Stable Isotope Analysis of Gryphaea arcuata Reveals the Prevalence of Humid Tropical Conditions During the Early Sinemurian of Normandy (Fresville), Northwestern France

1. Introduction

The Early Jurassic, particularly the Hettangian and Sinemurian, spanning from 201.4 to 192.9 Ma [1], records the recovery and evolution of ecosystems following the End-Triassic biotic crisis, one of the so-called major mass extinction events of the Phanerozoic [2,3,4,5,6,7,8,9,10]. The supercontinent Pangea, which began to fragment during the Triassic, continued to separate throughout the Jurassic [11]. However, the close proximity of all the corresponding present-day continents implies that floras and faunas shared many common biological properties. Reconstructed high atmospheric CO₂ levels [12,13], the presence of vegetation at high latitudes, and a general lack of evidence for polar ice sheets suggest that warm and more or less humid climates extended to high latitudes [14].

The Early Jurassic was a time of significant evolutionary change and diversification [15]. On land, it was characterised by the dominance of gymnosperms, including conifers, cycads and ginkgos ([6,7,8,9,10,11,12,13,14,15,16,17]. These plants formed extensive forests that supported diverse terrestrial ecosystems. Indeed, for example, the evolution of early dinosaurs continued, with notable groups such as coelophysid and ceratosaurian theropods, basal sauropodomorphs and thyreophoran ornithischians diversifying and occupying different ecological niches [18,19,20]. This period set the stage for the later dominance of dinosaurs in the Jurassic and Cretaceous periods [21]. Mammal-like reptiles (synapsids) and early true mammals were also present, although they remained relatively small and ecologically subordinate to the dinosaurs [22,23,24,25].

In the oceans, several new biological groups also emerged and diversified during this time. These groups were mainly scleractinian corals, foraminifera, radiolarians, psiloceratoid ammonites, ray-finned fish and marine reptiles [26,27,28,29,30,31,32]). The latter include giant predators such as ichthyosaurs, plesiosaurs and pliosaurs, whose homeothermy and endothermy are still the subject of intense physiological, ecological and geochemical research [33,34,35,36]. Among the bivalves and gastropods, the oyster-like genus Gryphaea, also known as “Devil’s Toenails” in the British folklore, was particularly abundant and serves sometimes as an index fossil for the Lower Jurassic along with ammonites [37,38,39]. Specifically, Gryphaea arcuata was a benthic organism that lived on the seafloor in shallow warm temperate marine environments [40,41]. Those bivalves thrived in the epicontinental seas that covered much of present-day Europe and parts of North America (Figure 1). These seas were characterised by warm to cool temperatures, moderate to low energy environments and muddy to sandy substrates, providing suitable conditions for Gryphaea arcuata ([2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43].

In the context of a Pangea that was just beginning to break up, the sedimentary marine deposits of western Europe (Figure 2) record the influence of both the Tethys (warm) and the Boreal (cool) domains [44]. The climatic context is therefore a crucial key to understanding the radiation of the marine species described above. It is important to mention that many temperature constructions have been made for the Jurassic and as far back as the Plienbaschian. However, only a few data are available for the Hettangian and the Sinemurian according to a compilation by [45], which represent the Early Jurassic immediately following the Triassic-Jurassic boundary. These marine temperatures, mostly between 15°C and 20°C, have been inferred from the oxygen isotope composition of a few teeth of demersal and pelagic fishes, Gryphaea sp., oysters and belemnites [40,45,46,47,48,49]. These marine temperatures appear to be much lower than those inferred from the early Plienbaschian and early Toarcian marine sediments [45]. Thus, the addition of more marine temperature data is crucial to understand whether or not the documented biological radiation took place in the context of gently warm Jurassic climate dynamics.

We therefore studied 75 invertebrate samples including 53 specimens of Gryphaea arcuata and 22 specimens of coexisting bivalves (11 Chlamys sp., 4 Plagiostoma sp. and 7 Pseudolimea sp.) from the Sinemurian (199.4 to 192.9 Ma; [1]) calcareous and marly marine sediments of the Fresville quarry (Cotentin Peninsula, Normandy, France). The co-occurrence of Arietites bucklandi (Sowerby 1818) and Gryphaea arcuata allowed this ammonite and this bivalve to be used as index fossils for dating these strata. Consequently, the sedimentary column is thus known to represent the basal Early Sinemurian covering a period of about 1.3 My [50].

