 |
|
Dramatic
view of the town of Randazzo and eruptive fissures emitting lava
flows that move towards the town, on 18 March 1981. This was one
of the most vigorous eruptions of Etna in recent decades, but fortunately
the activity weakened significantly after only two days, and the
lava flows stopped before reaching Randazzo. The main flows produced
by this eruption passed a few hundred m east of the town, destroying
farmland, fruit gardens, and dozens of isolated buildings.
Night view of lava flows running down Val Calanna, and toward inhabited
areas. The threatened town of Zafferana lies out of the photo towards
the right. 20 March 1992.
|
|
|
Although
generally considered a rather "harmless" volcano because devastatingly
explosive eruptions are very rare (with the strongest explosive activity
being commonly confined to the summit craters), Mt. Etna is a potentially
very hazardous volcano. The hazard is primarily from lava flows which
do not present a significant threat to human lives but a serious one
to property. The amount of damage to be expected from invations by lava
flows is strongly related to several variables among which the mass
eruption rate and the location of eruptive vents are the chief ones.
However, the unusually explosive flank eruptions in 2001 and 2002-2003
have revealed a previously unrecognized hazard: tephra falls, which
may affect much larger areas than lava flows. Less frequent events with
a much higher hazard potential are strongly explosive summit eruptions
and gravitationally induced sector collapse, evidence for both of which
has been found in recent years. All volcanic hazards at Etna are minor
if compared to the seismic risk in eastern Sicily.
Here
volcanic hazards are described according to different types of volcanic
activity and from highest to lowest degree of hazard of each eruption
type, starting with the most common types of activity (which are therefore
the most probable to occur in the future) and concluding with very rare,
but potentially very hazardous events. Note that there are gradations
among all types of activity.
I.
High-frequency, low-hazard events
I.1.
Summit eruptions
I.1.1.
Persistent, low-level Strombolian and/or effusive activity
The most frequent type of eruptive activity at Etna during the past
century has been the so-called "persistent" (that is, more
or less continuous and regular) activity at the summit craters, in particular
at the Northeast (NE) and Southeast (SE) Craters. Activity of this kind
is generally characterized by relatively small, discrete explosions
that eject incandescent fragments of lava to a few tens to a few hundreds
of meters high; this is called Strombolian
activity. The main products of Strombolian activity are bombs
and scoriae which accumulate around the eruptive vent(s) and build pyroclastic
cones (which are also frequently called "cinder" cones, a
somewhat misleading term). Minor ash emissions at times accompany the
Strombolian bursts. This activity does not constitute a significant
hazard to human property since the summit is not inhabited and not suitable
for agricultural activity. However, visitors to the summit area who
watch the activity at close range are exposed to the risk of falling
bombs and scoriae, which might cause serious injury or death.
The Strombolian explosions may or may not be accompanied by the emission
of fluid lava, usually from vents on the flanks or at the base of a
pyroclastic cone, at a low rate (0.1-1 cubic meters per second). Lava
emission may also occur without being accompanied by explosive activity
as in the spring of 1999 and in early 2001. Lava flows fed by the effusive
activity extend a few hundred meters to a few kilometers from their
source vents. They may interrupt the dirt roads in the summit area and
destroy monitoring equipment but are otherwise harmless since no lava
flow produced by this type of activity have ever extended into areas
of forest or cultivated land.
I.1.2.
Paroxysmal summit eruptions
Since the late 1970s much of the summit activity at Mt. Etna has been
characterized by short-lived but violent episodes of lava fountaining,
tephra emission and generation of fast-moving lava flows at high mass
eruption rates (however, not all of these events produced lava flows).
Since 1977, more than 170 such events have occurred, of which more than
120 took place between 1995 and 2001. These events are often called
paroxysms, paroxysmal
eruptions or paroxysmal eruptive
episodes. In comparison, less than 50 paroxysms were documented
during the period between 1900 and 1977, so it seems obvious that this
type of activity has become much more common in recent years, and therefore
it must be expected that summit activity in the near future will be
characterized by frequent paroxysms.
