) have shown that phytoplankton productivity in the Gulf of Tunis is low in autumn and rises substantially in winter. The availability of phytoplankton cannot therefore lie behind incrementsize variation in A. opercularis from the Gulf of Tunis. We suggest that the pattern there may be a reflection of strong thermal stratification in summer (Fig. 5), as a result of which seafloor ventilation was poor and aerobic consumption of dead plankton limited. With the breakdown of stratification and also greater wave energies associated with autumn storms the seafloor became oxygenated and the rich accumulation of detrital food particles was suspended, both conditions being conducive to growth of A. opercularis. If this explanation for Gulf-of Tunis forms is true then they constitute an ‘exception proving the rule’; that is, they bear out the notion, advanced initially for representatives from the southern North Sea, that disturbance of the seafloor (less frequent in the Gulf of Tunis than in the current-swept southern North Sea) is a fundamental control on growth of wild A. opercularis. If the explanation is true, it would also indicate that it is frequent suspension of benthic food particles (beneficial) rather than of sediment

Table 2 Maximum, minimum, range and mean of microgrowth-increment size for modern, wild A. opercularis from various contexts of tidal regime, sea-bed depth and temperature/depth profile in winter (February) and summer (August). MEAN (bold figures) = average for locality and column, for localities from which multiple specimens were measured. Depth information was only available for some localities. Temperatures were derived by interpolating between isotherms for mean monthly temperature (NOAA, 1994) and are shown in italic where somewhat uncertain (for shells from depths greater than 75 m an additional figure is shown, representing temperature at a depth close to that of the seafloor). Tidal ranges (means for spring tides): microtidal b2 m, mesotidal 2–4 m, macrotidal N4 m. MNHN = Muséum Nationale d'Histoire Naturelle, Paris; NHM = Natural History Museum, London; UD = University of Derby (School of Science). Microgrowth-increment height (mm)

Temperature (°C) Accession number

NHM 20080479/1 NHM 20080479/2

Macro

NHM 1887.9.25.70

Meso/macro

NHM 1988.3.25.45 UD 53350 NHM 20031101/1

Depth (m)

February 0m

August

50 m

75 m

0m

7

7

7

13.5

46

9

9

9

Meso

110–113

9

9

Meso/macro Meso

109

9.5 10

9 10

50 m

75 m

9 9 at 100 m 9 10 10 at 100 m

UD 53351 UD 53352

Meso/macro

NHM 20080483/1 NHM 20080483/2

Meso

UD 53353 UD 53354 UD 53355

Meso

MEAN Morocco (30°32′ N, 09°51′ W) Madeira/Porto Santo

MNHN IM-2008-1547 NHM 20080480

Modern, Mediterranean Malaga, Spain

MNHN IM-2008-1543

MEAN Landes, France MEAN La Coruña, Galicia, Spain

10

0.211 0.199 0.205

0.118 0.168 0.143

0.269 0.271 0.270

14

11

10

0.383

0.189

0.194

0.292

13.5

11

0.264

0.124

0.140

0.173

16 16

11.5 12

10 10 at 100 m 10 11 11 at 100 m

0.309 0.349

0.156 0.132

0.153 0.217

0.220 0.225

0.156 0.144 0.126 0.181 0.154 0.167 0.127 0.147 0.151 0.150 0.135 0.145 0.156 0.129

0.157 0.187 0.143 0.162 0.153 0.206 0.139 0.173 0.185 0.191 0.130 0.169 0.250 0.200

0.225 0.225 0.203 0.255 0.229 0.257 0.199 0.228 0.250 0.225 0.196 0.223 0.239 0.207

17 16 at 10 m

12.5

12

12

19.5

14

12

13

13

18

15

13

16 15

16 15

16 15

22 21

18 16

16 15

0.313 0.331 0.269 0.343 0.306 0.373 0.266 0.320 0.336 0.341 0.265 0.314 0.406 0.329

16

16

16

22.5 19 at 30 m

17

16

0.320

0.089

0.231

0.200

0.389 0.279 0.329 0.461 0.359 0.582 0.467 0.721 0.542 0.682 0.383 0.377 0.541

0.182 0.108 0.126 0.198 0.141 0.220 0.186 0.299 0.201 0.270 0.202 0.202 0.235

0.207 0.171 0.203 0.263 0.218 0.362 0.281 0.422 0.341 0.412 0.181 0.175 0.306

0.299 0.168 0.222 0.320 0.249 0.348 0.306 0.490 0.327 0.446 0.296 0.301 0.372

10

10

9

12

12

20–23

13

Meso Meso/micro

73

Micro

20–40

MEAN Western Gulf of Tunis, Tunisia

MEAN

MNHN IM-2008-1533 MNHN IM-2008-1534 MNHN IM-2008-1535

Micro

MNHN IM-2008-1537 MNHN IM-2008-1538 MNHN IM-2008-1539 MNHN IM-2008-1540 MNHN IM-2008-1541

Micro

50

Mean (raw data for midline)

0.329 0.367 0.348

MNHN IM-2008-1544 MNHN IM-2008-1545 MEAN La Franqui, Roussillon, France

Range (five-point averages)

8

NHM 20031101/2 MEAN Falmouth, Devon, England

Minimum (five-point averages)

9

MEAN Modern, Atlantic s.s. Loch Leven, Highland region, Scotland Cumbrae, Firth of Clyde, Scotland Cardigan Bay, Wales Western Approaches, Ireland/ England (50°38′N 08°04′W)

Maximum (five-point averages)

12.5

12.5

13

22

15.5

13.5

14

15

15

26

17

15

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

Modern, northern North Sea Firth of Forth, Scotland

Tidal regime

169

170

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

up to 7.5 m thick, principally comprising carbonate silts and muddy, fine-medium carbonate sands; the Sudbourne Member, up to 12 m thick, unconformably overlying the Ramsholt Member and principally comprising strongly cross-bedded, well sorted carbonate sands; and the Aldeburgh Member, up to 13 m thick, unconformably overlying the Ramsholt Member towards the north-east of the main onshore outcrop and principally comprising somewhat finer, less well sorted and less strongly cross-bedded carbonate sands than in the Sudbourne Member, hence perhaps a time-equivalent unit formed further offshore. The stratigraphic setting of the Formation is illustrated in Fig. 6.

Fig. 5. Microgrowth-increment (blue), and oxygen (red) and carbon (black dash) stableisotope data for a modern specimen of A. opercularis from 50 m water depth in the Gulf of Tunis, Mediterranean Sea (MNHN IM-2008-1537). See Fig. 4 for further details concerning symbols and axes. Monthly mean water temperature data for this location and depth (derived from NOAA, 1994) are included in boxes adjacent to appropriate sections of the δ18O curve. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

