Flooding the Santorini Caldera

Flooding the Santorini Caldera

The flooded caldera of Santorini volcano holds many secrets, buried beneath the ash and pumice of its last great eruption. In the Late Bronze Age, about 3600 years ago, an explosive eruption several times larger that of Krakatoa, 1883, wreaked devastation across this thriving island.

View, looking North, of the northern part of the Santorini caldera. The island of Thersia (left) is separated from the island of Thera (right) by a channel, that is now thought to have formed at the end of the Late Bronze Age eruption.

View, looking North, of the northern part of the Santorini caldera. The island of Therasia (left) is separated from the island of Thera (right) by a channel, that is now thought to have formed at the end of the Late Bronze Age eruption.

A great trading port, Akrotiri, was buried under metres of pumice; preserving for future generations a snapshot of Late Cycladic urban life.

Akrotiri ruins

Bronze Age street scene – Akrotiri, Santorini. Precursory earthquakes damaged the town. But there was time for the residents to clear up rubble (foreground), and rescue furniture (those are casts of bed frames in the street), before the town was buried in pumice.

Once the eruption was over and the dust had, literally, settled the geography of Santorini would have looked much like it is today: a group of three islands, forming a ring around a caldera occupied by the sea. The central ‘Kameni’ islands would not have begun to emerge for another thousand years, or more.

But when did the sea flood the caldera, and what were the consequences? It has long been suspected that the Santorini eruption triggered tsunamis, implicated in Late Bronze Age destruction along exposed coastlines around the Aegean. Observations of the Krakatoa eruption in August 1883 include harrowing accounts of the great sea waves that swept across low-lying areas of Java, Sumatra as this eruption reached a climax. These tsunami were triggered either by the convulsions of the sea-floor, as the island of Rakata collapsed into the newly-forming caldera; or, as now seems more likely, by the displacements of colossal volumes of water as the pyroclastic currents raced into the sea.

Krakatoa - before and after images.

Krakatoa – before and after: notice how much of the island of Rakata (labelled Krakatau) has disappeared in the eruption; and the large shoals of pumice that formed new islands north-west of Krakatau (Proceedings of the Royal Geographical Society).

But could tsunami have been triggered by the Santorini eruption; and how? Until now it has been thought that collapse of the caldera could have triggered a tsunami; but only if the caldera was already flooded by the sea. A growing body of geological and archaeological evidence suggests that the crater of Santorini volcano was not open to the sea, as it is today, at least at the beginning of the Late Bronze Age eruption. Large blocks of stromatolites, an algal concretion, and chicken-wire gypsum that were caught up in the eruption debris point to the presence of a muddy lake in the northern part of Santorini. Before the eruption, this would probably have been an area of hot, bubbling mud-pools, fed by volcanic heat and fluids from depth. Among the wall-paintings from Akrotiri is one, the flotilla fresco, that shows a city on a hill, with a river running by. Are these features that might have been found inside the caldera of the volcano?

Metre-wide block of stromatolite, thrown out during the LBA eruption

Metre-wide block of stromatolite, thrown out during the LBA eruption, now in the entrance to the Heliotopos hotel.

In a new paper in Nature Communications, we present some new evidence that may help to resolve some of these questions. High-resolution mapping of the seafloor shows the presence of a remarkable scour at the northern end of the deep channel that separates the islands of Therasia and Thera, below the town of Oia. This feature can be seen on older sea-floor maps; but not in such detail. Using sound-waves to look at the structures that this channel cuts through shows that the channel didn’t form until most of the material thrown out in the eruption had already been deposited. The shape of the scour will be familiar to anyone who has built dams of sand on a wet beach, and then let it fail. This shape shows that the channel formed quickly, and that the sea broke through and poured down a great waterfall into the caldera, but only after the eruption was over. Any tsunami waves that formed during the eruption must have been triggered by the entry of the pyroclastic currents into the sea.

Image of the northern 'breach' in the Santorini caldera, and the scour and channel that cut through it.

Image of the northern ‘breach’ in the Santorini caldera, and the scour and channel that cut through it. From Nomikou et al., 2016.

This new evidence, along with new ‘cosmic ray exposure’ age measurements showing that the cliffs of the northern part of Santorini are ancient and formed before the Late Bronze Age, may strengthen arguments that the pre-eruption interior of Santorini island was not filled with seawater as it is today. It doesn’t, however, have any direct connection to the events that led to the collapse of the major Late Cycladic culture on Crete, the Minoan, a few decades later. Disruption caused by the loss of a major trading hub and short-term damage caused by fallout of ash across the islands of the eastern Aegean might well have contributed to the eventual decline and fall of Minoan culture; but the later destruction of the Cretan palaces must have another cause. Earthquake, anyone?

