Italy's Volcanoes: The Cradle of Volcanology

Stromboli home page

Eruptive history Volcanic hazards References
Eruptive characteristics Geophysical monitoring Web sites
Latest news    

Volcanic hazards at Vesuvio



For a surprisingly long time, little thought was wasted for the hazards aspect of Vesuvio. One of the first comprehensive papers on volcanic hazards at Vesuvio is that by Barberi et al. (1983), encompassing hazards from lava flows, tephra falls, and glowing avalanches. Recent studies of the volcano and its products in the light of modern volcanology have led to a rewriting of the eruptive history of the volcano and reinterpretation of some of its most notable eruptions. It was only after these studies, in the early 1990's, that papers with alarming contents based on computer simulations were published in widely-distributed journals. One of the first of these was a paper in Nature by Barberi et al. (1990), considering the risk of tephra fallout in the case of a new eruption.

Recently, the hazard from ash plumes to aircraft at Vesuvio have been discussed by Macedonio et al. (1994). This special kind of hazards has gained much attention during the past ±15 years, after a series of near-fatal encounters of aircraft with airborne volcanic ash in Indonesia and Alaska, and its assessment for Vesuvio is certainly an important issue in considering a future eruption.

Pyroclastic flow hazards at Vesuvio were recognized first in the 1970's although as early as 1918, a courageous but overheard archeologist voiced assumptions that these phenomena might have occurred during the AD 79 eruption (Merrill 1918, 1920). Barberi et al. (1983) first considered this special kind of hazard and published maps of the areas at risk from pyroclastic flows and surges. However, it was only three years after the 1990 Nature paper by Barberi and his coworkers on fallout risk that computer simulations and reviews of the pyroclastic flow hazard were published in international journals (Scandone et al.1993b, Dobran et al. 1994). Since then, public discussion about the dangers of Vesuvio in Italy is vigorous, and the subject receives high media attention finally. However, everybody talks about corruption in Italy, but the country is far from getting that problem solved. In the case of Vesuvio, the backlog seems even more grave.

One of the major problems that volcanologists and local authorities have to deal with is the vast uncertainty concerning the time, type and magnitude of future activity of Vesuvio. Even though the volcano is being studied like never before and its eruptive history is being reconstructed in more and more detail, it is extremely difficult to tell what it will do next. It is clear that it has ended its most recent (1631-1944) cycle of activity, and it seems logical that the next eruption, whenever it will occur, will be the largest one since 1631.
There is virtually no hint to how long the current repose period will continue, a crucial factor for the magnitude of the next eruption. If the volcano is to maintain its past behavior, many centuries may pass before it becomes active again. An eruption after that will be inevitably devastating, both for its magnitude and for the presence of still more people around the volcano (if it is assumed that the population of the area will continue to grow at its present rate).


Following the examples mentioned above, the effects of all types of activity to be expected from Vesuvio must be considered. This is, from a) low-level Strombolian-Hawaiian (1895-99 and 1929 type) over b) Vulcanian (1906 and 1944 type) and c) sub-Plinian (AD 472 and 1631 type) to d) violent Plinian (AD 79 type) eruptions.

A) Low-level, mainly effusive summit crater activity
The first type of activity is the least probable after the long repose period since 1944. Should it nonetheless occur, it would primarily affect the summit and the active cone (assuming that the activity begins within the present summit crater). The first destruction would occur as soon as lava or pyroclastics reach the tourist facilities and other buildings on the crater rim. In the case of lava outflow on the flanks of the cone, the road and funicular leading to the crater would be blocked and destroyed. More vigorous outflow of lava such as occurred in 1929 would reach the base of the cone and flow even beyond, endangering the numerous isolated buildings on the lower flanks of the volcano as well as the uppermost parts of the towns around it. The fall of pyroclastics would not be a significant hazard factor except in the immediate vicinity of the crater.

