Clumped isotope analysis of zoned calcite cement, Carboniferous, Isle of Man

Sequential analyses of δ13C, δ18O and Δ47 values of calcite and dolomite deposited in millimetre‐sized cavities are reported from the Ronaldsway Member packstones, Isle of Man. The Ronaldsway brachiopods have δ13C values of ca +2.3‰ and δ18O values of ca −7.2‰; carbon is like predicted Carboniferous values, while oxygen values are more negative. The brachiopods show preserved microstructure but have marginal alteration and a streaky cathodoluminescence pattern. Crinoid ossicles have δ13C values of ca +2.3‰ and one with a δ18O value of ca −3.1‰, compatible with Carboniferous marine precipitates; three samples have δ18O values of ca −6.5‰ and are 18O‐depleted. Calcite stages 1 and 2 have δ13C values ca +3.2‰ and δ18O values ca −2.5‰, compatible with Carboniferous sea water. Stage 1 and 2 have non‐luminescent to orange CL zones. Stage 1 and early stage 2 contain red luminescent dolomite micro crystals generated during Mg calcite stabilisation. The Δ47 values for stage 1 and 2 cements indicate temperatures of 86 and 105°C that occurred after the stabilisation of Mg calcite. Stage 3–8 zoned cements preserve their original growth surfaces and their δ13C and δ18O values suggest precipitation during burial and exhumation. The Δ47 values of the brachiopods and crinoids indicate temperatures between 85 and 140°C indicating they were either recrystallised at high temperatures or affected by solid state reordering. To evaluate these alternatives two quantitative models, water–rock reaction and reordering models are performed. The allochems and cements are progressively altered by porewater towards the fluid‐buffered behaviour. The quantitative evaluation of calcite and dolomite solid‐state reordering suggests the elevated clumped isotopic temperatures are produced by interaction with hydrothermal fluids. This study improves understanding by applying previously untried techniques; further Δ47 data and quantifying elemental variations would help further interpretation but the poorly documented post‐depositional history is a drawback.


| INTRODUCTION
The oxygen isotopic composition of calcites has been widely used as a geothermometer (Emiliani, 1956;Epstein et al., 1951;Kim & O'Neil, 1997). Originally, the large sample size (>10 mg) needed for the oxygen isotope technique restricts its application. Advances in technology, however, has allowed increasingly smaller quantities to be analysed making isotopic variation within crystals possible. Perhaps the first study to document sequential δ 18 O changes within zoned calcite cement crystals in a carbonate rock was made by Dickson and Coleman (1980) on rocks from the Isle of Man. They interpreted variation in their data as due to postdepositional temperature changes. Individual components were identified by staining and were excised from thin sections by scraping with a dental tool. The δ 13 C and δ 18 O values of the powders were measured and used to construct a paragenetic sequence. However, the oxygen isotopic fractionation equation used in the Dickson and Coleman (1980) study, and in many others, has three unknowns only one of which, the δ 18 O value of the calcite, is determined. The δ 18 O value of the precipitating fluid and the temperature of precipitation is either guessed or assumed using additional information, such as burial history, fluid inclusion data, vitrinite reflectance or other temperature sensitive geochemical proxies. This situation changed with the advent of the clumped isotope geothermometer (Δ 47 and Δ 48 values) (Ghosh et al., 2006;Swart et al., 2021) which determines temperature directly, at the same time as the stable C and O isotopic measurements are made, assuming the calcite precipitates under equilibrium conditions. This paper re-examines material collected from the same locations as those used by Dickson and Coleman (1980). A similar paragenetic sequence has been established using staining and cathodoluminescence (CL) (Sippel, 1965). Clumped isotope measurements, however, show the original isotopic composition of these components in the Ronaldsway rocks have not been preserved and they contribute to a better understanding of the rock's diagenetic history.

