Development of new geochemical proxies for euykariotic algal response to climatic variations and application for investigating paleoceoanographic history on orbital timescales in the last 11 million years

Abstract

In order to accurately predict how organisms will respond to the rapidly changing climatic conditions, we need to understand how they adapted to past climate changes and why some of them became extinct. Marine phytoplankton are considered key climatic regulators, and among them, especially diatoms and coccolithophores, as it is through their photosynthetic activity that they modulate CO2 variability, while promoting carbon export from the surface ocean to the deep sediments via increasing ballasting efficiency. Diatoms are responsible for more than half of the global marine primary production, being the world¿s main biosolicificators, while coccolithophores are the main carbonate producers in the modern and past ocean, highly enhancing ballasting efficiency and carbon sequestration into the deep sediments. Given the reciprocal influence of diatoms and coccolithophores on climate, the geochemical analysis of their fossils can be used to trace the paleoceanographic history of certain oceanic regions, while providing information on their evolutionary response to past climate variations. In this thesis, the biominerals produced by coccolithophores and diatoms were used to apply the previously-calibrated coccolith Sr/Ca proxy to reconstruct paleoproductivity of the Atlantic sector of the Southern Ocean during the penultimate glacial cycle, to develop the new geochemical frustule B content proxy of diatom active bicarbonate uptake, and to develop and apply the coccolith Ca isotopic composition (¿44/40Ca) proxy to evaluate the mechanisms controlling coccolith Ca isotopic fractionation. Chapter I describes how the coccolith Sr/Ca paleoproductivity record of the Agulhas Bank slope (Atlantic sector of the Southern Ocean; core MD96-2080) during the penultimate glacial-interglacial cycle, provided information to resolve temporal changes in the westerlies¿ dynamics and clarify the role of obliquity in controlling their latitudinal displacement and strength. The productivity signal was successfully derived from coccolith Sr/Ca by applying a temperature and assemblage correction. The temperature effect was constrained using Mg/Ca sea surface temperatures from the foraminifer Globigerina bulloides from core MD96-2080. Phases of depressed productivity coincided with periods of stratification in the same core, indicated by high relative abundances of the coccolithophore Florisphaera profunda and with low relative abundances of the upwelling indicator G. bulloides in the nearby Cape Basin. These observations collectively suggest that productivity was regulated by upwelling throughout this region. We infer that, as in the present, periods of low productivity result from a more northerly position of the westerlies, potentially accompanied by subtropical front displacements and blockage of upwelling promoting easterlies during austral summers. The influence of westerlies in the study area appeared to not have only controlled primary productivity variation in the whole Agulhas Corridor, but also salinity and temperature variability of the whole Agulhas system, including the Indian and Atlantic sectors. Extreme cold events during obliquity minima, indicated by increased ice-rafted detritus (IRD) deposition on the Agulhas Plateau, and potentially driven by a greater summer sea ice extension, may have forced westerlies to lower latitudes, decreasing productivity in the Agulhas Bank slope and salinity in the South Indian subtropical gyre. An extension of the coccolith Sr/Ca productivity record of the Agulhas Bank slope, especially during extreme obliquity scenarios, may provide crucial information on the role of obliquity in modulating the westerly wind belt behavior, glacial CO2 sequestration and glacial terminations. Since future climatic conditions are predicted to resemble those of times previous to the Quaternary, investigating the paleoceanographic history on glacial-interglacial timescales as conducted in chapter I, may be insufficient to understand adaptations of primary producers to more extreme climates. Therefore, it appears necessary to conduct a careful investigation of the response of diatoms and coccolithophores, the two most important groups of primary producers, to typical ambient conditions of the Cenozoic, similar to those predicted for the future. Despite the importance of diatoms in regulating climate and the existence of large opal-containing sediments in key air-ocean exchange areas, most geochemical proxy records are based on carbonates. Among them, B content and B isotopic composition (¿11B) have been widely used to reconstruct pH from foraminifera and coral fossils. In chapter II, we assessed the possibility of a pH/CO2 seawater concentration control on B content in diatom opal to determine whether or not frustule B concentrations could be used as a pH proxy or to clarify their physiological responses to acidifying pH. This calibration study conducted in cultured diatoms of the species Thalassiosira pseudonana and T. weissflogii, shows that frustule B content may serve as an indicator of diatom active bicarbonate uptake, which increases under carbon limitation for photosynthesis at lower seawater CO2 concentrations and higher pH. Frustule B content was measured by both laser-ablation inductively coupled