Clay hydroxyl isotopes show an enhanced hydrologic cycle during the Paleocene-Eocene Thermal Maximum

The Paleocene-Eocene Thermal Maximum (PETM), occurring approximately 55.9 million years ago, was a period of abrupt global warming linked to a massive carbon injection into the ocean-atmosphere system. This is evidenced by a significant carbon isotope excursion (CIE). While evidence suggests substantial hydrological changes during the PETM, the precise nature and timing of these alterations have been difficult to ascertain due to a lack of suitable proxies. This study addresses this limitation by focusing on the isotopic composition of hydroxyl groups (OH-) in clay minerals. The North Sea Basin provides a highly expanded PETM section, ideal for detailed analysis. The researchers hypothesize that analyzing the isotopic composition of hydroxyl groups (δ²Hон and δ¹⁸Oон) within clay minerals, alongside bulk oxygen isotopes (δ¹⁸O bulk), will offer a more refined understanding of hydrologic variability during the PETM. This is because hydroxyl isotopes are less influenced by the compositional changes in clay minerals due to mixing or inherited signals, unlike bulk isotope values. Understanding the PETM hydrologic cycle is crucial for informing our understanding of present and future climate change responses. Previous studies have yielded conflicting evidence, with some suggesting increased terrestrial runoff in various regions, while others indicated increased aridity or heightened seasonality. Quantifying these hydrologic changes and their precise timing relative to the CIE is essential for improving climate models and predictions.

Existing literature on the PETM highlights the significant global warming event and associated carbon isotope excursion (CIE). Studies utilizing palynological data, biomarker isotopes, sedimentology, and paleosol analysis have indicated regional variations in hydrologic responses, including increased terrestrial runoff in some areas (e.g., Venezuela, Arctic Spitsbergen, New Zealand, North Sea, Arctic Ocean) and increased aridity or seasonality in others (e.g., southern Rocky Mountains, Spanish Pyrenees, Normandy). However, these studies often lack the precision and quantitative detail needed for a comprehensive understanding of the PETM hydrologic cycle and the timing of these changes relative to the CIE. The debate surrounding the origin of the increased kaolinite deposition during the PETM highlights the need for more robust proxies. Early interpretations attributed the increase to contemporaneous kaolinite formation, suggesting increased weathering under humid conditions. Later studies, considering the time required for regolith kaolinitization (estimated at one million years), proposed that the increase was due to kaolinite that formed before the PETM. However, this timescale refers to formation from unweathered bedrock and may not apply to the kaolinization of pre-existing clays within soil profiles. The authors intend to address these uncertainties.

The study analyzed 22 samples from the Sele Formation in the central North Sea Basin (well 22/10a-4). These samples consisted of <4 µm size fractions of sediments. Prior studies had already characterized the bulk organic matter δ¹³C and clay mineralogy of this section. The clay mineral assemblage comprised kaolinite, illite, illite-smectite (I-S), and chlorite. The researchers used Differential Thermal Isotope Analysis (DTIA) to measure the isotopic composition of hydroxyl groups (δ²Hон and δ¹⁸Oон) in the clay minerals. The DTIA method involves heating samples to release water molecules, enabling the analysis of hydrogen and oxygen isotopes in the released water. Before DTIA measurement, samples were treated with a sodium acetate/acetic acid buffer solution to remove carbonates. The δ¹⁸O and δ²H values were calculated by integrating the H₂O, δ¹⁸O, and δ²H traces for the hydroxyl peak and correcting for background and calibration. Multiple measurements were taken for each sample to assess analytical uncertainty. For bulk oxygen isotope measurements (δ¹⁸O bulk), samples were treated with HCl to remove carbonates and with H₂O₂ to remove organic matter before laser fluorination. The resulting O₂ gas was analyzed on a ThermoFischer Scientific MAT 253 isotope ratio mass spectrometer. To estimate the hydrogen isotope composition of clay mineral formation source water (δ²Hон-sw), mineral-specific hydrogen fractionation factors were used along with measured δ²Hон and clay mineral content. Calculations involved accounting for the proportional contribution of illite-smectite and kaolinite to the total hydroxyl water, utilizing their respective fractionation factors. The authors acknowledge the limitations of lacking reliable hydroxyl oxygen isotope fractionation factors for clays, making similar calculations impossible for δ¹⁸Oон. Correlations between isotope values and clay mineral proportions were analyzed using ordinary least squares regression.

