Cumbre Vieja and the San Juan eruption of 1949

The volcano Cumbre Vieja in the island of La Palma has been showing signs of unrest. The question on everyone’s mind is, will there be an eruption? Maybe, or maybe not. This is always hard to know.

The Spanish National Geographic Institute reports inflation, a total of 10 cm of deformation. As such it is evident that there is magma on the move under Cumbre Vieja, it has intruded underneath the volcano.

There have been multiple swarms of earthquakes since 2017 in Cumbre Vieja, a total of nine. Previous swarms were probably magma intrusions too, but which did not reach the surface. The recent swarm however is more shallow and more intense than its predecessors which raises the possibility that the outcome may be different.

The current swarm started on September 12. A total of 4530 earthquakes have been detected at depths of mainly around 10 km, although there are a few which have been very shallow. The swarm commenced under the summit of Cumbre Vieja, where a magma conduit probably exists which is supplying the intrusion. Earthquakes have propagated to the northwest. This probably represents the propagation of magma filled fractures, possibly sills, radially from the centre of Cumbre Vieja. However the earthquakes only show a but a blur of what is going on down there. The exact shape and pathways used by the intrusions cannot really be known with much precision. It is somewhat similar to the prelude to the eruption of El Hierro in 2011 which also seems to have commenced with a sill that later propagated a crack towards the seafloor.

Image from the NASA.

The location of the earthquakes suggest a possible eruption in the NW sector of Cumbre Vieja. However there is a factor of unpredictability. The exact path that the intrusion takes may or may not connect with the surface, such being difficult to know if there will or will not be an eruption . The precise location where the intrusion will breach the surface is also difficult to know. The fissure could open in the middle of a town, in a forest, or it could open underwater, which are wildly different situations with wildly different consequences.

We can know however the style that the next eruption of Cumbre Vieja will take, whenever and wherever it happens. To do this we must look at the past history of this volcano.


La Palma

La Palma is one of the Canary Islands. It was formed due to volcanic activity. The oldest rocks of the island are 3-4 million years old and belong to a submarine volcano. These submarine lavas are now found at heights of up to 1500 meters above sea level in the Barranco de las Angustias, in the old northern part of the island, which shows the enormous uplift that the island has undergone. Probably numerous sill intrusions have pushed the volcano upwards.

La Palma. From the NASA.

La Palma is shaped like an arrowhead. The northern part is formed by the old Taburiente volcano. Deep gullies dissect the ancient lava flows exposing the overlapping layers of volcanic extrusions and the frozen dykes and sills which cut through them. The volcanic edifice was destroyed by a series of giant landslides, the last of which took place around  560,000 years ago. Activity continued inside the landslide scarp until 530,000 years ago. Afterwards activity died out in the northern half of the island.

Large scarp formed due to erosion of Taburiente volcano. Some sills and dykes are visible on the left. From Wikimedia, by Zyance.

Volcanic activity in the southern half of the island has been ongoing for at least 125,000 years and has constructed another volcano known as Cumbre Vieja, or also simply as Dorsal Sur, “Southern Ridge”. It is a shaped like a ridge in a N-S direction. Despite being different edifices it seems that Cumbre Vieja is part of Taburiente’s structure. Taburiente had 5 subtle radial rifts. This is much better appreciated in submarine shield volcanoes which are often shaped like ridges or like three to six-pointed stars. Knowing well the shape of submarine volcanoes I can see that Taburiente displays the same five-pointed star structure, although being subaerial it is not so easily visible. The longest, dominant rift goes southward, known as Cumbre Nueva. It can be seen that Cumbre Vieja is the southern continuation of Cumbre Nueva.