The aim of our study is to evaluate the potential diagenesis of the fossils studied, the marine productivity of the water column in which they lived, and changes in seawater temperature. Our objectives will be achieved by determining and interpreting the stable carbon and oxygen isotope compositions of the bivalves together with their Sr and Mg contents. Temperature values and their changes will be discussed in relation to the oxygen isotope composition of Early Jurassic seawater, the paleogeography of Western Europe, lithology, faunal assemblages and the particular ecology of Gryphaea arcuata.

2. Materials and Methods

2.1. Geological Setting

At the beginning of the Jurassic, Pangea split to form several continental plates in response to a period of long-term extension [51,52]. A second-order transgression begins in the early Jurassic and continues until the middle of the Toarcian, with a change in the sedimentary regime [50,53]. A third-order transgression characterises the Early Sinemurian [50]. During the Sinemurian, the Fresville region was located at the edge of the carbonate platform at a palaeolatitude of around 35°-36°N after [54]) or around 31°-32°N after [55] (Figure 2). The transgression allowed the sea to invade the Fresville region from the eastern part of the platform, i.e. from the eastern edge of the Armorican massif [56]. This region constituted a vast gulf of the early Jurassic sea bordering the Armorican continent involving a calm sea with a muddy bottom and influences from the open sea [57].

Figure 2. Paleogeographic map of western Europe during the Sinemurian with the location of Fresville in northwestern Normandy, France. After [55]. Blue arrows: low-saline and cool southward currents; grey: land masses; dark grey: fluviatile/lacustrine environments; light blue: shallow marine environments; deep blue: deep epicontinental seas; dark blue: deep oceanic basin.

Figure 2. Paleogeographic map of western Europe during the Sinemurian with the location of Fresville in northwestern Normandy, France. After [55]. Blue arrows: low-saline and cool southward currents; grey: land masses; dark grey: fluviatile/lacustrine environments; light blue: shallow marine environments; deep blue: deep epicontinental seas; dark blue: deep oceanic basin.

The Fresville quarry is located in northwestern France, more specifically in the Cotentin Peninsula, department Manche, Normandy. This former limestone quarry, used for cement production until 1984, described in detail by [56], is located at “le Goulet”. It features a marl-limestone alternation belonging to the Early Sinemurian (Figure 3). This geological formation on this site rests on the Hettangian. Above this formation is a palaeosol of the same age as well as a detrital formation consisting of Cenomanian orbitoline sands. The change in lithology and the absence of Middle and Upper Jurassic deposits highlight variations in sea level and periods of emersion. The marl-limestone alternation can be seen over a height of around 20 m and contains an abundant and diverse [56] of marine invertebrates, as well as shark [57] and ichthyosaur remains, ichnofossils and plant remains.

Figure 3. Lower Sinemurian outcrops in the Fresville quarry, Normandy, France, highlighting the marine stratigraphic sequence with the numbering of alternating limestone, marly limestone and marl layers. See also Table 1.

Figure 3. Lower Sinemurian outcrops in the Fresville quarry, Normandy, France, highlighting the marine stratigraphic sequence with the numbering of alternating limestone, marly limestone and marl layers. See also Table 1.

Table 1. Stratigraphy, lithology, taxonomy, stable isotope ratios and trace element contents of the mollusc shells collected from the lower Sinemurian of Fresville, Normandy, France.

Table 1. Stratigraphy, lithology, taxonomy, stable isotope ratios and trace element contents of the mollusc shells collected from the lower Sinemurian of Fresville, Normandy, France.

The outcrop is bounded by two faults. The first is the result of an extensive tectonic regime during the breaking of Pangea in the Middle and Upper Jurassic. The second is the result of a compressional phase of Alpine age [1]. From a lithological point of view, the marl-limestone formation observed is characterized by a little variation in the content of clay minerals, which are illite and kaolinite. Biomicritic clayey limestone is characterised by a calcium carbonate content of less than 40 wt%. The difference between clayey limestone and marl depends on the calcium carbonate (CaCO₃) content, the limit of which corresponds to a content of 35 wt%. The outcrop has a monoclinal structure with a gentle slope of 3° to 5° to the south-east, attributed to subsidence [58].

In order to study the Early Sinemurian marine paleoenvironment recorded in the Fresville sedimentary sequence, the study was carried out in the south-eastern part of the quarry at geographical coordinates: N49°26’23.06 – W1°22’52.82’’ (Figure 3). The study material was collected in the lower part of the quarry, at an altitude of around 13 m above sea level. This part forms the first step of the quarry and constitutes the base of the Sinemurian [59]. The first step consists of alternating marly limestones and pyritic bluish-grey marls in the Bucklandi biozone covering the period comprised between 199.3 and 198 Ma, according to [50]. This biozone represents the first ammonite zone of the Sinemurian covering the period 199.3 Ma to 190.8 Ma [1], which is also characterized by the presence of Gryphaea arcuata while Gryphaea obliquata is always absent.

