Insights from observations and modelling into the evolution of superglacial lakes on the Greenland ice sheet

• Main Author: Leeson, Amber Alexandra
• Format: Thesis
• Language: English
• Published: University of Leeds 2013
• Subjects: Greenland; Ice Sheet
• Online Access: https://etheses.whiterose.ac.uk/5060/; https://etheses.whiterose.ac.uk/5060/1/amber_leeson_thesis.pdf

Supraglacial lakes (SGLs) form when runoff (meltwater + rain) pools in depressions on the Greenland ice sheet (GrIS). SGLs can collectively affect seasonal ice sheet flow rates when they drain episodically; although the net impact on flow speed is uncertain. In this thesis: 1) a new model of SGL Initiation and Growth (the SLInG model) is presented, 2) existing SGL observations are evaluated and combined to form a single optimised dataset, 3) these data are used to evaluate the model and 4) this model is used to investigate past trends in SGL evolution in south west Greenland. SLInG is a 2-dimensional transient hydrology model which routes runoff, which has been simulated using a regional climate model, over a digital elevation model (DEM) of the ice sheet surface. Water is routed using Darcy’s law for flow through a porous medium and Manning’s equation for open channel flow, and is allowed to collect in depressions in the DEM, thus forming SGLs. Observations of SGLs can be temporally sparse and variation in reported lake frequency can be significant between datasets. Three observational datasets of SGLs, automatically derived from satellite data, were found to omit a sizeable (29 to 48%) fraction of lakes identified manually. These datasets were combined using a hierarchical scheme, leading to a 67% increase in the number of lakes reported. By comparison with satellite observations, SLInG is found to be 19 times more likely to correctly predict the location, or absence, of a lake, than not. In addition, simulated and observed lake onset dates are highly correlated (r= ~0.8) and model estimates of the rate of growth of lake covered area are, on average, just 14% greater than observed values. SLInG was forced with 40 years of reanalysis data in order to investigate historical variation in the temporal evolution of SGLs. SLInG shows that SGLs have responded to recent dramatic changes in local climate by migrating inland by 150 m a.s.l. (3.75 m a.s.l. per year) during 1971-2010. This modelled trend is in good ...