A new scientific study investigates the interactions between the Icelandic volcano Eyjafjallajökull’s lava flow and the overlaying ice cap, revealing previously unknown subglacial lava-ice interactions.
Six years after the eruption, the volcano is revisited by the author of the study, Björn Oddsson, a geophysicist with Iceland’s Department of Civil Protection and Emergency Management. He and his team present the most up-to-date chronology of the events, reverse engineer the heat transfer processes involved, and discover a phenomenon which may invalidate previous studies of “prehistoric subglacial lava fields.”
Eyjafjallajökull (‘jökull’ is Icelandic for ‘glacier’) hit headlines in April 2010, as it spewed 250 million tonnes of ash into the atmosphere. The explosive event shook the West, as it took an unprecedented toll on trans-Atlantic and European travel, disrupting the journeys of an estimated 10 million passengers. It is only known to have erupted four times in the last two millennia.
The first hint that something major was about to happen in 2010 came as a nearby fissure — Fimmvörðuhálsa — to the northeast, began spouting lava in March and April 2010. Just as Fimmvörðuhálsa quieted, a “swarm of earthquakes” rocked the Eyjafjalla range, on April 13. The next day, Eyjafjallajökull started its 39-day eruption.
Over four and a half billion cubic feet (130 million m3) of ice was liquefied and vaporized as six billion gallons of lava spewed forth from Iceland’s Eyjafjallajökull stratovolcano. Flowing at distances up to 1,640 feet (500 m) each day, the lava poured down the northern slopes of the Eyjafjalla range, nearly halving the mass of the glacier Gígjökull, as it bored a channel underneath the ice.
Oddsson and co-authors Eyjólfur Magnússon and Magnus Gudmundsson have been on the leading edge of Eyjafjallajökull research, developing a comprehensive chronology of the subglacial processes at work in 2010. To complement their timeline, they developed a model demonstrating the probable interactions and volumes involved.
The eruption was exceptionally well-documented and studied in real-time by the world-class volcanologists and glaciologists who populate Iceland. Oddsson’s et al. paper relied on a previously uncombined series of datasets (i.e. synthetic aperture radar (SAR), tephra sampling, seismic readings, webcam footage) to develop an holistic model to explain the subglacial formation of the 3.2 km lava field.
In April 2010, magma began to rise to the surface — the “culmination of 18 years of intermittent volcanic unrest,” according to Freysteinn Sigmundsson and colleagues. The first outflow of lava rapidly began undermining the base layers of the Eyjafjallajökull ice cap, which was then around 656 ft (200 m) thick.
Over two billion gallons of meltwater was generated. Dammed by the surrounding glacier and rock, the water pooled within the caldera (a large cauldron-shaped volcanic crater). There, it was rapidly heated, building up the subglacial pressure under Eyjafjallajökull’s ice cap over two hours — mimicking a pressure cooker.
In the early hours of April 14, a “white eruption plume” broke through the overlying ice, ultimately ascending 3.1-6.2 (5-10 km) into the atmosphere. During the first three days of the eruption, a series of vast floods — “hyperconcentrated jökulhlaup[s]” — pulsed from under Gígjökull. The first jökulhlaup completely evacuated within half an hour, at up to 1.45 million gallons (5,500 m3) per second, according to Eyjólfur Magnússon of the University of Iceland.
The outpouring of this vast volume was the first indication of an enormous transfer of energy taking place beneath the Eyjafjallajökull ice cap. Oddsson and his team determined that over 45 percent of the heat from the eruption was expended melting the ice, based on inferences of the outflowing steam, tephra, water, and other materials.
Their paper presents a culmination of several decades-worth of research, providing a substantive advance on earlier research. For instance, in 1997 Stephen Matthews’s team estimated mass fluxes in ice, water, and lava based on steam plumes, and in 2002 John Smellie made inferences on the progress of a subglacial eruption on Deception Island, Antarctica. In 2015, Duncan Woodcock and his team provided a theoretical model for the processes, but Oddsson and his colleagues have succeeded in making firmer estimates of heat flux, at a far higher temporal resolution than ever before. It is an evolution of the working group’s 2012 study of Fimmvörðuhálsa, where similar approaches were applied.
Historically, jökulhlaups have directly claimed the lives of only seven Icelanders in the past 600 years. This rate is low, due to the preparedness of local emergency services, as well as the low population density and high level of understanding within the Icelandic population. According to a study led by Magnus Gudmunsson, most fatalities occurred near Grímsvötn — Iceland’s largest subglacial lake, situated in an active volcanic caldera.
Eight-hundred people were evacuated the day before the floodwaters barrelled down the Jökulsá and Markarfljót rivers.
Around 28 percent of the lava breached the northern caldera wall, and escaped under Gígjökull. Over one-and-a-half billion cubic feet (46 million m3) of Gígjökull’s ice mass was liquefied and evaporated as the lava flowed beneath the glacier.
As the lava was wasting the ice, it was being quenched by the ensuing meltwater. Four percent of the heat was lost to this water. A “lava crust” formed rapidly, insulating the rest of the lava, and preserving a high core temperature of over 1,832°F (1,000°C). This encrusted lava continued to flow nearly two miles (3.2 km) from the summit, underneath Gígjökull, melting the overlaying ice as it descended over the following two weeks.
Oddsson’s team explored the resultant lava field, characterised by a “rough, jagged and clinkery” surface, in August 2011 and 2012. Two distinct lava morphologies had formed on the northern slopes. The longer lava field extends of 1.6 miles (2.7 km). It formed as the lava was rapidly quenched by its interaction with the ice, and ensuing meltwater. It accounts for 90 percent of the lava which poured out under Gígjökull. A second layer poured out over the top. It formed a distinctly different rock-type as it cooled, as the overlaying ice had melted, and the water had all evaporated, or flowed downriver. Accordingly, the second lava layer cooled more slowly, losing its heat to the air.
This finding is important as it unveils the processes at work in 2010, as well as having implications for studies of “prehistoric subglacial lava fields.” Dr Kate Smith of the University of Exeter commented, “It is possible that lava-ice interaction in prehistoric eruptions has been underestimated,” as the evidence was obscured by successive layers of lava from the same event, which cooled in the air, rather than interacting with ice and meltwater.
Smith noted that this new observation is a “useful contribution to the body of work on volcano-ice interaction.” The investigation has affirmed and updated earlier glaciovolcanic investigations by David Lescinsky and Jonathan Fink of Arizona State University, outline in a seminal piece in 2000. Oddsson’s et al. findings corroborate the processes Lescinsky and Fink described, though their evidence for successive layering ”partly conceal[ing]” the record is a revelation.
This latest publication by Oddsson and his team establishes a comprehensive chronology of subglacial interactions, and reliable calculations of the heat transfer processes during the 2010 Eyjafjallajökull eruption. The paper emphasises the value of field observations of volcanic eruptions, especially from ice-capped calderas. It has shone a light on previously little-considered interactions, which has consequences for palaeoenvironmental and palaeoclimatic reconstructions. Overall, it is a valuable contribution to the ever-growing database of glaciovolcanic events, and emphasises the continued need for investigations of present and historic eruptions.