(detrimental) which accounts for the short-period fluctuation in microgrowth-increment size in wild modern forms from the southern North Sea. Similar explanations for growth-rate variation, invoking changes in lateral advection of food, have been applied to wild populations of the scallop Flabellipecten stearnsii (Clark, 2005) and the infaunal bivalve Arctica islandica (Witbaard, 1996; Schöne et al., 2003). 4. Stable-isotope and microgrowth-increment sclerochronology of Coralline Crag (mid-Pliocene) Aequipecten opercularis Johnson et al. (2000) reported stable-isotope data from two A. opercularis specimens from the mid-Pliocene Coralline Crag of Suffolk, eastern England. The δ18O profiles showed cyclical fluctuations similar to those seen in modern specimens but calculated temperature maxima and minima were rather lower than indicated by other evidence (e.g. the taxonomic composition of the biota). On the basis of initial microgrowth-increment data, taken to indicate higher temperatures, it was speculated that the isotopic composition of the shells had been altered diagenetically, albeit with the retention of a seasonal cyclicity. Williams et al. (2009) recalculated isotopic palaeotemperatures using slightly different values for δ18Owater but the discrepancy with other temperature estimates remained. In the following we review the context of Coralline Crag deposition (including other temperature estimates), evaluate the preservation of A. opercularis, provide extensive new isotopic and growth-increment information, and interpret this in the light of the understanding obtained from investigation of modern A. opercularis (Section 3) and of a test involving Arctica islandica. 4.1. Context 4.1.1. Location The Coralline Crag Formation has an essentially continuous, elongate outcrop extending 25 km south-west from near Aldeburgh, Suffolk (Fig. 6), accompanied by a number of small outliers located up to a further 55 km west–south-west of the main outcrop. The formation also extends beneath the North Sea for some 14 km north-east of Aldeburgh (Balson, 1992) and constitutes part of the sedimentary fill of the Southern North Sea Basin. 4.1.2. Stratigraphy The Coralline Crag is divided into three members (Balson et al., 1993; Balson, 1999): the Ramsholt Member, of widest occurrence and

4.1.3. Biotic and sedimentological evidence of environment The biota of the Ramsholt Member is far from restricted (see below) so the quiet-water environment evinced by sediment grainsize probably reflects a relatively deep offshore setting rather than a protected nearshore one. A depth in excess of 50 m has been inferred from the assemblage of foraminifera (Hodgson and Funnell, 1987). This is consistent with evidence of a well-developed thermocline supplied by the diverse coccolith flora (Jenkins and Houghton, 1987). The higher-energy environment represented by the coarser, strongly cross-bedded sediments of the Sudbourne and Aldeburgh members may simply reflect shallowing, but it could instead (or in addition) relate to increased tidality (and hence tidal-current strength) consequent upon development of a link between the Southern North Sea and Channel basins, thus causing greater exposure to Atlantic influence. Such a link across south-east England has been proposed by several authors (e.g. Funnell, 1996; Westaway et al., 2001), but others favour the palaeogeography represented in Fig. 6 for the duration of Coralline Crag deposition (Murray, 1992). All three members of the Coralline Crag contain a diverse, marine biota. Elements of the mollusc, bryozoan, ostracod, foraminiferal and dinoflagellate components are generally indicative of water temperatures higher for at least part of the year than in the presentday southern North Sea, which is a shallow, well-mixed (thermally uniform) water-body with a typical winter minimum temperature of 6– 7 °C and a typical summer maximum of 16–17 °C (Hickson, 1997, fig. 4.3). For example, warm-water molluscan elements in the Coralline Crag suggest summer temperatures in excess of 20 °C for three-four months (Raffi et al., 1985); species of the bryozoans Cupuladria and Metrarabdotos imply minimum (winter) seafloor temperatures of, respectively, 12 and 16 °C (Lagaaij, 1963; Cheetham, 1967); ostracod assemblages suggest a winter seafloor minimum of 11 °C and summer seafloor maximum of 18 °C (Wood et al., 1993; A. Wood, pers. comm. 2007; see also Wilkinson, 1980); and while the benthic foraminifer assemblage (Murray, 1987) implies cool-temperate conditions (at least six months below 10 °C according to the definition of Hall, 1964) the pelagic assemblage suggests temperatures between 10° and 18 °C over the course of the year (Jenkins et al. 1988); finally, elements of the pelagic dinoflagellate biota (Head, 1997, 1998) indicate conditions were warm temperate or even subtropical (no months cooler than 10 °C, four months above 20 °C according to Hall, 1964). In addition to these estimates based on the composition of the biota, Strauch (1968) derived a winter temperature of 13.5 °C and summer temperature of 24 °C from analysis of the size of the bivalve Hiatella. Furthermore, investigations of zooid-size variation in bryozoans have yielded estimates of summerwinter range of c. 7 °C (O'Dea and Okamura, 2000) and c. 8 °C (Knowles et al., 2009). The bryozoan-based estimates of annual temperature range are less than the modern figure of c. 9–11 °C for the southern North Sea but in accordance with most of the annual ranges for the Pliocene implied by constituents of the biota. The notable exception is the much greater annual range deduced from molluscs by Raffi et al. (1985), who, as well as determining a summer temperature substantially higher than now, inferred a winter temperature much like the present.

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

171

Fig. 6. Pliocene palaeogeography in the vicinity of the Southern North Sea and Channel basins (adapted from Murray, 1992, map NG1; closely stippled areas are land). Boxed area expanded in upper-left inset, showing principal onshore outcrop of the Coralline Crag Formation and the locations of sample sites (1 = The Cliff; 2 = Broom Pit; 3 = The Firs). Lowerright inset shows stratigraphical setting of the Coralline Crag Formation.

4.1.4. Age The Coralline Crag was formerly thought to have been deposited during the period of peak Pliocene warmth (3.29–2.97 Ma; Dowsett et al., 1996, 1999) but it is now taken to be somewhat older, no less than 3.4 Ma and possibly over 4 Ma (Williams et al., 2009; Wood et al. 2009). It nevertheless falls within an interval which is considered to have been globally warm (Dowsett, 2007; Naish et al., 2009). 4.2. Material, shell preservation and sampling We investigated nine shells, including the two analysed by Johnson et al. (2000). Specimens UD 53312–53315 were collected by PSB from the Ramsholt Member of The Cliff, Gedgrave (Fig. 6; Ordnance Survey grid reference TM 397486). Specimens UD 52795–7 and 53349 were collected by Mr J. Swan from spoil on the floor of Broom Pit, Gedgrave (Fig. 6; TM 406500); based on their preservation and our knowledge of faunal distribution at this location we are certain that they were derived from the Ramsholt Member rather than the overlying Sudbourne Member. Specimen UD 53316 was collected by PSB from the Sudbourne Member of The Firs, Sudbourne (Fig. 6; TM 431535). The adherent matrix on this was difficult to remove, requiring use of an air-abrasive tool to prepare the shell for isotopic sampling. This procedure damaged the outer surface, making it impossible to measure microgrowth increments. Aragonite preservation is excellent in at least the Ramsholt Member (Balson, 1983) so there was no reason other than the ‘anomalous’

isotopic palaeotemperatures previously obtained to suspect that the calcitic shells of A. opercularis might be altered. Scanning-electronmicroscope investigation of two shells confirmed pristine preservation of shell structure (Fig. 7B, C) but revealed in one that the outermost 0.5 mm contained narrow (?cyanobacterial) borings, occurring more frequently towards the outer surface. Some of these were empty; others had been thinly lined by dog-tooth calcite (Fig. 7D, E), in some cases remarkably iron-rich (up to 12 wt.% as determined by Energy Dispersive X-Ray Analysis of individual crystals). Isotopic sampling of the bored shell (and of six others; all analysed at NIGL) was to a depth of about 0.5 mm so the material analysed probably incorporated only a minute proportion of diagenetic calcite. Two additional shells (UD 52795 and 53349; analysed at IGF) were drilled more shallowly, to a depth of about 0.05 mm; the samples from these would have incorporated about 5% diagenetic material if the areas concerned were similarly bored. 4.3. Calculation of isotopic palaeotemperatures A value for δ18Owater is required in the isotopic temperature equation (see Section 2.1). Insertion of an appropriate value is obviously critical to accurately estimate temperature. Some previous workers (e.g. Krantz, 1990; Buchardt and Simonarson, 2003; Goewert and Surge, 2008) have assumed somewhat negative values for North Atlantic seawater in the Pliocene, but recent modelling reported by Williams et al. (2005), taking into account such factors as evaporation