At the time of our collecting visit, two sections were sampled: the lower part comprising layers 1 to 24 with a total thickness of 2.85 m and the upper part comprising layers 33 to 35 with a total thickness of 43 cm. These two lithological sections were separated by approximately 100 cm of outcrop, which was inaccessible for sampling due to the presence of a significant vegetation cover (Figure 3). For a detailed illustrated description of the sections currently visible in the Fresville quarry, see Desquesne (2023).

2.2. Sample Collection

The marine invertebrate fossil fauna found in the Fresville quarry is abundant (Figure 4). It consists of bivalves (e.g. Gryphaea arcuata, Chlamys sp, Plagiostoma sp), ammonites (Arietites bucklandi), gastropods, echinoderms, arthropods and brachiopods (Spiriferina walcotti); for a detailed illustrated list, see [56]. The fossil specimens used in this study were collected with the help of members of the Club de Géologie du Cotentin.

In this study, we mainly focus on the species Gryphaea arcuata whose low-Mg calcite shells [60,61,62] constitute a valuable material for seawater temperature reconstructions based on geochemical proxies. Indeed, low-Mg calcite is less sensitive to diagenesis as its solubility in aqueous fluids is much lower than aragonite [63,64,65,66].

The genus Gryphaea belongs to the Class Bivalvia, the Order Ostreida and Family Gryphaeidae. The chronostratigraphic distribution of Gryphaea extends from the Jurassic to the Cretaceous [1], and this genus was particularly abundant during the Sinemurian [1]. During the Hettangian and Sinemurian, the paleogeographic domain of Gryphaea extended between 33°N and 55°N latitude in western Europe [40].

According to [67] and [68], Gryphaea arcuata has an elongated, relatively thick, arched shell no more than 6 cm long. The hook is protruding and curved towards the upper valve. The shell of G. arcuata is highly inequivalve, meaning the two valves differ significantly in size and shape. The larger left valve is convex and strongly curved, resembling a claw or a crescent, while the smaller right valve is flat or slightly concave, serving primarily as a lid. Growth lines and ridges are prominent on the surface of the valves, indicating shell growth rhythms through time.

Gryphaea arcuata was living in the warm temperate shallow marine environments (0 to 50 m depth; [69]), fixed to the seabed, and feeding by filtering seawater [40]. They are considered to only thrive in a euhaline environment ([40]). Such marine environment with a normal salinity is suggested by the assemblage of invertebrate fauna that have been documented in the Fresville quarry: Gryphaea arcuata, ammonites and brachiopods [40].

2.3. Analytical Techniques

2.3.1. Sample Preparation

Samples were selected according to some recommendations given by [66], including the use of sub-mature to adult part of the shells to avoid isotopic disequilibrium with ambient seawater during the fast growth of the shell [70]. Careful examination of shell fragments obtained from smooth-textured shell surfaces was performed, then inspected under a binocular microscope. Only about two thirds of the collected samples in the field were kept for the cleaning protocol. Those selected samples underwent cleaning with double distilled water in an ultrasonic bath to remove marl particles from the shell without damaging it. To do this, each piece was placed in a beaker filled with distilled water, and the beaker was placed in the ultrasonic bath for 2 minutes. This step was repeated until the piece was completely clean. The second step was to take a shell sample from the surface by drilling the adult part close to the main growth axis using a Dremel™ equipped with a diamond-coated rotating head. The aliquots were finally ground to a powder (≈ 100 to 150 μm in size) in an agate mortar. The identification of the calcium carbonate polymorph was determined with Raman spectroscopy. Potential diagenesis has been discussed by interpreting the observed trends linking the δ¹³C values (‰ VPDB), δ¹⁸O values (‰ VPDB), Mg and Sr contents in ppm of low-Mg calcite shells.

2.3.2. Raman Spectroscopy

The Raman analyses were carried out at the Laboratoire de Géologie, École Normale Supérieure of Lyon. The Raman spectra of 58 samples were obtained during this study. Of these 58 samples, 56 bivalve samples, including 37 samples of Gryphaea arcuata, 7 Pseudolimea sp., 4 Plagiostoma sp. as well as 8 Chlamys sp distributed along the sedimentary sequence. Raman spectra were collected using a visible JY LabRam HR800 Raman spectrometer using a laser with a wavelength of 514.532 nm at ambient temperature and pressure. Each sample was measured at minima at one point. The spectra obtained cover the low and high frequencies, from 130 (130.071) to 1100 (1099.77) cm^-1. The acquisition time was equal to two acquisitions of 5 seconds each.