Selected references 

Heiken G, McCoy F, 1984, Caldera development during the Minoan eruption, Thira, Cyclades, Greece. Journal of Geophysical Research 89, 8441–8462

Manning SW et al., 2006, Chronology for the Aegean Late Bronze Age 1700-1400 B.C.,  Science 312, 565-569

Nomikou P et al., 2016, Post-eruptive flooding of Santorini caldera and implications for tsunami generation, Nature Communications 7: 13332, 8 November [open access]

Novikova T et al., 2011, Modelling of tsunami generated by the giant Late Bronze Age eruption of Thera, South Aegean Sea, Greece. Geophys J Int 186, 665-680.

Paris R, 2015, Source mechanisms of volcanic tsunamis, Phil Trans Roy Soc London A 373: 20140380

Pichler H, Schiering W, 1977, The Thera eruption and Late Minoan-IB destructions on Crete, Nature 267, 819 – 822


Thanks to NERC for funding for my work on Santorini; the organisers of NEMO 2016 and Heliotopos Conferences for a recent visit to Santorini; to Christos Doumas, Lefteris Zorzos, Clairy Palyvou and Maya Efstathiou for introductions to Akrotiri and the Late Bronze Age, and Floyd McCoy for discussions.

The smallest volcanic island in the world?

The smallest volcanic island in the world?
Oshima Ko Jima on May 4, 1805

Oshima Ko Jima on May 4, 1805. Sketch by Wilhelm Tilesius.

One of the delights of talking to children of primary school age is their disarming ability to ask really simple questions that demand straightforward answers, but leave you struggling to throw your academic caution to the wind. Even with the questions of the biggest, the smallest, the oldest and the youngest there are still different ways of (over)interpreting the question, that can leave you floundering. So for those of you looking for a clear answer to ‘what is the smallest volcanic island in the world?’, here is a suggestion of an answer from the early 19th Century. In 1805, Wilhelm Gottlieb Tilesius was engaged as the physician, naturalist and draftsman on the Russian ship Nadezhda, on the first Russian circumnavigation of the world. In May 1805, they were sailing across the sea of Japan and past the Matsumae peninsula in northern Hokkaido, heading for Kamchatka, when they came across two small volcanic islands ‘Oosima and Coosima’. Tilesius described Coosima (Oshima Ko Jima) as ‘perhaps the most diminutive volcanic island in the world’, noting that it was essentially a small pointed rock, which was incessantly smoking. The island was just ‘150 fathoms’, or 270 m, tall; lacking in vegetation and made of dark blue lavas with a weatherbeaten dark-red skirt around the base. Tilesius made all of his observations from on-board ship, as it took a quick tour around the island, and describes ‘light coloured smoke’ and ‘blue sulphur flame’ from the summit crater. Tilesius’ paper suggesting that this was the world’s smallest volcanic island soon found it’s way into Charles Daubeny‘s ‘Active and Extinct Volcanoes‘, and shortly afterwards in a number of popular periodicals of the time. Whether – with an area of 1.5 square kilometers – it really is the world’s smallest volcanic island; and whether it is actually ‘active’ are both moot points; in fact, so little seems to be known about the island that the Smithsonian catalogue still lacks a photograph of it.

Dr W Tilesius, 1820, On the Volcano called by the Japanese Coosima and situated in the neighbourhood of Cape Sangar, in the archipelago of Japan, Edinburgh Philosophical Journal, Vol III, no 6, October 1820, pp. 349-358.

Living with volcanoes, and learning from the past.

Living with volcanoes, and learning from the past.

November 13th, 1985, is a date that is still etched in my memory. This was the day that the Colombian town of Armero was submerged beneath a catastrophic flood of volcanic rocks, mud and water; a lahar that had swept down from the summit of the volcano Nevado del Ruiz, erupting about 40 kilometres away. For days, terrible scenes of anguish and despair filled our television screens, as rescuers struggled desperately to help the survivors, and recover the many thousands of victims. Thirty years on, and Colombia has one of the most sophisticated national volcano monitoring systems in the world, run from a network of observatories by the Servicio Geologico Colombiano (SGC). But what of the people of Armero; the survivors, and those who still live at the foot of the restless volcano, Nevado del Ruiz?