172

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

Fig. 7. SEM micrographs of shell material from the auricles of modern (A, UD 53356) and Coralline Crag (B, UD 53312; C–F, UD 53314) A. opercularis. A–C show equivalent preservation of internal foliated shell structure in the modern and Coralline Crag shells; D (same specimen as in C) shows narrow borings extending from the outer surface (left) and thinly lined with diagenetic calcite (seen enlarged in E); F shows borings from the inner surface (bottom right) of the same specimen which are devoid of diagenetic calcite.

and precipitation as well as a lower global ice volume than present, yields positive values in many situations. Dr. A.M. Haywood (University of Leeds, UK) has kindly supplied estimates for the specific location of the Coralline Crag using the models employed by Williams et al. (2005). These range from +0.1‰ (e.g. Atlantic-calibrated xbooa model) to + 0.5‰ (e.g. salinity-calibrated xatsb model). We have used these positive values in combination with the − 0.2 and −0.5‰ estimates of Buchardt and Simonarson (2003) for the extremes of Pliocene global average δ18Owater in order to generate a range of palaeotemperature estimates from our shell data. It should be noted that the effect of using lower/more negative values for δ18Owater is to reduce palaeotemperature estimates.

4.4. Results 4.4.1. δ18O profiles With the exception of the one shell from the Sudbourne Member, all δ18O values are in the positive range (Table 3). All profiles show a cyclical pattern (Figs. 8 and 9), broadly comparable to that seen in modern material (e.g. Fig. 4) but with a reduced amplitude and typically reduced wavelength. Based on the number of cycles of isotopic variation and a sensible estimate of the amount of time represented

by the earliest, unsampled part of shell ontogeny, it would seem that none of the shells was more than three years old at death. The unsampled earliest ontogeny of four of the nine shells (Figs. 8B, C and 9A, B) probably spans the first summer of life, when growth would have been faster than in later summers. Had they been obtainable, samples from this interval would have been less time-averaged than samples from later summers and might therefore have yielded lower δ18O values, corresponding to the most extreme summer temperatures. However, at least some shells with a profile spanning the first summer actually exhibit lower δ18O values in the second summer (Fig. 9D, F), showing that failure to sample the first summer does not necessarily lead to incorrect documentation of extreme summer δ18Oshell. It is possible that the first winter of life is represented in all profiles. As in the case of summer δ18O, several shells (Fig. 9C, E) exhibit more extreme (higher) values in later winters, hence any failure to sample the first winter would not necessarily lead to incorrect documentation of extreme winter δ18Oshell. A few of the samples from shells UD 52795 and 53349 were too small to run. This could have caused one or both seasonal extremes to be missed and thus may well explain the somewhat smaller ranges in δ18O (Table 3). Shallower sampling of these shells and analysis in a different laboratory from the others did not obviously impact the results. The maximum and minimum values (Table 3) for shell

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

173

Table 3 Minimum and maximum δ18Oshell values (‰) for Coralline Crag specimens with corresponding palaeotemperatures (°C) calculated for various estimates of the oxygen-isotopic composition of seawater (δ18Owater). UD = University of Derby (School of Science). UD number Ramsholt Member 52795 52796 52797 53312 53313 53314 53315 53349

Sudbourne Member 53316

δ18Oshell min /max

δ18Oshell range

Temperature δ18Owater = + 0.5a

Temperature δ18Owater = + 0.1b

Temperature δ18Owater = − 0.2c

Temperature δ18Owater = − 0.5d

+ 0.97 + 2.28 + 0.76 + 2.35 0.00 + 1.78 + 0.63 + 2.18 + 0.11 + 1.78 + 0.34 + 2.34 + 0.11 + 1.90 + 0.50 + 1.87

1.31

13.7 8.4 14.6 8.1 18.0 10.4 15.2 8.8 17.5 10.4 16.5 8.1 17.5 9.9 15.8 10.0

12.1 6.8 13.0 6.5 16.2 8.8 13.5 7.2 15.7 8.8 14.7 6.6 15.7 8.3 14.1 8.4

10.8 5.7 11.7 5.4 14.9 7.6 12.3 6.0 14.4 7.6 13.5 5.4 14.4 7.1 12.8 7.2

9.6 4.5 10.5 4.2 13.6 6.4 11.0 4.9 13.2 6.4 12.2 4.3 13.2 5.9 11.5 6.1

− 0.47 + 1.07

1.54

20.1 13.3

18.3 11.7

16.9 10.4

15.6 9.2

1.59 1.78 1.55 1.67 2.00 1.79 1.37

a, b: maximum and minimum δ18Owater estimates for the location of the Coralline Crag from modelling by A. M. Haywood (see Section 4.3 for details). c, d: maximum and minimum estimates for global average δ18Owater (Buchardt and Simonarson, 2003).

UD53314 (containing thinly lined borings) and shell UD53312 (not bored in the region investigated by SEM) are closely comparable. 4.4.2. δ13C profiles Some profiles show a cyclical pattern (Figs. 8B, C and 9B, C), in phase with but lower in amplitude than δ18O, while others show this and/or an overall trend to lower values (Figs. 8A, D, F and 9A). The mean value of δ13C (n = 187) amongst all shells is + 0.37‰, a figure closely comparable to that (+0.34‰) obtained previously from a smaller sample (Johnson et al., 2000). 4.4.3. Isotopic palaeotemperatures On the assumption that minimum and maximum values of δ18O provide a fairly accurate measure of summer and winter extremes of temperature (Sections 3.1, 4.4.1), all shells indicate a seasonal range in temperature (5.1–8.4 °C; Table 3) smaller than in the southern North Sea at present (typically 9–11 °C). Employing the highest value for δ18Owater (+0.5‰), the single shell from the Sudbourne Member indicates a summer temperature 3–4 °C higher than present (typical maximum 16–17 °C). However, all other shells (from the Ramsholt Member) indicate summer temperatures similar to present or lower — substantially so (by as much as 6–7 °C) if a lower value for δ18Owater (+0.1 to −0.5‰) is employed. Use of a lower δ18Owater value (even +0.1‰) yields winter temperatures which are at least in some cases equivalent to present (typical minimum 6–7 °C) and several degrees lower than present if the lowest value of δ18Owater (− 0.5‰) is used in the palaeotemperature calculation. 4.4.4. Microgrowth-increment profiles Profiles of microgrowth increments (Fig. 8) show cycles of variation similar in wavelength to δ18O (and δ13C) data but notably out of phase with δ18O (as plotted with the isotopic axis reversed) early in ontogeny. This is entirely different from the pattern (shortwavelength variation) seen in modern indigenous shells from the Southern North Sea Basin (Section 3.2.2); moreover, the range of variation and maximum size of increments is very much greater (compare Tables 1 and 4). Similar data were obtained previously from Coralline Crag shells using a different method of measurement, but the relationship with δ18O was not then recognised (Johnson et al.,

2000). In so far as a seasonal signal dominates, the pattern resembles that in cultured shells, but it differs notably by the autumn-throughwinter increase in increment size in early ontogeny. The later ontogeny of Coralline Crag shells shows in-phase cycles of variation in increment size and δ18O (as plotted with the isotopic axis reversed). 4.5. Interpretations 4.5.1. Integrity of isotopic record The cyclical patterns of δ18Oshell variation, the correlations between δ18O and increment size, and the very similar isotopic ranges of the shell that had thinly lined borings and the shell that was apparently not bored, all confirm the indications supplied by SEM investigation (Section 4.2): that the shell material preserves an original isotopic signature and that incorporation of small amounts of void-filling diagenetic calcite into samples has no significant impact on measured isotope ratios. 4.5.2. Annual seafloor temperature range reconstructed from δ18O evidence The somewhat reduced range of δ18O, and hence calculated palaeotemperature, in shells UD 52795 and 53349 can be plausibly ascribed to the failure of a few samples to run (Section 4.4.1). While probably providing a better indication of seasonal range, the maximum and minimum isotopic palaeotemperatures calculated for other shells may still not quite represent annual extremes because there were small unsampled zones between sample grooves. However, it is clear that the isotopic data point to a smaller annual range of seafloor temperature in the Coralline Crag Sea than in the modern southern North Sea. 4.5.3. Absolute seafloor temperatures reconstructed from δ18O data Being dependent on estimates of δ18Owater, absolute summer and winter temperatures are subject to more uncertainty than annual range. Given diverse other evidence that winter temperatures were the same or higher than now during Coralline Crag deposition (Section 4.1.3) it seems scarcely credible that winters were in fact colder and that isotopic temperatures based on the lowest estimate of δ18Owater (− 0.5‰; yielding minimum temperatures around 4 °C) are closest to reality.

