Abstract
Debris-covered glaciers are found in most of the world’s glacierized areas. The presence of debris on the glacier surface affects the ablation rate of the underlying ice. When the debris cover exceeds a thickness of a few centimetres, thermal insulation becomes dominant, thereby reducing the ablation of the underlying ice so that for a given climate forcing, debris-covered glaciers exhibit a significantly different response than that observed for debris-free glaciers. Understanding the response of debris-covered glaciers to a warming atmosphere is of critical importance for water resource management, assessing glacier-related hazards, and making sea-level predictions.
This thesis focuses on the development and application of numerical models that couple ice flow and debris transport. First, a depth-averaged flowline model is developed to study the effect of debris cover on transient evolution between steady states. Then, the effects of modifying potentially uneven debris distribution are implemented in the model in order to account for either patchy debris or terminal cryokarst features. Next, surface processes are incorporated into a Lagragian flow model so that the effect of their nonlinear feedbacks on glacier evolution may be studied. Finally, a depth-dependent model is used to examine variations in debris input and the resulting effect on the transient glacier behaviour.
For a retreating debris-covered glacier, the length response is shown to be strongly delayed compared to the volume response and that in general volume response times are much longer than for debris-free glaciers. It is found that periods of cold climate have a longer-lasting effect on the transient volume and particularly on the length of debris-covered glaciers than do periods of warm climate. Therefore, such glaciers tend to advance or stagnate in length in a fluctuating climate, and hence glacier length is not representative of climate but rather depends on the history of cold phases.
Patchy debris cover, found near the initial debris emergence, is found to largely negate any effects of enhanced thinning. When cryokarst features, such as ice cliffs and supraglacial lakes, are dynamically coupled to the ice flow in a way that is consistent with observations, it is shown to enhance both terminus thinning and the retreat rate. Incorporating these processes appears to be necessary in order to produce similar mass loss rates to those observed today.
Incorporating feedbacks from surface processes resolved at a smaller (sub-grid) scale is shown to reproduce the observed asymmetry of debris thickness profiles about ice crests and middle moraines. It is found that this asymmetry cannot be replicated without the addition of a melt rate correction due to surface orientation, hence it is important to include this in studies of small scale processes and to account for it when collecting observational data. However, it is shown that these processes do not have a strong effect on glacier evolution on a larger scale, as the complex feedbacks between the processes tend to cancel each other out.
Debris input location is found to make a significant difference to the debris distribution, fractional debris cover, and total debris volume on the surface. As these features are critical for determining a glacier’s transient response to a warming climate, the implication is that steady state glacier extent is a poor indicator of a debris-covered glacier’s transient behaviour. A further result is that for certain glaciers with a temperature-dependent debris source, the ongoing retreat phase associated with the current warming trend may eventually reverse and turn into an advance, on a timescale of several centuries. A simplified version that captures much of the englacial dynamics and behaviour without using a computationally expensive depth-resolving model, is shown to be reasonable effective, especially for gradually varying climates.
The main finding of this thesis is that there exist numerous interactions and feedbacks between the governing physical processes that are not present for debris-free glaciers, resulting in a more complex transient behaviour. On the small scale, nonlinear feedbacks between surface processes lead to spatially asymmetric debris cover that has a small but observable effect on overall glacier dynamics. On the larger scale, debris-covered glaciers have several characteristic transient response timescales, with length response lagging noticeably behind volume response. Glaciers of similar steady state extent may exhibit substantially different responses to climate forcing depending on the extent and distribution of surface debris, as well as the presence of ice cliffs and supraglacial ponds. It is therefore necessary to better understand these processes and more accurately represent them in modelling studies in order to reproduce the recent and future evolution of debris-covered glaciers.
Ferguson, James Christopher