| GEOLOGIAL SETTING
The samples used for this study have been collected from the Ronaldsway Member of the Derbyhaven Formation and are of Carboniferous age (Dickson et al., 1987) exposed in Turkeyland quarry (54.088437°N, −4.607540°W) and archived samples from a coastal outcrop at Ronaldsway (54.085151°N, −4.607178°W). The coastal outcrop is now concealed beneath an extension to the runway for the Isle of Man Airport (Figure 1). The Ronaldsway Member is composed of centimetre thick layers of calcareous illitic claystone that separate metre thick, graded grainstones deposited on a mid-ramp setting from high-energy storm events (Chadwick et al., 2001). The irregular upper surface of the Ronaldsway Member in Turkeyland quarry is a hardground overlain by shales of the Skillicore Member. The analysed samples come from the middle of the Ronaldsway Member, 10 m below its top.
Packstones from the Ronaldsway Member often have a depositional framework constructed from millimetre-sized crinoid ossicles and centimetre-sized brachiopods and mollusc shells; a fine sand-sized sediment accumulated on the upper surfaces of this framework composed of carbonate peloids, micrite, quartz, feldspar and muscovite grains ( Figure 2). The sampled packstones contains dispersed corals (Heterophyllia, Siphonodendron and Michelinia), bryozoa, ostracods, Konickopora and foraminifera. Shelter pores, millimetres in size occur within the framework, some were enlarged by the dissolution of aragonitic shells. These shelter pores contain the largest, chemically zoned calcite cement crystals; their variations in ferrous iron content are revealed by stain colours from mauve to purple, to blue with increasing iron content ( Figure 2). These zoned calcite cement crystals are the focus of this study.

| Petrology
The alizarin red S/potassium ferricyanide acidified staining solution (Dickson, 1965) was applied to thin sections and examined using the petrographical light microscope.
Colour CL images were collected using a custom-made cold-cathode instrument. An accelerating potential of 26 kV generated a gun current of 450-600 mA with an air chamber pressure of 0.01-0.05 Torr. The chamber sits on a Nikon Optiphot microscope connected by a c-mount to an Optronics Magnafire peltier digital camera. Monochrome CL images were collected using a Centaurus collector mounted on Quanta 650F, Field Emission Gun Scanning Electron Microscope (FEGSEM) using a 15 kV beam and a working distance of 13 mm.

| Clumped isotopes
The 23 samples analysed for clumped isotopes (Δ 47 ) (ca 1 mg) were obtained by scraping two stained thin sections, using a tungsten steel needle, under a binocular microscope; one section from Turkeyland Quarry and the second a few metres away from a coastal outcrop (Figure 1).
Clumped isotope measurements were performed at the Godwin Laboratory for Palaeoclimate Research, University of Cambridge, United Kingdom using a Thermo Scientific MAT 253 mass spectrometer. Between 6 and 15 aliquots (ca 100 μg) of each sample were reacted with orthophosphoric acid at 70°C using a Thermo Scientific Kiel IV carbonate device. The cryogenic trapping system of the Kiel device was modified by adding a Porapak trap that is cooled with two Peltier elements to remove organic compounds and isobaric contaminants prior to isotopic measurements (Petersen et al., 2016;Schmid & Bernasconi, 2010). This trap is cooled to ca -12°C during each run and baked out for at least 1 h before the next run. For each replicate, the initial beam intensity of m/z 44 was around 20 V and decreased to ca 12 V over the course of eight cycles. Pressure sensitive negative backgrounds on the rare isotopologue masses were determined before each run by performing peak shape scans on all masses at different intensities on m/z 44 (25, 20, 15 and 10 V). These data and conversion of the background corrected raw data into the Intercarb carbon dioxide equilibrated scale (I-CDES) (Bernasconi et al., 2021) were processed using the Easotope software (John & Bowen, 2016).
The δ 13 C and δ 18 O values of the carbonates were generated at the same time as the clumped isotope analyses and all data for the carbonate components are reported relative to Vienna Pee Dee Belemnite (V-PDB) using the conventional notation.
The oxygen isotope composition of the precipitation water is reported relative to Vienna standard mean ocean water (V-SMOW) and was calculated using δ 18 O values of the analysed calcite, Δ 47 derived temperatures and the calcite-water oxygen fractionation equations of Kim and O'Neil (1997)

