Seeing a hill rise up in front of your eyes where there was a lake just weeks before, and seeing it happen close to a major city, that would be something. Perhaps the Neapolitans could have stayed away – after all, the people living close-by were being pummelled with pumice while the new hill grew and they had every reason to be somewhere else. But whilst the locals fled leaving their possessions behind, people from Naples flocked towards the eruption to get a better view. None of them knew about Pompeii – that town had been lost and forgotten about – and neither would they suffer its fate. But this was an experience worthy of Naples. (It was also the only eruption with a Macchiavellian connection, but that is another story.)
Naples is known for Vesuvius, a sleeping giant which shows no indication of re-awakening. Vesuvius is like the other version of Sleeping Beauty, the princess which every prince should be desperate to stay well away from. This princess breaths fire; she is like the nameless dragon which killed and was killed by Beowulf. Tolkien turned it into Smaug, sleeping on the dwarf hoard, with Bard given the role of Beowulf. Its prince needs the guile of the Prince of Machiavelli and let sleeping princess-dragons be. (Remember how Maleficent transformed into a dragon in order to stop the prince waking up the sleeping beauty Aurora? Naples is like that.)
But there is more to Naples than the dragon of Vesuvius that will inevitably one day curtail the encroaching city. To the west lies an older, and larger dragon, the ‘Ancalogon the Black’ of Naples. The town of Pozzuoli lies at her heart. These are the Phlegrean Fields – Campi Flegrei, a region that once burned Europe with her fire. The dragon sleeps – underneath the most densely populated caldera in the world. Over a million people could be affected if any prince dared to wake her.
The Phlegrean Fields form a nested caldera, meaning there have been several caldera-forming episodes which created slightly different but overlapping calderas. There were in fact two major eruptions. The older caldera dates from 40,000 years ago and was formed in the Campanian Ignimbrite eruption. It is around 12 by 16 km across, but only parts of the rim survive. The eruption was probably mid-VEI7 but the precise volume is quite uncertain. It may even have been two separate eruptions. The younger caldera is from 15,000 years ago and erupted the Neapolitan Yellow Tuff. It was smaller (low VEI-7) and only partly obliterated the older caldera.
(This history is not universally accepted. It has also been argued that the Campanian caldera was much smaller and that the current caldera is mostly due to the younger eruption. )
Sometime after the younger eruption, the caldera flooded. The original collapse had left the caldera near sea level. But the eruption happened during a time of rapid sea level rise as the ice age drew to a close, and as it rose the sea transgressed into the caldera. By 6000 years ago sea level had risen by around 100 meters. For much of this time, the caldera was a marine bay, with an estimated depth of 100-150 meters at the centre. But there was also resurgence in the caldera which over time balanced the sea level rise. While this was going on, the region alternated between wet and dry, depending on which rose up faster, the sea or the land.
During this time, the dragon of Campi Flegrei had not fully fallen asleep, or perhaps she had restless dreams. At times, a new eruption or explosion would leave its mark on the caldera. These eruptions were very much smaller than the big ones (most were VEI3 or 4) but they left the caldera pockmarked with cones and craters. There have been some 70 eruptions in the caldera since the Neapolitan collapse. These were monogenetic, meaning each came from a separate magma intrusion in a new location. At first, the eruptions occurred mainly close to the caldera rim. Many of these earlier eruptions happened below sea level. Around 8000 years ago the caldera went quiet. About 6000 years ago eruptions resumed with 20 eruptions over a period of 2500 years. The new eruptions occurred mainly around an uplifted block within the caldera. And 3500 years ago the caldera went quiet again. There is only one eruption known after this: Monte Nuovo.
The uplifted block is on-land near the coastal town of Pozzuoli. The well known sulphur field, Solfatara, is also associated with this block. Another locus of activity exists off-shore, at the centre of the Bay of Pozzuoli. Both sites, near the city and in the centre of the Bay, show tectonic earthquake activity. It is all in the faults. There is a bit of folding in the region, with one side (the anticline) near Pozzuoli and the other side (the syncline) in the Bay. The uplift has been focussed on the anticline whilst the syncline is a location for magma upwelling.
