Mathieu Daëron

High-precision measurements of carbonate oxygen-17 compositions are increasingly used for studying paleo-environments, hydrology, diagenesis, or biomineralization processes. Calibrating triple oxygen isotope equilibrium fractionations between calcite and water as a function of temperature provides a critical baseline for many applications. Here we report new observations constraining these fractionations between a suite of natural and synthetic carbonates and their parent waters, obtained using VCOF-CRDS, a new spectroscopic method providing ppm-level metrological linearity. Our observations on slow-growing natural calcites collected from Laghetto Basso and Devils Hole and those of our synthetic carbonates that closely approach δ18O equilibrium are in good agreement with the latest theoretical models of calcite/water equilibrium fractionations between 8 °C and 200 °C. Data/model agreement is maximized when assuming that reference materials IAEA-603 and NBS18 have Δ17OVSMOW values of −0.068 ‰, and −0.042 ‰ respectively. These estimates are consistent with some but not all of the observations reported so far by other groups. A sensitivity analysis shows that varying these parameters can strongly bias estimates of equilibrium 17θ[calcite/water], whereas reformulating the equilibrium calibration in terms of chemically homogeneous comparisons (Δ17O[carbonate/IAEA-603] − Δ17O[water/VSMOW]) is entirely unaffected by Δ17O[IAEA-603/VSMOW] but sensitive to the Δ17O difference between IAEA-603 and NBS18. We conclude that pinning down the relative Δ17O values of NBS18 and IAEA-603 is the most pressing short-term need for paleo-reconstructions. We also recommend systematically normalizing carbonate Δ17O measurements using two carbonate reference materials with sufficiently different δ18O, and fully describing these normalization choices when reporting new Δ17O measurements.

Antarctic ice cores and ocean-sediment records preserve evidence for an increase in the amplitude of glacial-interglacial cycles at around 430 ka, known as the Mid-Brunhes Transition (MBT). However, similar evidence from non-polar terrestrial environments is rare, casting some doubt on the global extent of this transition. Here we present a multi-proxy speleothem record from Corchia Cave (Alpi Apuane, Italy) that spans the MBT. It comprises a stacked d18O and d13C time series from multiple stalagmites anchored in time by U-Th and U-Pb ages; and trace element, 87Sr/86Sr, and d18O and d13C profiles from a subaqueous calcite deposit (CD3) that has grown continuously from 970 ka to the present. We anchored the CD3 record to the chronology of a stalagmite stack by synchronisation of their respective d18O and d13C profiles. CD3 is well suited to this study because it yields a suite of proxies from just a single specimen that covers multiple glacial-interglacial cycles either side of the MBT. In particular, its d13C profile provides a reference for comparing the amplitude of glacial-interglacial temperature changes at Corchia to globally integrated ice-volume (LR04 benthic 18O/16O stack) and greenhouse gas (ice-core CO2 and CH4)time series. The CD3 temperature record builds on a previous trace element study, which revealed that the Mg/Ca in this speleothem is strongly influenced by mineralisation temperature (a proxy for external air temperature at the cave site). This is supported by subsequent clumped-isotope palaeothermometry. We thus developed a continuous palaeotemperature time series for CD3 extending to ~650 ka via a Mg-D47 transfer function. The temperature profile reveals compelling evidence for a shift in glacial-interglacial amplitude across the MBT. Temperatures during the interglacials of MIS15e, 15a and 13a are lower in Corchia compared to those of MIS11c, 9e, 5e and the Holocene; temperatures during MIS7e and 7c are the exception, only reaching the levels of the pre-MBT interglacials. Minimum glacial temperatures for MIS16 and 14 are warmer in Corchia than those of the subsequent glacial maxima, and the MIS12 and 6 glacials are the coldest of the last 650 kyr. All of these patterns are consistent with existing global ice-volume and greenhouse gas records but provide a rare and important terrestrial perspective. This finding confirms previous assessments that the MBT was global in extent.

The phenomenon of polar amplification causes high-latitude warming to exceed the global mean and enhances seasonal variations in temperature and precipitation. As such, it complicates the understanding of both modern and past global warming impacts in polar regions. Characterizing the climate change in polar regions is further hampered by limited data coverage and persistent challenges in interpreting local paleoenvironmental archives.To help address these knowledge gaps, we present bulk and seasonally resolved stable oxygen (δ¹⁸O) and clumped isotope (Δ₄₇) data from exceptionally well-preserved bivalves from the Early Jurassic North Pole (eastern Siberia), spanning the late Pliensbachian icehouse-Toarcian hothouse transition. Upper Pliensbachian Harpax specimens show pronounced seasonal δ¹⁸O variability of 1.5-6 ‰, corresponding to apparent temperature ranges of 6-26 °C assuming invariant seawater δ¹⁸O. In contrast, seasonally resolved clumped isotope data do not yield a statistically significant seasonal temperature difference (1.4 ± 1 °C). Together with a low mean Δ₄₇ temperature of ~3.5 ± 1 °C, this discrepancy implies that the large δ¹⁸O amplitude reflects strong seasonal variability in seawater δ¹⁸O, likely driven by enhanced precipitation, and/or meltwater input in a relatively proximal sedimentary setting with near-freezing mean annual temperatures.By contrast, Toarcian Dacryomya jacutica specimens show reduced δ¹⁸O variability of 1-2 ‰ (~4-8 °C) and a larger, statistically significant seasonal Δ₄₇ temperature difference of 5.9 ± 1.3 °C, with a mean Δ₄₇ temperature of ~10.3 °C. The agreement between δ¹⁸O and Δ₄₇-derived seasonality indicates a limited contribution of seawater δ¹⁸O variability during the Toarcian and points to a shift toward more distal sedimentary conditions. Collectively, these results provide one of the first quantitative constraints on Arctic temperature seasonality under greenhouse climate conditions in deep time.