2.3.3. Trace Elements

Measurements were performed at the Ecole Normale Supérieure de Lyon, France. Sr and Mg concentrations were obtained by dissolving about 10 mg of carbonate powder in 10 ml of HNO₃ 2%. For all samples, pairs of aliquots were prepared with 10 times dilution along with a fixed amount of Scandium and Indium that were added to correct concentrations from instrument drift. Solutions were analysed using a quadrupole ICP-Mass Spectrometer (i-CAP-Q ICP-MS), for minor and trace elements, respectively. The reproducibility of measurements was assessed through the analysis of the carbonate standard CCH1 [71].

2.3.4. Carbon and Oxygen Stable Isotopes

Measurements were performed at the “Laboratoire de Géologie de Lyon”, Université Claude Bernard, Lyon, France. Stable isotope ratios were determined by using an auto sampler MultiPrep^TM system coupled to a dual–inlet GV Isoprime^TM isotope ratio mass spectrometer (IRMS). Aliquot size was about 300 μg of calcium carbonate. All aliquots were reacted with anhydrous oversaturated phosphoric acid at 90°C during 20 minutes. Oxygen isotope ratios of calcium carbonate are computed assuming an acid fractionation factor 1000lnα(CO₂–CaCO₃) of 8.1 between carbon dioxide and calcite [72]. All sample measurements were duplicated and adjusted to the international references NIST NBS18 (δ¹⁸O_VPDB = -23.2‰; δ¹³C_VPDB = -5.01‰) and NBS19 (δ¹⁸O_VPDB = -2.20‰; δ¹³C_VPDB = +1.95‰). External reproducibility is close to ±0.1‰ for δ¹⁸O and ±0.05 for δ¹³C (2σ).

3. Results

3.1. Raman Spectra

Raman analyses of the shells studied are given in Table 1 and, for example, two spectra of Gryphaea arcuata are shown in Figure 5. All bivalve shells are made of low-Mg calcite that was readily identified from the low frequency part of the spectra, and especially from the position and splitting of the symmetric bending mode which occurs as a single peak at 712 cm^-1 in calcite [73]. The peaks obtained at 156 cm^-1 and 282 cm^-1 are attributed to the translation and rotation mode known as the lattice mode [74]. The strongest band for calcite, located at 1087 cm^-1, is an internal mode resulting from the symmetrical stretching of the carbonate ion, known as the stretching mode.

3.2. Sr and Mg Contents of Gryphaea arcuata and Other Associated Bivalves

Sr and Mg contents of Gryphaea arcuata range from 335 ppm to 6161 ppm and from 160 ppm to 2076 ppm, respectively (Table 1). The distribution of both set of data does not follow a normal distribution according to a Shapiro-Wilk test (Table 2). However, the data are statistically positively correlated according to a Spearman test (r = 0.572) which is adapted to data not following a normal distribution (Table 3). The linear regression model highlights the presence of outliers for Sr (44124, 4318 and 6161 ppm) and for Mg (2076 ppm) that correspond to the highest contents measured in the sub-mature parts of the shells. A Mann-Whitney U test was also applied to the Sr and Mg values of Gryphaea arcuata depending on the mineralogy of the host sediment to investigate its possible influence of the isotope composition of the bivalve shells. The hypothesis H₀ stating similar quantitative distributions in marls-limestones and marls-marly limestones was thus verified for Sr contents with acceptable p-values of 0.523 and 0.786, respectively. The same hypothesis and groups were also verified for Mg contents with an acceptable p-value of 0.440 for marls-limestones while a very low p value of 0.062 is observed for marls-marly limestones, respectively. Sr and Mg contents of Chlamys (n = 11) range from 1197 ppm to 4666 ppm and from 576 ppm to 1629 ppm, respectively. Both variables are not significantly correlated according to a Spearman test (r = -0.158; Table 4). The linear regression model highlights the presence of outliers for Sr (3232 and 4666 ppm). For the two remaining studied bivalves, the very restricted number of samples precludes the use of any kind of statistical test. Both bivalve genera exhibit large ranges of trace elements. Indeed, Sr and Mg contents of Pseudolimea (n = 7) range from 613 ppm to 4627 ppm and from 440 ppm to 1343 ppm, respectively, while those for Plagiostoma (n = 4) range from 335 ppm to 2000 ppm and from 160 ppm to 1070 ppm, respectively (Table 5).