| Calibration equation
For conversion to temperature, a previous calibration (Breitenbach et al., 2018) generated at the University of Cambridge was used and combined with standards (1) 1000 ln a calcite − water = 18.03 10 3 ∕T − 32.42 (2) 1000 ln a dolomite − water = 3.14 10 6 ∕T 2 − 3.14 F I G U R E 1 Maps showing the location of samples. Inset shows the area occupied by the Castletown Carboniferous in green to the south of the Isle of Man; area of the red box is enlarged as the main map around Ronaldsway airport, Isle of Man. Stratigraphic boundaries shown by thin black lines, faults by thick black lines and sample sites by red numbers (after Dickson et al., 1987). produced at the University of Miami, but also analysed at the University of Cambridge (Table 1, Figure 3). Calibrations between temperature and Δ 47 and Δ 48 values for these samples measured at the University of Miami have been previously published (Staudigel et al., 2018;Swart et al., 2019Swart et al., , 2021. As the data published in Breitenbach et al. (2018) were corrected using values for the ETH standards published by Meckler et al. (2014), these values have been updated using the newly published values for these standards (Bernasconi et al., 2021). A new calibration has been produced combining these values together with the University of Miami standard, reacted at 70°C. As all materials have been analysed at 70°C no acid fractionation factors have been applied and an equation specifically designed to be applicable at a reaction temperature of 70°C has been developed (Equation 3).
Adjusting this to a reaction temperature of 90°C, using a constant acid fraction factor of 0.032‰ yields Equation 4. Although the intercept of this equation is slightly higher than that reported by Swart et al. (2019), the equations are statistically identical at the 95% confidence limits.

| Analytical data
The analytical data are given in Table 2.

| Brachiopods
Two brachiopods were analysed; a chonetid with a characteristic millimetre thick corrugated fibrous shell and an abraded pseudopunctate productid shell a few millimetres thick. In plane light, the original distinctive microstructure of these shells is well preserved at the scale of the light microscope. However, both shells stain pink with blue margins indicating the shells are peripherally composed of ferroan calcite and they both show streaky, orange-coloured CL indicating alteration. They have low δ 18 O values (−8.8 and − 8.6‰) and positive δ 13 C values (+2.2 and + 3.1‰); their Δ 47 values are 0.423 and 0.427‰.
Transmitted plane light image of stained thin section from Ronaldsway Member, Derbyhaven Formation (Viséan), Isle of Man. The large crinoid (LC) has geopetal sediment resting on its upper surface that includes a smaller crinoid (SC); both crinoids have syntaxial overgrowths with stages numbered 1 (oldest) to 8 (youngest). Crinoid LC has an orange, mauve and blue patchy stain indicating the presence of varying amounts of ferrous iron. Cement stages 1 and 2 around crinoid LC; stain orange (nonferroan calcite) with irregular mauve patches (slightly ferroan calcite). Two large syntaxial crystals dominate the cement fill to the pore between the geopetal sediment and a bivalve mould above.
T A B L E 1 Samples used in the calibration presented in this paper. (1) Samples analysed by Staudigel et al. (2016) and Swart et al. (2019). (2) Samples analysed by Breitenbach et al. (2018) and Δ 47 values corrected using revised values for the ETH 1-4 standard by Bernasconi et al. (2021).

| Crinoids
The internal pores in the Ronaldsway crinoid ossicles are filled with a variety of materials, some are filled with micritic sediment, some are filled with non-ferroan calcite cement and some are filled with late ferroan calcite cement ( Figure 2). The crinoid stereom in all four analysed ossicles stains orange (non-ferroan calcite); the pore cement in crinoid 6b also stain orange but some of the pore cements in crinoids 5a, 6a and 6d stain blue (ferroan calcite). The stereom of the Ronaldsway crinoids is poorly defined. The four analysed crinoid ossicles have similar δ 13 C values