Topography of La Palma. Note the northern volcano Taburiente which is cut by deep gullies and a central erosional crater, and the southern volcano Cumbre Vieja which is dotted by young volcanic cones. There is a bow-shaped ridge connecting both volcanoes, this is Cumbre Nueva, the ancient rift zone of Taburiente, partly destroyed by a landslide. From


The main magma erupted in La Palma, as well as in the Canary Islands, is basanite, which is relatively fluid, but not as much as say Hawaii. The fluidity is comparable to the more frequently active Mount Etna in Sicily. The magma is not so fluid that all of it would flow away upon landing on the surface, but it is not so viscous that it is entirely blasted into light pumice and ash carried away by the wind. The eruption style is known as “violent strombolian” or “violent hawaiian” depending on whether it produces explosions or sustained fountains. It is the middle ground between the blazing rivers of lava and the billowing columns of ash. This style is ideal for producing pyroclastic material that rains around the fountain, rapidly constructing a mountain around the vent, known as a scoria cone. These conical mounds of ejecta are everywhere over Cumbre Vieja. Because the volcano doesn’t have any central vent that erupts repeatedly, then it makes a new fissure each time it erupts. The pyroclastic material rapidly oxidices. This gives the terrain various hues ranging from black to red, which together with the abundant canarian pine trees gives the characteristic landscape of Cumbre Vieja.

Desertic volcanic landscape near the southern point of the island. San Antonio volcano visible in the centre formed in the 1677 eruption, The brighter cone to the left of San Antonio is Teneguia, which formed in 1971. From Wikimedia by Tony Hisgett.

Other magma types present in Cumbre Vieja are the tephrite and phonolite groups which are more silicic and viscous. They are  present in trace amounts making small lava domes. A small volume of phonolite was emitted in 1585 producing tiny cryptodomes and domes, although the eruption was mainly basanitic.

Types of volcanic rocks depending on silica and alkali content. From Wikimedia by Woudloper.

Cumbre Vieja last erupted in 1971, 1949, 1712, 1677, 1646, and 1585. It is the most active volcano in the Canary Islands. Eruptions have taken place at intervals of 20-60 years. The exception being the remarkable 237 years long dormancy between 1712 and 1949. Why did this happen? It is possible that the volcano follows cycles of more frequent eruptions separated by long dormancies. Another possibility is that the enormous 6-year long eruption of nearby Lanzarote Island, occurring in 1730, induced a long dormancy in Cumbre Vieja.

It would not be unexpected that now, 50 years after the last eruption, there was a new one.

The eruption of 1949

The eruption that took place in 1949 is an interesting example of a typical Cumbre Vieja eruption.

Swarms of earthquakes had been frequent since 1936 and leading until the eruption. The morning of June 24 some fumes were noticed, and soon afterwards a towering black column of ash was rising hundreds of meters, if not more, into the sky. A new volcano had formed along the crest of Cumbre Vieja. The fissure had opened a small distance north of the highest point of the ridge. The vent is known as Duraznero.

During the following days Duraznero continued to erupt, belching out ash and rocks. Earthquakes frequently rocked the nearby communities and steaming fractures opened in the  ground around Duraznero. Magma must have been making its way into growing fractures. Over the days the erupting fissure progressively grew to a length of 500 meters and developed 5 main vents, of which Duraznero 2, at the southern end, was the most active, creating a 170 meter-wide crater. The activity was entirely explosive but of a low intensity that must have been little more than a slight annoyance to the local population. The erupted lava was tephrite. Earthquakes were more impactful, they damaged houses, cracked roads, and occasioned rockfalls. On July 6 the ash was carried downwind over the island of Tenerife where it wrapped around the summit of El Teide in a menacing black cloud.

Fissures of Duraznero. Image by KrisNM.

On July 8 a stream of lava came out from a new location known as Llano del Banco, 3 kilometres north of Duraznero, and from the other end of a system of cracks that had opened up. It did so quietly with no explosive activity whatsoever. The lava must have been degassed by Duraznero, gone into cracks, and found an outlet at a lower elevation from Llano del Banco. The lava erupted was tephrite, same as that of the earlier phase of the eruption. The initial fissure died out at about the time the new vent opened.

It is common for eruptions of Cumbre Vieja to have some vents which are dominantly explosive while others are effusive. In the eruptions of 1646, 1677 and 1712 it also happened that the vents which opened at the highest elevations had explosive activity and built large cones of scoria, while other fissures opened at lower elevations, sometimes offrift, and even at sea level, producing solely lava flows. The eruption of 1949 shows how the process works. A vent that is high up degasses the magma and then it is carried laterally through fractures towards openings downslope from which it emerges effusively.

Lava descended in fiery tongues from Llano del Banco down the flanks of the mountain. People were being evacuated as the flow headed for populated areas. It took 10 hours for the lava to reach the main road of the south of La Palma. Later that day the flow had destroyed 20 structures, including houses, cellars, and barns.

