Island on Fire Read online

  Donora’s noxious smog is not an exact analogue for Laki’s volcanic haze, but it does show the kind of health hazards people face when breathing in a heavy load of particulates for days on end. So too does another great pollution disaster, one which may be more directly relevant to the Laki experience. In the winter of 1952, London experienced its own Donora moment – the so-called Great Smog. On 5 December, a thick haze descended over the city, where weather patterns trapped it for the next four days. But unlike Pennsylvania, where the steel plants did the damage, London was doomed by people simply going about their everyday business.

  Early December 1952 was bitterly cold, and Londoners had been burning more coal than normal. The low-grade domestic coal used at the time released copious amounts of sulphur dioxide in the smoke. Air pollution measurements made by the London County Council found concentrations of sulphur dioxide as high as 3,830 micrograms per cubic metre. The World Health Organization defines the safety threshold for sulphur dioxide exposure as 20 micrograms per cubic metre for an entire day, and 500 micrograms for a ten-minute peak.

  Yet in December 1952 people continued to go outside, covering their noses and mouths with masks or handkerchiefs. The streets were so dark that cars had their headlights on at noon. A performance of La Traviata began on time at Sadler’s Wells, only to be halted after the first act because of smog inside the theatre.

  Within days people began to die. They would continue to die for months to come, long after the smog had dissipated. The weekly number of extra deaths in Greater London (the number of deaths exceeding those recorded during the same time period the previous year) peaked at about 4,500 for the week ending 13 December 1952. Total mortality rates for the month were an astonishing 80 per cent higher than the previous year, and 50 and 40 per cent higher for January and February 1953.

  The Great London Smog of December 1952 had residents wearing face masks to protect themselves from the sulphureous pollution.

  A 1954 report by the UK Ministry of Health suggested that most of these people may have died not from air pollution, but from influenza. But a 2004 analysis reviewed flu reports from the time and calculated that only a fraction of the excess deaths could be explained in this way. Also, analyses of lung tissue from people who died during the Great Smog found soot and other particle types in the lungs. And a review of autopsy records at the Royal London Hospital revealed that deaths from chronic obstructive pulmonary disease between December 1952 and February 1953 were double those of corresponding months in other years.

  The haze in the summer and autumn of 1783 would have been far, far worse than the Great Smog of 1952. Whereas the London fog persisted for only four days, the Laki fog enshrouded the continent almost continually between June and September or October – even, in some cases, December. For months, Laki disgorged millions of tonnes of acid aerosols daily. Sulphur dioxide concentrations across Europe would have surely passed critical thresholds for human health time and again.

  Reports from the time are eerily similar to those of modern pollution events. The stench of sulphur hung in the air, and people struggled to breathe. In Champseru, France, a ‘pestilence’ of the throat lingered for ten months, long after the Laki fog had dispersed. In the Netherlands, Sebaldi Brugmans reported that people suffering from asthma had a particularly hard time during the Laki haze. Environmental historian John Grattan has suggested that this may have been the first time anyone explicitly linked bad air quality to worsening asthma symptoms.

  Most of the respiratory symptoms described across Europe in 1783 can be explained by the sulphur, chlorine and fluorine that Laki produced. And it’s more than possible that Laki could again kill hundreds of thousands across Europe.

  Anja Schmidt is a rising star in volcanology. A slight and intense woman, she grew up in East Germany and began her career in information technology, but soon saw that opportunities would be greater – and perhaps more fun – if she returned to university and studied volcanoes instead. Today she is an atmospheric modeller at the University of Leeds, using computer simulations to test how pollutants and other materials move around in the atmosphere.

  Sitting at a picnic table outside one of Iceland’s famous hot dog stands, Schmidt tells us how she became interested in Laki. She had been awarded a student fellowship at Leeds that would let her pursue any topic of interest. She wanted to use the university’s climate model to study how volcanic eruptions might affect the atmosphere. For that, she needed data on historical eruptions, and the eruption of Laki – whose emissions had been painstakingly calculated by scientists such as Thor Thordarson and Steve Self – provided exactly what she required.