3.3. Carbon and Oxygen Isotope Ratios of Gryphaea arcuata and Other Associated Bivalves

Both δ¹³C and δ¹⁸O values for mollusc shells, especially Gryphaea arcuata, which was sampled throughout the sequence, show no significant temporal trend when plotted against the stratigraphic sequence, however, the lowest δ¹⁸O values and the highest δ¹³C values are located in the middle part of the sequence (Figure 6 and Figure 7). The dataset (Table 1) for Gryphaea arcuata is characterized by δ¹³C and δ¹⁸O values ranging respectively from 0.57‰ to 3.79‰ (mean = 2.52±0.72‰) and -3.16 to -0.67‰ (mean = -1.72±0.43‰). A significant positive correlation between δ¹³C and δ¹⁸O is observed according to both Pearson and Spearman tests (Table 3; Figure 8). The statistical analysis of data also reveals that both δ¹³C and δ¹⁸O values follow a normal distribution after the removal of a few outliers (Table 2). Indeed, two of the lowest values measured for δ¹³C (0.64 and 0.84 ‰ VPDB) and two of the highest δ¹⁸O (-0.78 and -0.67) appear as outliers in the regression linear model. Consequently, those values will not be considered further in the calculation of the living water of these bivalves. It is also worthy to note that both δ¹³C and δ¹⁸O values are not significantly correlated to either Sr or Mg contents (Table 3; Figure 9 and Figure 10). Mann-Whitney U tests were also applied to the δ¹³C and δ¹⁸O values of Gryphaea arcuata depending on the mineralogy of the host sediment to investigate its possible influence on the isotopic composition of the bivalve shells. Mean values, generally close to each other, are reported in the Table 5 depending on the lithologies that were distinguished as three groups: limestones, marly limestones and marls. The hypothesis H₀ stating similar quantitative distributions in marls-limestones and marls-marly limestones was thus verified for δ¹⁸O values with acceptable p-values of 0.305 and 0.397, respectively. The same hypothesis and groups were also verified for δ¹³C values with an acceptable p-value of 0.949 for marls-marly limestones while a very low p value of 0.082 is observed for marls-limestones, respectively. Both δ¹³C and δ¹⁸O values of Chlamys are not significantly correlated to either Sr or Mg contents (Table 4; Figure 9 and Figure 10) but are significantly correlated between themselves (Table 4; Figure 8). For the two remaining studied bivalves, the very restricted number of samples precludes the use of any kind of statistical test. δ¹³C and δ¹⁸O values of Pseudolimea (n = 7) range from 1.39‰ ppm to 3.23‰ and from -2.30‰ to -1.57‰, respectively, while those for Plagiostoma (n = 4) range from 1.32‰ ppm to 2.22‰ and from -2.46‰ to -1.79‰, respectively (Table 5).

4. Discussion

4.1. Paleoecology of Gryphaea arcuata

The paleoecology of Gryphaea arcuata can be summarized on the basis of previous studies published by [40,61,67,68,75]. Gryphaea arcuata, commonly known as the “devil’s toenail,” is a well-studied fossil oyster from the Lower Jurassic. This species is particularly notable for its distinctive, curved shell shape. Indeed, Gryphaea arcuata is characterized by its asymmetrical, strongly curved shell with a prominent umbo. The left valve is larger and more convex, while the right valve is flatter and fits snugly into the left valve. These morphological features are adaptations to its sessile lifestyle, allowing it to anchor itself to the substrate in a stable manner. As a suspension feeder and benthic organism, Gryphaea arcuata filtered plankton and organic particles from the water column. This feeding strategy indicates that Gryphaea arcuata played a crucial role in the marine food web, contributing to the cycling of nutrients within its ecosystem.

Gryphaea arcuata, like many bivalves, lived in shallow marine environments characterized by euhaline waters and at depths most likely comprised between the limit of storm wave action (< 15 m) and the lower limit of the euphotic zone, comprised between 100 m and 200 m, depending on the seafloor topography and water turbidity. This bivalve thrived in the epicontinental seas that were characterized by relatively warm to subtropical temperatures. At the margins of the Tethys marine basin in western Europe, it has been documented in the latitude range 33°–50°N, restricted to moderate to low energy environments, and muddy to sandy substrates. [40] estimated on the basis of the stable isotope ratios of their sub-mature shells that this bivalve lived in marine waters whose temperature was comprised between 14°C and 21°C.