The larger region in which Campi Flegrei is embedded shows extension, in the form of a 30-km wide graben which runs from northwest to southeast. Over the past 150,000 years, there may have been as much as 750 meters of subsidence within this graben. The volcanic activity is likely related to this extension.
Ups and Downs
The Phlegrean Fields show signs of the sleeping dragon underneath. It belches not fire but steam, in the form of numerous fumaroles (especially in Solfatara, one of the eruption sites of 5000 years ago). It also moves up and down over the years, as if breathing. The ups and downs are called bradyseisms (literally: ‘slow movements’) and they are characteristic of large non-erupting calderas. Although magma may be involved, much of the height changes are caused by underground movement of heated water.
A notable episode of fast changing levels occurred between 1960 and 1990, when the town of Pozzuoli was uplifted by 3 meters. In the Netherlands, such a change (hopefully up rather than down) would have been world news! Since 1990 the town has been slowly going down and more recently up again, but nothing like the exciting days of the 1980’s. The underground water that is causing (some of) this has two origins. There is water present at a depth of 2 km which comes from the Bay of Pozzuoli, while another water reservoir at 8 km depth is much older and is associated with a magma reservoir. The 2-km layer is associated with a cap rock which is poorly permeable, so that water can collect below it. The fast episodes of the 1960’s and 1980’s were related to changes in the shallow reservoir. The slower changes over the past two decades come from the deeper one.
There is nothing unusual about the recent changes. Ground levels here have long been changeable. The Roman pillars at the old market of Pozzuoli show that the village was well above sea level in their days, 2000 years ago. But damage done to the pillars by marine organisms show that at later times they were submerged by more than 7 meters! There is one drawing from the Middle Ages showing the pillars sticking out from the sea, with the market square itself below the water line.
Hot water is of course in demand, and Campi Flegrei has plenty of that. Agnano, a large crater in the Phlegrean Fields, was known for its thermal baths already in Roman times. But this did not last: it became a crater lake in the Middle Ages because of the subsidence. The lake was drained again only in the 19th century.
There is other evidence for height changes. Out in the Bay of Pozzuoli are the remains of two Roman roads, now submerged: in Roman times, the coast line was not where it is now. Some, but not all, places recovered from the deep subsidence of the Middle Ages. There is a Roman harbour now 12 meters under water! No – no Dutch person would have build a town here.
After the Roman era, the region subsided at an average rate of 1.7 cm per year. That sounds acceptable, but it amounts to 1.7 meters per century! At times the subsidence went even faster. There were a few reversals, such as in the period 600-700, in a period of perhaps 200 years prior to the Monte Nuova eruption, and in the period since 1950. The town of Pozzuoli ended up being partly submerged midway through the period but close to its old level after the 1538 eruption. But some places along the coast line did not fare as well, as shown by the submerged Roman roads.
More detailed studies have used the Serapeum pillars mentioned above, by carbon dating the marine bore holes. When the market was first built, the floor was 13 meters above sea level. Around 200 AD, a new floor was constructed, raised by 2 meters above the previous one. Was this because of intruding sea water? Another restoration took place in 394, so at that time the market was not yet flooded. By 540 AD (from the carbon dating) the pillars were submerged by 7 meters. This indicates rapid subsidence in the 5th century, by as much as 15 cm per year. Such rapid change can only come from migration of underground water. In fact, during 1986 a similar rate of subsidence was reached in Pozzuoli, and this was attributed to loss of pressure in the water system that feeds the fumaroles of Solfatara.
Those fumaroles of Solfatara have been long known. Strabo (roughly the year zero) wrote “above the city (Pozzuoli) lies the Forum of Hephaestus, a plain shut in all round by exceedingly hot ridges, which in numerous places have fumaroles that are like chimneys and that have a rather noisome smell; and the plain is full of drifted sulphur.” There are even two reports of a volcanic explosion in Solfatara in 1198 . But these reports are from centuries later: no document around the time describing Solfatara mentions an explosion. It has been suggested these reports actually refer to the earthquakes in 1498 (during the run-up to the eruption) and that the date of 1198 is a typographical error. The 1538 eruption of Monte Nuova is therefore considered as the first eruption at Campi Flegrei for 3000 years.