Over the past year, researchers from the University of East Anglia and the ‘STREVA‘ project have been working with the SGC and a filmmaker, Lambda films, to collect oral histories, to explore what people recall from that fateful day, and to learn more about how people live with the volcano today. The result is three short films: beautifully shot, tremendously moving, and well worth a few minutes of your time.


The first volcanic eruption to be photographed?

The first volcanic eruption to be photographed?

In the digital era of instant communication, breaking news of volcanic eruptions usually arrive image-first. This year, spectacular eruptions of Calbuco (Chile), Fuego (Guatemala) and Etna (Italy) have all made it into the end-of-year ‘top tens‘, in glorious multicolour detail. But when was the first photograph taken that captured one instant during a volcanic eruption? And which was the first such photograph to make it into print?

One example may be the April 1872 eruption of Vesuvius, Italy. This short and destructive eruption was one of the most violent paroxysms at Vesuvius during the 19th century. The eruption was quickly documented by Luigi Palmieri – Director of the Vesuvius Observatory from 1852 – 1896. His report of the eruption contains a dramatic line drawing of Vesuvius in eruption on 26th April, which the caption implied was a sketch based on a photograph taken from Naples.

Vesuvius in eruption, April 26, 1872.

Vesuvius in eruption, April 26, 1872. Original caption ‘from a photograph taken in the neighbourhood of Naples”. (Palmieri and Mallet, 1873).

Some years later, John Wesley Judd (1881) noted that ‘on the occasion of this outburst [the 1872 eruption], the aid of instantaneous photography was first made available for obtaining a permanent record of the appearances displayed at volcanic eruptions‘. Judd published a woodcut of a photograph as Figure 5, with no further details relating to its origin; but the image is clearly of the same event and from a fairly similar location to that depicted by Palmieri.

Vesuvius, April 1872. Woodcut image.

Vesuvius, April 1872. Woodcut image, originally published as Fig. 5 in Judd (1881).

A very similar image – most likely a photograph from the same sequence seems to have later become a ‘stock’ volcano photograph; appearing as the frontispiece to Edward Hull‘s ‘Volcanoes past and present’ (1892), as Plate 1 in Bonney‘s ‘Volcanoes’ (1899), and even later as a repainted, colour plate in a popular science magazine (Thomson, 1921). 

Eruption of Vesuvius, 1872-3. Frontispiece in Hull (1892).

Eruption of Vesuvius, 1872-3. Frontispiece in Hull (1892). Original caption ‘From a photograph by Negrettti and Zambra’.

Vesuvius 1872 from 'The outline of science', Thomson (1921). Original caption

Vesuvius 1872 from ‘The outline of science’, Thomson (1921). Original caption ‘from a photograph by Negretti and Zambra’.

Both Hull and Thomson credited the photograph to ‘Negretti and Zambra‘, a company specialising in optical, photographic and meteorological instruments, and photographic materials – including lantern slides. A plausible candidate for the original photographer could be Giorgio Sommer, who ran a studio in Naples. Some of his collections of photographs of Vesuvius from this eruption can be found in archives including Luminous Lint and elsewhere. As an indication of the wider circulation of these images at the time, another similar image can be found as a glass plate in the archives of Tempest Anderson; a British opthalmologist and inveterate traveller and photographer of volcanoes in the late 19th Century. Anderson’s scientific volcano photography included documenting the aftermath of the devastating eruptions of the Soufriere, St Vincent, in 1902, some images of which were published in his 1903 illustrated book ‘Volcanic Studies’.

But are these action shots the first ‘instantaneous’ images of an explosive eruption? A quick search reveals a few albumen prints of steaming volcanoes from the latter parts of the 1860’s (including Etna in 1865, by Sommer; Nea Kameni, Santorini, Greece in 1866; and an image of Kilauea that perhaps dates from 1865). There are also other images of the April 1872 eruption, although taken from a rather different and less revealing location. So perhaps Judd was right – or do any readers have any other suggestions?

Cited references and further reading. 

Anderson, T (1903) Volcanic Studies. John Murray, London.

Bonney, TG (1899) Volcanoes: their structure and significance. John Murray, London.

Hull, E (1892) Volcanoes: past and present. Walter Scott, London.

Judd, JW (1881) Volcanoes: what they are and what they teach. Kegan Paul, London.

Palmieri, L (1873) The eruption of Vesuvius in 1872. With notes, and an introductory sketch .. by  R. Mallet. Asher and Co., London.

Thomson, JA (1921) The outline of science, George Newnes Ltd., London.