  ‘I don’t want to do a study with no implications for society,’ Schmidt says. So she tackled the question of what might happen if a Laki-like eruption were to go off tomorrow. Give the Leeds computer model a certain number of factors – how much material the volcano ejects, for how long, and at what altitude – and it can create a prediction of where that material might spread. For Laki, Schmidt programmed the model to erupt the same sort of stuff that Laki produced in 1783–84, but under today’s atmospheric conditions. Her goal: to see how air pollution from such an eruption would spread across Europe today.

  Schmidt focused on the spread of particles smaller than 2.5 micrometres. This size, known as ‘PM 2.5’, is a standard epidemiological measure for particle sizes that cause respiratory distress. High concentrations of particles smaller than PM 2.5 can lead to breathing problems, especially in children and the elderly. This is true whether the particles are made of sulphur dioxide, soot, or any other material.

  Schmidt’s paper, titled ‘Excess mortality in Europe following a future Laki-style eruption’, makes for depressing reading. During the first three months of the eruption, volcanic particles would cause air pollution loads in central, western and northern Europe to double. For 36 extra days the PM 2.5 air quality standard would be above the recommended guidelines from the World Health Organization. As a result, Schmidt’s team calculated, some 142,000 people would die. Take another look at that number: 142,000 deaths.

  Surprisingly, her work received relatively little press when it was published in the September 2011 issue of the Proceedings of the National Academy of Sciences. Perhaps people were inured after the drumbeat of coverage of the 2010 Eyjafjallajökull eruption. Or perhaps the devastation from another Laki just seemed a little too hypothetical. In January 2012, however, the UK’s National Risk Register for the first time listed volcanic eruptions as something the country needs to prepare for. The list cited Schmidt’s study.

  Still, the question remains: what could governments actually do if such an eruption occurred?

  CHAPTER NINE

  The Next Big Bang

  How worried should we be?

  NEARLY TWO DECADES BEFORE the mountain blew, it quivered. First came the earthquakes – small ones of magnitudes one, two and three, detectable by seismometers but not anything that farmers going about their business in Eyjafjallajökull’s shadow would have noticed. Then the volcano began to move. Global-positioning stations resting on its flanks lifted ever so slightly higher, as if it were drawing in a deep breath.

  After years of grumbling and spitting, the volcano began to awaken for real. In late March and April 2009, quakes began rattling beneath Eyjafjallajökull, as magma moved upward in its belly. Seismologists listened as the tremors became bigger and more frequent. By March of 2010 the volcano was shaking loudly. Its eastern flank began bulging upward as magma gathered beneath it.

  Eyjafjallajökull finally roared to life on 20 March. Fire fountains exploded out of a rocky ridge just east of the mountain’s ice-covered summit. It was one of the prettiest eruptions in years. During the day, the dusky light of a northern spring noon highlighted the spectacular orange bursts, while at night the cool green glow of the northern lights shimmered overhead. The eruption, on a flank known as Fimmvörduháls (Five Cairns Pass), was happening on a popular trekking path. Hikers now hired j
eeps or quad bikes to run them up the icy trails and drop them within metres of the eruption.

  Icelanders call this sort of event a ‘tourist eruption’, and enterprising operators quickly started running daily trips from Reykjavík. Vehicles were soon jockeying for position up and down the icy ridge, while police tried to corral the crowds into strict safety zones. It didn’t always work: two people died of exposure after they drove too far from the eruption and their car ran out of fuel.

  Had the volcano stopped at that point, Eyjafjallajökull would have been not much more than a footnote in the history books. But on 12 April lava ceased fountaining out of the barren ridge of Fimmvörduháls. A day and a half later, everything went to hell.