Prelude to the eruption
Often the signs of an impending eruption are recognized very easily – but only in hindsight. Volcanic 20:20 foresight only comes after the fact. So it was here. The old records show indications that something had been brewing but these signs could not have been recognized as such at the time.
Recent studies indicate that slow re-emergence of land lost after the Roman period started after 1250. Around 1300, new buildings (churches, actually) appear at Pozzuoli in a place which previously was submerged. The inflation was probably ‘only’ a few meters. This period may also be when the old market square with the Roman pillars resurfaced. It was the first indication of a change at Campi Flegrei since the deflation had begun a millennium earlier.
The first direct sign of re-activation started in 1470. Between 1470 and 1472 there was an earthquake swarm, strong enough that damage to buildings in Pozzuoli is recorded. More gas was emitted by nearby Solfatara, sufficient to kill trees. After 1472, it appears things quieted down again, although a few earthquakes continued to be strong enough to cause damage.
Around 1500, another slow change was becoming obvious: the land was rising. This was clear because the shoreline was receding. New land became available for people to use. That did not go unnoticed by people in need of new houses because of the damage done to the old ones. A declaration in 1503 was made that the new land belonged to the University of Pozzuoli. Another declaration regarding ownership of new land was issued in 1511, so apparently the sea was still withdrawing further. Between 1250 and 1536, the promontory at Pozzuoli had risen by 14 meters. The coast further west had inflated by 5-8 meters.
In between these declarations, there were strong earthquakes in 1505 and 1508, causing significant damage in Pozzuoli. Again, it quieted down apart from a damaging earthquake in 1520.
Things suddenly turned for the worse in September 1536 when the earthquakes became much more frequent and more severe. In February 1537, reports state that many houses in Pozzuoli had been damaged, some severely so. People began to leave the town. At the same time Solfatara became smellier.
On April 20, 1538, an earthquake was felt in Naples strong enough to cause fear. Activity continued for the next months, and the earthquakes became progressively stronger and more frequent over the summer. This peaked on September 14, only to restart on September 20. By September 22, people began to sleep outdoors, away from the ruined houses. The shaking continued and became near continuous by September 27. In distant Naples, 5 to 10 events were felt per day. Even for the Neapolitans, this got a bit much.
On 28 September, people notice a ‘dissection’ of the sea. Just off-shore, a ridge is uplifted to 4.5 meters above sea level. It isolates a bit of water which is covered in dead fish. The remaining locals see an opportunity: the region’s famous entrepreneural spirit regarding ownership was alive already! The fish are quickly collected and taken to Naples to be sold at the market. Perhaps the fish had been killed by sulfur being released into the water. Around this time, wells in Pozzuoli (several kilometers away) begin to fill with water. The land here may have begun to deflate as the activity moves further down the coast.
During the night, the earthquakes continue. Hardly a house is left undamaged and many are close to collapse. People sleep outdoors.
Now something strange happens. In the morning of September 29, along the coast near to where the ridge had formed, over a period of a few hours, a region around 200 meters across deflates by some 4 meters. The hole begins to fill with water. Around noon the deflation reverses and now a bulge forms in the same place. It continues to grow during the afternoon. Cracks appear in the bulge, and water comes out some of them. The height of the bulge is not recorded, but the inflation must have been at least 1-2 meters per hour.
Shortly after sunset on 29 September, 1538, a vent opens around 4 kilometers west of Pozzuoli and expels flames which rise from the sea. The eruption is near the town of Tripergole. This town has been known for its hot-water baths since at least the 1300’s and possibly since Roman times: Cicero’s house was supposedly here. The town is now lost and the exact location is not known. Over the next half hour, the flames move towards the coast and to Tripergole, destroying its castle and hospital. A strong earthquake occurs and a new vent opens at the top of the bulge. This signals the start of the main eruption. The majority of the eruption will come from this vent.