174

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

Fig. 8. Microgrowth-increment (blue), and oxygen (red) and carbon (black dash) stable-isotope data for A. opercularis from the Ramsholt Member of the Coralline Crag Formation (Pliocene), Suffolk, England. See Fig. 4 for further details concerning symbols and axes. (A)–(C): UD 52795–52797, respectively; (D): UD 53312; (E) UD 53314; (F): UD 53349. Note that increment data for UD 52796 and 52797, obtained by a different method, were accidentally transposed in Johnson et al. (2000, fig. 5), delaying recognition of the out-of-phase increment/δ18O relationship in Coralline Crag shells described herein. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Adopting less extreme estimates for δ18Owater (−0.2‰ and + 0.1‰) still yields winter temperatures comparable to present in some cases. Only for the highest estimate (+0.5‰) are all winter minima higher than now, and then by only about 2 °C in a number of cases. In no case does the winter minimum temperature approach that signified by the presence of the bryozoan Metrarabdotos (Section 4.1.3). Using the highest estimate for δ18Owater yields temperature maxima for Ramsholt-Member shells which are either similar to summer values in the present-day southern North Sea or somewhat lower (even in shells without significant sample gaps, e.g. UD 52796 and 53312; Table 3). Given the evidence from the microbiota of deposition at a depth greater than 50 m and of water-column stratification (Section 4.1.3), it seems likely that these values, even if accurate for the seafloor, are underestimates of summer surface temperature. This view is supported by the higher maximum temperature obtained from the Sudbourne-Member shell (UD53316), representative of a bettermixed and perhaps shallower water body.

4.5.4. Surface temperatures reconstructed from δ18O and microgrowth-increment data The notion of summer water-column stratification during RamsholtMember deposition is borne out by the large, long-period (seasonal) fluctuation in increment size and out-of-phase relationship with temperature (as inferred from δ18O) early in ontogeny, both of which features occur in modern shells from the seasonally strongly stratified setting of the Gulf of Tunis (Section 3.2.3). Water-depth for the RamsholtMember shells was probably the same as, or more than, that of the investigated Gulf-of-Tunis shells (50 m) so it is reasonable to suppose that the summer seafloor-surface temperature difference was at least as great: some 9 °C (Fig. 5). Adding this amount to the isotopically determined seafloor temperatures gives summer values in excess of present summer surface temperatures in the southern North Sea, even for seafloor temperatures based on a δ18Owater value of −0.5‰ (Table 3). For seafloor temperatures based on higher values of δ18Owater this procedure yields summer surface temperatures well over 20 °C (up to

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

175

Table 4 Maximum, minimum, range and mean of microgrowth-increment size for A. opercularis from the Ramsholt Member (Coralline Crag Formation). MEAN (bold figures) = average for column. UD = University of Derby (School of Science). UD number

52795 52796 52797 53312 53314 53349 MEAN

Microgrowth-increment height (mm) Maximum (fivepoint averages)

Minimum (fivepoint averages)

Range (fivepoint averages)

Mean (raw data for midline)

0.510 0.499 0.610 0.680 0.411 0.341 0.509

0.191 0.171 0.170 0.210 0.135 0.120 0.166

0.319 0.328 0.440 0.470 0.276 0.221 0.342

0.311 0.301 0.366 0.449 0.260 0.211 0.316

typically less in Coralline Crag shells than in modern shells from the southern North Sea. ‘Scope for growth’ (Bayne and Newell, 1983) is, however, substantially reduced in the later ontogeny of bivalves — a consequence of such factors as negative allometry in filtration rate and progressively greater diversion of surplus energy resources to gamete production (Johnson et al., 2007) — so it could well be that for ontogenetically older Coralline Crag shells the negative effects of declining temperature on growth outweighed the benefits of an autumn/winter increase in food availability. Not only would this lead to an in-phase relationship between temperature and increment size, it would also contribute to a reduction in annual growth in later ontogeny. A further factor contributing to relatively low annual growth in the later ontogeny of Coralline Crag A. opercularis may have been the longer duration of growth breaks, implied by more prominent ‘growth rings’ than in modern shells (compare Figs. 4 and 8). In Coralline Crag A. opercularis ‘growth rings’ are often preceded by declines in microgrowth-increment size and hence probably reflect the same fairly long-term (rather than traumatic) change in environment (e.g. the inferred summer reduction in benthic oxygen levels).

Fig. 9. Oxygen (red) and carbon (black dash) stable-isotope data for A. opercularis from the Ramsholt (A, B) and Sudbourne (C) members of the Coralline Crag Formation (Pliocene), Suffolk, England. Note that the y-axes are reversed. (A): UD 53313; (B): UD53315; (C): UD 53316. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

27 °C). In the Gulf of Tunis winter seafloor temperature at 50 m is little different from that at the surface (Fig. 5) so the winter temperatures derived from Ramsholt-Member shells can be taken to represent surface temperatures. Since the seasonal range in isotopically determined seafloor temperature is greater than 5 °C for all Ramsholt-Member shells (Table 3) it may be inferred that the seasonal range in surface temperature was at least 14 °C, i.e. much more than the present seasonal range in the southern North Sea (9–11 °C). 4.5.5. Cause of temperature/increment relationship in later ontogeny While the ‘Gulf-of-Tunis’ model (autumn oxygenation of the seafloor and resuspension of detrital food) can account for the out-ofphase relationship between temperature and microgrowth-increment size in the early ontogeny of Coralline Crag A. opercularis, it does not explain the in-phase relationship in later ontogeny or the fact that annual growth (as indicated by the wavelength of δ18O profiles) is

4.5.6. Interpretation of δ13C data It seems likely that in-phase variation with δ18Oshell reflects temperature-controlled fluctuations in benthic respiration (Section 3.1). In view of the evidence from microgrowth increments of high winter metabolism in A. opercularis one might perhaps have expected a closer correlation of δ13C with increment size than with δ18O. However, A. opercularis would have been only one of the contributors to benthic respiration, which may well have been dominated by other organisms (e.g. bacteria) whose respiration varied with temperature. It is unclear whether ontogenetic trends to lower δ13C reflect any environmental change; they may result from changing physiology with age (McConnaughey and Gillikin, 2008). However, the low mean value (0.37‰) relative to the pre-industrial Holocene mean (0.79‰; Section 3.1) probably reflects a higher atmospheric CO2 content (Johnson et al., 2000). This is consistent with evidence from leaf-stomatal density (Kürschner et al., 1996) and δ13C of marine organic matter (Raymo et al., 1996). 5. Test of isotopic palaeotemperature estimates for the Coralline Crag Although δ18O of A. opercularis shell is in principle a reliable indicator of temperature, and specimens from the Coralline Crag appear to be well preserved, we considered it prudent to investigate another species to see whether the results from A. opercularis would be corroborated. The infaunal bivalve Arctica islandica is a common associate of A. opercularis in the Coralline Crag. The isotope sclerochronology of modern examples has been intensively investigated, and the species provides accurate temperature information (e.g. Schöne and Fiebig,