On July 10 lava cascaded over a cliff into the ocean. From this day on the entry of lava into the sea became continuous, and a lava delta was gradually constructed. Cloud of steams rising over the waters were illuminated by the convoluted streams of incandescent rock.

A new change in the eruption took place on July 12. The composition of lava erupted from Llano del Banco changed from tephrite to basanite. It became less silicic. At a similar time a new vent opened 400 meters north of the initial vent of Duraznero in the location known as Hoyo Negro. Black cauliflowers of ash pierced with flashes of lightning rose rhythmically from the Hoyo Negro vent. It erupted various magma types including basanites, tephri-phonolites and phono-tephrites. Once again the vent uprift was explosive while the vent downrift was effusive. The basanitic magmas must have released their gas into the explosions of Hoyo Negro and then come out laterally through the opening in Llano del Banco.

Hoyo Negro projected bombs to a distance of 1 kilometre from the vent snapping the trees and setting portions of the pine forest on fire. Clouds of ash frequently dusted the western part of the island. The explosions excavated a 400-meter wide crater on sloping ground. This created a spectacular 200-metre cliff against the higher side of the slope, which exposed the many layers of ejecta painted in a variety of colours.

Hoyo Negro. Image by Rafael Medina.

A raging stream of lava continued to issue from Llano del Banco and cascade towards the coast. Despite erupting continuously for 18 days the vent produced no distinguishable ejecta, and shows how the gas had been entirely removed from the melt before erupting. The ground above the fissure collapsed among loud noises, the rocks fell into the stream and were carried away, a length of 150 meters of rock above the conduit was eroded away and disappeared leaving behind a deep chasm in the forest.

On July 22 the activity of Hoyo Negro was down to a solfatara. Llano del Banco was also dying down. By July 26 the eruption had fully stopped.

Early on the morning of July 30 the eruption suddenly resumed. Duraznero and Hoyo Negro exploded simultaneously. An hour later fluid basanite lavas emerged from the location of Duraznero 1 and poured into an old crater where it formed a lava lake which then overflowed and formed a narrow stream of lava that rapidly sped down the steep slopes of Cumbre Vieja, cutting the road of Santa Cruz de la Palma, and nearly reaching the sea after 11 hours of advance, when the eruption came to a stop. This was the last episode of the 1949 eruption.

The flow of July 30, although of rapid advance, it was fed at a rate of only 10 m3/s, which is very low. It was also similar to the mean eruption rate of Llano del Banco, which was approximately 14 m3/s. The explosive activity was of little volume so it probably does not change the overall numbers too much. As such the eruption of 1949 was of very low intensity, in both its effusive and explosive counterparts. Slow eruptions are typical of the Canary Islands. Such low intensity eruptions do not pose much of a hazard to the people, in fact no one died in the 1949 eruption, despite 120 houses or so being destroyed, and people having approached the eruption in order to view it. This doesn’t mean that the hazard is inexistent.

If someone stands very close to the vents he/she could be asphyxiated by the noxious gasses or may be impacted by a lava bomb or by lightning. Rarely when lava flows reach steep slopes they collapse into blistering landslides resembling small-scale pyroclastic flows that could potentially kill someone. Conditions around volcanic eruptions can change suddenly in unpredictable ways and become hostile to humans. Safety is not guaranteed.

If Cumbre Vieja erupts in the future it will probably resemble the 1949 eruption in many ways: an earthquake prelude to the eruption that may deal damage to structures, unpredictable opening of fissures, some vents producing mainly explosive activity while others feeding mainly streams of lava that destroy human properties, and also the likely entry of lava into the sea.

Of course if the current earthquake swarm will culminate in an eruption or not cannot be known for sure. Swarms before the 1949 eruption occurred as early as 1936 and did not culminate in eruption until 13 years later.


San Martin volcano, formed in a 1646 eruption. Image by Rafael Medina.


Interesting links

Eruption of 1949 (in spanish).

IGN news (spanish).

GRAFCAN visor (includes geologic and topographic maps).


1,260 thoughts on “Cumbre Vieja and the San Juan eruption of 1949

  1. La Palma’s “Ragnarolo”…
    Any ideas on the height of the higher lava fountains, please? (Pref in feet as I’m an old Brit). I’m guessing on 600+ feet.

