UBC Theses and Dissertations
Isotopic identification of subglacial processes Maxwell, Michael
A comprehensive stable isotope study of basal ice and debris layers in two Yukon surging glaciers suggests an isotopically variable basal freezing cycle. Trapridge and Backe Glaciers, St. Elias Range, Yukon, Canada, have parallel basal debris layers that extend for hundreds of metres along marginal ice faces and in meltwater tunnels. The isotopic compositions of samples at 1-5 cm intervals from vertical ice cores in basal ice exposures, have the following characteristics: 1. The δO¹⁸ and δD values of ice immediately above a debris layer are higher than the values immediately below a debris layer (up to 3‰). 2. The δO¹⁸ and δD values vary between debris layers. The most common trend is an increase in δO¹⁸ and δD with decreasing height above the bed. 3. Clear ice layers, which are frequently immediately above debris layers, have higher δO¹⁸ and δD values than surrounding ice. 4. Across core lengths of 0.5-2.2 m, there is no significant overall isotopic shift from top to bottom of a core. Along a 100 m longitudinal tunnel, there is no significant isotopic change in the average of samplings at the entrance, middle and exit of the tunnel. Trapridge and Backe Glaciers are subpolar surging glaciers that surge on a cycle of 40-50 years. They are approximately 5 km long and 1 km wide with a lower ablation zone that is frozen to the bed and an upper accumulation zone that has basal ice at the pressure melting temperature. The basal ice is comprised of parallel debris layers ranging from less than 1 cm to 40 cm in thickness, and ice layers from less than 1 cm to 70 cm in thickness. The debris sequences are 5-25 m thick, accounting for up to one half of the glacier depths in the ablating ice zones. Microparticle and cation concentrations are much higher in basal ice than in bulk (precipitated) ice. The isotopic, chemical, microparticle, and layer characteristics of the debris sequences do not suggest a shearing mechanism of incorporation. I propose accretion by a basal freezing process which is isotopically variable. Incorporation of a debris layer is associated with a change in composition of the water supply or a change in the freezing mechanism, thereby causing an isotopic shift. The degree of fractionation during basal accretion will be affected by freezing rates and mixing in the water reservoir. An annual isotopic composition cycle in basal water caused by mixing with surface water, or an annual freezing rate cycle caused by variable basal water supply and heat flow, can produce the isotopic variations between debris layers. In addition, ice segregation processes such as primary and secondary heaving may affect the fractionation process, particularly during ice segregation from a sediment substrate. Heavy isotope relationships between δD and δO¹⁸ were used to examine the basal freezing fractionation mechanism. The δD and δO¹⁸ relationships had straight line approximations with slopes 7.53 ±0.39 and 6.75 ±0.28 for Trapridge Glacier_bulk and basal ice respectively, and 7.31 ±0.13 and 6.27 ±0.17 for Backe Glacier bulk and basal ice. A majority of samples (all but 5 of 200 samples) plotted to the right of the meteoric line of slope 8.04 ±0.34 that was determined in this study. The bulk ice has been isotopically alterred during snow evolution processes such as percolation exchange. Backe and Trapridge Glaciers consist of significant proportions of superimposed ice, suggesting that the bulk ice slope represents isotopic shifts in snow with a range of initial isotopic compositions, rather than isotopic evolution along the bulk slope line. The isotopic composition of the basal water supply during basal freezing determines the significance of the basal ice slope. Spatial or time variable basal isotopic compositions imply that the basal ice slope represents samples that have been shifted from a range of original compositions also, rather than a time-evolving fractionation from an average basal isotopic reservoir. The hypothesis of a basal accretion process with dependence upon mixing and variable freezing rates discourages the use of the heavy isotope slope as an indicator of a single basal fractionation process (i.e. constant fractionation from an isotopically constant reservoir). The heavy isotope relationship may still be useful to identify bulk and basal ice in a glacier where differences in slope have been characterized.
Item Citations and Data