176

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

6.1. Did summer-only warming occur throughout Coralline Crag deposition?

Fig. 10. Oxygen (red) and carbon (black dash) stable-isotope data through the third annual increment of a specimen of Arctica islandica from the Coralline Crag Formation (Pliocene), Suffolk, England (Institute of Geosciences, University of Mainz BRS-AJ1-CCD1R; analysed at IGF). Note that the y-axis is reversed. Indicated maximum and minimum temperatures from δ18O data were derived using the equation of Schöne et al. (2005) and a value of + 0.1‰ for δ18Owater. The blue triangles represent major growth breaks marking the boundaries of the annual increment. These breaks evidently formed in the autumn, as in modern Ar. islandica (Schöne et al., 2005). Although precise locality details were not available for the specimen, its unbroken and otherwise pristine state (aragonite preservation) provide convincing evidence that it was derived from the Ramsholt Member. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2009). Fig. 10 shows results from a right valve (almost certainly from the Ramsholt Member), obtained according to the methods described in Section 2.1 (analysis at IGF). The maximum and minimum isotopic palaeotemperatures shown, calculated for δ18Owater = +0.1‰, are remarkably similar to those from A. opercularis for the same δ18Owater value (Table 3). Comparable data have been obtained from other Coralline Crag Ar. islandica (E.M. Harper reported in Williams et al., 2009). We therefore conclude that the isotopic determinations of seafloor temperature based on A. opercularis (Sections 4.5.2, 4.5.3) are sound. It is worth noting that Ar. islandica is a component of the CelticBoreal province (Raffi et al., 1985), with a thermal maximum of 16 °C in the North Sea at present (Witbaard and Bergman, 2003). Therefore its mere presence alongside A. opercularis in the Coralline Crag is an indication that the ‘cool’ isotopic palaeotemperatures derived from the latter are probably accurate. The shape of the δ18O profile from the Coralline Crag Ar. islandica specimen (‘sinusoidal’) is more reminiscent of that of modern specimens that lived above rather than below the thermocline-depth (Schöne and Fiebig, 2009). If indeed the Ar. islandica specimen, and cooccurring A. opercularis in the Ramsholt Member, lived at this shallow depth then isotopic estimates of summer palaeotemperature would have to be taken as a measure of surface values. The isotopic palaeotemperatures concerned (similar to or lower than modern summer surface temperatures in the present-day southern North Sea, even for δ18Owater = +0.1‰; 4.4.3) are, however, scarcely credible for a greenhouse world, so we maintain the view (Sections 4.5.3, 4.5.4) that they represent sub-thermocline values from a sea whose summer surface temperature was much higher. This interpretation is open to test (Section 8).

6. Implications for climatic variation and controls in the mid-Pliocene (and future) From a climatological standpoint the most interesting conclusions from the present work are that during Ramsholt-Member deposition the annual range in sea-surface temperature was much higher than at present in the southern North Sea, and that this was brought about by elevation of summer temperature (Sections 4.5.3, 4.5.4). Two important questions, considered below, arise from this.

The evidence of summer-only warming comes from the Ramsholt Member and strictly speaking only applies to those times when the investigated A. opercularis individuals from this unit were alive. While the molluscan assemblage supports the notion of summer-only warming, a large body of other assemblage (and individual-size) data provides compelling evidence of elevated winter (as well as summer) temperatures, by as much as 10 °C (Section 4.1.3). The isotopic composition of the single investigated A. opercularis shell from the Sudbourne Member also indicates year-round warming (for δ18Owater ≥−0.2‰; Table 3). One is therefore led to the conclusion that during Coralline Crag deposition the climate in the Southern North Sea Basin fluctuated between states of summer-only and year-round warming relative to present conditions. Much of the currently available data is not even localised to stratigraphic member within the Coralline Crag so it is difficult to formulate a view on the duration and frequency of intervals of summer-only and year-round warming. However, our reading of existing information, including the observation of horizons relatively rich in the cool-water bivalve A. islandica, is that both circumstances existed during deposition of the Ramsholt Member. 6.2. What was the cause of summer-only (and year-round) warming? Global climate models for the time of peak mid-Pliocene warmth (c. 3 Ma) indicate both summer and winter warming at high latitudes, with a tendency for warming to be greater in winter; the same result has been obtained from a regional model for the North Atlantic area (models reviewed in Haywood et al., 2007). It would seem, therefore, that some factors not considered in these models must be invoked to explain intervals of summer-only warming in the southern North Sea area at the slightly earlier time of Coralline Crag deposition. At present, north-west Europe enjoys markedly warmer winter conditions than at equivalent latitudes in eastern North America because of heat transport to the area by the Gulf Stream/North Atlantic Drift (GS/NAD) current system (O'Hare et al., 2005). Summer temperature is not much affected by the GS/NAD. Under circumstances of global warming, and with no diminution in oceanic heat-supply, one would expect both winter and summer temperatures in north-west Europe to be raised above current values, just as climate models for the time of peak mid-Pliocene warmth show. If, however, oceanic heatsupply were reduced, summer temperatures would remain high but winter temperatures might fall to values much like those at present, producing a situation of summer-only warming, as inferred above for intervals during Coralline Crag deposition. With palaeogeography essentially as now, the effect of Pliocene global warmth would probably have been to enhance GS/NAD strength (Haywood et al., 2007), a view supported by proxy evidence of heat distribution in the North Atlantic at the time of peak mid-Pliocene warmth (Dowsett et al., 2009). One must therefore look to changes in palaeogeography to account for intervals of summer-only warming during Coralline Crag deposition. One possibility is that such intervals relate to the presence of a land-bridge between England and France, preventing ingress from the south-west of waters ultimately originating from the (probably enhanced) GS/NAD. The existence of such a feature is considered likely by some (Section 4.1.3), but it is not clear that its replacement by an area of shallow sea connecting the Channel and Southern North Sea basins would increase exposure to warmer waters sufficiently to cause year-round warming in the latter area. A second possibility is that the newly emergent Panama Isthmus was periodically breached, reducing the vigour of the Gulf Stream. A number of such breaches are recognised in the interval corresponding to Coralline Crag deposition (Schmidt, 2007) and the coeval occurrence of cool-water molluscs at locations south of Cape Hatteras, eastern USA (Sunken Meadow Member of the Yorktown Formation;