The eruptive history of Vesuvius is documented in Scandone et al., 2008, and listed in the Smithsonian Institution Global Volcanism Programme pages.

About this blog.

I am a volcanologist based in Oxford, UK, with an interest in the stories of past eruptions. My blogs tend to focus on volcanoes – contemporary, recent or ancient. There will be quite a lot of ‘historical volcanology’ in my posts over the next few months, as I am curating an exhibition on volcanoes with Oxford’s Bodleian Libraries, which will open in Spring 2017. I am delighted to have joined EGU blogs, and hope that my posts may find some interested readers!


Volcanoes of the Ethiopian Rift Valley

The great Rift Valley of Ethiopia is not only the cradle of humankind, but also the place on Earth where humans have lived with volcanoes, and exploited their resources, for the longest period of time. Perhaps as long ago as 3 Million years, early hominids began to fashion tools from the volcanic rocks from which the Rift Valley was floored, including basalt and obsidian.

IMG_7747_stitch Riftview

View into the Main Ethiopian Rift Valley, on the descent from Butajira to Ziway. Aluto volcano in the centre distance.

The Ethiopian Rift Valley is just one part of the East African Rift system – the largest active continental rift on Earth. While the Ethiopian rift hosts nearly 60 volcanoes that are thought to have erupted in the past 10,000 years, there is only very sparse information about the current status of any of its ‘active’ volcanoes. There are historical records for just two or three eruptions along the MER: 1631 (Dama Ali), and ca. 1820 (Fantale) and (Kone). In contrast, the Afar segment of the rift includes one volcano known to have been in eruption almost continuously since 1873 (Erta Ale), and several other volcanoes that have had major recent or historical eruptions (Dubbi 1400 and 1861; Dabbahu, 2005; the Manda Hararo Rift, 2007, 2009; Dallafilla, 2008, and Nabro, 2011). So are the volcanoes of the MER simply declining into old age and senescence? Or do they continue to pose a threat to the tens of millions of people who live and work the land across this vast region?

IMG_7836_stitch Shalla

Panorama of Lake Shala, part of which fills the huge caldera of O’a volcano.

To address this question, and others, the NERC funded RiftVolc consortium is carrying out a broad-scale investigation of the past eruptive histories, present status and potential for future activity of the volcanoes of the Central Main Ethiopian Rift. This spans eleven known or suspected centres, several of which have suffered major convulsions of caldera-collapse and eruption of great sheets of ignimbrite across the rift floor in the distant past. The first challenge is to piece together the eruptive histories of these volcanoes over the past few tens of thousands of years, and this is something that starts with fieldwork designed to detect the traces of these past events in the rock record. The RiftVolc field team, led by post-doctoral researcher Karen Fontijn, and with doctoral students Keri McNamara (Bristol) and Ben Clarke (Edinburgh) and masters students Amde and Firawalin (AAU) are spending the next five weeks completing a rapid survey of the volcanic ash and pumice deposits preserved within the rift.

IMG_8378_stitch CorbettiEast

Panorama of the eastern caldera of Corbetti volcano.

The first challenge is to identify the tell-tale clues that the sequences of young rocks, soils and sediments contain volcanic deposits. Close to the volcano, we might expect an individual large explosive eruption to leave both thick and coarse deposits; but go too close to the volcano, and there may be so much volcanic material that it can become hard to identify the products of single significant eruptions, as opposed to the ‘background noise’ of smaller but more frequent eruptions.


Karen and Keri examining the young pyroclastic rocks of Corbetti volcano

Out on the flanks of the volcano and beyond, we have the vagaries of geological preservation (did the pumice or ash land somewhere where it would then stay unmodified?) and erosion and weathering (removing or modifying the evidence) to contend with. Each of these factors will depend not only on the nature of the original deposit (how thick it was; what it was made from); but also on the environment in which the deposit formed (on a lake bed? the open savannah?  a forest? On a slope, or not?), and on the subsequent history of that environment (did the lake dry out, or continue to fill with sediment? Did the pumice become stabilised in the grass land; or did it get blown or washed away? How quickly did the soil and vegetation recover after the eruption? How deeply has weathering penetrated in the intervening millennia since the eruption?). Lots of questions to ponder!


Shelly fossils from an ancient lake deposit, interbedded with pumice layers from Aluto volcano

Our long-term goal is to better understand what sort of hazards the Rift’s volcanoes pose to those who live on and around them. There are, of course, much greater immediate challenges to communities in the region linked to the competition for the natural resources (water, land) in this region; but the rapid development of geothermal prospects in the Rift does mean that we need to pay closer attention to the state of the volcanoes that are the source of the geothermal heat.