B) Vulcanian-type, effusive-explosive eruptions
This is the most probable type of activity to be expected now, after several decades of inactivity, although its size should be greater than that of any eruption after 1631. Eruptions of a comparable magnitude occurred most recently in 1906 and 1944, causing widespread destruction and numerous deaths. The main destructive factors were lava flows (on the S flank in 1906 and on the NW flank in 1944) and tephra falls (in the N to NE sector in 1906 and on the E side in 1944), the latter accounting to most of the deaths (up to 500 in 1906 and about 25 in 1944).
It is not clear whether lava flows will occur in the next eruption. If so, they will very likely cause destruction due to the dense urbanization of the entire Vesuvian area. The only area protected from lava flows is the sector encompassing the NNW to ENE sides of the Soma-Vesuvio edifice, an area protected by the rim of the Somma caldera. The areas of destruction will strongly depend on whether lava erupts from the summit or from lateral vents, and on which side and elevation such lateral vents are located. In the case of an eruption after at least several decades of repose, magma effusion might occur at a high rate and produce a particularly large volume of lava. On the other hand, explosive fragmentation of all magma might prevent any effusive activity.
The most significant hazard during a medium-sized eruption is from fallout tephra. Barberi et al. (1990) presented computer modelling of the distribution of airfall tephra and probable collapse of buildings from tephra load was calculated for an eruption occurring after 45 years of inactivity. Assuming that up to 80 x 10^6 m^3 of magma had accumulated in Vesuvio's reservoir since 1944, they warned that most buildings would collapse under the weight of tephra in the belt of towns around the volcano. As far as Napoli, up to 20% of all roofs would cave in.
The heaviest damage is to be expected in the area statistically most threatened due to the prevailing wind directions, that is, the sector encompassing the N to E flanks of the volcano. Towns such as Somma Vesuviana, Ottaviano and San Giuseppe have repeatedly suffered heavy fallout from Vesuvian eruptions, mostly in 1631, 1737, 1779, 1794, 1822 and 1906. In 1906, the majority of victims died when stricken by large blocks or bombs, or when roofs caved in (more than 100 died that way in the church at San Giuseppe).
Avalanches of tephra may occur on the flanks of the active cone and extend to its base but little beyond, if the next eruption is of similar size to the 1906 and 1944 events. In the case of more vigorous eruption, larger and more mobile avalanches (pyroclastic flows) may form and extend well beyond the base of the cone, into inhabited areas. Damage from such events should be expected to be considerable, not to speak of effects for any humans staying in the zone affected by the flows.
Less serious effects of an eruption of this size would affect large areas around the volcano, disrupting all traffic and production in one of the major population centers of the Mediterranean.

C) Sub-Plinian, predominantly explosive eruptions
An eruption occurring after many decades of repose could well assume proportions similar to the devastating events of AD 472 or 1631. The effects of the latter have been described on a separate page and may serve as an example of how Vesuvio will likely behave in the future. However, the population has since had a manyfold increase and the area around Vesuvio is getting more and more industrialized. The impact of a 1631 size eruption would thus be much more dramatic. Apart from the damage it would cause to private property and agriculture, an eruption of this size would paralyze one of the major economic centers of the Mediterranean and disrupt life over large areas. The number of persons displaced or otherwise affected would range up to one million or more.
The eruption would be predominantly, if not purely, explosive, thus risk from lava flows is to be considered neglegible. The main hazards are from pyroclastic flows, surges, airfall, and lahars. In recent computer models, pyroclastic flows were calculated to arrive in populated areas only 5-7 minutes after the onset of eruption column collapse, depending on the size of the eruption (Dobran et al. 1994). Except in smaller to medium sized eruptions, these flows (at least their more dilute portion, the surges) would even be capable of surmounting the barrier of the Somma ridge, as they apparently did in the A.D. 472 eruption. The distribution of the devastating avalanches would strongly depend on the topography as the dense pyroclastic flows follow valleys and ravines. Thus, much of this material would be funnelled by the boundaries of the Somma barrier, on the NW and E sides of the active cone. In the W, SW, S and SE sectors, no significant topographic obstacles are there to halt the flows, the youthful flanks of the volcano being fairly smooth in those areas.
Widespread damage is to be expected from airfall tephra in the area of main fallout. Lahars are a major hazard during and for some time after a sub-Plinian eruption, being capable to cause disruption much farther from the volcano than pyroclastic flows and surges.
After more than 50 years of repose, this is the type of eruption most likely to be expected. Based on the recent eruptive history of Vesuvio, the present repose period may extend much longer, but given the changing behavior of the volcano, such an event may as well occur very soon. A violent explosive eruption occurred in AD 512, after only 40 years of little or no activity. Even though not much is known about that event, one contemporaneous report accurately describes pyroclastic flows damaging and burying vegetation.