  At 10.29 p.m. on 13 April, quakes started shaking directly underneath the snow-capped summit of Eyjafjallajökull, just to the west of the tourist eruption. For two and a half hours they went off like gunshots. At one point, as magma gushed upward, quakes were coming nearly once a minute in the five kilometres of rock directly below the summit. At around 1.15 a.m. on 14 April, the quakes started slowing down and a low-frequency tremor began reverberating through the top of the mountain. Icelandic seismologists later concluded that at this point the magma had emerged and begun melting the mountain’s ice cap. Every second, fire from within the Earth was melting some 300 to 500 cubic metres of ice.

  Because of the risk of meltwater floods, officials began evacuating residents from the south side of the volcano, and then the east and north. The first jökulhlaup rushed down the north side of the mountain in the early morning of 14 April. Today you can see the traces of its fury if you take a jawrattling ride up the valley of the Markarfljót river. Your driver will barrel past tranquil farmhouses, through icy streams, and across the great grey valley floor before coming to a halt on Eyjafjallajökull’s north flank. The mountain looms above you, dark and ominous, its ice-capped peak usually enshrouded in white clouds. Descending towards you is a massive finger of ice, the glacier known as Gigjökull. The ice is deep blue at its heart, white closer to the surface, and dusted everywhere with the grey-black ash of the eruption.

  Before April 2010, this frozen hulk reached down the side of Eyjafjallajökull and plunged into a beautiful glacial lagoon, where icebergs studded the jewel-blue water. All that changed in an instant when the volcano erupted and meltwater began roaring down. The flood coursed down the mountainside and spilled into the river valley so powerfully that it washed away the tranquil lagoon. Even the gauge put there to measure such an outburst was obliterated as a slurry of ice, rock and water rushed down the Markarfljót valley. A second gauge, downstream, recorded the flood moving at 2,500 cubic metres per second. Levees built for just such a moment held fast, channeling most of the flow away from farms. Thwarted, the churning waters rushed out through the valley and into the sea to the south. A second flood followed the next day, powerful enough to overtop the protective levees.

  But that was just the water from Eyjafjallajökull. There was also, of course, the ash. As soon as earthquakes began rumbling beneath the ice-capped summit, scientists knew ash might be a problem. When hot magma meets cold ice, it generates steam that fragments the magma into zillions of tiny pieces. That’s volcanic ash, most of which measures just a few millimetres across. Because the particles are so light, they can be hurled high into the atmosphere by the force of the eruption and then carried long distances by the winds.

  Eyjafjallajökull began pouring out its infamous ash sometime in the early morning of 14 April. Icelandic officials first spotted it at 5.55 a.m. that day, from a survey plane sent to monitor the eruption. Those aboard saw a white plume rising through the cloud bank covering the mountain’s summit. It was the first glimpse of the ash that would rewrite European rules on dealing with natural hazards.

  Initially, the big challenge was to figure out how much ash was coming from Eyjafjallajökull and where it was going. Radar signals are used to obtain this information, but in April 2010 Iceland had just one weather radar system in operation, at the Keflavík complex, which sprawls on a peninsula west of Reykjavík. Once a United States air base, Keflavík is now the country’s main commercial airport and thus a natural place to keep its main weather radar. But between Keflavík and Eyjafjallajökull lies a mountain range that blocked much of the radar’s line of sight. For more than a quarter of the time that the volcano was erupting, the ash plume was too low for the Keflavík radar to see it. (Officials have since added a second radar in eastern Iceland, plus two mobile radars that can be driven where needed to track ash plumes.)

  On that first day, the ash plume started out white and darkened throughout the afternoon as more and more material was spat out. By 6.30 p.m. it was nearly black and shot through with lightning. Westerly winds carried the ash eastward at first, towards northern Norway, where airspace closed that evening. The next day, the plume spread far wider and curled south, covering the airspace for other parts of Scandinavia as well as the United Kingdom.