Fresh road cut section through an obsidian lava dome and drape of pyroclastic rocks, on the way to Urji volcano and Corbetti’s new geothermal power plant

Aside from the volcanoes, the main Ethiopian Rift and its lakes make for a spectacular environment to work in. Despite receding lake levels and failing rains this year, there are vibrant patches of forest and a host of exotic birds and animals to enjoy.


Pair of little bee eaters, Lake Awassa


Flock of flamingo, Lake Chitu


Marabou stork, Lake Ziway





Camels eating Prickly Pear cactus, Corbetti


Ethiopian Fish Eagle, Lake Ziway

Acknowledgements. RiftVolc is a NERC-funded collaborative research project. Many thanks to our Ethiopian collaborators at the Addis Ababa University School of Earth Sciences and the Institute of Geophysics, Space Science and Astronomy (IGSSA) for hosting us and facilitating the joint field campaign; and to Ethioder for providing field vehicles and excellent drivers.

The Mexico City earthquake, 19 September 1985

As a volcanologist based in the UK, I am in the privileged position of rarely being affected by the natural events that I study. And, although I have worked for extended periods of time in earthquake-prone regions, I have never experienced anything more than the gentle nudge of a small tremor.

Thirty years ago, shortly after 7 am local time on 19 September 1985, Mexico City was struck by a large earthquake. The epicentre was near the Pacific coast, and the main event had a magnitude of 8 – just a little less than the September 2015 Illapel earthquake in Chile. Listening in the UK to BBC reports, it soon became clear that there was a terrible disaster unfolding – even though the conurbation was hundreds of kilometres from the epicentre, there was considerable damage to hundreds of buildings across the city. Later analysis revealed an unusual feature of the damage – most of the buildings that collapsed were 5 – 20 stories high; while the majority of both shorter, and taller, buildings were undamaged. This feature arose from the local geology – much of the city is built across old lake deposits. The layer of lake sediments acted to amplify the ground motion as the seismic waves passed by, leading to strong accelerations, and a characteristic period of around 2 seconds. Buildings with a natural period of vibration close to 2 seconds would have begun to resonate, and eventually fail catastrophically, as the strong shaking continued for several minutes. The result of this was a shocking pattern across the city of collapsed buildings cheek by jowl with others that showed no signs of damage.

Among the many thousands who lost their lives were several British language students, who had just arrived in Mexico City on their way to their ‘year abroad’ teaching assignments across Latin America. One, Helen Cawthray, was a linguist at St Catharine’s College, Cambridge; a contemporary and friend of mine. Another, Susan Mell, was a linguist at St Anne’s College, Oxford, where I now teach Earth Sciences. A poignant reminder not only of the random element of disasters, but also of their global reach. As we look back on the thirty years that have passed, the question remains: have the lessons been learned? Or might it all happen again?

Simple memorial to Sue Mell, undergraduate at St Anne's College, Oxford.

Maria Graham, and a large earthquake in Chile, 1822

Maria Graham, and a large earthquake in Chile, 1822

As news comes in of another very large earthquake in Chile – the third magnitude 8 earthquake along Chile’s Pacific margin in the past six years – this is a stark reminder of the destructive potential of these extreme natural events. These days we are used to the rapid, or near-real-time diffusion of news as these events unfold – in this case, as the tsunami ran along the Chilean coast, and propagated across the Pacific ocean. First indications are that the region that ruptured during the earthquake was a large section of the subduction zone plate boundary where the Nazca tectonic plate is sliding beneath the South American plate; close to an area that previously ruptured in great earthquakes in 1943, 1906, 1880 and 1822.

Map of historical rupture zones of large Chilean earthquakes. Source: United States Geological Survey.

Map of historical rupture zones of large Chilean earthquakes. The red dots show the locations of the 16 September 2015 earthquake and early aftershocks. Source: United States Geological Survey.

One of the earliest first-hand accounts in English of a large earthquake in this part of Chile comes from the writings of Maria Graham, who was living near Valparaiso in 1822; close to the source of the great earthquake of 19th November 1822, and within the region that was most hard hit by the event. Her journal – published in 1824 – records the event in great detail; and in particular, describes the dramatic coastal uplift that occurred as an immediate consequence of the event. A version of her report was later read to the Geological Society of London, where it caused a good deal of interest and, later, controversy.

View from Maria Graham's house. Bodleian libraries, Oxford, 4° R 56 Jur.