D) Plinian (AD 79 type) eruptions
This is the worst type of activity to be expected, but probably only after a repose period of at least several centuries. If Vesuvio is to stay quiet for that long a period, many generations will live in its shadow without being harmed by it. The reawakening will inevitably cause a tremendous disaster, even if evacuations save the lives of the Vesuvian area residents. Eruptions of that size have ravaged the municipal area of Napoli in the past (as in the 3750 BP "Avellino" eruption, the latest Plinian event before AD 79).
Hazards from eruption of this type would be much the same as those described in the preceding paragraph but of a significantly larger scale. Thus, heavy tephra falls would occur over much greater distances and pyroclastic flows would reach farther from the volcano. Pyroclastic surges have covered all of the present urban area of Napoli attaining thicknesses of at least 0.1 m and up to 2 m during the late stage of the "Avellino" eruption. In the case of a repetition of that event, damage to the city would exceed human imagination. It would depend on timely warnings and evacuations if lives would be saved. An unexpected eruption with a size ranging between 1906 and AD 79 "Pompei" events would kill 15,000 to 20,000 people (1993b). Were pyroclastic surges to reach Napoli early in such an eruption, that figure could be much higher.


Preventing a disaster of dimensions that would baffle all imagination depends essentially on the occurrence of premonitory phenomena, their correct interpretation, timely warnings and evacuations. The latter steps in turn depend heavily on contingency planning and smooth and accurate communication between scientists, authorities, and the public. The issue of mitigating volcanic desasters has been widely publicized in the past decade, and there have been major achievements which resulted in successful warnings and evacuations at Pinatubo in 1991 and Rabaul in 1994. In the case of Vesuvio, the issue seems to be more complex, though. First, Vesuvio is far more densely populated than any other volcano on Earth, and the area around it is a major economic and cultural center. Second, communication in an emergency situation would involve many more, and possibly in part counteracting, groups of people than in the cases named above, as well as the media.

Starting with premonitory phenomena, there seems to be a certain likelihood that an imminent eruption could be foreseen and be warned of. The major historical eruptions of Vesuvio have been preceded by increased seismicity and other phenomena which, given the sophisticated monitorig equipment now installed on the volcano, would surely be well registered. However, not all such "premonitory phenomena" are necessarily followed by eruptive activity. It would therefore be difficult for the scientists monitoring such phenomena to decide whether they are genuine forerunners of eruption or not. Certainly, the extreme danger from any eruption at Vesuvio would justify an alarm even in the case of doubt. However, a false alarm and evacuation without an eruption would have severe consequences, first regarding the credibility of the volcanologists (and authorities), and second, socio-economic.
The disruption of business and social life and the displacement of hundreds of thousands of people for a period of undetermined duration would probably cause ravaging controversies all across the country. Worse, in the case of another alarm, maybe only a few years after the first, false one, volcanologists would encounter much more difficulty to convince the authorities, which in turn would encounter similar problems convincing the public. In other words, would anyone leave after a false alarm?

The same question applies, though, for the case of a warning without a preceding false alarm. Logistically, the evacuation of at least 600,000 people would be a problem almost impossible to solve even in an industrialized country like Italy. A contingency plan prepared recently by a special commission of scientists assumed that warning could be given up to 20 days before an eruption (see a news article taken from Nature). The commission further assumed that 600,000 people could be evacuated within a week, using trains and buses. Who ever has been in the Napoli area or has even only used trains or buses anywhere in Italy will probably imagine that this would be an extremely difficult task.


Copyright © Boris Behncke, "Italy's Volcanoes: The Cradle of Volcanology"

Page set up in January 1996, last modified on 4 March 2000
Hosted by VolcanoDiscovery