| Cements
The largest cement crystals that show eight stained zones are present as optically continuous overgrowths on crinoid ossicles (Figure 2). The internal structure of the stage 1 and 2 cement crystals is only revealed on CL images (Figures 4 and 5). Stage 1 luminesces a dull brown colour with irregularly shaped non-luminescent patches. Stage 2 has multiple bright orange coloured CL zones a few micrometres wide that alternate with dull orange wider zones tens of microns wide. 'Swarms' of red luminescing dolomite microcrystals are abundant in zone 1, a few occur in early stage 2 and none are present in late stage 2 or subsequent cement zones (Figure 4). Stages 3-8 stain either orange, mauve or deep blue ( Figure 2). The sharp contacts between these zones mark the position of former crystal faces with a dogtooth habit. In longitudinal section these crystal zones have an acute shape, in transverse section they have an irregular hexagonal shape indicating the crystal form involved was the acute scalenohedron. The orange, mauve and blue stained zones shown on Figure 2 correspond to white, grey and non-luminescent zones on Figure 5. The superior resolution of the CL images shows a complex internal pattern that is not revealed by staining; the fine zonation in some areas (e.g. stage 5 on Figure 5) is disrupted by many fluid inclusions. These areas were avoided when sampling.
The δ 13 C values for stages 3-6 decrease from +3.2 to +1.6‰ and the δ 18 O values decline from −6.2 to −11.6‰ ( Figure 6A). Stages 7-8 show a reversal of this pattern with the δ 13 C values increasing from +1.6 to +1.8‰ and δ 18 O from −10.2 to −9.7‰ ( Figure 6A). The Δ 47 values for stages 3 to 8 show a loop pattern ( Figure 6B) that is different from the δ 13 C and δ 18 O values for the same stages ( Figure 6A).

F I G U R E 4
Thin section image of cement filling a triangularshaped intergranular pore between inclined brachiopod shells above and rectangular crinoid below, Ronaldsway Member, Derbyhaven Formation, Isle of Man. (A) CL image, crinoid has patchy black to dull orange coloured luminescence with coarser reticulate pattern at its centre, remnant of ordinal stereom structure. Cement seeded on crinoid starts with irregular patchy non-luminescent areas followed by fine orange-coloured zigzag zones and distal non-luminescent cement. Red luminescent dolomite crystals a few millimetre in diameter occur in the early cement. (B) Sketch to show distribution of main components, cement divided into three stages. F I G U R E 5 SEM CL stitched images of calcite cement crystals at the centre of a grainstone pore from the Ronaldsway Member, Carboniferous limestone, Isle of Man. The inclined central crystal is sliced parallel to its c-axis; zones 2-8 are numbered. Zones 3-5 show this crystal had an acute termination during growth. Zone 2 has complex zigzag shaped minor zones whose terminal surface is overgrown discordantly by zones 3 and 4. The speckled area around the acute termination of zone 5 contain many fluid inclusions; some of the inclusions empty, they were intersected by the plane of the thin section and their fluid escaped.

| Brachiopods
The Ronaldsway brachiopods have mean δ 13 C values = +2.3‰ (n = 4; σ = 0.88; from Table 2 and Dickson & Coleman, 1980) close to the predicted values of Saltzman and Thomas (2012) for Carboniferous brachiopods that inhabited shallow tropical sea water. The Ronaldsway δ 18 O values average −7.2‰ (n = 4; σ = 0.21); much lighter than the range for Carboniferous brachiopods predicted by Grossman (2012) that is between −1‰ and − 3‰. Six well-preserved Carboniferous brachiopods from the deeply buried Bird Spring Formation, USA are reported by Shenton et al. (2015) to have average δ 13 C values = +2.7‰ (σ = 0.5) and average δ 18 O values = −1.1‰ (σ = 0.4), both compatible with their expected original values. The Ronaldsway brachiopods have retained their δ 13 C values from the time of the shells they grew but their δ 18 O values have been severely altered.

| Crinoids
The Ronaldsway crinoids were composed of High-Mg calcite (HMC) and contain pores that were filled with cement of different ages, their C and O isotopic compositions are more complicated than the brachiopods. The earliest cement contains dolomite microcrystals indicating it was also composed of HMC while stage 2 and younger cements are composed only of calcite.

