Uncovering the Potential of Utilising Terrestrial Biogenic Markers in Ice Cores as Proxies to Past Environmental Conditions

Abstract

Ice cores provide an un-matched high-resolution archive into paleoclimate reconstruction that other environmental archives (e.g. coral, sediment cores, and tree rings) do not. Limited research has focussed on terrestrial organic matter within ice cores. However, given the known relationships between terrestrial organic matter and the environment, there is a potential that investigating these compounds within ice cores could uncover a wealth of information. Secondary organic aerosols (SOAs) and their biogenic volatile organic compound (BVOC) precursors are intrinsically linked to atmospheric chemistry, climate control, and land use. These relationships have the potential to inform our knowledge on the reconstruction of past biospheres. Ten target compounds were established for this study: SOA-markers of isoprene; 2 methylerythritol and 2-methylglyceric acid, SOA-markers of α-pinene and/or β-pinene; cis pinonic acid, 3-methyl-1,2,3-butanetricarboxylic acid (3-MBTCA), pinolic acid, cis-norpinonic acid, nopinone, keto-pinic acid, and pinic acid, and a biomass burning marker; levoglucosan. This thesis details the development and optimisation of a highly sensitive, high-performance liquid chromatographic mass spectrometry (HPLC-MS) method for the quantification of the ten target compounds that has achieved limit of detections (LODs) at sub-ppt levels. This method was successfully tested on a subsection of an Antarctic ice core, detecting and quantifying seven SOA markers that have never before been found in Antarctic ice cores. The highly sensitive method was then implemented on a full, Mount Elbrus (Russian Caucasus Mountain) ice core, producing a complete 1300-year timeseries of all ten target SOA-markers. Through comparison with past land use data, reconstructed temperature records, and known sociological events, it was uncovered that this data can indicate and confirm several characteristics of past environments. In this thesis, past forest management will be discussed as a direct mechanism through which humans have unintentionally altered the aerosol composition of the atmosphere with a focus on the difference between pre- and post-industrial periods. Past events can be reconstructed through this data, such as; the medieval warm climate phenomenon, mass population increase in the Ottoman Empire, and the industrial revolution. This is the first time that a continuous 1300-year-old timeseries of SOA-markers in an ice core has been presented, and the valuable nature of these results promote the analysis of SOA-markers to become ‘the norm’ during future ice core analysis campaigns. Detection and quantification of the target SOA-markers in this investigation lays the foundation for understanding the information that can be gathered from past SOA records. Exploration of the suitability of these SOA-markers for reconstruction of past oxidant concentrations is one way in which these archives are utilised in this thesis. Understanding the oxidative capacity of past atmospheres, specifically how it altered with changing climate, could be vital for past biosphere reconstruction. Several methods have attempted to evaluate past concentrations of atmospheric oxidants (hydroxyl radical, nitrate radicals, and ozone) but with limited success. This thesis presents the development of a novel combined experimental and modelling method to predict past oxidant concentrations. Atmospheric box modelling is implemented to combine with physical ice core results in utilising SOA-markers to predict the ozone concentrations of past atmospheres. Through box modelling, changes in the concentration ratio between two monoterpene-derived SOA-markers (pinic acid and pinonic acid) were observed and quantified with differing steady state ozone concentrations. The physical ice core ratio between these two acids was quantified and corelated to the modelling results to predict ozone concentrations of past atmospheres. When compared to recent measured ozone data, the results are encouraging with trends being replicated and similar quantities achieved.