Supra-glacial Lakes on the Greenland Ice Sheet

Image by Greenpeace's Nick Cobbing

Background - Climate Change & the Greenland Ice Sheet

Global mean surface temperatures have increased by 0.6 ± 0.2°C over the 20th century. During this time period tide-gauge measurements show that there has been a 15 ± 5 cm rise in sea level, 7.5 ± 7 cm of which is likely to be as a result of  the Greenland Ice Sheet (GIS) and the Antarctic Ice Sheet (AIS) losing mass. Climate models predict a further warming of between 1°C and 3.5°C during the next century with the possibility of increased warming thereafter if greenhouse gas emissions are not significantly reduced. The sea level rise resulting from further ice sheet mass loss induced by climate warming is likely to have a significant impact upon numerous communities that currently inhabit low-lying areas. The increased fresh water input into the Atlantic Ocean resulting from GIS mass loss may also affect thermohaline circulation (large-scale ocean currents) which in turn could provoke further changes in climate. In order to make informed policy decisions regarding future cuts in greenhouse gas emissions it is important that we can predict the consequences of future climate warming. This in turn requires that we have an accurate understanding of how the GIS and the AIS will respond to climate change.

Changes in the mass of the Greenland Ice Sheet result from an imbalance between mass gain (predominantly due to accumulation from precipitation) and mass loss (mainly due to melting and calving). Mass is predominantly lost over the summer when warmer temperatures cause surface melting at lower altitudes on the GIS. The GIS melt season spans, on average, June through to September with peak melting usually occurring late July / beginning of August. The amount of melt is higher than found on the AIS with melting occurring, on average, over 50% of the GIS surface. Calving occurs where an outlet glacier meets the sea and the flowing nature of the glacier causes pieces of ice to break off, thus producing icebergs. The rate at which a glacier calves is dependent upon the velocity of the glacier, with faster moving glaciers losing more mass. Thus, when considering the effect of climate change upon the GIS, we must not only consider the mass loss directly due to increased ice melt, but also any possible impact this increased melting may have upon ice sheet velocity. This is referred to as the dynamic response of the ice sheet to climate change.

Calving Glacier
(Image from National Snow and Ice Data Centre)

Until recently it was believed that the dynamic response of the GIS to a warming climate would be slow (in the region of hundreds of thousands of years). Since heat only diffuses very slowly to the lower layers of the ice sheet, any increases in surface temperature could only slowly lead to enhanced rates of basal ice deformation or sliding with associated increases in ice flow and calving.

Recent measurements have shown that higher elevations of the GIS are roughly in mass balance. However, at lower areas at the ice sheet margin there has been a significant increase in mass loss, with the ice sheet losing ice at a rate of approximately 50km³ per year (corresponding to a thinning of approximately 5cm per year and a sea level rise of 0.13mm per year). Parts of the southern ice sheet are thinning at a rate of over 1m per year. This rapid increase in mass loss cannot be accounted for purely by increased surface melt and suggests that rising temperatures are also affecting ice velocity.

Recent findings from close to the equilibrium line of western Greenland indicate that it is indeed a melt-induced increase in ice sheet velocity, and thus calving, that is responsible for the additional mass loss. It has been suggested that this acceleration may be due to enhanced basal sliding caused by the lubricating effect of surface melt water penetrating to the glacier bed. This coupling between surface melt and ice dynamics could provide a mechanism for the rapid dynamic response of ice sheets to climate warming and may be the cause of the significant thinning observed at parts of the southern GIS. In order to accurately predict the effect of future climate change upon the GIS it is important that models take into account this dynamic response. This requires an understanding of the link between surface melt and ice dynamics.

Project Description

This project investigates the seasonal evolution of supra-glacial lakes at the margins of the Greenland Ice Sheet. These lakes form on the surface of the ice sheet, near the ice sheet margins, as a result of summer melt. They typically form below 1500 meters altitude and can grow to several square kilometres in size. During the melt season these lakes have been observed to drain through features, such as crevasses, on the lake bed. These drainage events have the potential to rapidly supply water to the ice sheet bed and thus may play a critical role in linking surface melt to ice dynamics. The role of supra-glacial lakes in the hydrological cycle is therefore of considerable interest. In this project we combine satellite remote sensing and in-situ meteorology to investigate seasonal changes in the volumes of water stored and released by supra-glacial lakes on the western GIS.

               
Supra-glacial lakes formed over crevasse fields
(images by me, http://www.whoi.edu/oceanus/index.do & Greenpeace's Nick Cobbing).     

Link to video of supra-glacial lakes (takes a while to load but it's worth the wait!)

Back to top

Study Area & Methods

We studied two sites at the western margins of the Greenland Ice Sheet - around 'Swiss Camp' (69^o57'N, 49^o31'W) and at Russell Glacier (~67N, ~48W). An intial survey of 10 coarse (250 meter) resolution images taken throughout 2001 showed a clear seasonal variation in the number of Supra-glacial lakes present in the vicinity of Swiss Camp. We then selected 2 high (30 meter) resolution images, seperated by a 25 day interval, showing the Swiss Camp site during the height of summer melt. We also surveyed 2 high (15 meter) resolution scenes of the Russell Glacier site taken on similar dates, thus allowing us to conduct a latitudinal, altitudinal and temporal comparison of lake distribution. We used meteorological data to model the volume of melt water produced in the lakes' catchment and compared this to lake area changes we observed. This allowed us to estimate the average lake depth at each date, the volume of water draining from the surface, the resulting sub-glacial water depth (if the water were to drain to the ice-sheet base) and the average rate of drainage from each lake.

Overview of study areas and image coverage.

3d visualisation of study areas and image coverage.
Red dots indicate meteorological stations.

Back to top