Iceland can do surprises. Its main activity is in a limited number of volcanoes which can erupt anytime, with frequencies ranging between once per century to twice per decade. These are the usual suspects, including Grimsvotn, Bardarbunga, Askja, Katla, Hekla, Krafla. (Grimsvotn is the only one not ending in ‘a’.) The beasts can go small or big, and do fire eruptions or explosions – they are like a box chocolate, as Forrest Gump famously said. In the current decade, we have added the Reykjanes peninsula to the list. There are other volcanoes in Iceland, but their eruptions are much less frequent. An eruption from a volcano with a dormancy of many centuries always comes as a surprise – as we have seen with the original Fagradalsfjall eruption.

Icelandic volcanoes are associated with a number of rift zones. (The number is 9, to be precise, including the submarine Reykjanes Ridge.) These zones are some tens of kilometers wide but can run along the length or breadth of Iceland. Within the zones are the individual volcanic centres (30 in total), typically consisting of a central volcano (there are 23 of those), often with its own fissure swarms (19) which can extend for 100 km or more. Not all locations have central volcanoes: the Reykjanes peninsula, for instance, can erupt anywhere and does not revisit an old cone (other than in different episodes in the same eruption). On the other hand, a few volcanic centres have two central volcanoes.

Eruptions in different volcanic centres are mostly independent events, but as the eruptions can last for years to decades (such as in the Krafla fires – this can be seen as a single eruption or a series of related eruptions), two different systems may still erupt at the same time, largely by coincidence. There are suggestions for phases of enhanced activity across Iceland, but we have not enough data yet to know. Give it another thousand years.

Of the 8 on-land volcanic zones, the eastern volcanic zone (EVZ) is by far the dominant one. Around 70% of the known eruptions (not counting the Reykjanes Ridge where no good data exists) have occurred here. Second on the list is the northern zone (NVZ) with 12%, followed by the western zone (WVZ) with 11%. The outliers are the also-rans: the Reykjanes Peninsula (RV) accounts for 4%, and the Öræfajökull Volcanic Belt (OVB) and the Snæfellsnes Volcanic Belt (SVB) for 1% each. Finally the Mid-Iceland Belt is almost a goner with 0.1%, making one wonder whether it exists. Still, like Father Christmas, eruptions can show up anywhere. A 1% chance is not zero.

The Snæfellsnes Volcanic Zone (SVZ) is the odd one out in Iceland. It is disconnected from the other zones. The peninsula traces an ancient transform zone, 7 million years ago, which was abandoned when the transform zone moved south to form the Reykjanes peninsula. But althogh there is no longer any transform movement or rifting, it did not go fully extinct, or perhaps it re-activated in the past million years. The last known eruption here, mentioned in the Book of Settlement, was 1000 years ago.

The SVZ contains four distinct volcanic systems. From west to east, these are Snaefelljökull, covering about 600 km², Lysuskard (150 km²), Ljósufjöll (1,800 km²) and finally Snjofjöll. Snaefelljökull may have had as many 30 eruptions in the holocene, Lysuskard (around the mountain range labelled Snaefellness on the map) had two and Ljósufjöll some 20. Snjofjöll has not erupted in that time and may be extinct: it is normally left off the list. Snaefelljökull is surrounded by lava fields of a variety of ages but the youngest age is 1750 years. (One field may be a bit younger.) Ljósufjöll has a similar number of lava fields but they are not well dated. However, the report from the Book of Settlement confirms that the most recent eruption of the region was here, around 960 AD.

Based on averages, this area is now overdue an eruption. But of course, volcanoes don’t do schedules and any argument of an ‘overdue eruption’ hardly ever works. However, there is still some seismic activity on the peninsula. Micro-earthquakes up to M1 have been found under the southern flank of Snaefelljökull, at a depth of 9-13 km. These have been attributed to a local magma chamber, albeit one that seems happy to remain where it is.

Ljósufjöll

The Ljósufjöll volcanic system extends for some 100 km, from the north coast above Sejafell to the start of the peninsula. At the northwestern end, the lava field called Berserkjahraun is located. The southeastern end point is at a crater near the town of Bifröst (yes, that does exist). There is a central volcano, 1km high, which has had rhyolitic eruptions, but it has had only a few eruptions since the ice age. The central volcano has also produced basaltic tephra. It is an ancient region: the oldest dates for silicic lava are over 1 million years. (In Iceland, silicic eruptions are associated with central volcanoes and their flanks. Rift eruptions are normally basaltic.)

The Ljósufjöll eruptions in the holocene have mainly been on fissures away from the central volcano. The eruptions here have made use of older faults which allowed magma to come up and formed short fissures and small craters. These are not rifting events: rifting has long since moved elsewhere in Iceland, but the faults are still there.

