Iceland has ice. Glaciers cover 10% of its landscape, including its highest volcanoes. Of its frequent eruptors, only Hekla is (almost) ice free. Katla, Bardarbunga, and Grimsvötn, which together account for the large majority of eruptions, are all hidden underneath ice sheets, which gives problems studying the volcanoes themselves. But a more serious issue is that ice on a volcano can be a dangerous combination. Even minor eruptions can generate catastrophic melt water floods which destroy everything in their path. These are not just flash floods. Think of it as a tsunami coming down the mountain. Several volcanoes in Iceland are frequent offenders. But the infrequent ones may be more dangerous than the usual suspects.
Öræfajökull is one of those infrequent erupters. It is Iceland’s tallest volcano, and is the fourth tallest stratovolcano in Europe (after Elbrus, Etna, and Beerenberg). The name gives an indication of its nature: Jökull means glacier. The mountain itself has no single name (the original name was Hnappafel), only its ice cover does. This is a wild land with a glacier that towers over its mountain.
After the settlement of Iceland, the fertile plains on the southwest of Öræfajökull were called Litla-Hérað and although isolated, the area was densely populated with twenty or more farms and several churches. (The name ‘Hérað’ was used for a wealthy parish, so Litla-Hérað was a ‘small major parish’.) On the southeast side of Öræfajökull was Breiðamörk, also well-off, where around 900 AD, the settler Þórður Illugi built a farm at Fjall. The farm did well, acquired a church (or the church acquired the farmstead – both happened at times), and a large number of other farmsteads came. At that time the glaciers were much smaller than they are now, and between the glaciers and the coast was a prairie-like landscape with lush grass, copses of trees, and rivers. It must have seemed like heaven.
And than the disaster of 1362 struck. Afterwards, the land was left empty for over 40 years. When people returned, they called it ‘Öræfi’, meaning wilderness. The glacier took its name from this. Today, the fluvial sand desert of Skeiðarársandur borders where once the paradise of Litla-Hérað was.
The glacier-covered mountain of Öræfajökull dominates the skyline here. Its caldera, 14 km2 in areas, is filled with ice, 550 meter deep. Around the caldera rim are the individual peaks of the volcano: Hvannadalshnúkur , Rótarfjallshnúkur , the twin peaks of Hnappar, Sveinstindur and Sveinsgnípa. Hvannadalshnúkur is the tallest, at 2110 meter, and is the highest mountain in Iceland (albeit by a margin of only ten meter). The mountain is 20 km across at its base near sea level, making it rather steep. A quarter of its base area is covered by the glacier. The volume of the mountain is around 370 km3, which makes it the largest volcano in Iceland by volume as well as height.
Öræfajökull has another claim to fame: it has Iceland’s highest amount of snowfall. Each year, over 10 meter of snow accumulates at the summit. The glacier did not get there by accident! The rain in Iceland comes mainly from the south (the north is drier), and Öræfajökull is close to the sea – it captures quite a lot of it. At sea level the weather is not quite as wet, and there the climate is mild – for Iceland!
The current mountain formed when two older, adjacent volcanoes merged. This was during the ice ages, when the glaciers hemmed in the eruptions. They kept the lava close to the eruption sites, and forced the mountain to build upward rather than outward. This is why the mountain is both tall and steep: it was made that way by the containing ice.
The largest explosive eruption in Iceland since the settlement occurred with little warning. Whether the mountain was known to be volcanic is not known, but it had been quiet ever since the settlement. The explosion in 1362 blew out 2.5 km3 of rock, or 10 km3 of tephra. Of this, 2km3 fell on land, mainly on the southeast of the country. The western parts of Iceland were protected by the westerly wind of the day of the eruption. The debris caused damage 70 km away, but the wind blew a lot of the tephra out over the ocean. Six jökulhlaups came out from the various glacial outlets, mostly to the southwest. The Litla-Hérað region never fully recovered its vegetation: the debris from the eruption is still evident in many places. It is among the least recognized volcanic disasters of the last millennium.
There are descriptions of the eruption in several old documents, but few are contemporary with the events. One of the oldest is from the annals from the Skálholt bishopric, and date from about 30 years after the eruption, within living memory. The description gives an impression of the devastation:
‘Fires appeared in three places to the south. They kept on from June to autumn. With a great deal of noise. All of Litla Hérað destroyed & much of Hornafjörður & the Lón district. Roads to the Alþing also destroyed. Knappafell glacier surged into sea. Up to thirty fathoms deep. With landslides, mud & dirt that levelled gravel plain. Kirk & parishes at Hof & Rauðilækur all razed. Sands piled up on plain. Icebergs stranded on them. Houses could hardly be seen. Ash fell in north of land. Could be traced all the way. Accompanying this, pumice seen in western fjords. Ships could hardly sail because of it [& even more widespread to the north]’
The report of ash in the north refers to the fact that this document was written in the north of Iceland. Thus, it describes a local impact and does not imply that ash did not fall elsewhere.