    • I have seen trajectory calculations of lava bombs, 10 secs = 1608 ft and 14 secs = 3,153 ft and a National Geographic article said 5,000 ft. (which I would take with a grain of salt)

    • Not sure I’d put my feet anywhere near glowing lava.

      (Sorry, couldn’t resist. The opportunity was too good.)

      Looks like about 600-1000, I’d say, depending on the vigour of the fountain.

      • Try double that, there are a lot of people timing the lava bombs falling and many are getting heights over 500 meters. Teneguia had fountains of such intensity, it is basically a strombolian eruption but continuous.

        According to GVP Teneguia took nearly 2 weeks before there is any mention of fluid lava in the bulletin reports and that 2 weeks was characterized by activity very similar to what we see now. So it coudl be some time before the lava field really goes, and before lava reaches the ocean. On the other hand I expect we will get quite a massive cone, it is already sizable after just a few days, it is on a slope but must be getting close to 100 meters tall.

  2. Any guesses on how high the lava bombs are going
    when they make it to the top of the screne on the live stream ..

  3. Hi Clive ..

    I didn’t see your post ..
    But we were both wondering the same thing :o)

  4. Any speculation on what’s causing these longer lulls at Fagradalshraun lately? Clearly it wasn’t a one-off caused by something partly blocking the conduit …

    • Probably at some depth under the cone (unknown but probably within a km for the upper end) there has formed a magma chamber that has to pressurize over that time now. Early on it was fountaining every 10 minutes because there was no chamber, over time as it has expanded so episodes get bigger in volume and further apart in duration. I woudl guess that magma chamber has a connection to the lava filling Geldingadalir too, which is why we saw vents form there. It might even be that lava is just part of the chamber now, in which case over time the area migth just cave in and reforms into a big disorganised field of vents or even a small caldera type structure.

      Pu’u O’o used to go a month between fountains, and the fountains would only last a day, like a mini flood basalt really. Etna is much the same, ranging from hours up to years between fountains, but all the craters are semi-continuously active. I expect Fagradalshraun will end up like this too, eventually, but it might take at least several years for that, Pu’u O’o already had a big magma chamber to feed from before it even formed, and there was also abundant magma storage in the ERZ, combined to make a real monster of a lava geyser, there is nothing like that yet at Fagradalshraun.

      • Interesting on the possibility of a magma chamber.

        As far as I know (so I may well be wrong) we’ve not seen a magma chamber form in the modern era. So, could what we’re seeing be how they do so (in some cases, anyway)?

        My own (admittedly unsubstantiated) theory is that the pattern we’re seeing on the drumplots indicates ongoing (though intermittent) activity over the last couple of days. That, coupled with a few persistent bright spots on the slope down to Natthagi makes me wonder if there’s occasional lava injections under the area we saw the new vents. It wouldn’t be easy to see on siesmo; no outgassing that deep due to pressure at that depth.

        As I recall, many thought that’s what happened the last time there was a “hiatus”; it was a silent and sneaky eruption.

        The one clue I do look for is moss fires; those show a rise in lava field level. Unfortunately, with all the rain lately, that’s not a very good indicator even if the level is rising.

        I suspect that the volcano is up to something sneaky this time too. Either that, or it’s just waiting for a sufficient quantity of experts to declare the eruption over. Or, maybe it doesn’t want to share the spotlight with La Palma. 🙂

        • I think personally it will take something pretty significant to close the open conduit. Kilauea and Ambrym were terminated by rifting. I dont know but shields in Iceland are usually also in rifts, would seem possible they end this way too. The fact all the large shields I have seen have collapse craters at their summit is telling.

          On Reykjanes it could be a rifting event at Krysuvik or Svartsengi that kills off Fagradalshraun, they are likely all connected, Krysuvik just being the location of a magma chamber allowing for fast eruptions. So potentially if we get a magma chamber reaching a larger size over time then we could see eruptions of similar intensity at Fagradalshraun, possibly even resulting in its termination. Perhaps this sequence took place at all the others in the early Holocene, and Fagradalsfjall is just catching up.