[40] shown that the modification of shape in Gryphaea arcuata was influenced by climatic conditions, nutrient availability and water oxygenation. Two distinct morphotypes have been identified, although it is possible that a continuum of forms exists, as intermediate forms have also been documented. The first environment, characterised by tropical temperatures and high relative humidity along with a nutrient-rich ecosystem, was colonised by slow-growing Gryphaea arcuata, which exhibited the following characteristics: small, wide and thin shells. The second environment was inhabited by Gryphaea arcuata, which exhibited long, large, thick, and narrow shells with a rapid growth rate. These shells were observed to thrive in a cooler environment that aligned with the optimal requirements for Gryphaea arcuata.

4.2. Diagenetic Alteration of the Studied Samples

All the bivalve shells initially composed of aragonite were transformed into calcite, clearly demonstrating the action of diagenetic processes that may have deeply modified their geochemical composition following deposition. Indeed, despite a lower number of analysed samples relative to Gryphaea, they exhibit lower minimal, mean and maximal δ¹⁸O values (Table 5). In the case of Sr contents, only Plagiostoma and Pseudolimea show lower contents than Gryphaea. The trace element compositions and stable isotope ratios of these samples will therefore not be considered as potential proxies for their marine palaeoenvironment. Moreover, interpreting such lower δ¹⁸O values would lead to the calculation of overestimated marine temperatures [76,77,78,79]. Gryphaea arcuata, which originally consisted of low-Mg calcite (Carter, 1990), offers some interesting geochemical criteria in favour of a good preservation of its stable isotope ratios.

Indeed, a statistical analysis of the data reveals a negligible influence of the host lithology (limestone, marl-limestone, marl) on the values of δ¹⁸O, δ¹³C, Mg and Sr contents. It should be noted, however, that there are some exceptions with the samples F45, F49 and F67, which are particularly rich in Sr and hosted either in limestones or marly limestones (Table 1). It can thus be observed that there is a minimal or non-existent influence of lithological control on the values of δ¹³C, δ¹⁸O, Mg and Sr. This implies that these sedimentary alternations most likely do not represent cycles resulting from internal (volcanism) or external (orbital astronomical) factors. It can be concluded that the observed variations in the chemical composition of the sediments are not the result of climatic forcing, but rather of early diagenetic processes. These processes involve the movement of fluids containing calcium, strontium and magnesium, which occur during the dissolution of aragonite and the precipitation of calcite in chemically supersaturated zones [80,81,82,83]. It seems plausible that these processes occurred during the compaction of sedimentary layers linked to hydrostatic pressure. It can thus be postulated that these lithological alternations represent chemical diffusion fronts, which would explain why they do not reflect the original differences in nutrient availability (δ¹³C proxy) or water temperature (δ¹⁸O proxy). Consequently, the early Sinemurian may have been a period of relative environmental stability in this region of the Earth. Furthermore, there are two arguments in favour of a non-existent or minor role for late diagenesis involving interactions between these marine sediments and crustal fluids ultimately of meteoric origin.

In the case of late diagenesis, platform sediments that emerge following a sea level fall resulting from glacio-eustatic or tectono-eustatic processes, are subject to water-rock interactions involving freshwater of meteoric origin. These waters are distinguished by markedly low Sr and Mg concentrations [84,85,86,87] relative to marine waters and oxygen isotope compositions exhibiting a depletion in ¹⁸O compared to SMOW. Indeed, for latitudes between 30° and 35°, the δ¹⁸O values of meteoric waters should be between -6‰ and -4‰ (VSMOW) after a correction of data from [88], which considers the Earth without ice caps. In this framework, the δ¹⁸O of the oceans would have been -0.75(±0.10‰) instead of 0‰ on the VSMOW scale. This value is calculated considering the melting of about 2.6x10¹⁹ kg of continental ice [89], with a mean δ¹⁸O value of about -40‰ VSMOW [90], into 1.38x10²¹ kg of liquid oceanic water. Furthermore, carbon isotope compositions of fresh waters are influenced by the input of dissolved atmospheric carbon and the oxidation of soil organic matter, as evidenced for example by speleothems [91,92] or soil carbonates [93]. Consequently, this dissolved inorganic carbon (DIC) exhibits δ¹³C values that can be as low as -14‰ on the VPDB scale in the case the source of carbon is the oxidation of C3 plants [83]. On the contrary, DIC displays distinctly positive values (0.5‰ < δ¹³C DIC < +2.5‰ VPDB for an atmospheric pCO₂ of ≈ 330±10 ppm and its δ¹³C value of -7.5±0.2‰ prevailing during the 1971-1978 GEOSECS sampling program), reflecting primary productivity in the photic zone of the water column [94]. As a consequence of this late diagenesis, the values of δ¹⁸O, δ¹³C and Sr and Mg contents tend to linearly decrease, as evidenced by the bivariate diagrams (Figure 8, Figure 9 and Figure 10). In the case of the Fresville study site, however, such trends were not observed, except for Chlamys, for which a sufficient number of samples made it possible to test the significance of a linear correlation between the δ¹³C and δ¹⁸O variables (Table 4; Figure 8). The δ¹³C and δ¹⁸O data distributions for the Gryphaea samples follow normal distributions once the few outliers are removed from the data set. This observation supports the idea that diagenesis has had little or no effect on the isotopic compositions because biomineral-water interactions generally result in highly asymmetric data distributions, as has been shown by numerical simulations and data sets in the case of biogenic apatites [95].