| Calcite cements
The δ 13 C and δ 18 O values of stage 1 and 2 cements are within the estimated range of values usually accredited to carbonate precipitated from Carboniferous sea water. Stage 1 and 2 cement show complex internal structure by CL (Figure 4) and, along with their seed crinoids, contain dolomite microcrystals that were generated when their original Mg calcite stabilised to calcite and dolomite (Dickson, 2001;Meyers & Lohmann, 1978;Walker et al., 1990). This reaction, despite being water catalysed, did not change their isotopic composition and once completed, the products isotopic composition remained unchanged through subsequent burial and exhumation. Dickson and Coleman (1980) interpreted the cement zones in the Ronaldsway cement as a continuously evolving succession of events. However, a large drop in δ 18 O values (3.7‰) occurs between stage 2-3 that is associated with a discontinuity surface shown on Figure 5. The multiple, zig-zag zones of stage 2 are overlain by a single acute  Kim and O'Neil's (1997) equilibrium equation is on the left ordinate. Purple asterisks = brachiopods, yellow diamonds = crinoids and red spots = calcite cement zones. The brachiopod and crinoids are numbered as on Table 2 and the cement stages numbered from the oldest 1 to the youngest 8. euhedral surface of stage 3. This surface separates by a gap of unknown duration during which the sediment's pore water could have evolved.
The lowering of δ 18 O values found though zones 3-6 ( Table 2) was explained by a rise in temperature on burial by Dickson and Coleman (1980). An increase in δ 18 O values is found between stages 6-8 that were not previously sampled. A similar reversing pattern in δ 18 O values has been found in cement from the oil leg of a Cretaceous reservoir, that Cox et al. (2010) interpreted as due to temperature changes on burial and later uplift.

| Post-cement formation
Most of the millimetre-sized pores that once existed in the Ronaldsway samples are filled with calcite cement, however, in some cases, calcite is overlain by saddle dolomite, dickite, bitumen and rarely fluorite cements. Dickson and Coleman (1980) interpreted the δ 18 O values of saddle dolomite (−8.7‰; n = 4) to indicate precipitation at ca 90°C assuming a δ 18 O VSMOW value of 0‰. Sample 15 (Table 2) is a typical saddle dolomite; its euhedral crystals have curved faces and internally its' crystals show undulose extinction and curved cleavage planes. Samples 16 and 17 are inclusion rich, xenotopic dolomites that replace former carbonate. Hendry et al. (2014) report a more comprehensive study on dolomite associated with faults and fractures throughout the Castletown Carboniferous outcrop. They distinguished three types of dolomites, characterised by their δ 13 C and δ 18 O values, strontium concentrations and fluid inclusions; their type 2 dolomite is equivalent to the Ronaldsway saddle dolomite. Their types 2 and 3 dolomites were interpreted to have precipitated from saline brine at temperatures ranging from 98 to 223°C.

| Clumped isotopes
The Δ 47 values of the calcitic allochems (brachiopods, crinoids) and calcitic and dolomitic cements all indicate that the samples have either been recrystallised at high temperatures (Table 2) and/or been reset because of solidstate reordering during burial at elevated temperatures. For example, the brachiopods have a calculated temperature of 125°C while the crinoids have temperatures between 85 and 140°C. The cements have similar ranges of temperatures (86-105°C) while the dolomites have a Δ 47 value derived temperature close to the blocking temperature (85-150°C). These contrast with a range of temperatures from 80 to 300°C, calculated from fluid inclusions (Hendry et al., 2014).