At the time of writing of this annals, the area of Litla Hérað had not yet been resettled. People began to move back in shortly after, but this was 40 years after the event. After the eruption, the area was left not just depopulated, but also uninhabitable.
The Flatey Book written during the last decades of the fourteenth century, adds another description:
‘Volcanic eruption in Knappafellsjokull with darkness so intense that roads could not be seen at midday, and the entire Litlaherad was devastated.‘
The ‘roads’ of course refers to a poorly visible path or track, not an A-road! It was so dark that the track could not be picked up – not complete darkness, perhaps, but darker than twilight.
The ‘Annals of Oddi’, dating from around 1580, report another tradition, albeit from 5-10 generations later:
‘Anno 1366: Fires arose in Litla Hjerad and destroyed the whole region: there once had been 70 farmsteads: no single living being left except one olde woman: and mare’ The year that is stated is not exactly correct, but it must refer to the 1362 event. The later stories about the death toll of the 1362 eruptions and about the two survivors come from this, but this detail is not backed up by the earlier report. The total population of the area has been estimated at a few hundred people, based on the number of farms, so this would imply a survival rate of less than 1%.
Ísleifur Einarsson, around 1700, wrote down the oral tradition from the area itself, including the story of the survivors. In this tradition, the old woman was now a (presumably male) shepherd:
According to the folk of Öræfi, from what they have heard [and from those who had heard it from their predecessors, and they from theirs, by word of mouth], the Öræfi district has been wiped out twice. [Some say three times]. On one occasion, the shepherd at Svínafell, by the name of Hallur, had been herding the sheep home and the women had begun milking them, when all at once a huge crack came in Öræfajökull so that they were truly amazed. Then after a little while, another crack came. And then the shepherd said: it would not be wise to wait for the third. He had then fled up into the Flosahellir cave – it is up on the mountainside to the east of Svínafell farm – and then the third crack came in the aforementioned glacier, and it all burst open and so much water and so many rocks gushed right out of it and through every gully that all the folk and livestock in Öræfi perished, except for this shepherd and one horse with a blaze on its forehead’
These stories give rather little to go on. Their emphasis is on the effect of the floods (the jökulhaups), and the surging glaciers. Ash is only mentioned as falling towards the north. However, this description must be incomplete. The floods were enormous and devastating, but the damage was from more than just water. I will come back to this later. The lack of survivors has meant part of the real story was lost, and the reports could have been based on later exploration of the damage, rather than reports from eye witnesses. The layering of the tephra deposits found by geologists have suggested that the explosive event was brief, a day or less; the reports that the eruption lasted until the autumn may refer to minor later eruptions, or just refer to ice or permafrost geysering from the heat.
This was a much smaller eruption, which produced about 0.2km3 of ash. It came after 450 years of quiet. The first sign of trouble was a series of felt earthquakes, on Sunday 7 August 1727, beginning during the church service. For a while things calmed down, but early on Monday the shaking became severe enough to do damage inside the farmsteads.
At 9 o’clock a huge crack in the mountain side appeared and jökulhlaups burst out from beneath several of the glacier outlets. They destroyed the fields, but only three people were reported killed. The flood waters bore boulders, pumice stones, ash and icebergs. The ash fall made everything dark for three days, but it cleared up on the fourth day, after which there was little further ash fall. Volcanic ash covered the farmlands of Svínafell and Skaftafell, and there was no hay that summer in the districts nearest the eruption site. Sandfell, the main settlement, was missed by two landslides, coming down on either side!
Niels Horrebow described the jökulhlaup:
‘The water flowed from the mountain between two farmhouses called Hoff and Sandfeld , which lie not above six miles from the foot of the mountain, and about the same distance from each other. It spread itself beyond these houses in the flat country, and washed through the lower house and dairy, and carried off all the milk, butter, &c. the people saved themselves by getting on the tops of the houses. The water did not rise so high, and only filled the inside. Numbers of the cattle from both farms were carried off, and some of them were afterwards found parboiled. It ran along the valleys and emptied itself into the sea; but the stones, sand and earth, carried away by the current were not any way equal in quantity, nor rushed on so violently as the former, though greater damages were sustained. No cattle were lost in the other eruption, nor any field of grass destroyed, the same being entirely demolished by former inundations: but the stream of the eruption we are now talking of, passed over fine fields, and destroyed the better part of the cattle that were grazing in them’
In another version, rather than the cattle it was one of the missing people that ended up boiled.