    • “RE: “Live Webcam .. and fountains are very tall now”

      I’m confused as to what we are watching. This image and angle of view, from the north, appears to be the vent of interest during the first 24 hours of the proceedings. Then a second, lower down the slope opened and gained interest, particular with respect to its lava flow. The angle of view coming from the south. Am I correct in this observation?

  5. Seriously? CLEARLY VISIBLE shock waves and that typical high pitched detonation sound (at least from the first boom @22:12 local time)
    Did they forget to evacuate their dynamite storage hut from there?

    Seriously, this is supposed to be only gases under pressure in the magma. No chemical reactions.
    How would this be able to make such distinct noises?
    And yet it does. And it obviously didn’t for the first time, talk Krakatoa 1883.
    I’m slightly puzzled…

    • If the pressure is high enough, that can lead to very impressive bangs and there is always the possibility of phreatic explosions.

      I love this video of Mt Tavurvur. Not a particularly big explosion in the scheme of things but gives rise a to a great Wilson cloud.

      • I know that video. I really thought it was fake, no longer though 😮
        Absolutely stunning. Is there papers about that phenomenon with respect to volcanoes?

        <> –
        That means that something in these volcanoes must have moved supersonic. Could that be due to just the hot gases rapidly decompressing from say 500 bars to atmospheric pressure when the magma bursts?

    • A shockwave is just air being pushed so hard it goes supersonic. A chemical explosive creates a gas that is the density of a solid which obviously isnt stable so pushes out so fast it achieves this. A nuclear explosion is basically just heating up a piece of metal until it undergoes a BLEVE explosion but the amount of energy is so high it is in practice like putting a piece of the core of the Sun on the ground and letting it go…

      Volcanoes are also BLEVE explosions, the majority of the power is from the dissolved water even in dry eruptions, except in very rare cases of rapid mantle ascent where CO2 can do this aswell. If you put wet rocks in a fire they can explode, now make it 1 million times as big.


  6. This video, just published but from two days ago, shows considerable ash build up nearer the vent. Some of those roofs must be close to collapse. I also shows a secondary vent, I think.

    • The lava flows are still moving?
      With this kind of Viscous lava and high fountains .. You gets very viscous flows

      So far no fluid fresh Basanite at all have emerged .. Only old basanite or perhaps its even Tephrite thats erupting now?

      • Not sure if these vents are still going but the first lava erupted at Llano del Banco was viscous too, maybe not too different from this. It seems the change is very sudden, and maybe marked by change in the eruption style too. The fluid lava took a while to show up in 1949 and 1971, it is a waiting game.

        It is a bit like at Kilauea in 2018, the question if it would erupt new lava. Many had been sceptical but it was pretty obvious, it seems the same here. It will be quite a shock I expect, the massive blow torch fountain might die down but then fast flowing pahoehoe rushes down. It would not be expected so long after the start.

    • Cheers Andrew great video – of course not so great for home owners.

  7. Bushcraft Bear has done a little update from his trip into Los Llanos to shop. Shows nicely how people are getting on with their lowlife so close to the volcano, but also how much ash is about. Also has a couple of nice distance shots of the volcano and it’s mini column / lava fountains.

    I like his videos because he shows us what is going on, not just a fixed shot.

  8. 390 buildings destroyed (+40 compared to 22 September)
    The lava flow covers 180,1 ha (+14 ha in 11h)
    14 km of roads destroyed

  9. I have never heard so much constant noise since yesterday from this Volcano . It has sounded like pulsations as if its a heart beating and now the sound is different like very large breaths its breathing out or waves crashing on a shore.

    They seem to be only seconds between each deep noise.

  10. An, um, enthusiastic visualisation of the possible ocean entry.
    I don’t speak Spanish but caught enough scientific words and bits like French to conclude the narration is in the description, which Google translates well enough.

    • If it hits the ocean it will be very very steamy.. a spectacular sight .. since its not a smooth crust surface. It will be spectacular To see it crash over the 100 m tall seacliffs! huge orange hot blocks swallowed by the ocean and thick steam formimg pyrocumulus and pheratic explosions in the water

    • Gives a new meaning to the term “trapdoor volcano” with that lava generation sequence…

    • RE:”New post is up.”

      I trust that’s where we’ll be following the conversations from now on.

Comments are closed.