In many carbonate archives, Δ47 signatures appear to be primarily driven by crystallization temperatures, with little evidence for other influencing factors, implying that 13C and 18O isotopes are effectively (re)distributed among carbonate isotoplogues in accordance with thermodynamic stability during or just before mineralization. This is not the case for all types of carbonates, but appears to hold true for biocarbonates such as bivalves, gastropods, or planktic foraminifera. For historical reasons, things are not as clear-cut when it comes to benthic foraminifera, a particularly important source of information on past marine environments at the scale of the Cenozoic and beyond. In hope of fostering productive discussions, we revisit this issue with a focus on the following questions:What is the current body of evidence from modern/recent observations?How much do the various Δ47 calibrations currently applied to foraminifera differ?Is there any practical difference between Δ47 calibrations based exclusively on modern/recent foraminifera and "composite" calibrations based on many different types of carbonates?What should be the foraminifer Δ47 community's next steps to try and resolve these issues?

Reconstructing past climate dynamics on terrestrial environment remains a major challenge in paleoclimate research. Improving our understanding of how continental ecosystems responded to abrupt climate oscillations is essential for assessing future climate impacts on terrestrial environments and human societies. While ice-core and marine archives document large-scale and rapid climate variability, the links between climate and continental surface processes remain poorly constrained. Identifying robust climate proxies in continental sedimentary records is therefore crucial.Fossil earthworm calcite granules preserved in loess–paleosol sequences have recently emerged as promising archives of past climate conditions, providing insights into temperature and precipitation during the last glacial period in Western Europe. However, the climatic interpretation of these proxies requires a robust calibration based on modern earthworm calcite granules to better constrain the environmental and biological parameters controlling granule formation, such as temperature, soil moisture, and litter composition.Here, we present an experimental calibration approach using modern earthworms (Lumbricus terrestris) reared under controlled environmental conditions. Soil temperature and food sources were systematically varied to assess their influence on granule production and isotopic signatures. Calcite granules were analysed for δ¹⁸O and δ¹³C, while δ¹³C was also measured in soil organic matter and litter. For the first time, clumped isotope (Δ₄₇) measurements were performed on earthworm calcite granules, allowing direct temperature estimates independent of past soil-water δ¹⁸O.This experimental approach provides new constraints on vital effects and isotopic fractionation in earthworm calcite granules and improves their use as quantitative paleoclimate proxies. Our results complement previously established empirical relationships between (i) the oxygen isotopic composition of meteoric water, granules, and temperature, and (ii) the δ¹³C of litter and the δ¹³C of granules, strengthening the potential of earthworm calcite granules for reconstructing past terrestrial climate dynamics.

Rough surfaces known as stylolites are common geological features that are developed by pressure solution, especially in carbonate rocks, where they are used as strain markers and as stress gauges. As applications are developing in various geological settings, questions arise regarding the uncertainties associated with quantitative estimates of paleostress using stylolite roughness. This contribution reports for the first time a measurement of the temperature at which pressure solution was active by applying clumped isotopes thermometry to calcite cement found in jogs linking the tips of the stylolites. This authigenic calcite formed as a redistribution of the surrounding dissolved material by the same dissolution processes that formed the extensive stylolite network. We compare the depth derived from these temperatures to the depth calculated from the vertical stress inversion of a bedding parallel stylolite population documented on a slab of the Calcare Massiccio formation (early Jurassic) formerly collected in the Umbria-Marches Arcuate Ridge (Northern Apennines, Italy). We further validate the coevality between the jog development and the pressure solution by simulating the stress field around the stylolite tip. Calcite clumped isotopes constrain crystallization to temperatures between 35 and 40 °C from a common fluid with a δ¹⁸O signature around −1.3‰ SMOW. Additional δ¹⁸O isotopes on numerous jogs allows the range of precipitation temperature to be extended to from 25 to 53 °C, corresponding to a depth range of 650 to 1900 m. This may be directly compared to the results of stylolite roughness inversion for stress, which predict a range of vertical stress from 14 to 46 MPa, corresponding to depths from 400 to 2000 m. The overall correlation between these two independent depth estimates suggests that sedimentary stylolites can reliably be used as a depth gauge, independently of the thermal gradient. Beyond the method validation, our study also reveals some mechanisms of pressure solution and the associated p,T conditions favouring their development in carbonates.

