It is now an established hypothesisthat subglacial lakes hold uniquemicrobial life and contain important

records of past climate change in the sediments on their floors.Consequently, there are importantscientific reasons underpinning plansfor their exploration. The issue facedtoday concerns which lake will be the first to be explored. The largestsubglacial lake, Lake Vostok, hasreceived widespread media andscientific interest as a likely candidate.

However, more recently it has becomeapparent that a smaller lake might beexplored before Lake Vostok.

The choice for exploration of the UK-led consortium is called LakeEllsworth: a 10 km-long subglacial lakein West Antarctica. Lake Ellsworth wasdiscovered by a technique known asradio-echo sounding, which works in a manner similar to seismic sounding,only using radio waves instead ofsound waves. Subglacial lakes areeasy to identify using this technique,

as the style of reflection from a lake’ssurface, being flat, bright and smooth,is distinct from an ice-rock contact,being undulating, faint and rough (see right). Lake Ellsworth is locatedbeneath 3.5 km of ice in WestAntarctica, and is one of only fourlakes outside East Antarctica, wheremore than 100 lakes have been found(including Lake Vostok).

Exploration of subglacial lakes involvesthe major technical challenge ofsomehow getting through the iceabove and then into the water. Thereare two methods available. One is todrill through the ice, the other is tomelt the ice. Drilling (or coring) the iceis unlikely to be appropriate for lakeaccess because, to ensure that thecore does not freeze against the coldice, antifreeze must be pored into thecore. Such antifreeze (usually akerosene-based liquid), if it ever getsinto the lake, would be a majorcontaminant, and would underminethe scientific credibility of theprogramme: who would trust theresults of microbial detection if theequipment used had not passedcontamination controls? This leavesmelting the ice as the only feasiblemethod. There are two ways of meltingdown into a subglacial lake. One is touse a ‘thermo probe’, which melts ➜

2 RESEARCH REVIEW • DECEMBER 2004GEOGRAPHICAL SCIENCES • THE EXPLORATION OF LAKE ELLSWORTH: AN ANTARCTIC SUBGLACIAL LAKE

In 1994 scientists met at the University of Cambridge todiscuss the exploration of lakes beneath the ice sheets onAntarctica. Ten years later a consortium of UK scientistsmet at Bristol University to plan the very first expedition.Professor Martin Siegert of the Bristol Glaciology Centre,School of Geographical Sciences, was there.

The exploration of Lake Ellsworth:an Antarctic subglacial lake

Lake Ellsworth is located beneath3.5 km of ice in West Antarctica

➜ the ice as it travels down the icecolumn, which then freezes above it.Energy could be supplied to the probevia a tether which unwinds from theprobe. The disadvantages, that 3.5 kmof wire has to be stored in the probeand that the probe cannot return to theice surface – thus disallowing samplereturn – mean that this technique isnot well suited to subglacial lakeresearch. This leaves hot-water drilling;a technique similar to ice coring, butusing hot water to melt a hole withinthe ice. The problem with hot-waterdrilling is that no-one has ever drilledthrough more than about 2 km of ice.Thus, its proposition requires thelargest hot-water drill to be designed.The major advantage that lakeEllsworth has over other lakes, from ahot-water drilling perspective, is that

the ice in West Antarctica is likely to bemuch warmer than in East Antarctica,thus making hot-water drilling mucheasier. Moreover, the major advantageof hot-water drilling over other accesstechniques is that it is the only onethat allows samples to be returnedfrom the lake to the ice surface.

The UK plan for exploration of LakeEllsworth is in three stages. The first isto undertake the detailed geophysicalreconnaissance of the lake. This willinvolve radio-echo sounding to mapthe lake surface, and seismic soundingto measure the water depth. Aproposal, led by the University ofBristol, has been submitted to theNatural Environment ResearchCouncil. The plan is to undertake thesurvey in the Antarctic summer of2006-07. Stage two is to develop the

technology required for lake access,measurement and sampling. Stagethree is to undertake the fieldwork toexplore the lake (see above). Theselast two stages are the focus of theBristol-led plan for subglacial lakesresearch.

The exploration plan involves droppinga probe, tethered to the surface,through a borehole excavated by hot-water drilling, to the lake. The probewill contain a series of instruments,including sophisticated equipment todetect microbial life. The probe will belowered through the water column,measuring and sampling the waterbody as it does so, to the lake floor. At this point it will obtain a sample ofthe lake-floor sediment and return to the ice surface. Samples will thenbe scrutinised in a variety of UKlaboratories.

The Bristol-led plan for subglaciallakes research is the first of its kind. If successful, it will pave the way forother similar environments to beexplored, both in Antarctica and on icy planets and moons elsewhere in

the solar system. Consequently, thedetection of life in Lake Ellsworth maybe seen in future as a step to thediscovery of extraterrestrial life. ■

www.ggy.bristol.ac.uk/ellsworth

3RESEARCH REVIEW • DECEMBER 2004 GEOGRAPHICAL SCIENCES • THE EXPLORATION OF LAKE ELLSWORTH: AN ANTARCTIC SUBGLACIAL LAKE

The Bristol-led plan for subglaciallakes research is the first of its kind

The exploration of Lake Ellsworth, involving hot-water drilling through the iceand the deployment of a ‘hydrobot’ probe into the lake, which will measureand sample water and lake-floor sediment

The detection of Lake Ellsworth using radio-echo sounding