mass spectrometry (LA-ICPMS) and secondary ion mass spectrometry (SIMS/ion probe). For both species, the results obtained by both methods show that frustules grown at higher pH have higher B contents and higher Si requirements per fixed carbon. Though several factors interact to determine the structure of the phytoplankton community, the observed lower diatom Si requirements at lower seawater pH suggest that a future more acidic ocean may produce a decrease of carbon export in Si-rich regions as diatoms would be less silicified and ballasting efficiency would be lower, while shifting the phytoplankton community of Si-depleted regions from small-cells to large diatoms, which would overall increase the biological pump efficiency. Cellular B uptake appears to occur via active borate uptake through non-selective bicarbonate transporters, wherewith borate uptake increases with the higher bicarbonate demand and higher borate/bicarbonate ratios at more basic seawater and lower CO2 concentrations. The mechanism of B transport from the site of uptake to the site of silica deposition remains unknown, but may occur via silicon transport vesicles, in which borate may be imported for B detoxification and/or as part of a pH regulation strategy either through Na-dependent borate/Cl- antiport or borate/H+ antiport. B deposition in the silica matrix may occur via substitution of a borate ion for a negatively charged SiO- formed during silicification. The results from this calibration study show that measurements of B content might reveal the varying importance of active bicarbonate acquisition mechanisms of diatoms in the past in response to carbon limitation. Non-published preliminary results of B content in fossil diatoms suggest strong B contamination of sedimentary opal, as detailed in section general discussion and implications, which could potentially be used advantageously to obtain more precise measurements of frustule ¿11B. Further work is needed to improve the precision of frustule B content measurements, determine the timing of the B contamination in sedimentary frustules, and conduct precise measurements of frustule ¿11B, which may be used to trace seawater ¿11B variability, a parameter that has been shown to be uncertain in the past and that is required to accurately apply foraminifer ¿11B to reconstruct seawater pH. Coccoliths are key contributors to the global carbonate sink. Therefore, changes in their Ca isotopic fractionation could affect seawater ¿44/40Ca, playing an important role in the global Ca cycle with implications for seawater chemistry and climate. Despite this, a quantitative interpretation of coccolith Ca fractionation and a clear understanding of the mechanisms driving it are not yet available. In chapter III, the CaSri-Co model developed in this thesis was used to simulate the new and published ¿44/40Ca, Sr/Ca and carbon isotopic composition (¿13C) results of cultured coccolithophores of the species Emiliania huxleyi, Gephyrocapsa oceanica, and Calcidiscus leptoporus, showing that calcification rate, Ca retention efficiency and the characteristics of the solvation environment are the main drivers of coccolith Ca isotopic fractionation and Sr/Ca variability. Higher calcification rates, higher Ca retention efficiencies and lower solvation-desolvation rates increase coccolith Ca isotopic fractionation and Sr/Ca. While coccolith Ca isotopic fractionation was especially sensitive to changes in the solvation-desolvation rates, the main factor producing coccolith Sr/Ca variations was found to be Ca retention efficiency. The application of the model to combined results of ¿44/40Ca and Sr/Ca of sediment coccoliths of two size fractions (3-5 and 8-10 µm) from Site 925 in the Western Equatorial Atlantic, show that the previously described adaptations of coccolithophores to carbon limitation (i.e. decrease in coccolith thickness and carbon isotopic fractionation) may have concomitantly occurred with an increase in Ca retention efficiency during the CO2 decline period over the last 11 Ma, potentially also in response to carbon limitation. Further studies are necessary to clarify the relationship between seawater CO2 concentrations and Ca retention efficiency in coccolithophores. The comparison between coccolith and foraminifer ¿44/40Ca from the same site evidences the additional coccolith Ca isotopic fractionation variability produced by biomineralization processes that are unique to coccolithophores. Our model shows that changes in solvation-desolvation rates due to seawater structure strength variability, potentially produced by seawater cooling and/or modification of the solvation environment via cellular exudates, may contribute to the observed variability of Ca isotopic fractionation in the sedimentary record. The potential of coccolith Ca isotopic fractionation to produce changes in seawater ¿44/40Ca, assuming a significant contribution to the global fractionation factor, underscores the importance of taking into account coccolith-based studies in the modelling of the past and future Ca cycle. Moreover, since seawater may have been more variable before the time interval here studied (i.e. 11 Ma), an extension of the here presented coccolith ¿44/40Ca and Sr/Ca combined records back in time may provide information to understand how adaptations of coccolithophores¿ biomineralization and photosynthetic machineries to variable climate may have affected their Ca isotopic fractionation and hence, the global Ca cycle.