View from Maria Graham’s house. Bodleian libraries, Oxford, 4° R 56 Jur

Excerpts from Maria Graham’s ‘Journal of a residence in Chile, during the year 1822; and a voyage from Chile to Brazil, in 1823’,  London, 1824

November 20th, 1822.

At a quarter past ten [in the evening], the house received a violent shock, with a noise like the explosion of a mine. I sat still.. until, the vibration still increasing, the chimneys fell, and I saw the walls of the house open.. We jumped down to the ground, and were scarcely there when the motion of the earth changed from a quick vibration to a rolling like that of a ship at sea. The shock lasted three minutes. Never shall I forget the horrible sensation of that night. [Back in the house] I observed that the furniture in the different rooms .. Had all been moved in the same direction, and found that direction to be north-west and south-east.

Mr Cruikshank has ridden over from old Quintero: he tells us that there are large rents along the sea shore; and during the night the sea seems to have receded in an extraordinary manner, and especially in Quintero Bay. I see from the hill, rocks above the water that never were exposed before.

On the night of the nineteenth, during the first great shock, the sea in Valparaiso bay rose suddenly, and as suddenly retired in an extraordinary manner, and in about a quarter of an hour seemed to recover its equilibrium; but the whole shore is more exposed and the rocks are about four feet higher out of the water than before.

View of Quintero Bay

View of Quintero Bay, drawn by Maria Graham. Bodleian libraries, Oxford, 4° R 56 Jur.

December 9th, 1822.

in the evening I had a pleasant walk to the beach with Lord Cochrane; we went chiefly for the purpose of tracing the effects of the earthquake along the rocks. On the beach, though it is high water, many rocks with beds of muscles remain dry, and the fish are dead; which proves that the beach is raised about four feet at the Herradura. Above these recent shells, beds of older ones may be traced at various heights along the shore; and such are found near the summits of some of the loftiest hills in Chile.

In her journal accounts, Graham went on to speculate that repeated earthquakes could be responsible for the general elevation of land, and the building of mountains, in places like the Andes; themes that were later taken up by Charles Lyell, and then Charles Darwin – who was in Chile 13 years later, where he experienced the 1835 Concepcion earthquake firsthand.


IRIS Special Event Page, Illapel – great collection of resources on the Illapel earthquake

United States Geological Survey – Illapel earthquake information

The Maria Graham Project, Nottingham Trent University

Profile of Maria Graham on TrowelBlazers

Energy Poverty and Geothermal Energy Futures

Energy Poverty and Geothermal Energy Futures

Ethiopia is one of the most impoverished nations in the world, in terms of the number of people who live without access to electricity. The World Energy Outlook reported that in 2014, 70 million people in Ethiopia, or 77% of the population, have no access to electricity. Ethiopia is also one of the more volcanically-active regions of the world, with 65 volcanoes or volcanic fields that are thought to have erupted within the past 10,000 years – though few of these volcanoes have been studied in any detail; and fewer still are closely monitored.


Geothermal power plant infrastructure at Aluto volcano, Ziway, Ethiopia.

One benefit of this abundance of young volcanoes is that the geothermal energy potential of the region is significant – offering the potential of accessible and renewable low-carbon energy. Further south, along an extension of the Great Rift Valley, Kenya is already taking steps to exploit geothermal energy, with an installed capacity by December 2014 of 340 MW and an ambition to increase this by at least an order of magnitude within the next 15 years. In Ethiopia, current capacity currently stands at around 7 MW – provided by the Aluto Langano Power Plant, which was the first operational geothermal power plant in the country. In Ethiopia, as in Kenya, there is considerable ambition to develop geothermal power further – with the several volcanic centres identified as having the potential to supply 450 – 675 MW by 2020. In a country where per-capita electric power consumption is just 52 kWh (100 times less than that in the UK), that’s a lot of new energy.


Entrance to the Ethiopian Electric Power Corporation Aluto Langano Geothermal power plant, in the centre of Aluto volcano, Main Ethiopian Rift Valley.

All of this interest in young volcanoes as potential sources of ‘clean energy’ provides a significant opportunity for geoscientists to try and find out a little bit more about their eruptive past, and their potential for future activity; and to work out where the hot fluids and gases that provide the geothermal prospect are stored within the crust.

PP systems soil - CO2 measuring equipment

Using a PP systems respiration chamber to measure the escape of CO2 from the ground surface across the volcano.