4.3. The Marine Paleotemperatures, Water Depth and Seawater δ¹⁸O Value

Ostreids belong to the phylum of bivalve molluscs that are known to secrete shell calcite or aragonite in oxygen isotope equilibrium with the ambient water [70,96,97,98,99]. Ostreids therefore record fluctuations in the physico-chemical parameters of the aquatic environment in which they grew. The conventional calculation of marine palaeotemperatures from the δ¹⁸O of biominerals is an inherently indeterminate problem from a mathematical standpoint. Two possible approaches can be taken: either an assumption must be made about the δ¹⁸O of the water in the living environment, or it must be determined by an independent method. The theory based on ‘clumped isotopes’ is a sophisticated one, as the measurements of mass 47 (∆₄₇) allow for the calculation of an absolute temperature, thus enabling the determination of the water δ¹⁸O value. However, this method requires specific calibrations contingent on the taxa under investigation, as evidenced by a recent study [100] conducted on brachiopods whose shells are composed of low-magnesium calcite. Indeed, at a given temperature, the measured values of ∆₄₇ are distinct from those measured for foraminifera [101,102,103,104], inorganic calcites precipitated in the laboratory [105,106,107] cave calcites [106,108] and bivalve calcites [109]. In the absence of precise calibration, uncertainties in temperature may be as high as ±3°C to 5°C, which renders them of limited interest.

In the present case study, it is possible to estimate the δ¹⁸O of seawater based on a pair of data acquired on phosphate remains (teeth) of a plesiosaur associated with a shark tooth collected close to the village of Crussol in the French Ardennes [33]. The two fossils were deposited in marine sediments of Sinemurian age at a latitude comparable to that at which the remains of the Fresville invertebrate faunas were deposited, indicating that they originated from the same water mass (Figure 2). [36] have shown that plesiosaurs partially thermoregulated their bodies, with a body temperature of approximately 32°C determined at latitudes between 30° and 35°N [36]. This data can be utilized to calculate the δ¹⁸O value of the Sinemurian water mass using equation (1), adapted from [110]:

$$\mathrm{T} = \mathrm{A} \text{–} \mathrm{B}({\delta }^{18}{\mathrm{O}}_p\text{–} {\delta }^{18}{\mathrm{O}}_{bw})$$

(1)

with A = 117.4; B = -4.50; δ¹⁸O_p = 18.8‰ (VSMOW) and δ¹⁸O_bw = δ¹⁸O_w + ε¹⁸_bw-w, ε¹⁸_bw-w being the oxygen isotope enrichment factor between body water (bw) and ambient water (w). Thus, a δ¹⁸O_w value of -0.98 (‰ VSMOW) is calculated from formula (2) as follows:

$${\delta }^{18}{\mathrm{O}}_w= {\displaystyle \frac{A-T+B\left({{\delta }^{18}{\mathrm{O}}_p-ϵ}_{bw-w}^{18}\right)}{B}}$$

(2)

The temperature (T) of surface marine waters is obtained from the tooth δ¹⁸O value (19.7‰ VSMOW) of Sphenodus, which was most likely a pelagic neoselachian shark according to its morphological features [111]. Here, we highlight that his sample has been collected from the same sedimentary bed as that of Plesiosaurus. So, using equation (1), in which the value of δ¹⁸O_sw is substituted for δ¹⁸O_bw, we obtain a temperature of 24°C for the Sphenodus living marine environment.

The seawater δ¹⁸O value is then input into a calcite-water isotope fractionation equation [112] in order to calculate the average temperature of the water in which the Gryphaea lived. The temperature of 17°C is lower than that of surface water, which is in accordance with the hypothesis that these invertebrates lived in a benthic environment. In the upper part of the water column (≤ 100 m), the vertical temperature gradient is between 0.1°C and 0.2°C per metre for latitudes close to 30° [113]. Therefore, a temperature difference of 7°C between pelagic and benthic organisms can be equated to a depth difference comprised between 35 m and 70 m.

