Using the boron isotope-pH proxy to investigate CO2-climate coupling in the geological past

Brown, Rachel (2024) Using the boron isotope-pH proxy to investigate CO₂-climate coupling in the geological past. University of Southampton, Doctoral Thesis, 259pp.

Record type: Thesis (Doctoral)

Abstract

Over the last 7 million years, the Earth has undergone major long-term cooling culminating in the development of continental scale ice sheets in the northern hemisphere around 2.5 million years ago. This cooling is believed to be associated with a reduction in the potent green house gas carbon dioxide (CO₂). It is well documented that anthropogenic global warming and climate change is driven by humanity’s emissions of CO₂ (and other greenhouse gasses) from fossil fuel burning, cement making and deforestation. However, various feedbacks in the Earth’s climate system (i.e., ice sheet albedo, water vapour, etc.) introduce a large amount of uncertainty regarding the expected warming for a given future emission scenario. Studying the relationship between natural CO₂ change and climate in the past enables us to investigate the role of these feedbacks and is a vital way in which we can narrow the uncertainty in the predictions of the magnitude of future climate change. To this end, in this thesis, the boron isotopic composition (δ¹¹B) of planktic foraminifera are used to reconstruct CO₂ across two periods of dramatic global climate change: the Late Miocene around 6 million years ago, and the Plio-Pleistocene Transition that occurred 2.6 million years ago. In both cases the records presented are at the highest temporal resolution to date, they thus provide unique insights into the role of CO₂ in determining the thermal evolution of the Earth and reveal valuable constraints on climate sensitivity and its dependence on background climate state. Although the δ¹¹B-(pH) CO₂ proxy is emerging as a powerful tool, the applicability of the existing δ¹¹B-pH calibration for G. ruber to deep time, and hence its reliability as a CO₂ proxy, has recently been questioned. The first aim of this thesis was therefore to investigate the “vital effects” of G. ruber using a combination of carbon, oxygen, and boron isotope (δ¹³C, δ¹⁸O and δ¹¹B) data from across a range of test size fractions. These data show that there is no meaningful difference in G. ruber vital effects or depth habitat from the Plio-Pleistocene Transition to the recent. The robustness of the G. ruber CO₂ reconstructions for the Pliocene are demonstrated through a comparison of G.ruber-derived δ¹¹B_borate, pH and CO₂ estimates from the same samples using another planktic foraminifera (T. trilobus). From the similarity of these results, and the wider study on G. ruber “vital effects” and habitat, it is concluded that there was no meaningful change in G. ruber vital effects at least since the Plio-Pleistocene Transition. Consequently, previous and future records generated with this foraminifera are likely to be accurate. Armed with this improved understanding of the proxy, a new high-resolution δ¹¹B record from across the Plio-Pleistocene Transition is presented that suggests that CO₂ dramatically decreased by ~100 ppm at ~2.8 Ma, and was repeatedly below 280 ppm, the proposed threshold for the development of northern hemisphere ice sheets, from ~2.75 Ma until ~2.4 Ma. From ~2.4 Ma, CO₂ returned to Pliocene-like levels of ~370 ppm, though there is no evidence of the removal of northern hemisphere ice sheets, suggesting that the northern hemisphere ice sheets display a hysteresis-type behaviour (Chapter 4). To gain a quantitative understanding of the role of CO₂ decline in Plio-Pleistocene cooling, a comprehensive compilation of sea surface temperature data for the last 4 million years is also constructed (Chapter 5). A combination of this record of "global" sea surface temperature data with the long-term CO₂ data confirms that Plio-Pleistocene cooling was largely driven by CO₂ decline amplified by the ice-sheet albedo feedback and augmented by the emergence of the ice feedback possibly as the result of crossing a key tipping point at ~2.5 Ma (Chapter 5). To further investigate the relationship between CO₂ and climate, a high-resolution δ¹¹B-CO₂ record during the Late Miocene (5–7 Ma) is presented(Chapter 6). This is an interval where previous studies have described ade coupling of CO₂ and climate change. The new record presented in Chapter6 reveals that, in contrast to the long-term picture, a transient CO₂ excursion to a minimum in CO₂ of ~300 ppm is coincident with a global minimum in temperature at ~6 Ma. A compilation of sea surface temperature data for the time-period reveals that CO₂ was well-coupled to changes in global temperature and displayed similar climate sensitivity to that of the Pleistocene. These new records and sea surface temperature compilations illustrate that cooling over the last 7 million years, and the development of large northern hemisphere ice sheets, was associated with substantive and often relatively rapid reductions in atmospheric CO₂. The data presented highlight the important role climate feedbacks (such as changes albedo associated with global vegetation cover and land ice) play in modulating the relationship between CO₂ and the broader climate change. To this end, it is evident that on short time scales (<100,000 yrs) CO₂ and climate appear consistently related, however this relationship is not always linear and is modified by tipping points and a state dependence to climate sensitivity. While these features are not currently incorporated into climate models, this study illustrates the power of the palaeo climate record to further our understanding of how the climate system operates.