Graduation Year


Document Type




Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department


Major Professor

Jason Gulley, Ph.D.

Committee Member

Matthew Covington, Ph.D.

Committee Member

Timothy Dixon, Ph.D.

Committee Member

Glenn Thompson, Ph.D.


Glaciology, Moulins, Subglacial, Lake drainage


Since the 1990's the Greenland Ice Sheet (GrIS) has been losing mass at an accelerating rate in response to climatic warming and is currently the largest terrestrial contributor to sea-level rise. While ice sheet models agree the GrIS will continue losing mass throughout the century, there are significant uncertainties associated with future sea-level rise contributions. Predicting the GrIS's response to future climate warming scenarios is limited by gaps in our understanding of the links between ice sheet hydrology and dynamics. Meltwater produced on the ice surface flows within supraglacial streams that deliver it to crevasses or moulins—vertical conduits extending from the ice surface to the ice sheet's bed. When the rate of meltwater delivery to moulins exceeds the hydraulic capacity of the moulin-connected subglacial drainage system, meltwater will temporarily backup within moulin shafts, increasing water pressures at the bed which can increase sliding speeds. Despite the central role of moulins in connecting supraglacial and subglacial hydraulic systems, little is known about their role in coupling hydrology and sliding on the GrIS.

This dissertation uses several new data sets acquired within the ablation area of Sermeq Avannarleq in the western GrIS to further our understanding of the hydraulic systems that influence sliding within the GrIS ablation area. First, I investigate whether delays in the timing of meltwater delivery for more extensive, higher-elevation catchments could explain the previously progressively later timing of peak daily ice velocity observed with increased elevation and distance from the ice sheet's margin. We measured meltwater delivery to moulins, moulin water level, and the ice velocity response for two moulins at different elevations. Our results show that differences in the timing of meltwater delivery caused peak moulin water level to consistently occur later in the day at our higher-elevation moulin, lagging behind peak pressure at the lower elevation site by 1–3.25 hours. However, delays in the timing of meltwater delivery to moulins and the timing of peak moulin water level did not entirely account for the delays in ice velocity at our higher-elevation site. These observations indicate that there are non-local controls on sliding at higher elevations.

Next, I reassess whether the size and location of internally drained catchments (IDCs)—the area on the ice sheet's surface that drains via supraglacial stream networks into a terminal moulin—are static features. We document significant interannual variability in the flow paths of the highest-order streams within two mid-elevation catchments. Snow-infill of the previous year's incised streams over the winter created snow plugs that diverted flow away from the catchment's terminal moulin during the subsequent melt season. Instead, the supraglacial streams incised through drainage divides, altering the size, shape, and number of IDCs in this part of the ablation area.

Finally, I investigate the importance of channelization relative to the connection of isolated cavities in controlling seasonal ice deceleration within the GrIS ablation area. I use observations of the ice-dynamic response to a passing subglacial floodwave following the rapid draining of several supraglacial lakes. We argue that the reduction of daily minimum sliding speeds can be explained by the transient connection of isolated cavities rather than the growth of subglacial channels. These observations build upon previous work to show that changes within poorly connected parts of the subglacial drainage system can explain slowdowns previously attributed to increased channelization.