| Water-rock reaction
The lowering of the δ 18 O values of carbonate cements through the zonal sequence was first interpreted as due to rising temperature during burial (Dickson & Coleman, 1980). Such alteration can be assumed to be a consequence of changes in the water-rock (W/R) ratio with the increasing burial temperature. To examine this assumption, a water-rock reaction model such as used by Banner and Hanson (1990) can be applied using a range of W/R ratios and a pore fluid of −1.5‰, typical of Carboniferous sea water (Grossman, 2012). The result of this model also captures the δ 18 O of the fluid (Figure 7). For the allochems (brachiopods and crinoids), the narrow range of W/R ratios indicates a relatively closed diagenetic environment and therefore, the elevated temperatures are mostly attributed to the closed system recrystallisation during the burial stage (Figure 7). With regard to the cements, the higher W/R ratios are associated with the late cements (Stage 3-8), while the early cements (Stage 1-2) are associated with low W/R ratio. This varied range of F I G U R E 7 Cross plot of Δ 47 derived temperatures compared to δ 18 O values of the parent fluids. The solid lines represent the modelled W/R ratios using the model from Banner and Hanson (1990). The sea water δ 18 O value (−1.2‰) for the Carboniferous is taken from Grossman (2012). The labels in the cement samples represent the growth generation in Figure 2 and Table 2. W/R ratios indicate that the diagenetic environment of late cements (Stage 3-8) is gradually buffered by the burial pore fluid (Figure 7). In contrast, two dolomites show elevated temperatures relative to the burial temperatures estimated by Newman (1999) (ca 120°C) and therefore support the occurrence of hydrothermal dolomitisation. However, the δ 18 O values in the fluid calculated from one of the saddle dolomites is much lower than the values for the allochems and the other two dolomite samples. As there is no difference in δ 13 C and δ 18 O values amongst all three dolomites, the negative δ 18 O of the fluids may be attributed to a lower formation temperature.

| Solid-State reordering
During the burial and subsequent exposure of carbonates, the reordering of C-O bonds in the solid crystal lattice can be driven by the increasing and decreasing formation temperatures in the absence of water-rock reactions (Henkes et al., 2014;Lloyd et al., 2018;Passey & Henkes, 2012). This takes place without loss of petrographic characteristics. The final temperature, that is locked in after formation, uplift and cooling, is termed the apparent blocking temperature and can be estimated using various models such as the 'transient defects/equilibrium defect model' (TD model) (Henkes et al., 2014) and/or the 'paired diffusion model' (PD model) (Stolper & Eiler, 2015). Based on numerical calculations using these models, the threshold needed to alter the Δ 47 values of calcites and dolomites by solid-state reordering is ca 115°C for a period of approximately 110 years for calcite and ca 200°C over a similar period for dolomite. To evaluate the signals of clumped isotopic temperatures in dolomite and calcite possibly resulting from solid-state reordering, the Stolper and Eiler (2015) reordering model was applied.
While at the present time the sedimentary history of the Isle of Man is unknown, the Δ 47 values can help define the burial history by examining the Δ 47 values relative to the known blocking temperatures for calcite and dolomite. However, it should be noted that large portions of the geological record are missing from the succession with Quaternary glacial deposits directly overlying the ca 350 m thick platform carbonates of the Castletown Carboniferous succession. Sediments belonging to the missing epochs are present in the offshore Irish Sea Basin, but their rate of thinning onto the island and the extent of erosive episodes are difficult to estimate due to poor well control (Chadwick et al., 2001). Kilometres of Namurian and Westphalian sediments accumulated in subsiding basins of late Carboniferous age around the Isle of Man, their deposition ended with Variscan tectonic activity that caused basin inversion and regional uplift. The burial history of the offshore drilling site in the east Irish Sea may serve as an analogue to evaluate the influence of solid-state reordering on clumped isotopes. The maximum burial depth of the Carboniferous formation is around 3,750 m corresponding to the 1.0% value of vitrinite reflectance (R o ) (Newman, 1999). A geothermal gradient of ca 30°C/ km can be calculated using vitrinite reflectance and the model of Sweeney and Burnham (1990). Correspondingly, the maximum burial temperature is assumed to be no F I G U R E 8 Modelling blocking temperatures in dolomite and calcite using the paired diffusion model. The temperatures recorded by the calcite and dolomite are indicated by the symbols on the right-hand side of the graphs. (A) The blocking temperatures produced by the burial history of Newman (1999). The solid black line shows the burial temperature used in the model and the blue and red lines the mean temperatures predicted to be recorded by the calcite and dolomite minerals. The x-axis represents the time from sample deposition (348 Ma) until the present (0 Ma). (B) The modelled blocking temperatures depend on the different burial temperatures that the samples experienced. All the dashed lines represent the upper and lower standard error of the temperature estimates. more than 120°C. However, the modelling results show that it is not possible to produce the measured Δ 47 values under these conditions ( Figure 8A). In order to fit the Δ 47 derived temperatures, the maximum burial temperature needs to be elevated to 260°C ( Figure 8B). Such an increased range of temperatures is supported by the fluid inclusion data of Hendry et al. (2014) which reports values as high as 300°C. Therefore, either the burial history is incorrect, or the Δ 47 values were produced by interaction with hydrothermal fluids that penetrated through the strata (Hendry et al., 2014).

