The incidence of glacier surging in Svalbard: evidence from multivariate statistics

H Jiskoot, Paul Joseph Boyle, T Murray

    Research output: Contribution to journalArticlepeer-review

    Abstract

    Surge-type glaciers switch between long periods of slow flow and short periods of fast flow. The spatial distribution of surge-type glaciers is markedly non-random. The clustering of surge-type glaciers in certain glaciated regions and their complete absence from other regions, suggests that environmental factors control surging. To identify factors that control surging, the relationship between surging, glacial variables and geological environment was tested using logit regression models. In the analysis we used 132 surge-type and 372 normal glaciers identified from the literature and aerial photographs in the High Arctic archipelago of Svalbard. We calculated that approximately 13% of the glaciers of Svalbard are of surge-type. In the analysis we used 13 variables collected from glacier inventories and geology maps. Variables that are significantly associated with surging in Svalbard are glacier length, surface slope and the lithology of the underlying bedrock. Multivariate analyses show that long glaciers overlying shale or mudstone with steep surface slopes have the highest probabilities of surging. Surge-type glaciers are significantly less likely to overlie rocks older than Devonian. These findings do not support Kamb's linked cavity theory of surging but lend some support to soft-bed surge theories. (C) 1998 Elsevier Science Ltd. All rights reserved.

    Original languageEnglish
    Pages (from-to)387-99
    Number of pages13
    JournalComputers and Geosciences
    Volume24
    Publication statusPublished - May 1998

    Keywords

    • glacier surging
    • logit regression model
    • multivariate statistics
    • Svalbard
    • GLIM
    • FLOW
    • MECHANISM

    Fingerprint

    Dive into the research topics of 'The incidence of glacier surging in Svalbard: evidence from multivariate statistics'. Together they form a unique fingerprint.

    Cite this