Predicting subglacial lake locations and meltwater drainage pathways in Antarctica, Greenland and North America

Add to your list(s) Download to your calendar using vCal

• Stephen J. Livingstone (University of Sheffield)
• Wednesday 16 October 2013, 16:30-17:30
• Scott Polar Research Institute, main lecture theatre.

If you have a question about this talk, please contact Poul Christoffersen.

There is increasing recognition that subglacial lakes act as key components within the ice sheet system, capable of influencing ice-sheet topography, ice volume and ice flow. At present, much glaciological research is concerned with the role of modern subglacial lake systems in Antarctica. Another approach to the exploration of subglacial lakes involves identification of the geological record of subglacial lakes that once existed beneath ice sheets of the last glaciation. Investigation of such palaeo-subglacial lakes offers significant advantages because we have comprehensive information about the bed properties, they are much more accessible and we can examine and sample the sediments with ease. However, their identification in the geological record remains controversial.

We therefore present a simple diagnostic approach based on the Shreve equation, for predicting and investigating likely (palaeo-)subglacial lake locations. Data on the current topography and seafloor bathymetry, and elevation models of the ice and ground surface topography from data-calibrated glaciological modelling are used to calculate the hydraulic potential surface at the ice-sheet bed. Meltwater routing algorithms and the flooding of local hydraulic minima allow us to predict subglacial routeways and lakes respectively.

Discovered subglacial lakes beneath the Antarctic Ice Sheet present an opportunity to verify the model using the BEDMAP2 dataset. Using a lake threshold of 5 km2 we identify 12,767 subglacial lakes occurring over 4% of the grounded bed and are able to recover >60% of the discovered subglacial lakes. Applying the same approach to the Greenland Ice Sheet produces 1,607 potential subglacial lakes, covering 1.3% of the bed. These lake localities will make suitable targets for radar surveys attempting to find subglacial lakes.

Finally, we apply the Shreve equation to the North American Ice Sheet to try and predict likely palaeo-subglacial lake locations. Given that specific ice surface elevations of the former North American Ice Sheet are only inferred from modelling, and thus contain significant uncertainty, we utilise results from an ensemble of models to examine where on the bed subglacial lakes are likely to have occurred. Predictions are calculated at discrete time-slices through deglaciation to assess the temporal variability and persistence of subglacial lakes and drainage networks. These lake likelihood predictions could usefully form targets for detailed field investigations.

This talk is part of the Scott Polar Research Institute - Polar Physical Sciences Seminar series.

This talk is included in these lists:

• AUB_Cambridge Seminars
• Climate Science Seminars within Cambridge
• Department of Geography
• Scott Polar Research Institute - Polar Physical Sciences Seminar
• Scott Polar Research Institute, main lecture theatre
• seminars

Note that ex-directory lists are not shown.