“In the year 1538 on the day of San Geronimo (28 September) a great earthquake was felt in the said city [Pozzuoli] [The vent] opened to approach the castle of Tripergola and shattered everything and ruined it, and then filled it with sand, with stones and made a new mountain there in 24 hours where it can still be seen today“. Antonio Russo, inhabitant of Tripergole, 30 July 1587 (https://www.archeoflegrei.it/il-villaggio-di-tripergole/)
Such contemporary reports make it clear that the first vent was submarine. The eruption started 50 to 100 meters off the coast, presumably at the end of a dike. After the initial break-out, the rift contracted towards its source, passing through Tripergole. This initial rift is not visible now, and is presumably buried by Monte Nuova. This indicates it wasn’t very long, and also shows that the bulge was not far from the sea shore.
But where was it? The bulge had formed on the land between Lake Averno, Mount Barbaro (a crater rim 2 km east-northeast of Lake Averno) and the Bay of Pozzuoli. Tripergole was on a rise near Lake Lucrino, and the first vent was out at sea.
The map shows the current lay of the land. Lake Lucrino was well known in Roman times. It was separated from the sea by a sea wall and was used as a harbour. A canal ran to Lake Averno, and this lake at one time provided safe shelter for the Roman fleet. The outlines of the lake were changed by the new hill. The first vent was in the bay but avoided Lake Lucrino. It may have been south or southeast of Monte Nuova. That would put Tripergole underneath the southern side of Monte Nuova, and suggests the first rift extended between the summit and the sea.
At the start of the eruption, the plume rises to about 5 km high, producing smoke (both black and white), incandescent plumes, stones, a large amount of pumice and (notably) wet ash or even mud. The wetness was probably initially related to the marine eruption, and later to the pool that had formed at the location of the bulge during the few hours of inflation: it had a direct connection to a water reservoir. Boreholes have shown that the early ejecta were some 7 meters thick and were deposited mainly on the southern side of the current mountain: this may show the initial rifting from sea to shore.
The eruption causes chaos. To the south, the area is covered in the mud and large stones. A large amount of pumice collects in the Bay. Pozzuoli, to the east, also is covered in mud and smaller stones. After about an hour, Naples receives a covering, first from dry ‘sand’ and later from wet ash, at times almost liquid. A warm wind comes from the direction of Pozzuoli.
The eruption continues overnight. Pozzuoli is now covered in 25 cm of debris. Naples receives a few cm, with the fall-out becoming less wet during the night. The ash falls reach as far as 100 km to the east, blown by the westerly wind. Close to the eruption, cracks form in the ground, one of which runs to Lake Averno. Two warm springs form, one on the shore of Lake Averno and the other in between it and the main vent. The spring at the lake produces salty water, the other one fresh water. There may also have been a secondary eruption vent between Monte Nuova and Lake Averno.
During the day of Monday, Sept 30 and throughout the following night, the eruption continues with large explosions, expelling large stones and wet ash. People from Naples, always curious, try to see the new crater, but travel beyond Pozzuoli is impossible. We have several documents describing the events of this day. One of these is by Francesco del Nero, written on Sept 30. Del Nero was the brother-in-law of Niccolo Macchiavelli – but Macchiavelli’s Prince (“Everyone sees what you appear to be, few experience what you really are”) would have known better than to wake the dragon of Pozzuoli! By the end of the day, the new hill has reached a height of 140 meters. Reports describe that many birds are killed by the ash and sulphur.
On Tuesday, Oct 1, the eruption begins to grow less. For the next two days, the explosions lessen and the fall-out slowly ceases. This quietening phase is interrupted on Thursday Oct 3, when a large explosion occurs at around 3pm, dropping pumice as far away as 8km (based on a report from a ship at that distance). A new layer of 2-3 meters of grey ash is deposited on Monte Nuovo. Different from the previous ejecta, this ash is dry. Whatever water reservoir was interacting with the eruption has either run out or lost its connection to the mountain.
On Friday Oct 4, Francesco Marchesino visits the devastation in Pozzuoli. He reports “there were not ten houses…that were not either destroyed, or completely or partially ruined to the ground, and without a citizen and such was the devastation that not a single stone remained in the place where the master had placed it mason”. Half the Cathedral had collapsed and all the gardens were “covered in ash“.
