A buried saltwater reservoir could help color "Blood Falls," seeping from the end of East Antarctica's Taylor Glacier. A decade of observation has revealed lakes and streams hidden beneath Antarctic ice. Image credit: Peter Rejcek/Antarctic Photo Library/NSF

A buried saltwater reservoir could help color "Blood Falls," seeping from the end of East Antarctica's Taylor Glacier. A decade of observation has revealed lakes and streams hidden beneath Antarctic ice. Image credit: Peter Rejcek/Antarctic Photo Library/NSF

A broad swath of the East Antarctic ice sheet collapsed like an icy puddle on a sunny day, leaving behind a 10-mile-wide crater—evidence of a massive subsurface flood captured by two NASA satellites in 2007 and 2008.

It was one of a suite of observations over the past decade that has irreversibly altered scientific understanding of Antarctica’s subsurface lakes. Instead of static, isolated pockets of water trapped within the ice layers, the new picture—gradually revealed by space and airborne instruments, ground-based GPS and increasingly sophisticated computer models—is one of intricate and dynamic hydrology. Rapidly flowing subglacial streams channel water from one hidden lake to another, sudden drainage causes massive collapse, and the meltwater speeds the slide of mighty glaciers toward the sea.

The big picture was put together by a group of researchers specializing in oceans and ice and published in The Royal Society of London’s “Philosophical Transactions A” in August 2016. It shows significant progress in the understanding of Antarctica’s under-ice lake systems, promising to sharpen the accuracy of computer modeling on a continental scale; this could allow scientists to forecast more precisely alterations in an ice sheet under pressure from a changing climate.

“Ten years ago, we didn’t realize subglacial lakes were so active,” said Helen Fricker of the Scripps Institution of Oceanography, a member of NASA’s Sea Level Change Team and lead author of the research group’s review article. “We didn’t realize water moves so quickly from one lake to another. We didn’t know it was a dynamic system, where water could transfer so quickly.”

Seeing through the ice

The review article summarizes the “transformational decade” of research since the discovery of these active subglacial water systems.

The zone of subsurface water flow, where ice meets bedrock, is difficult to measure, requiring radar, radio-echo sounding, and seismic surveys that can penetrate the ice sheet—one that is, on average, 2.2 kilometers (1.3 miles) thick. The ice surface also must be monitored for elevation changes that signal the filling and draining of lakes, and the movement of water along channels cut through bedrock.

The measurements are made using satellites, aircraft, and instrument arrays on the surface.

Such measurements, though challenging, are critical to gain a clearer understanding of the deeply hidden plumbing driving ice-sheet evolution.

And the stakes are high. Insulation, high pressure and heat from Earth’s interior melt enough of the ice sheet’s base to produce an estimated 65 gigatons of water per year, much of it deposited in subglacial lakes. These come in two forms: those remaining stable and inactive for tens to thousands of years, and those that are active on scales of less than a decade.

These active lake systems prevail on the Antarctic coast and are mostly found under rapidly flowing outlet glaciers, as well as ice streams; they drain up to 90 percent of the water flowing into the ocean from the ice. They are, in fact, a controlling factor, reducing friction between bedrock and ice and speeding ice flow.

The drainage of an Antarctic subglacial lake, Cook E2, left a crater on the surface, captured with synthetic aperture radar readings from CryoSat-2; colors highlight ice surface. Image credit: Malcolm McMillan, University of Leeds and the UK Centre for Polar Observation and Modelling.

The gaping crater formed in East Antarctica between 2007 and 2008 is among the most impressive effects reviewed by the research team. Over a 20-month period, flooding drained 4.9 to 6.4 square kilometers (1.8 to 2.5 square miles) of water from Lake Cook E2, causing an ice-surface elevation change of 65 meters (213 feet). At its peak, the floodwaters moved at a rate of 160 cubic meters (5,650 cubic feet) per second. This subglacial flooding event was twice as big as any ever recorded, amounting to about 10 percent of Antarctica’s yearly under-ice meltwater. Its dimensions were mapped using data from CryoSat-2, in its synthetic aperture radar-interferometric mode, and ICESat’s synthetic aperture radar.

A ground-based GPS array, meanwhile, was established over Subglacial Lake Whillans in West Antarctica—the first array of its kind—after a large number of interconnected lakes was discovered in the lower Whillans Ice Stream on the Siple Coast. The lakes were distributed across three catchment basins, from north to south. After initial installation in 2007-2008 and an expansion in 2010, the stations gathered data until 2015, providing an eight-year time series at a far higher frequency of sampling than possible with the periodic passage of satellites.

Folding subglacial systems into broader ice modeling

Further satellite measurements allowed researchers to create a 10-year record of subglacial lake activity, including draining and filling, changes in lake shape, movement of water from place to place and recurring patterns.

In the Recovery Ice Stream in East Antarctica, ICESat data confirmed direct hydrological links between active subglacial lakes, with surface elevation dropping over upstream lakes as it rose over lakes downstream.

Such measurements allowed other research teams to create computer models that faithfully reproduced volume changes observed in the Willams-Mercer subglacial lake system in West Antarctica from 2003 to 2009.

A decade of game-changing observations of water flow beneath Antarctic ice is bringing the science community closer to a long-desired goal: achieving a complete enough understanding of these processes to include them in broader ice-sheet models. That, in turn, could yield far deeper insight into likely future states of an ever-changing ice sheet.

“There’s a lot of work to do, still,” Fricker said. “We don’t understand how this subglacial water system affects ice sheet dynamics. We’re closer to understanding that than we were 10 years ago; we just need data. We need continued monitoring from satellites.”

Read the paper, "A decade of progress in observing and modelling Antarctic subglacial water systems"

Read related story, "Studies offer glimpse of melting under Antarctic ice"