At Aluto volcano, work by Will Hutchison using imagery from an aircraft survey (to identify young faults and fractures), and a ground-based survey of where (natural) carbon dioxide is seeping out of the volcano at the present day, has helped develop a cartoon ‘model’ for this volcano. Our current view is that Aluto volcano currently leaks quite small amounts of heat and gas to the surface; mainly along long-lived fractures and faults, some of which have origins older than the volcano itself. Inside the volcano, fluids are trapped under layers of impermeable rock – perhaps two to three kilometres below the surface – where they are heated by the warm rocks of the volcanic hearth.

Hutchison et al 2015 figure


Planned drilling campaigns on Aluto, and on the neighbouring volcano and geothermal prospect, Corbetti, should eventually fill in some of the gaps in our geological knowledge; and help to transform the energy futures of some of the millions of people who live along the Ethiopian Rift valley.


Hutchison, W., T. A. Mather, D. M. Pyle, J. Biggs, G. Yirgu, 2015, Structural controls on fluid pathways in an active rift system: A case study of the Aluto volcanic complex. Geosphere, 11, 542-562, DOI:10.1130/GES01119.1 [Open Access]

Kebede, S., 2012, Geothermal Exploration and Development in Ethiopia: Status and Future Plan, in: Short Course VII on Exploration for Geothermal Resources, 14 pp.

Data Sources 

World Bank – Electric Power Consumption

World Energy Outlook, Africa

Volcán Calbuco: what do we know so far?

Around midday on April 24, 2015, the Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA’s Terra satellite acquired this natural-color image of the ash and gas plume from Calbuco volcano in southern Chile.

Image of Calbuco volcano on April 24, 2015, from NASA’s Earth Observatory. Natural colour image from the Moderate Resolution Imaging Spectroradiometer (MODIS) on the Terra satellite. The narrow plume of ash and gas is being blown to the East, away from Calbuco and towards the town of San Carlos de Bariloche.

Detailed assessments of what happened during the April 22-23 eruption of Calbuco, Chile, are now coming in from the agencies responsible for the scientific monitoring of the eruption (SERNAGEOMIN) and for the emergency response (ONEMI). The volcano is well monitored and accessible, and as a result there has been a great deal of high quality information, and imagery, made available very quickly. In addition, there is a wealth of satellite remote sensing data, which together allow us to collect up some basic statistics about the scale of the eruption. So here are some summary statistics for now:

  1. This was the first explosive eruption of Calbuco since a small eruption that lasted 4 hours on 26 August 1972. In the intervening 42 years, there was an episode of strong ‘fumarole’ emission in August 1996, but no recent signs of unrest.
  2. The eruptions of 22-23 April began with little prior warning, and both formed strong, buoyant plumes of volcanic ash that rose high into the atmosphere – captured in some of the most amazing video and timelapse footage of an eruption anywhere in the world. The first eruption started at 18:05 (local time)  on 22 April; the ash column rose to 16 km, and ejected 40 million cubic metres of ash in about 90 minutes. The second eruption began after midnight (01:00 local time, on 23 April), with an ash column that rose to 17 km, and ejected 170 million cubic metres of ash over 6 hours. Based on the volume of material erupted (0.2 cubic km), and the eruption plume height, the combined phases of the eruption can be classified as a VEI 4 event, with an eruption magnitude of 4.4 – 4.6 (depending on the assumed density of the deposits).
  3. The Calbuco eruption was the third large eruption in this region of Chile in the past 10 years; but 4 or 5 times smaller than the eruptions of Chaiten (2008) and Puyehue Cordon-Caulle (2011).
  4. At its greatest extent, the ash cloud covered an area of over 400,000 square kilometres, affecting a population of over 4 million people in Chile and Argentina (modelled using CIESIN). Ash fallout was reported from Concepcion, on the Pacific coast, to Trelew and Puerto Madryn on the Atlantic coast of Argentina.
  5. The magma involved in the eruption was a typical andesite/dacite, containing volcanic glass and crystals of plagioclase and amphibole, with minor quartz and biotite.
  6. The SO2 gas release from the eruption was substantial – around 0.2 – 0.4 Million tonnes – probably some way short of the levels needed to have a significant impact on the climate system.
  7. The eruption was accompanied by dramatic pulses of lightning (a common feature in volcanic eruptions), and easily visible from space.
  8. At least 6500 people were evacuated as a result of the activity. The nearby town of Ensenada was badly affected by thick pumice and ash deposits, and lahars pose continuing hazards in the drainages that run off Calbuco, and into nearby lakes (Llanquihue, Chapo). The eruption has strongly affected some of the salmon fisheries in the region. Downwind, air transportation has been disrupted in Chile, Argentina and Uruguay.
  9. At the time of writing, SERNAGEOMIN note that the seismic activity has diminished somewhat , but the volcano remains at a red alert.