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

Williams et al., 2009) could well reflect a diminution in Gulf-Stream strength. Modelling by Lunt et al. (2008) has shown that closure of the Panama Isthmus in the Pliocene results in a substantial increase in northward oceanic heat transport compared to the ‘open’ condition. A third possibility is that global warmth in the Pliocene was not in fact associated with permanently elevated GS/NAD flow but, rather, with a pattern of fluctuating flow, independent of breaches in the Panama Isthmus. This adds a further scenario to the several differing ones that have been proposed for the state of the GS/NAD under the influence of future global warming (Dowsett et al. 2009). 7. Summary of principal conclusions (1) The early ontogeny of Aequipecten opercularis provides an essentially faithful and full record of seasonal temperature variation in the form of δ18Oshell. (2) Variation in size of microgrowth increments in wild forms is principally a reflection of the availability of suspended detrital food. (3) Oxygen-isotopic evidence from A. opercularis, corroborated by evidence from Arctica islandica, indicates that the seasonal range of seafloor temperature during deposition of the midPliocene Ramsholt Member (Coralline Crag Formation; Suffolk, UK) was less than at present in the adjacent southern North Sea. Winter seafloor temperature was probably about the same as at present while summer seafloor temperature was probably lower. (4) The pattern of early ontogenetic variation in microgrowthincrement size in Ramsholt-Member A. opercularis suggests intense thermal stratification of the water column in summer, with the implication that the seasonal range in surface temperature was higher than in the southern North Sea at present. (5) While the climate was warmer than now in summer at those times during Ramsholt-Member deposition when A. opercularis was present, evidence from other elements of the biota indicates that this state alternated with year-round warming. (6) The cause of fluctuations in climate state may have been variation in heat supply through the Gulf Stream/North Atlantic Drift, a situation that may also obtain in the globally warmer future. 8. Further work The issues raised in Section 6 could be addressed through the following research. (1) A high-resolution study of temporal variation in seasonal water temperature and hydrography through Ramsholt-Member deposition: (combining biotic-assemblage with sclerochronological evidence) to test for the existence of fluctuations between summer-only and year-round surface-warming. (2) An isotopic study of A. opercularis from age-equivalent but shallower-water strata in Belgium (Lillo Formation, Luchtbal Sands Member; De Schepper et al., 2008) to test whether these yield the expected evidence of higher (supra-thermocline) summer temperatures. (3) A full sclerochronological study of various bivalves, including A. opercularis and Ar. islandica, from somewhat later strata in Belgium, spanning the period of peak mid-Pliocene warmth (Lillo Formation, Oorderen Sands Member; De Schepper et al., 2008), to test for fluctuations in oceanic heat supply after full establishment of the Panama Isthmus. (4) An equivalent study of bivalves from the whole mid-Pliocene interval on the eastern seaboard of the USA to identify similarities and differences in temperatures north and south of Cape Hatteras, and either side of the Atlantic, and hence to test more rigorously for fluctuations in oceanic heat supply. A good deal of temperature data exists for the Pliocene of the eastern USA

177

from assemblage (e.g. Hazel, 1971; Ward et al., 1991) and isotopic (Krantz, 1990; Jones and Allmon, 1995; Goewert and Surge, 2008) analysis. However, there is scope for much more of the latter and for combining isotopic with microgrowth-increment analysis. (5) An equivalent study of bivalves, including Ar. islandica, from the Pliocene Tjörnes Beds of Iceland, building on the ‘wholeshell’ isotopic work of Buchardt and Simonarson (2003) and serving to test conclusions based on the combined data from either side of the Atlantic. As well as this further work on fossil material there is also a need to test the growth-control model proposed for wild A. opercularis: this could be accomplished by laboratory experiments involving manipulation of oxygen level, turbidity, temperature, and benthic and planktonic food supply. Acknowledgements We are grateful to Serge Gofas (Universidad de Málaga, Spain) for the supply of modern A. opercularis from the collections of the Muséum National d'Histoire Naturelle, Paris, and to Virginie Héros for repeated extensions of the loan. Kathie Way kindly facilitated the loan of modern material from The Natural History Museum, London and Guillermo Roman (Instituto Español de Oceanografía, La Coruña, Spain) generously supplied locally caught specimens. Matt Hunt, Steve Taylor, Graham Souch and Steve Hodson (University of Derby) assisted, respectively, with specimen preparation, photography, scanning-electronic-microscope investigation and cartography. We thank Liz Harper, Alan Haywood, Tanya Knowles, Greg O'Hare, Paul Taylor and Adrian Wood for useful information and discussion, and Harry Dowsett and Donna Surge for careful and constructive reviews of the manuscript. Isotopic analysis at NIGL was carried out with the assistance of Hilary Sloane under award IP/573/0998 to ALAJ. ALAJ gratefully acknowledges support from the Research-Inspired Curriculum Fund of the University of Derby (RICF) and the Alexander von Humboldt Foundation (resumption of Theodor Heuss Research Fellowship, University of Mainz). AB is grateful for support from RICF and the BGS University Funding Initiative (BUFI) S157 (www.bgs.ac. uk/research/bufi/home.html). BRS kindly acknowledges support by the German Research Foundation, DFG (SCHO 793/4) — this is Geocycles publication number 625. References Balson, P.S., 1983. Temperate, meteoric diagenesis of Pliocene skeletal carbonates from Eastern England. Journal of the Geological Society 140, 377–385. Balson, P.S., 1992. Tertiary. In: Cameron, T.D.J., Crosby, A., Balson, P.S., Jeffery, D.H., Lott, G.K. (Eds.), The geology of the southern North Sea. HMSO, London, pp. 91–100. Balson, P.S., 1999. The Coralline Crag. In: Daley, B., Balson, P.S. (Eds.), British tertiary stratigraphy. Geological Conservation Review Series, vol. 15. The Joint Nature Conservation Committee, Peterborough, pp. 253–288. Balson, P.S., Mathers, S.J., Zalasiewicz, J.A., 1993. The lithostratigraphy of the Coralline Crag (Pliocene) of Suffolk. Proceedings of the Geologists' Association 104, 59–70. Bayne, B.L., Newell, R.C., 1983. Physiological energetics of marine molluscs. In: Saleuddin, A.S.M., Wilbur, K.M. (Eds.), The Mollusca, vol. 4. Academic Press, New York, pp. 407–515. Physiology, Part 1. Broom, M.J., Mason, J., 1978. Growth and spawning in the pectinid Chlamys opercularis in relation to temperature and phytoplankton concentration. Marine Biology 47, 277–285. Buchardt, B., Simonarson, L.A., 2003. Isotopic palaeotemperatiures from the Tjörnes Beds in Iceland: evidence of Pliocene cooling. Palaeogeography, Palaeoclimatology, Palaeoecology 189, 71–95. Buddemeier, R.W., Maragos, J.E., 1974. Radiographic studies of reef coral exoskeletons: rates and patterns of coral growth. Journal of Experimental Marine Biology and Ecology 14, 179–200. Cheetham, A.H., 1967. Paleoclimatic significance of the bryozoan Metrarabdotos. Transactions of the Gulf Coast Association of Geological Societies 17, 400–407. Clark II, G.R., 2005. Daily growth lines in some living Pectens (Mollusca: Bivalvia), and some applications in a fossil relative: time and tide will tell. Palaeogeography, Palaoclimatology, Palaeoecology 228, 26–42. Coplen, T.B., Kendall, C., Hopple, J., 1983. Comparison of stable isotope reference samples. Nature 302, 236–238.

