School of Geography Studentship
1st October 1995 - 30th September 1999
Glacier surging is an internally triggered oscillatory flow instability in which abrupt increases in flow velocity are accompanied by a downglacier transport of ice and often a marked frontal advance (Meier and Post, 1969). The exact mechanism(s) of surging and surge trigger(s) are still largely unknown as are the factors controlling surging. This PhD research explores and quantifies the relations between surge-type glaciers and glacial and environmental characteristics by isolating factors that discriminate surge-type glaciers from normal glaciers, hence the controls on surging. These controls are then used to verify proposed surge mechanisms. Further, variations in surge behaviour are used to distinguish between groups of surge-type glaciers and surge behaviour.
Through multivariate logit analysis, a set of glacial, mass balance related, geological, and thermal attributes of a glacier population in Svalbard were tested on the prevalence of surging. Long glaciers with steep slopes overlying fine-grained lithologies younger than Devonian and with orientations in a broad arc from NW to SE are shown to be more likely to exhibit surge behaviour. Further, polythermal regime and elevation span were found to be strongly related to the likelihood of surging. Residual analysis revealed a number of previously unidentified surge-type glaciers, as well as surge-type glaciers with uncommon characteristics. Glaciers in the Yukon Territory, Canada, were also analysed: long glaciers had increased surge probabilities. Fowler�s index, an indication of a linked cavity surge mechanism, was only found to be significantly increasing the likelihood of surging for shorter glaciers. The findings of the logit analysis suggest that in Svalbard, Kamb�s linked-cavity surge theory is not supported, but surges probably take place through a soft bed mechanism, with a possible thermal control. No significant relation between subglacial geological boundaries or mass balance attributes and surging was found. Possible causes of glacier length and substrate controls on surging are subglacial debris composition, longitudinal stress distribution and hydrological instability.
The surge history and behaviour of Sortebræ (68º45' N, 27º05' W), East Greenland was analysed and compared to surges in other regions. Sortebræ�s western flow unit surged between 1933 and 1943 and the glaciers main flow unit surged between 1991 and 1995. Vertical and horizontal movements, as well as surface features of the 1990s surge, were analysed in detail using multi-model photogrammetry. The surge affected an area of 335 km2 and resulted in an ice volume displacement of 18.6±0.4 km3, causing a surplus calving flux of 2.3±0.1 to 5.9±0.4 km3 a-1. In less than 2 years the tidewater front advanced 4-5 km. Sortebræ has a quiescent phase of at least 60 years and a surge phase of 2-4 plus years. The behaviour of this surge and others in East Greenland suggests that the surge mechanism in this region resembles that of Svalbard, rather than other regions. Observations of interacting flow units of Sortebræ suggest that restriction of outflow could be a major control on surging.