Antarctic Subglacial Lakes

Subglacial lakes are bodies of liquid water that lie beneath the Antarctic ice sheet, in the interface between the ice and the bed material. With ice acting as an insulator, typical levels of geothermal heat flux are sufficient to warm the ice base to the pressure melting point, despite surface temperatures tens of centigrade below freezing. Subglacial water flows under the combined forces of gravity and the pressure of the ice above and can pond in subglacial hollows and troughs, forming lakes¹.

Like any other continent, Antarctica’s bed morphology involves a complex system of mountains, valleys and lowlands. Under certain circumstances, water can fill entire subglacial troughs leading to giant lakes, such as the 280 km long Lake Vostok – one of the world’s top ten largest freshwater lakes in terms of depth, surface area and volume². Such lakes are located near ice divides and may contain undisturbed water, and sediment dating back millions of years. Towards the ice sheet margins, subglacial water gathers into networks of distributed and channelized systems³, lubricating the beds of the fast-flowing ice streams⁴. Here, water can collect in lakes, filling a bed depression to the level at which it discharges downstream and subsequently refills^{5, 6}.

There are a wide variety of subglacial lake systems beneath the Antarctic ice sheet, ranging from large stable lakes at the ice sheet centre (e.g. Lake Vostok), to smaller stable lakes scattered across the ice bed (e.g. Lake CECS⁷ and Lake Ellsworth), and the small hydrologically “active” lakes predominantly located close to the ice-sheet margin (e.g. Lake Whillans⁵) (Figure 1).

The first subglacial lake observation was made in 1969 by airborne radio-echo sounding (RES) beneath Sovetskya Station in central East Antarctica – the flat and continuously bright radio-wave reflections unmistakably as a consequence of deep (>10 m) freshwater beneath the ice^{8, 9}. Since then, RES has been used to detect >200 lakes across the Antarctic continent¹⁰. While RES is a good determinant of pooled basal water, it cannot be used to measure lake depths since radio-waves are absorbed in water for all but the shallowest and purest water bodies¹¹. Instead, seismic sounding is appropriate to make direct measurements of water depth. To date, the bases of only four lakes have been measured successfully by seismics; Lake Vostok (~1000 m deep)¹², Lake Ellsworth (~160 m)¹³, Lake Mercer (~15 m), and a lake at South Pole (~30 m)¹⁴. For the larger lakes, such as Lake Vostok, where the cavity of water is substantial, gravity measurements can be used to infer the shape of the lake bed, which when used in conjunction with seismic data can reveal the bathymetry at a macro-level¹⁵. In the future electromagnetic geophysical techniques offer the potential to create multi-dimensional images of subglacial lakes and provide insight on their salinities^{9, 16}, as trialled in the austral summer of 2018/19 by the US Antarctic program on Lake Mercer.

Active subglacial lakes can be resolved using satellite-altimeter-derived measurements of ice-surface elevation change; during filling the ice surface rises by several meters and, similarly, during discharge it lowers^{5, 17}. Satellite altimetry has been used to delineate over 120 so-called hydrologically ‘active’ lakes, including Lakes Whillans and Mercer in West Antarctica¹⁸. However, often RES data from the same sites do not reveal ‘classic’ looking lake reflectors, potentially due to the water filling into a complex series of connected smaller hollows rather than a single basin, and sometimes in unexpected locations such as the lee side of subglacial obstacles to ice flow¹⁹. In some cases, water has been shown to exit one lake and fill into another several hundred kilometres away, forming a temporary river between the two. Between 1996 and 1998, the flux of water flowing between two active lakes in East Antarctica was estimated to be similar to the River Thames in London¹⁷.

After Lake Vostok was identified as a giant lake beneath central East Antarctica in 1996, subglacial lakes have been the subject of significant media and scientific attention. As unique environments, isolated from the rest of the planet for hundreds of thousands of years, they are hypothesised to be habitats for unusual, specially-adapted microbes and recorders of ancient climate change, beginning in time where ice cores end. These hypotheses are fully testable if subglacial lakes are accessed and sampled.

SCAR became involved in guiding plans for subglacial lake exploration in 2000, ensuring international dialogue and exchange of scientific findings and plans. Key to their involvement was an appreciation that these pristine subglacial lake environments should be protected and preserved during access sampling and in situ experiments. SCAR first commissioned a Group of Specialists, which then became a formal scientific research programme named Subglacial Antarctic Lake Environments – SALE. Separately, the US National Research Council organised a review of subglacial lake exploration, to understand how scientific ambition could be met while ensuring environmental protection²⁰. Following this, SCAR developed a Code of Conduct on subglacial access, which was accepted at the 2011 ATCM (held in Buenos Aires) and in revised form in at the 2017 ATCM (in Beijing). It explains the scientific basis for what cleanliness means, and details how this requirement can be achieved during in situ measurement and sampling²¹.

The 2011-12 and 2012-13 Antarctic seasons were pivotal for subglacial lake exploration. Three exploration programmes were conducted, with varying degrees of success. First, in February 2012, a Russian team extended the Vostok ice core to penetrate into Lake Vostok²². By allowing lake water to freeze into the borehole and reactivating the corer the following seasons, a ‘sample’ of lake water was recovered, albeit not using sterile procedures. Second, in December 2012, a UK team failed to activate a purpose-built hot-water drill to access and sample Lake Ellsworth in West Antarctica²³. Third, in January 2013, a US team, also using a clean hot-water drill²⁴, successfully sampled the active subglacial Lake Whillans, demonstrating that Antarctic subglacial lakes contain microbial life²⁵.

In 2015, an international meeting was held at the UK Royal Society’s Chicheley Hall to share lessons on the various deep-drilling missions, and to identify future plans and ambitions²⁶. Subsequently, in December 2018 and January 2019, Lake Mercer was accessed and the water column and sediments were sampled successfully by the US programme. The Lake Mercer science team hypothesized that microbial transformations in this lake will be driven to a large extent by relict marine organic matter deposited during a past climate scenario (J.C. Priscu, personal communication). Samples from the Mercer project are currently being analyzed and results are forthcoming. While many lessons have been learnt on how to conduct deep drilling missions, major scientific questions remain unanswered. Direct sampling of other lakes, such as deep, hydraulically stable Lake Vostok, will be required to address these questions.