The priest at Hof, Pastor Jón Þorláksson, had to flee for his life but survived. His stories were later written down by a traveler, Ólafur Ólafsson:
‘When I was below the middle of Flögujökull looking up at the summit, it seemed to me the glacier grew higher and swelled out one moment and then collapsed and sank inwards the next. Nor had I been mistaken, for it soon came to pass what this sight boded. The following morning, Monday 8 August, we all not only often felt tremendous earth tremors, but also heard a horrifying cracking sound as loud as thunder. In all this commotion, all loose objects in the home were falling down, and everything seemed most likely to collapse, both the buildings and the mountain. … At 9 in the morning, three tremendous cracking sounds were heard, much louder than the previous one, and accompanying them were jökulhlaups or eruptions, the final one the worst, carrying away all the horses and other livestock in its way in one fell swoop. After which the glacier itself slewed down onto the lowland, like molten metal poured out of a melting-pot. It was so high once it had surged down onto the level ground that I could see only a tiny part of Lómagnúpur promontory above it, about the size of a bird. Once this had happened, water began to pour out to the east of the glacier and destroyed what little grassland that was left there. … The situation then changed yet again, the glacier itself surging forwards, bearing icebergs all the way to the sea, though the largest remained right at the foot of the mountain. The next thing, the air was filled with fire and ashes, with ceaseless cracking and crashing; the cloud of ash was so thick that it caused there to be no difference between night and day; the only visible light was coming from the edge of the fires that had arisen in five or six gashes in the mountain. For three continuous days, the parish of Öræfi suffered from these fires, floods and ash falls. It is not at all easy to describe all this exactly as it really was, for all the earth was black from pumice cinders and it was risky to go outside because of glowing stones that rained down from above, so that one had to hold buckets and tubs above one’s head for protection.’
Jón Þorláksson later commented that the floodwaters had been so hot that the horses shied from them.
It appears that the eruption itself took place within the caldera (as it did in 1362). The jökulhlaups came before the main explosions. The melt water came from a fissure, on the western side, about 1100 meter above sea level. Six or seven columns of flame were seen within the fissure, and smoke was visible until the following spring, but no evidence of any flow of lava.
A more detailed reconstruction can be found in http://www.vedur.is/gogn/vefgogn/jokulhlaup/haettumat/oraefajokull_markarfljotsaurar/bok_en/chapter_II.pdf
Water and pyroclastics
The reports and papers suggest that the damage in both eruptions came from jökulhlaups and glacial surges. But for the 1362 eruption, it seems very unlikely that only water and ice caused damage. This was a VEI6 with 10km3 of ash, tephra and pumice; some of the ash and pumice reached Europe. Such large eruptions devastate their immediate surroundings. There must have been major effects of the explosion itself. (As an aside, eruptions from Öræfajökull have a significant fraction of rhyolite eruptions, and this may be the most explosive volcano in Europe.)
In fact, Dave McGarvie (https://ounews.co/science-mct/science-environment/icelands-tallest-volcano-awakening/) mentions that there are massive ash and pyroclastic deposits visible in the landscape. This explains why the land was left empty for so long, and why it became a waste land. The ash and tephra had covered the land and killed all vegetation. The farms were wealthy before, but after resettlement, the farming was poor, and the land had lost its fertility. In 1362, there appears to have been a sequence with, in order, tephra and ash, followed by pyroclastic flows, followed by pumice. Jökulhlaups were the afterthought. In 1727 the jökulhlaups came first and did almost all of the damage, but the subsequent ash was still significant, in spite of the eruption being puny compared to 1362.
Interestingly, archeology of the farmsteads of 1362 suggest that they had been abandoned, together with their valuable content, before the ash and pumice from the main eruption destroyed them in 1362. When the buildings collapsed under the weight of ash, they were already empty. People had time to flee. Did they make it to safety? That we don’t know. They may well have been killed by the rain of hot pumice.
Another point to make is the effect of the ice. At the time of the 1362 eruption, glaciers had begun to grow but probably were still small compared to later. In 1727, the glaciers were much larger, and were still growing as the Little Ice Age moved towards its peak. Farmsteads were being destroyed by the advancing ice. The two eruptions occurred with very different sizes of the glaciers and this may have affected the flow direction and severity of the jökulhlaups. Nowadays, glaciers are rapidly diminishing but are still large. It may not be easy to predict how the floods would flow under current ice conditions and we should be careful with extrapolating from the two known eruptions. The past is not always a good indication of the future.
An interesting question is where the 10km3 of tephra in 1362 came from. The caldera where the explosion occurred is 14km2 in size and up to 500 meters deep, giving a maximum volume of 7km3. The volume will be a bit less if the caldera is bowl-shaped. Now the tephra is considerably expanded compared to the rock that was blown up. The dense rock equivalent (DRE) is 2.5km3. This means that the caldera is larger than the eruption volume, or at worst similar in size. Part of the caldera may in fact come from the 1362 eruption. However, the numbers leave no room for any mountain peak on top of the caldera. It appears likely that there was already a depression, but that it was enlarged by the explosion. The rim of the caldera was probably already there at the time, but perhaps not fully in its current form. Of course, an excavated hole may have been filled in by lava after the eruptions: how can we tell, beneath half a kilometer of ice? But Iceland’s volcanoes are a changeable lot. Big eruptions can re-design their shapes.
Albert, November 2017
The main source used for this post is Helgi Björnsson, The Glaciers of Iceland, published in 2017.
Also well worth reading is Sigurdur Thorarinsson, THE ORlEFAJOKULL ERUPTION OF 1362, pubished in 1958. It is available at http://utgafa.ni.is/Acta-Naturalia-Islandica/Acta-Naturalia-Islandica-II-2.pdf