October 5 and 6 continue the decline of the eruption. Naples being Naples, people flock in to see the new hill and to have a look in the mouth of the crater. The people scattered across the slopes are caught out by a sudden explosion, at 3 or 4pm in the afternoon. There are again reports of wet ash in the area, as far as Naples, but on Monte Nuovo this last layer has been found only on the southern slope where it reaches a thickness of 25 meters. This seems to have been a collapse event: the reports from Naples should be taken with caution. 24 people die, either from suffocation from the gas emissions or from the explosion itself. These are the only known fatalities of the eruption.
After this, there is no further activity. Smoke rises for a few weeks and abundant fumaroles deposit sulphur, but the eruption has now ended and the princess-dragon returned to her lair for a well-earned nap.
Although not a large eruption, it built a substantial hill. Monte Nuovo is 140 meters tall and covers a bit over 1 km2. The eruption volume was 0.05km3, a VEI-3. The eruption column may have reached around 5 km high. This would be a minor eruption in most places, especially in a location that had twice done a VEI-7. Indeed, it was a well observed curiosity but not something to panic a Neapolitan, although the locals suffered loss. The village of Tripergole was destroyed for ever and it is still buried under the mountain, but this was not seen as a great loss. It was just a part of life in Naples.
After the eruption, deflation set in. Over the next two years, Pozzuoli sank by some 6 meters while the shore near Monte Nuovo went down by a similar amount.
But it almost wasn’t the end. Trasati et al (2023, Geophysical Research Letters) present archeological evidence indicating that inflation resumed two years after the eruption. Over the next 40 years, Pozzuoli regained 1 meter in height, before deflation resumed. Models indicate that this inflation came from the same source (location and depth) as the inflation before 1536. Was magma still flowing in, and was this the onset of another eruption, but one that failed? Or was this just Campi Flegrei doing what it does best – changing elevation?
Where did the eruption come from? Models suggest that the post-1250 inflation had come from a magma reservoir located some 5 km deep in the Bay of Pozzuoli, south of the town. This may have been close to the centre of the NYT caldera. The volume of this intrusion was around 1 km3. The magma was not particularly hot, at some 840 C. In 1536, an intrusion occurred towards or under Monte Nuovo, at a depth of 3-4 km and with a volume of 0.3 km3. The rapid bulging before the eruption can be modeled as an intrusion of order 500 meters deep, of volume of 0.05km3. The eruption of wet ash suggests a significant amount of water may have been present at this depth. Studies of the crystallization of the ejected magma indicates that the ascent from 4 km to this 500-meter deep chamber occurred in only hours to days. The rapid bulging was caused by this ascent.
It should be stressed how unusual this eruption was. The area had not erupted for 3000 years. The previous eruptions had come in phases, separated by decades, but this was a stand-alone eruption, a singular event. It was also on the small side: previous eruptions had ranged from 0.01-1 km3 and this one was closer to the lower limit (more like a Jabberwock than a full-framed dragon, perhaps). The feeding magma was also rather cool. Clearly, it was a marginal eruption. If the magma hadn’t found a route to Monte Nuovo, perhaps it would have given up and failed.
And, as in uffish thought he stood,
The Jabberwock, with eyes of flame,
Came whiffling through the tulgey wood,
And burbled as it came!
(Lewis Carroll: Jabberwocky)
The location was also slightly unusual. It was close to the original caldera rim and most eruptions here took place more than 5000 years ago. (However, the most recent eruption from neighbouring Lake Averno was also one of the youngest eruptions, dated to 3700 years ago. So one could also argue that with two of the most recent eruptions located here, and both being of similar (small) size, there is a pattern developing.)
A recurrence in the near future is therefore not too likely, at least if we only look at the two eruptions in the past 4000 years. But the activity in the region since 1960, with inflation, earthquake episodes and increased emissions at Solfatara may be a warning sign. Based on 1538, one could argue that an eruption in the next 300 years is not implausible. If the run-up to the 1538 eruption is typical, we would get much more warning than we have had so far, though, so anything in the next few years would be less likely. But we don’t know how typical this was.
Should we prepare for an eruption? (“How to train your dragon“?) The question becomes, for what precisely? Large or small? Where? During the unrest of the 1980’s, the population of Pozzuoli (72,000 people) was evacuated – but to a place also inside the Campi Flegrei caldera! That attitude to safety too is Naples.