| Solid-State reordering or water-rock interaction?
Based on the preceding discussion it is apparent that the Δ 47 values of the carbonates are consistent with either solid-state reordering and/or recrystallisation. However, based on present knowledge of the burial history of the Isle of Man, it seems that the sediments were not buried to sufficient depths to produce the Δ 47 values because of solid-state reordering. Therefore, unless the burial model is incorrect, the data suggest alteration of carbonates in the presence of a relatively low W/R ratio under known burial conditions, perhaps influenced by hydrothermally sourced fluids.

IMPLICATIONS
The Dickson and Coleman (1980) paper on the Isle of Man represented one of the first attempts to use micro sampling to take a sample apart, rather than using bulk samples, in order to understand the complex diagenetic history of a formation such as the Ronaldsway Formation. However, despite the revolutionary nature of that study, the authors still struggled with the task of solving two unknowns (temperature and the δ 18 O value of the fluid) when only one variable (the δ 18 O value of the calcite) was measured. This problem was addressed by combining the clumped isotope method, which allowed direct measurement of temperature, with use of the Kiel device for clumped isotope analysis, that allowed samples of significantly smaller size to be analysed. This paper therefore revisits and revises the conclusions of the original paper.
The understanding of the origin and diagenesis of components that form the Ronaldsway packstones has been improved through the integration of CL petrography and clumped isotope analysis with existing and newly measured C and O isotope data. The brachiopod skeletons are texturally preserved, at the scale of the light microscope, but their patchy staining for ferrous iron, streaky CL response and their low δ 18 O values (−7.2‰; n 4; σ 0.21) all indicate they have been altered. Their Δ 47 values are also incompatible with a sea water precipitate as they indicate temperatures ca 125°C and a δ 18 O VSMOW value of +10.0‰. Despite this overwhelming evidence for alteration, their δ 13 C values are like those assigned to contemporary unaltered brachiopods.
The sampling of fossil echinoderms for analysis usually results in a mixture of skeleton and intraskeletal pore fillings; these components are often difficult or impossible to distinguish making interpretation difficult. The pores in some Ronaldsway crinoids were filled with marine Mg calcite cement that retain their marine stable isotope values while others were filled with younger ferroan calcite causing their δ 18 O VSMOW values to be lowered. The Ronaldsway crinoid Δ 47 values all indicate temperatures and δ 18 O VSMOW values that are incompatible with skeletal formation.
The Ronaldsway calcite cement is divided into early stages 1 and 2 and later stages 3-8, despite coming from the same syntaxial crystals. The early stages 1 and 2 cement precipitated as Mg calcite from marine-like water; it precipitated rapidly, stabilised to a mixture of calcite and dolomite and was heated during burial. These three processes could singly or jointly have changed their isotopic values; however, their δ 13 C and δ 18 O values apparently were retained while their Δ 47 values changed.
The later stage 3-8 cements are the only Ronaldsway component to preserve their original growth surfaces. Variation in their δ 13 C and δ 18 O values suggest precipitation during burial and exhumation. The temperature and δ 18 O value of the formation water assessed from their Δ 47 values are compatible with later dolomite and dickite cements. The temperatures calculated from the late Ronaldsway cement Δ 47 values (between 100 and 140°C), however, places them in the window which can be interpreted as alteration by either recrystallisation or solidstate reordering.
Overall, the importance of the framework of burial history on constraining the interpretation of geochemical proxies and deciphering the diagenetic evolution in carbonates is highlighted. The integrated petrographicgeochemical investigation in different carbonate minerals (calcite and dolomite) is necessary in the analysis of diagenesis and palaeoenvironment.

ACKNO WLE DGE MENTS
We thank the two anonymous reviewers who helped to improve the manuscript.

CONFLICT OF INTEREST
There is no conflict of interest.