The study revealed a significant decrease in clay OH hydrogen isotope values (δ²Hон) of approximately 8‰ at the PETM onset, coinciding with a rise in kaolinite content. This indicates enhanced rainfall and weathering, suggesting an intensified hydrologic cycle. The δ²Hон values returned to pre-PETM levels well before the end of the CIE, implying short-lived hydrologic changes relative to carbon cycle perturbations. In contrast to δ²Hон, δ¹⁸Oон changes appeared more closely related to clay composition and δ¹⁸O bulk, with weaker correlation to hydrologic changes. Bulk clay δ¹⁸O values showed a positive correlation with illite-smectite oxygen content and a stronger negative correlation with kaolinite oxygen content than δ¹⁸Oон. The estimated hydrogen isotope composition of clay mineral formation source water (δ²Hон-sw) closely tracked the measured δ²Hон, but with greater variability. Analysis of correlations between isotope values and clay mineral proportions suggested that compositional effects were weak, implying that δ²Hон variations reflect changes in source water isotopic composition rather than clay composition. The observed decrease in δ²Hон-sw was consistent with n-alkane δ²H data from the Normandy Vasterival section, further supporting the interpretation of changes in source water. The authors argue that the observed changes in δ²Hон are best explained by clay formation and alteration during the PETM rather than erosion of pre-existing deposits. The increase in kaolinite is interpreted as resulting from the rapid transformation of 2:1-type phyllosilicates (smectite and illite) to 1:1-type clays (kaolinite) within soil profiles, rather than slow regolith kaolinization. The 8‰ VSMOW decrease in δ²Hон at the PETM onset is interpreted to reflect a decrease in the δ²H of rainfall, potentially due to the 'amount effect' (a negative correlation between rainfall amount and isotopic composition) and increased precipitation intensity. The δ²Hон values returned to pre-PETM baseline before CIE termination, indicating short-lived increased rainfall intensity. The study also notes that the δ²Hон decrease seems to precede the CIE, potentially suggesting hydrologic changes not solely driven by carbon excursion, although this could be influenced by complex factors in CIE onset.

The findings support the hypothesis that the PETM was associated with an intensified hydrologic cycle. The observed decrease in δ²Hон at the PETM onset, coinciding with increased kaolinite and low-salinity dinoflagellate abundances, strongly indicates increased rainfall and runoff. This aligns with previous interpretations suggesting that global warming during the PETM caused an intensified hydrologic cycle and northward migration of storm tracks. The short-lived nature of the hydrologic changes, indicated by the rapid return of δ²Hон to pre-PETM values, highlights the dynamic interaction between the carbon cycle and the hydrologic system. The study contributes to our understanding of the PETM by providing quantitative evidence of the magnitude and timing of hydrologic changes using a novel proxy (clay hydroxyl isotopes). The results demonstrate the potential of clay hydroxyl isotopes as a valuable tool for reconstructing past hydrologic conditions, particularly during periods of rapid climate change. The observed pre-PETM decrease in δ²Hон may reflect increased terrestrial input due to lowered sea levels. This indicates that hydrologic changes were not solely a response to the CIE but may have been influenced by other factors, such as tectonic activity or sea-level changes.

This study successfully used a recently developed method (DTIA) to measure δ²H and δ¹⁸O of hydroxyl groups in clay minerals to reconstruct past hydrological changes during the PETM. A significant decrease in δ¹⁸Oн coinciding with the CIE was observed, indicating increased precipitation possibly caused by increased tropical cyclone activity in response to global warming. Concomitant increases in low-salinity dinoflagellates support this interpretation. Hydrologic changes were short-lived compared to the CIE. These results imply increased kaolinization, weathering intensity, and humidity in response to warming events. Future work should focus on improving constraints on hydrogen isotopic fractionation factors for clay hydroxyl groups to refine paleoprecipitation δ²H variability estimates. Further studies in other PETM sections are needed to determine the geographical extent of these changes. The use of clay hydroxyl group isotopes as a proxy shows promise for improving our understanding of paleoclimate, provided compositional changes in clay mineralogy are well characterized.

The study acknowledges several limitations. The lack of reliable hydroxyl oxygen isotope fractionation factors for clays prevented similar calculations for δ¹⁸Oон. While the authors argue that organic matter contamination has minimal impact, more research is needed to fully quantify this effect. The interpretation relies on assumptions about the relative contributions of illite-smectite and kaolinite to hydroxyl water and their fractionation factors, which might need refinement. The interpretation of the pre-PETM decrease in δ²Hон as solely reflecting increased terrestrial input could be challenged by other factors. Finally, the findings are based on a single, geographically limited section, necessitating similar studies in other PETM sections to ascertain the broader applicability of the conclusions.