We can probably assume that if something is brewing, there will be a few years of notice. The best thing to do is to already write the policy on how to handle such warning signs. Once the event begins, there will be too many competing interests to do anything. It is as with any crisis: action taken before it begins is much easier to implement. Once a crisis is developing, the main call will be to delay action. That is human nature: ask Machiavelli. (“How we live is so different from how we ought to live that he who studies what ought to be done rather than what is done will learn the way to his downfall rather than to his preservation.” ― Niccolo Machiavelli, The Prince.)
But remember that over 150,000 people live inside the caldera and a million people live within reach. If a Monte Nuovo eruption were to recur, the eruption would directly affect over 100,000 people and bury the houses of perhaps 10,000 of them. There is no such thing as a minor eruption inside a major city.
And there is one lesson everyone should take to heart. Don’t buy sea front property in Pozzuoli.
“Even dragons have their endings” (Tolkien, The Hobbit).
The best way to end is by re-telling Henrik’s vision of a future Campi Flegrei eruption, taken from https://www.volcanocafe.org/devil-may-care-campi-flegrei-ndvp-3/
It is a fine spring evening and tonight, Napoli are hosting AS Milan, a contest that will have a great bearing on the eventual destination of the Scudetto. It is also the run-up to the local elections and tonight, the Mayor entertains a very important gathering of political backers. Over at the Osservatorio Vesuviano INGV, the seismologist on duty is much alarmed by the series of earthquakes , apparently centred near Pianura, 10 km from the city centre and close to the margin of both the CI and NYT calderas. He calls the Direttore and by the time she arrives, the instruments indicate that the ground has begun to inflate, possibly as much as 50 cm as the instruments located at Solfatara, more than 4 km away, indicate 12 cm. The direttore immediately calls the Mayor but is told that he is unavailable. It is not until nearly midnight before she gets through to him.
Over at Stadio San Paolo, no-one has noticed anything as the greatest quakes were no more than M3.3 in strength and in any case, the Neapolitans are too exited. By the time the game ends, Napoli have won three goals to one and the streets are filled with a jubilant throng. Barring an improbable upset away at Siena, the Scudetto will return to Napoli!
Signor Sindaco is not best pleased by the call from Signora Direttori. He too, as well as his backers, is caught up in the jubilant mood and he has no time for unwanted news – it’s time to turn sporting success into political coin – and in any case, if we are to be honest, he is not fit to deal with serious matters effectively at present. He hangs up on the Direttore with the promise of a meeting at 10 a.m. next morning.
During the night, INGV staff have established that the ground is indeed inflating rapidly over a large area centred on the suburb of Pianura, that it inflates at a rate of almost 50 cm per hour which seems to be accelerating and that by four a.m., the ground has already risen more than four metres and it looks as if an eruption may be imminent. Armed with irrefutable data, the Direttore finally manages to get some response and together with the Chief Commissioner decides to begin an evacuation on their own initiative.
At 7 a.m., inflation stands at almost seven metres. Evacuation of the Pianura and Astroni suburbs has begun but it collides with the morning rush. In spite of the efforts of the Carabinieri, by 8 a.m. the roads in and out of Naples are completely congested. The Mayor, woken at five, has made a public appearance and urged calm as well as full compliance with the directions given by the Carabinieri, but few Neapolitans hear this. Most are on a high from last night’s victorious game against AS Milan and furious at the congestion of the traffic.
At 8:53 a.m. there is a series of large earthquakes that most people feel and three minutes later, an ominous cloud rises over where Parco Attianese used to be, one that resembles an acacia tree but is dark, grey, billowing and from which lightning bolts streak towards the ground. It is a cloud that rises higher and higher in the air as well as spreads rapidly in every direction with great walls of darkness through which a glow of lightning can occasionally be seen.
A description of the eruption can be found at
Details of the eruption used in this post are described in De Vito et al. Bulletin of Volcanology (1989), 49, 608 and in Guioboni and Ciucarelli Bulletin of Volcanology (2010) 73, 655
The magma system is described in Di Vito, et al., Sci Rep (2016) 6, 32245
Albert, October 2023