Copy the page URI to the clipboard
Barnes, Piers Robert Fitzgerald
(2002).
DOI: https://doi.org/10.21954/ou.ro.0000e7e7
Abstract
The distribution of natural impurities in polar ice is of considerable relevance to processes occurring in ice sheets and to the study of ice cores used to determine past climate. In this thesis a reliable method is developed using the scanning electron microscope equipped with a cold stage to directly locate and identify, with x-ray analysis, impurity in situ, The technique uses sublimation to collect impurities on the ice surface. Specimens of cold polycrystalline snow and ice from cores drilled in both Greenland and Antarctica are examined. The distribution is revealed to be heterogeneous and primarily dependent on the bulk ice impurity content In some samples dust particles are distributed preferentially at grain boundaries and are observed pinning grain boundaries. Soluble impurities are found dissolved in the lattice, at grain boundaries and coating bubbles Significant quantities of sulphuric acid are found in veins at crystal triple junctions only when sufficient acid is present in the ice to also coat the surrounding grain boundaries. Thus a smaller fraction resides in veins than previously thought. Factors such as grain size, fabric orientation, ice temperature and age are also likely to play a significant role in determining impurity distribution. The response of electrical conductivity to chemical impurity in the Dome C (Antarctica) ice core is quantified and is consistent with these observations. It is modelled by the presence of sulphuric acid at grain boundaries and possibly veins. A large proportion of the salt chloride in the core must be incorporated in the lattice to explain the measured conductivity. Limited diffusion of chloride and sulphate ions, but not sodium, is observed in the Dome C core causing broadening of peaks. Models reliant on normal grain growth to stimulate solute movement can explain this. This work reconciles differences between previous location studies, presents a consistent but not well-quantified description of impurity distribution and addresses possibilities of post-depositional solute movement.