  This trajectory triggered a regional crisis. After the 1982 near-disaster, when a British Airways plane nearly crashed after flying through Indonesian ash, the International Civil Aviation Organization decided that no flights should go through any airspace where concentrations of volcanic ash are greater than zero. So as soon as Eyjafjallajökull began erupting, Icelandic scientists started sending their observations of the plume’s location to the London Volcanic Ash Advisory Centre (VAAC). There are nine VAACs worldwide, each responsible for monitoring and forecasting volcanic ash plumes in their assigned region.

  Icelandair names all its planes after the island’s volcanoes, but surely didn’t foresee the irony in this particular name.

  Forecasters at the VAAC used computer simulations to predict where the Eyjafjallajökull plume might spread from day to day, depending on weather conditions and how much ash the volcano was ejecting. Day after day, those forecasts blocked out huge regions of airspace all across Europe. Aviation authorities followed the zero-tolerance rule for ash and shut down airports from Heathrow to Oslo. (Ironically, Keflavík itself remained open, since it lay west of the spreading plume.)

  Some 100,000 flights were cancelled in the week after the eruption began, stranding ten million passengers. Only then did aviation officials begin reevaluating their definition of what was safe to fly in. Under huge pressure to get planes back in the air, governments started allowing flights to take off. In the end, the airspace closures cost Europe billions of euros because of lost passenger revenues and delays to air cargo. The consequences stretched as far as Kenya, where women were laid off by the floral industry after their flowers were left rotting at airports.

  Eyjafjallajökull gave its final belch on 17 June 2010. All told, the volcano disgorged nearly 480 million tonnes of ash over more than seven million square kilometres of Europe and the North Atlantic. Less than 0.02 per cent of the erupted ash made it to mainland Europe – and yet even that modest amount caused utter chaos.

  By most measures, Eyjafjallajökull wasn’t an exceptional eruption. On the VEI scale of an eruption’s power, it ranked a modest 3. Both Katla in 1918 and Grímsvötn in 2011 spat out more rock and ash than it did. What made Eyjafjallajökull so disruptive was a combination of three things: it erupted ash almost constantly for 39 days; its ash grains were relatively small, and so could be lofted a long way; and the winds carrying the ash just happened to be blowing from the north and northwest most of the time the mountain was erupting.

  Some fifteen eruptions of Eyjafjallajökull’s size go off in Iceland each century. But what might happen if a larger eruption were to occur tomorrow? What about, say, a Laki-style event? It would produce a lot less ash than Eyjafjallajökull did in 2010, but a lot more deadly gas. There’s obviously no certainty here, but we can speculate on how such an eruption might unfold – and how it might change our world.

  It all begins as small tremors start shaking south-central Iceland. At first the earthquakes are too tiny for anyone to feel, but the Icelandic Meteo
rological Office’s seismic network picks them up, and a swarm of dots now show up on the IMO website. Let’s say they appear just west of the Laki crater row, in the famous ‘fire districts’.

  A blogger picks up on the activity and writes an excited post about how Laki is reawakening. Almost immediately the IMO issues a statement correcting him on some factual errors, but the post is enough to draw the attention of the world’s media. Newspapers and websites start running dire-sounding stories about an impending eruption, and the IMO website keeps crashing as people try looking up the activity for themselves. But gradually the earthquakes become less frequent, and then they disappear. It looks like the whole thing might have been a false alarm.

  The volcano is only teasing. Nine years later, the small trembling of quakes reappears and suddenly accelerates its drumbeat. The quakes are clustering along a line that runs from the southwest to the northeast towards the huge Vatnajökull ice cap – plot the line on a map and it is nearly dead centre between the rift that opened in the year 939, in the Eldgjá eruption, and in 1783, as Laki.

  Now slightly concerned, IMO scientists deploy more seismometers and global-positioning instruments to this sparsely settled area. Just in time, it seems: the newly installed GPS monitors discover that the western edge of the line of quakes is beginning to rise, and quickly. Magma must be on the move. The earthquakes are speeding up, and occurring at shallower depths.