4.4. Implications on the Regional Paleoclimate and Hydrological Budget

The calculated seawater δ¹⁸O value close to -1‰ (VSMOW) is lower than that calculated in section 5.1 for a world without polar caps consisting of freshwater, which was -0.75‰ (VSMOW). A mass balance calculation indicates that this value of -1‰ corresponds to a mixture of 92‰ marine water with 8% fresh water. This mass balance is predicated on the assumption that the δ¹⁸O value of meteoric waters at these latitudes should be comprised between -6‰ and -4‰ (VSMOW), subsequent to correction for the oceanic reservoir, which was depleted by 0.75‰ relative to the present value of 0‰ (SMOW reference). The addition of 8% freshwater has a minimal impact on the salinity of marine water, with a change of approximately -1 g.L^-1. This preserves the euhaline character of the water mass, which is a defining feature of the presence of Gryphaea arcuata and ammonites like Arietites bucklandi. The presence of Gryphaea at depths of approximately 50 m is not unexpected. This average depth falls within the euphotic zone, provided that the water column is not significantly affected by turbidity and is sheltered from storm waves. Additionally, some modern and native European and Pacific oysters have been observed at depths down to 80 m [114,115]. A seawater δ¹⁸O value of approximately -1‰ (VSMOW) for the Lower Sinemurian at latitudes of 30-35°N indicates a positive water balance (precipitation/evaporation > 1), suggesting that a relatively humid climate must have prevailed over land. Assuming a mean temperature of 24°C, the climatic regime would have corresponded to what is currently designated a humid subtropical climate.

5. Conclusions

A geochemical study of the shells of molluscs present in the Sinemurian marl-limestone alternations at Fresville (Normandy, France) initially demonstrated that the low-Mg calcite shells of the oyster Gryphea arcuata exhibited resilience to the diagenetic processes inherent to marine platform deposits. In the case of the bivalves Pseudolimea, Plagiostoma and Chlamys, the polymorphic transformation of their shells from aragonite to calcite, as evidenced by Raman spectroscopy, serves to highlight the significant alterations that have occurred in their original geochemical compositions as a result of diagenetic processes. The trace element contents and stable isotope ratios of these fossils are unsuitable as proxies for marine paleoenvironmental conditions due to their lower δ¹⁸O values, which could lead to an overestimation of marine paleotemperatures. In contrast, the shells of Gryphaea arcuata, which were originally composed of low-Mg calcite, retained well-preserved stable isotope ratios. This conclusion was reached on the basis of the observation that there was a minimal lithological influence on their δ¹³C, δ¹⁸O, Mg, and Sr values. Furthermore, there was a lack of linear trends in geochemical markers that are generally the result of mineral-water interactions. Additionally, the normal distribution of δ¹³C and δ¹⁸O values in Gryphaea samples suggests that there were relatively steady environmental conditions during the Early Sinemurian. It was thus concluded that minimal late diagenesis involving meteoric waters, which typically results in a decrease in δ¹³C, δ¹⁸O, Sr, and Mg values due to freshwater interactions, occurred at the Fresville site. The combination of their δ¹³C and δ¹⁸O values with data from the literature concerning teeth (fish and Plesiosaurus) enabled the estimation of the δ¹⁸O of Sinemurian seawater (-1‰ VSMOW or -1.27‰ VPDB) in this region, located at approximately 30-35°N, as well as the temperatures of the surface waters (T = 24°C) and the benthic domain (T = 17°C) inhabited by molluscs. It is notable that the surface water temperatures of 24°C are slightly higher than those previously proposed in the literature (T = 15°C to 20°C). However, they are still within the range that could be expected for such latitudes during a greenhouse effect period with an atmosphere that is likely to have had a higher concentration of CO₂ than that of the Quaternary. The estimated average temperature of 17°C for Gryphaea arcuata is consistent with the temperature range that is optimal for the survival of these benthic molluscs, which inhabit a depth of approximately 50 m. In addition, ¹³C-enriched carbon isotope compositions of Gryphaea arcuata calcite shells relative to the VPDB marine reference indicate a marine environment characterized by a high rate of marine productivity. The seawater δ¹⁸O value of approximately -1‰ (VSMOW) for the Lower Sinemurian at latitudes of 30°N-35°N indicates a positive water balance (precipitation/evaporation > 1), which suggests that a relatively humid climate must have prevailed over land. Assuming a mean temperature of 24°C for surface marine waters, the climatic regime would have corresponded to a humid subtropical climate, as observed nowadays in the southeastern United States, northern Argentina, southeastern China, or eastern Australia, according to the Köppen-Geiger climate classification.