178

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179

De Schepper, S., Head, M.J., Louwye, S., 2008. Pliocene dinoflagellate cyst stratigraphy, palaeoecology and sequence stratigraphy of the Tunnel-Canal Dock, Belgium. Geological Magazine 146, 92–112. Dowsett, H.J., 2007. Pliocene sea-surface temperature and the PRISM paleoclimate reconstructions. In: Williams, M., Haywood, A.M., Gregory, F.J., Schmidt, D.N. (Eds.), Deep time perspectives on climate change: marrying the signal from computer models and biological proxies. The Micropalaeontological Society, Special Publication. The Geological Society, London, pp. 459–480. Dowsett, H., Barron, J., Poore, R., 1996. Middle Pliocene sea surface temperatures: a global reconstruction. Marine Micropaleontology 27, 13–25. Dowsett, H.J., Barron, J.A., Poore, R.Z., Thompson, R.S., Cronin, T.M., Ishman, S.E., Willard, D.A., 1999. Middle Pliocene paleoenvironmental reconstruction: PRISM2. USGS Open File Report, pp. 99–535. http://pubs.usgs.gov/of/1999/of99-535/. Dowsett, H.J., Chandler, M.A., Robinson, M.M., 2009. Surface temperatures of the MidPliocene North Atlantic Ocean: implications for future climate. Philosophical Transactions of the Royal Society, London A367, 69–84. Funnell, B.M., 1996. Plio-Pleistocene palaeogeography of the southern North Sea Basin (3.75–0.60 Ma). Quaternary Science Reviews 15, 391–405. Goewert, A.E., Surge, D., 2008. Seasonality and growth patterns using isotope sclerochronology in shells of the Pliocene scallop Chesapecten madisonius. Geo-Marine Letters 28, 327–338. Goodwin, D.H., Schöne, B.R., Dettman, D.L., 2003. Resolution and fidelity of oxygen isotopes as paleotemperature proxies in bivalve mollusk shells: models and observations. Palaios 18, 110–125. Grossmann, E.L., Ku, T.-L., 1986. Oxygen and carbon isotope fractionation in biogenic aragonite: temperature effects. Chemical Geology (Isotope Geoscience Section) 59, 59–74. Gruffydd, Ll.D., 1981. Observations on the rate of production of external ridges on the shell of Pecten maximus in the laboratory. Journal of the Marine Biological Association of the United Kingdom 61, 401–411. Hall, C.A., 1964. Shallow-water marine climates and molluscan provinces. Ecology 45, 226–234. Hallmann, N., Burchell, M., Schöne, B.R., Irvine, G.V., Maxwell, D., 2009. High-resolution sclerochronological analysis of the bivalve mollusk Saxidomus gigantea from Alaska and British Columbia: techniques for revealing environmental archives and archaeological seasonality. Journal of Archaeological Science 36, 2353–2364. Haywood, A.M., Valdes, P.J., Hill, D.J., Williams, M., 2007. The mid-Pliocene warm period: a test-bed for integrating data and models. In: Williams, M., Haywood, A.M., Gregory, F.J., Schmidt, D.N. (Eds.), Deep time perspectives on climate change: marrying the signal from computer models and biological proxies. The Micropalaeontological Society, Special Publications. The Geological Society, London, pp. 443–457. Hazel, J.E., 1971. Paleoclimatology of the Yorktown Formation (Upper Miocene and Lower Pliocene) of Virginia and North Carolina. Bulletin du Centre de Recherches de Pau (France), Societé Nationale des Petroles d'Aquitaine 5, 361–375 supplement. Head, M.J., 1997. Thermophilic dinoflagellate assemblages from the Mid-Pliocene of Eastern England. Journal of Palaeontology 71, 165–193. Head, M.J., 1998. New goniodomacean dinoflagellates with a compound hypotractal archeopyle from the late Cenozoic: Capisocysta Warny and Wrenn, emend. Journal of Paleontology 72, 797–809. Heilmayer, O., Brey, T., Storch, D., Mackensen, A., Arntz, W.E., 2004. Population dynamics and metabolism of Aequipecten opercularis (L.) from the western English Channel (Roscoff, France). Journal of Sea Research 52, 33–44. Hickson, J.A., 1997. Stable isotope profiles in shells of the bivalve Aequipecten opercularis (L.) as environmental indicators. Ph.D. Thesis, Univ. Derby, UK. Hickson, J.A., Johnson, A.L.A., Heaton, T.H.E., Balson, P.S., 1999. The shell of the Queen Scallop Aequipecten opercularis (L.) as a promising tool for palaeoenvironmental reconstruction: evidence and reasons for equilibrium stable-isotope incorporation. Palaeogeography, Palaeoclimatology, Palaeoecology 154, 325–337. Hickson, J.A., Johnson, A.L.A., Heaton, T.H.E., Balson, P.S., 2000. Late Holocene environment of the southern North Sea from the stable isotopic composition of Queen Scallop shells. Palaeontologia Electronica 3 (2). http://palaeo-electronica.org/2000_2/scallop/ issue2_00.htm. Hodgson, G.E., Funnell, B.M., 1987. Foraminiferal biofacies of the early Pliocene Coralline Crag. In: Hart, M.B. (Ed.), Micropalaeontology of carbonate environments. British Micropalaeontological Society, Special Publication. Ellis Harwood, Chichester, UK, pp. 44–73. Jenkins, D.G., Houghton, S.D., 1987. Age, correlation and palaeoecology of the St Erth Beds and the Coralline Crag of England. Mededelingen van de Werkgroep voor Tertiare en Kwartaire Geologie 24, 147–156. Jenkins, D.G., Curry, D., Funnell, B.M., Whittaker, J.E., 1988. Planktonic foraminifera from the Pliocene Coralline Crag of Suffolk, Eastern England. Journal of Micropalaeontology 7, 1–10. Johnson, A.L.A., Hickson, J.A., Swan, J., Brown, M.R., Heaton, T.H.E., Balson, P.S., Chenery, S., 2000. The Queen Scallop Aequipecten opercularis: a new source of information on late Cenozoic marine environments in Europe. In: Harper, E.M., Taylor, J.D., Crame, J.A. (Eds.), The evolutionary biology of the Bivalvia. The Geological Society, London. Special Publications 177, 425–439. Johnson, A.L.A., Liquorish, M., Sha, J., 2007. Variation in growth-rate and form of a Bathonian (Middle Jurassic) oyster in England, and its environmental implications. Palaeontology 50, 1155–1173. Jones, D.S., Allmon, W.D., 1995. Records of upwelling, seasonality and growth in stableisotope profiles of Pliocene mollusk shells from Florida. Lethaia 28, 61–74. Knowles, T., Taylor, P.D., Williams, M., Haywood, A.M., Okamura, B., 2009. Pliocene seasonality across the North Atlantic inferred from cheilostome bryozoans. Palaeogeography, Palaeoclimatology, Palaeoecology 277, 226–235.