Taking the pulse of a large volcano: Mocho-Choshuenco, Chile

Taking the pulse of a large volcano: Mocho-Choshuenco, Chile

As the recent eruptions of Calbuco and Villarrica in southern Chile have shown, the long arcs of volcanoes that stretch around the world’s subduction zones have the potential to cause widespread disruption to lives and livelihoods, with little or no warning. Fortunately, neither of these eruptions has, so far, led to any reported loss of life – but the consequences  of these eruptions for the communities living within reach of the ash plumes and beyond will continue to play out for months or years into the future.


The young cone of Mocho volcano, southern Chile, which may have erupted as recently as 1937. Mocho-Choshuenco volcano is one focus of our ongoing work in the region.

We have been working in southern Chile for a few years now, helping to extend what is known about past explosive eruptions at some of the region’s most active volcanoes. In this part of Chile, the written records of past eruptions only extend back a few hundred years – at the most – so most of our work has involved digging into the geological records of the region, to try and piece together the fragmentary stories of past eruptions. This can be slow and painstaking work, both in the field and in the laboratory, but is always exciting when things start to come together.

Field sampling on Mocho-Choshuenco volcano: deposits of the ‘Enco’ eruption.

This week, Harriet Rawson has published her first major scientific paper on the volcanic eruption history of Mocho-Choshuenco over the past 18,000 years. The 18,000 year timescale spans the volcanic activity that has taken place since the end of the last ice age; and we can be fairly confident that by visiting hundreds of sites around the volcano, we have found most of the ‘major’  eruptions, and many of the ‘moderate’ explosive eruptions from this volcano over this time period. The results of Harriet’s work are summarised in the picture below – which shows the timing, composition and sizes of eruptions through time. Just for context, the March 3 eruption of Villarrica was small (10 million cubic metres of ash, or magnitude 3 on the y axis), with a composition represented by an orange colour (so a bit like the Mocho eruptions around 4,000 years ago); while the April 22 eruption of Calbuco was moderate (210 million cubic metres of ash, or magnitude 4.5 on the y axis), and a composition in the green to pale blue range (like the eruption around 2000 years ago).


The record of explosive volcanic eruptions at Mocho-Choshuenco volcano over the past 18,000 years (from Rawson et al., 2015). The x axis shows time, as ‘thousands of years before present’, based on radiocarbon dating of flecks of charcoal preseved within the deposits. The y-axis shows the ‘size’ of the eruption, in terms of the eruption magnitude, which is a logarithmic scale of erupted mass or volume of ash and pumice. The coloured curves represent the age and erupted composition of the volcanic events that have been recognised – with the ‘peak’ of the curve showing the best estimate of the eruption age, and size. The width of the curve gives an indication of the uncertainty in the timing of the eruption. The cartoon parallel to the x-axis shows how regional climate and ice cover at the volcano are thought to have changed over the same time period.

In many ways, this work is just the start of the forensic process of understanding how this particular volcano works and of the threats it might pose for the future;  but it is also a critical piece of the jigsaw in terms of understanding the pulse of the volcanic arc, and crossing the gap between the geological past, and the volcanic present.


The wonderful ‘Salto Huilo Huilo‘ in the Huilo Huilo ecological reserve, at the foot of Mocho-Choshuenco.


This work has been funded primarily by the UK Natural Environment Research Council, and represents the outcome of many years of collaborations with colleagues from the Chilean Geological Survey, SERNAGEOMIN, with field work in the region supported by numerous colleagues and assistants, and with the support of CONAF and Reserva Huilo-Huilo.


K Fontijn et al., 2014, Late Quaternary tephrostratigraphy of southern Chile and Argentina, Quaternary Science Reviews 89, 70 – 84. [Open Access]

H Rawson et al., 2015, The frequency and magnitude of post-glacial explosive eruptions at Volcan Mocho-Choshuenco, southern Chile. Journal of Volcanology and Geothermal Research, doi:10.1016/j.jvolgeores.2015.04.003 [Open Access] Datasets available on figshare.

DM Pyle, 2015, Sizes of volcanic eruptions, Chapter 13 in Sigurdsson et al., eds, Encyclopedia of Volcanoes, 2nd edition, pp 257-264. doi:10.1016/B978-0-12-385938-9.00013-4


Get every new post on this blog delivered to your Inbox.

Join other followers: