Understanding marine ice stream retreat using numerical modelling and geophysical data

Currently there is major uncertainty surrounding the contributions of the Greenland and Antarctic ice sheets to future sea level rise. The majority of the ice discharged by these ice sheets into the surrounding oceans is by drainage of fast flowing 'marine ice streams' which have their bases below present sea level and which flow at speeds of a few hundred metres to several kilometres per year. Recent satellite observations show that some of these marine ice streams are experiencing rapid change in the form of thinning and retreat of their margins. These ice streams are a major focus for current glaciological research, but up until now this research has mostly concentrated on either theoretical studies of ice stream retreat and/or observations of modern ice streams. Such studies typically only provide a snapshot of an ice stream's life-cycle, limited to the last few decades. Current attempts to determine what controls the retreat of these ice streams remain inconclusive. This problem is compounded by the fact that existing large-scale ice sheet models, used for future sea level predictions in the 'Intergovernmental Panel on Climate Change' 2007 report, are not able to incorporate such rapid changes in ice streams. Understanding the controls of ice stream retreat and predicting their future behaviour will be crucial, if the scientific community intends to quantify future sea level rise with any degree of certainty. This study aims to determine the controlling factors on marine ice stream retreat by combining numerical modelling with marine-geophysical observations of ice stream retreat. We will use observations from three well documented palaeo-ice streams in Antarctica (i.e. ancient ice streams recorded by geological and geophysical evidence) that underwent contrasting styles and rates of retreat following the last glacial maximum and ranging from rapid through episodic to slow. The numerical model that we will use is specifically developed to allow us to simulate ice stream retreat. The marine geophysical and geological observations will provide the physical setting for the numerical model and, crucially, the retreat history of the different ice streams for comparison to the model simulations. To our knowledge, a robust comparison and validation of a numerical marine ice stream model using such data has never before been attempted. The inter-disciplinary approach taken in this project provides the urgently needed link between observational records of ice stream retreat and numerical modelling. It will advance our understanding of the factors that control such retreat and also provide crucial insights for developing ice sheet models, and thus contribute to improving the ability of the scientific community to predict future sea level rise.