Foraminiferal isotopes are widely used to study past oceans, with different species recording conditions at different depths. Their δ¹⁸O values record both seawater oxygen‐18 and temperature according to species‐specific fractionation factors, while their Δ₄₇ signatures likely depend only on temperature. We describe an open‐source framework to collect/combine data relevant to foraminiferal isotopes, by constraining species‐specific oxygen‐18 fractionation factors (¹⁸α) based on culture experiments, stratified plankton tows or core‐top sediments; compiling stratified plankton tow constraints on living depths for planktic species; extracting seawater temperature, δ¹⁸O, and chemistry from existing databases for any latitude, longitude, and depth‐range; inferring calcification temperatures based on the above data. We find that although ¹⁸α differs between species, its temperature sensitivity remains indistinguishable from inorganic calcite. Based on > 2,600 observations we show that, although most planktic δ¹⁸O values are consistent with seawater temperature and δ¹⁸O over their expected living depths, a sizable minority (12%–24%) have heavier‐than‐predicted δ¹⁸O, best explained by calcification in deeper waters. We use this framework to revisit three recent Δ₄₇ calibration studies of planktic/benthic foraminifera, confirming that planktic Δ₄₇ varies systematically with oxygen‐18‐derived temperature estimates, even for samples whose δ¹⁸O disagrees with assumed climatological conditions, and demonstrating excellent agreement between planktic foraminifera and modern, largely inorganic Δ₄₇ calibrations. Benthic foraminifera remain ambiguous: modern benthic Δ₄₇ values appear offset from planktic ones, yet applying equilibrium Δ₄₇ calibration to the Cenozoic benthic foraminifer record of Meckler et al. (2022, https://doi.org/10.1126/science.abk0604) largely reconciles it with δ¹⁸O‐derived temperatures, with discrete Δ₄₇/δ¹⁸O discrepancies persisting in the Late Paleocene/Eocene/Plio‐Pleistocene.

Least-squares regression methods are mathematically powerful, conceptually and computationally simple, and widely used in many fields. However, none of the commonly-used flavors of least-squares regression, such as York regression or Generalized Least Squares (GLS), take into account the full set of covariances between all observed (x,y) values. Here we describe the Omnivariant Generalized Least Squares (OGLS) method to fit a model of the form y=f(x), accounting for the full error correlation structure of the (x,y) data, based on a first-order linear propagation of the uncertainties in all variables into errors in y residuals, followed by minimizing the vector of y residuals with respect to the Mahalanobis norm defined by its covariance matrix. This approach may be described as a generalization of both York regression and GLS. It is mathematically exact for straight-line fits, and is also suitable for many non-linear models. Here we describe the principles of OGLS regression and discuss its properties, caveats, and practical use, and provide two consistent open-source implementations in Python and R. To illustrate how various fields of geochronology and stable-isotope geochemistry may benefit from this new method, we discuss how OGLS may specifically apply to ⁴⁰ Ar/ ³⁹ Ar dating and how it provides robust mathematical evidence that Δ ₄₇ carbonate calibrations in the recently defined I-CDES metrological scale are statistically indistinguishable, effectively solving long-standing methodological discrepancies.

Foraminiferal isotopes are widely used to study past oceans, with different species recording conditions at different depths. Their δ18O values record both seawater oxygen-18 and temperature according to species-specific fractionation factors, while their Δ47 signatures likely depend only on temperature. We describe an open-source framework to collect/combine data relevant to foraminiferal isotopes, by constraining species-specific oxygen-18 fractionation factors (18α) based on culture experiments, stratified plankton tows or core-top sediments; compiling stratified plankton tow constraints on living depths for planktic species; extracting seawater temperature, δ18O, and chemistry from existing databases for any latitude, longitude, and depth-range; inferring calcification temperatures based on the above data. We find that although 18α differs between species, its temperature sensitivity remains indistinguishable from inorganic calcite. Based on > 2600 observations we show that, although most planktic δ18O values are consistent with seawater temperature and δ18O over their expected living depths, a sizable minority (12–24 %) have heavier-than-predicted δ18O, best explained by calcification in deeper waters. We use this framework to revisit three recent Δ47 calibration studies of planktic/benthic foraminifera, confirming that planktic Δ47 varies systematically with oxygen-18-derived temperature estimates, even for samples whose δ18O disagrees with assumed climatological conditions, and demonstrating excellent agreement between planktic foraminifera and modern, largely inorganic Δ47 calibrations. Benthic foraminifera remain ambiguous: modern benthic Δ47 values appear offset from planktic ones, yet applying equilibrium Δ47 calibration to the Cenozoic benthic foraminifer record of Meckler et al. (2022) largely reconciles it with δ18O-derived temperatures, with discrete Δ47/δ18O discrepancies persisting in the Late Paleocene/Eocene/Plio-Pleistocene.