The volcanic system changes appearance between west and east. In the western region, the volcanic system is narrow and well-defined: the width is only a few kilometers. In the east, the volcanic region becomes much wider, with a spread of some 20 kilometers. The western eruptions have been much larger than the eastern ones. In the east, eruptions tend to be in the valleys rather than the hills. This includes the two valleys of two valleys, Hnappadalur and Hitardalur, both south facing with lakes at the top. One may expect that eruptions here have a phreatic component. The fissures tend to be short and the lava volumes limited. The magma is alkali-basaltic, from the mantle with only a small amount of crust melted into it. The eruptions formed scoria cones.

A number of the eruption products have been dated to be between 100,000 and 200,000 years old. Four eruptions are known to be younger than 4000 years, which includes the lava with the most worrying name in Iceland, the Berserkjahraun. These eruptions were between 500 and 1500 years apart. Their basaltic composition shows that the magma flows can bypass the central magma chambers which fed the rhyolitic eruptions. The basaltic magma rose from the mantle, but whether it came up underneath the central volcano or elsewhere is not known.

The 960 AD eruption

The most recent eruption around 960 AD formed the so-called Rauðhálsahraun. The lava covered around 13 km² within the valley of Hnappadalur. The volume of the eruption may have been up to 0.1 km³. According to the Book of Settlement, the eruption took place in Eldborg crater, but this is wrong. Eldborg is a beautifully complete ring near the coast which erupted some 5000 years ago but not since. The 960AD (within a few decades) occurred further north, up the valley.

The Rauðhálsahraun flowed from (and is named after) the Rauðhálsar scoria cone, at the eastern end of the lava field. The cone is 80 meters high and measures 500 by 300 meters. Within the crater is a smaller crater which is over 20 meters deep, presumably from an individual explosion during the eruption. There is an ash layer associated with the eruption which extends about twice as far as the lava, and which reaches a thickness 10 to 50 cm. The eruption started with a phreatic phase, before the lava began to cover the land.

The eruption occured during the earliest settlement of Iceland. This is clear from the peat, which changes colour from black to brown from below to above the ash layer, a colour changes that has been attributed to human activity. The Book of Settlement mentions a town (or farm) called Hripi occupied by a old man called Thorir, which was burned down by the lava. The Rauðhálsahraun covered vegetated land, and it is plausible that a farm existed there. The date assigned to the eruption, 960 AD, was derived base on the assumption that Thoris was one of the original settlers, most of whom arrived between 890 and 910 AD. After the eruption the land was no longer farmable and was forgotten. The story was therefore associated with Eldborg, impressive and much closer to the coast where it was still easily accessible.

A new swarm

This is not a likely location for the next Icelandic eruption. Carl, in the Mammoth guide to Icelandic volcanism (essential reading to find out about those Icelandic volcanoes you never heard off) gives it the lowest risk ranking of 1. Still, there is just a little brewing of something there.

Since August, there has been some earthquake activity in the region of Ljósufjöll. The swarm is located southeast of Rauðhálsahraun, between the lakes called Haleiksvatn and Grjótárvatn. This is still well within the eastern zone, in the region where the eruptions have been widespread although generally small. The earthquakes have reached magnitude 2 to 2.7, and are at a depth of around 15 km.

The activity is not new: it seems to have started in May 2021 when it culminated with a M3 event in July that year, before diminishing. In August this year the activity picked up again, increasing to some 100 registered earthquakes during October. As of this moment (17 Dec 2024), there have been 43 earthquakes in the past 7 days.

The activity seems to be on a deep fault. There is no indication that magma is involved. But in this region, the faults are an intrinsic part of the volcanic activity: the magma uses these faults to find a way to the surface. Although the activity is tectonic, the duration suggests continuing stress that is not being resolved easily.

We do not know what the precursor activity to an eruption in this region will look like. Because of the lack of shallow magma chambers, it is expected that deep earthquakes will play a role, especially if long-lasting. The build-up may be intermittent, and may last months to years or even decades. We just don’t know what to expect.

If an eruption were to occur, we can expect it be minor and tourist-friendly, although small phreatic explosions are not unlikely so the tourists should not get too close for comfort. Luckily, although the region is accessible, it is quite a walk from the main road.

But this gets ahead do the facts. So far this is just a long-lasting, minor earthquake swarm. It is just another region to keep an eye on.

Albert, December 2024

The post has made use of the information in the Catalogue of Icelandic Volcanoes, and the paper by Haukur Johanneson, 1977. Náttúrufræðingurinn, 47, 129-141, kindly translated (approximately, at least) with help of giggle.