Krantz, D.E., 1990. Mollusk-isotope records of Plio-Pleistocene marine paleoclimate, U.S. Middle Atlantic Coastal Plain. Palaios 5, 317–335. Kürschner, W.M., Van der Burgh, J., Visscher, H., Dilcher, D.L., 1996. Oak leaves as biosensors of Late Neogene and Early Pleistocene paleoatmospheric CO2 concentrations. Marine Micropaleontology 27, 299–312. Lagaaij, R., 1963. Cupuladria canariensis (Busk) — portrait of a bryozoan. Palaeontology 6, 172–217. Lunt, D.J., Valdes, P.J., Haywood, A., Rutt, I.C., 2008. Closure of the Panama Seaway during the Pliocene: implications for climate and Northern Hemisphere glaciation. Climate Dynamics 30, 1–18. McConnaughey, T.A., Gillikin, D.P., 2008. Carbon isotopes in mollusk shell carbonates. Geo-Marine Letters 28, 287–299. Murray, J.W., 1987. Benthic foraminiferal assemblages: criteria for the distinction of temperate and subtropical carbonate environments. In: Hart, M.B. (Ed.), Micropalaeontology of carbonate environments. British Micropalaeontological Society, Special Publication. Ellis Harwood, Chichester, UK, pp. 9–20. Murray, J.W., 1992. Palaeogene and Neogene. In: Cope, J.C.W., Ingham, J.K., Rawson, P.F. (Eds.), Atlas of palaeogeography and lithofacies: The Geological Society, London, Memoir, vol. 13, pp. 141–147. Naish, T., 55 others, 2009. Obliquity-paced Pliocene West Antarctic ice sheet oscillations. Nature 458, 322–329. NOAA (US Department of Commerce, National Oceanic and Atmospheric Administration, 1994. NODC (Levitus) World Ocean Atlas: Ocean Temperature: Monthly Long Term Mean. http://www.cdc.noaa.gov/cgi-bin/db_search/DBSearch.pl?Dataset=NODC +(Levitus)+World+Ocean+Atlas&Variable=Ocean+temperature, 1994 accessed 18 September 2009. O'Dea, A., Okamura, B., 2000. Intracolony variation in zooid size in cheilostome bryozoans as a new technique for investigating palaeoseasonality. Palaeogeography, Palaeoclimatology, Palaeoecology 162, 319–332. O'Hare, G., Johnson, A., Pope, R., 2005. Current shifts in abrupt climate change: the stability of the North Atlantic Conveyor and its influence on future climate. Geography 90, 250–266. O'Neil, J.R., Clayton, R.N., Mayeda, T.K., 1969. Oxygen isotope fractionation in divalent metal carbonates. Journal of Chemical Physics 51, 5547–5558. Owen, R., Richardson, C., Kennedy, H., 2002. The influence of shell growth rate on striae deposition in the scallop Pecten maximus. Journal of the Marine Biological Association of the United Kingdom 82, 621–623. Proudman Oceanographic Laboratory, 2009. POLTIPS 3. Coastal Tidal Prediction Software. Natural Environment Research Council, UK. Raffi, S., Stanley, S.M., Marasti, R., 1985. Biogeographic patterns and Plio-Pleistocene extinction of Bivalvia in the Mediterranean and southern North Sea. Paleobiology 11, 368–388. Raymo, M., Grant, B., Horowitz, M., Rau, G.H., 1996. Mid-Pliocene warmth: stronger greenhouse and stronger conveyor. Marine Micropaleontology 27, 313–326. Reid, P.C., Lancelot, C., Gieskes, W.W.C., Hagmeier, E., Weichart, G., 1990. Phytoplankton of the North Sea and its dynamics — a review. Netherlands Journal of Sea Research 26, 295–331. Richardson, C.A., 2001. Molluscs as archives of environmental change. Oceanography and Marine Biology: an Annual Review 39, 103–164. Richardson, C.A., Taylor, A.C., Venn, T.J., 1982. Growth of the queen scallop Chlamys opercularis in suspended cages in the Firth of Clyde. Journal of the Marine Biological Association of the United Kingdom 62, 157–169. Romanek, C.S., Grossman, E.L., Morse, J.W., 2002. Carbon isotopic fractionation in synthetic aragonite and calcite: effects of temperature and precipitation rate. Geochimica et Cosmochimica Acta 56, 419–430. Schmidt, D.N., 2007. The closure history of the Central American seaway: evidence from isotopes and fossils to models and molecules. In: Williams, M., Haywood, A.M., Gregory, F.J., Schmidt, D.N. (Eds.), Deep time perspectives on climate change: marrying the signal from computer models and biological proxies. The Micropalaeontological Society, Special Publications. The Geological Society, London, pp. 427–442. Schöne, B.R., Fiebig, J., 2009. Seasonality in the North Sea during the Allerød and Late Medieval Climate Optimum using bivalve sclerochronology. International Journal of Earth Sciences 98, 83–98. Schöne, B.R., Surge, D. (Eds.), 2005. Looking back over skeletal diaries — high-resolution environmental reconstructions from accretionary hard parts of aquatic organisms: Palaeogeography, Palaeoclimatology, Palaeoecology, vol. 228, pp. 1–192. Schöne, B.R., Lega, J., Flessa, K.W., Goodwin, D.H., Dettman, D.L., 2002. Reconstructing daily temperatures from growth rates of the intertidal bivalve mollusk Chione cortezi (northern Gulf of California, Mexico). Palaeogeography, Palaeoclimatology, Palaeoecology 184, 131–146. Schöne, B.R., Oschmann, W., Rössler, J., Freyre Castro, A.D., Houk, S.D., Kröncke, I., Dreyer, W., Janssen, R., Rumohr, H., Dunca, E., 2003. North Atlantic Oscillation dynamics recorded in shells of a long-lived bivalve mollusk. Geology 31, 1037–1040. Schöne, B.R., Fiebig, J., Pfeiffer, M., Gless, R., Hickson, J., Johnson, A.L.A., Dreyer, W., Oschmann, W., 2005. Climate records from a bivalved Methuselah (Arctica islandica, Mollusca; Iceland). Palaeogeography, Palaeoclimatology, Palaeoecology 228, 130–148. Strauch, F., 1968. Determination of Cenozoic sea-temperatures using Hiatella arctica. Palaeogeography, Palaoclimatology, Palaeoecology 5, 213–233. Taylor, A.C., Venn, T.J., 1978. Growth of the queen scallop, Chlamys opercularis, from the Clyde Sea area. Journal of the Marine Biological Association of the United Kingdom 58, 687–700. Ward, L.W., Bailey, R.H., Carter, J.G., 1991. Pliocene and Early Pleistocene stratigraphy, depositional history, and molluscan paleobiogeography of the coastal plain. In: Horton, J.W., Zullo, V.A. (Eds.), The geology of the Carolinas: Carolina Geological

A.L.A. Johnson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 284 (2009) 164–179 Society fiftieth anniversary volume. University of Tennessee Press, Knoxville, TN, pp. 274–289. Westaway, R., Maddy, D., Bridgland, D., 2001. Flow in the lower continental crust as a mechanism for the Quaternary uplift of south-east England: constraints from the Thames terrace record. Quaternary Science Reviews 21, 559–603. Wilkinson, I.P., 1980. Coralline Crag Ostracoda and their environmental and stratigraphical significance. Proceedings of the Geologists' Association 91, 291–306. Williams, M., Haywood, A.M., Hillenbrand, C.D., Wilkinson, I.P., 2005. Efficacy of δ18O data from Pliocene planktonic foraminifer calcite for spatial sea surface temperature reconstruction: comparison with a fully coupled ocean-atmosphere GCM and fossil assemblage data from the mid Pliocene. Geological Magazine 142, 399–417. Williams, M., Haywood, A.M., Harper, E.M., Johnson, A.L.A., Knowles, T., Leng, M.J., Lunt, D.J., Okamura, B., Taylor, P.D., Zalasiewicz, J., 2009. Pliocene climate and seasonality in North Atlantic shelf seas. Philosophical Transactions of the Royal Society. London A367, 85–108.

179

Witbaard, R., 1996. Growth variations in Arctica islandica L. (Mollusca): a reflection of hydrography-related food supply. ICES Journal of Marine Science 53, 981–987. Witbaard, R., Bergman, M.J.N., 2003. The distribution and population structure of the bivalve Arctica islandica L. in the North Sea: what possible factors are involved? Journal of Sea Research 50, 11–25. Wood, A.M., Whatley, R.C., Cronin, T., Holtz, T., 1993. Pliocene palaeotemperature reconstruction for the southern North Sea based on Ostracoda. Quaternary Science Reviews 12, 747–767. Wood, A.M., Wilkinson, I.P., Maybury, C.A., Whatley, R.C., 2009. Neogene. In: Whittaker, J.E., Hart, M.B. (Eds.), Ostracods in British stratigraphy. The Micropalaeontological Society, Special Publications. The Geological Society, London, pp. 411–446. Zarrad, R., Missaoui, H., Alemany, F., Salah, R.M., Garcia, A., Ridha, M., Othman, J., Amor, E.A., 2006. Spawning areas and larval distributions of anchovy Engraulis engrasicolus in relation to environmental conditions in the Gulf of Tunis (Central Mediterranean Sea). Scientia Marina 70, 137–146.