Ocean circulation and melting beneath the ice shelves of the south-eastern Amundsen Sea

Sea levels around the world are currently rising, threatening coastal populations with flooding and increased erosion, and evaluating the future threat requires an ability to forecast changes in sea level. To do this we must understand what is happening to the Earth's great reservoirs of freshwater, and whether or not they are slowly draining into the ocean. The largest of these reservoirs by far is the Antarctic Ice Sheet, which contains 70% of all the freshwater on the planet, and we know that parts of the ice sheet are thinning. The fastest changes are happening near the edge of the ice sheet, where it flows into the sea in a place called Pine Island Bay, and the speed of the changes has taken scientists by surprise.

Pine Island Bay is geographically the far south of the Pacific Ocean, and the image of warmth that this conjures up is not entirely misplaced. The air temperatures never rise above freezing, but beneath the cold surface of the sea, water temperatures rise as high as 1 degree Celcius, well above the freezing point. Pine Island Glacier is a vast river of ice that flows out into Pine Island Bay, carrying as much water as the River Rhine in frozen form. The last 60 km of the Glacier floats on the waters of Pine Island Bay, and the bottom melts so intensely that half of the ice carried in the glacier is lost within the space of 30 years. It is not hard to understand that warm water causes rapid melting, but what do "warm" and "rapid" really mean? If we change the water temperature by a small amount, by how much will the melt rate change? And critically, what might cause the ocean temperature to change?

To find the answers to those questions we must make measurements of the water temperature beneath the glacier, and simultaneous measurements of the rate at which the base of the glacier is melting into the ocean, but to do so is enormously challenging. The glacier is between 300 m and 1 km thick, so it is difficult to access its base. The key is cutting-edge technology, in the form a robotic submarine capable of diving beneath the ice, making measurements along a pre-defined track, then returning to the surface with the data, and a set of rugged, autonomous radar systems that can left on the glacier's surface throughout the Antarctic winter precisely measuring the rate at which the thickness of the ice changes.

The robot submarine has been designed and built by NERC engineers and has already proved itself on preliminary missions beneath Pine Island Glacier in 2009. The radar systems will be developed as part of this project. They will combine a well-known radar technique, FMCW radar, with careful measurement of the phase of the return echoes to establish the position of unique features in the image, such as the bottom of the glacier, with very high precision of the order of 1 mm over a 1 km range. Four of these radar instruments will be left on the surface of Pine Island Glacier, engineered to allow year-round autonomous operation and monitoring of the gradual change of ice thickness with time.

Armed with the data from these new instruments we will use a computer model that describes the flow of water within the remote cavern beneath the glacier and in the sea to the north of it. Using this model we will determine how heat that is transported into the cavern by ocean currents is used to melt the ice shelf and what impact changes in the climate of this part of Antarctic will have on the ocean currents and resulting melt rates. This information will allow others to assess with greater certainty how future climate change will impact the glaciers of Pine Island Bay and hence how this remote part of the world will influence the future coastlines of places such as Holland and East Anglia.

The results of this research, and those of others who make use of the datasets that will be generated, will contribute towards our understanding of the drivers of ice sheet and climate variability that will inform environmental policy-makers through the IPCC report series.

A dissemination article will be prepared for publication in the "Projects" journal, distributed to stakeholders throughout Europe.

Results will also be disseminated to post-graduate students and early career scientists through teaching.

Summaries of important and newsworthy outcomes will be distributed to the national and international press via the BAS Press Office.

Beneficiaries of the technical developments of the study, from outside the academic community, will include those in the radar community, both industrial and defence related. Specifically, defence related researchers in the MoD, DSTL and Qinetiq, and major industry players such as Thales, and radar SMEs such as Guidance Microwave Ltd. The UCL Sensors, Systems and Circuits Group, led by the UCL investigator, has extensive contacts within these organisations. In addition, the results of the study will be of interest to the UCL Environment Institute and the (UCL) Jill Dando Institute of Security and Crime Science.

Although the response of ice shelves to ocean forcing are fundamental to future changes in the ocean's climate, the direct impact of the research will be felt more quickly in the spinoffs that will result from the development of the phase-sensitive FMCW radar technique deployed in this work:

1. The radar instrument will be designed in a modular fashion to allow the system to be enhanced so that glaciologists will be able to observe the motion of deep internal layers over grounded ice. The present system has been used successfully to observe motion of shallower layers; the new system will raise the prospect of a major improvement in performance.

2. Imaging of oil deposits and other reservoirs of commercial interest.

3. Commercialisation of the radar system, through SMEs (eg Guidance Microwave Ltd.) We anticipate a demand for the basic phase-sensitive FMCW radar system within the glaciological community of around 100 units. High power variants will increase that number. The number of systems likely to be required for use in other geophysical applications is largely unknown.