The main hazard from paroxysmal eruptive episodes is constituted by
heavy falls of pyroclastics, also named tephra. These products encompass
all types of fragmented volcanic rock (in contrast to coherent lava
flows), from fine-grained material (ash) over medium-sized (lapilli)
to large fragments (bombs and blocks). If these ejecta have a high degree
of porosity they are also called scoriae (or scoria), whatever their
grain size. Large fragments as a rule fall close to the erupting vent(s)
and therefore may affect only people who are too close (mostly because
they do not recognize the danger or are surprised by the sudden onset
of a paroxysm). Lapilli-sized scoriae and finer material is frequently
carried toward inhabited areas located downwind of the volcano. Heavy
tephra falls in those areas can disrupt traffic, damage cars (especially
windshields) and window panes, and even cause injuries since scoria
fragments are highly abrasive. Furthermore plants may suffer from such
rains of scoriae, although in the long term they have a powerfully fertilizing
effect. In the case of particularly strong paroxysms occurring during
dry periods, forest fires may be caused by the fall of very hot pyroclastics,
as happened during a powerful explosive episode at the Voragine in July
1960.
The most memorable sequence of tephra falls in recent time occurred
in the spring of 2000 when the SE Crater generated more than 60 paroxysmal
eruptive episodes. Heavy showers of millimeter to centimeter-sized scoriae
repeatedly occurred in a sector encompassing the NE, E and SE flanks
of Etna, and cleaning efforts became an obsession to the inhabitants
of the affected towns and villages. Abundant quantities of lapilli and
ash not only left continuous deposits on roads making driving an adventure,
but they also clogged drainage systems and entered homes.
The effects of strongly explosive paroxysmal eruptive episodes can be
even more far-reaching and endanger air traffic. Since Etna lies in
one of the areas of major business and touristic interest, civil aviation
may be affected by tephra falls and eruption plumes; furthermore a large
U.S. military air base (Sigonella) lies about 40 km S of the volcano.
The existence of this risk, which had previously not been recognized
at this volcano, was dramatically demonstrated on 26 April 2000 when
an airplane with more than 100 passengers and staff encountered falling
lapilli while flying underneath an eruption plume emitted from the SE
Crater shortly before. The pilot made an emergency landing at Catania
airport and no one was injured, but the aircraft suffered significant
damage from abrasion by impacting scoriae; it was pure luck that this
indicent did not develop into a tragic accident. As a direct consequence,
flights were re-routed from the E side of Etna (which is most commonly
used but also the sector most exposed to the passage of tephra plumes)
to the safer W side as long as paroxysmal eruptive episodes continued
at the SE Crater.
I.1.3.
Phreatic or phreatomagmatic activity
Etna is not particularly well known for a type of activity that consists
of the explosive interaction of magma (or hot rock in general) with
external water, but it seems that this is more common than previously
thought, especially at the summit craters. Water may seep into the conduits
of Etna during heavy rain falls or due to the melting of ice and snow
(which occurs after each snow fall near the summit craters because the
ground there is hot) and come into contact with hot rock at depth. If
the conduit is blocked, vapor generated by this encounter will generate
increased pressure, until the material blocking the conduit is explosively
removed. Alternatively, the uprise of fresh magma through a water-soaked
conduit will lead to similarly explosive interactions. If all the ejected
material is composed of old rock without any fresh magma fragments,
the activity is called phreatic;
if a mixture of old rock and fragments derived from fresh magma is ejected,
it is called phreatomagmatic. A
further variation on the theme is the obstruction of one of the summit
craters by a pond of dense, solidifying lava; gas pressure beneath this
"cork" may accumulate and lead to its explosive disruption
without fresh magma being involved, leading to phreatic activity. The
products are blocks of old rock that may reach dimensions of several
meters, lapilli-sized fragments of old rock and ash and minor amounts
of fresh magmatic ejecta whose proportion may vary from zero to about
50%. What makes phreatic or phreatomagmatic explosions particularly
treacherous is their sudden onset, the almost complete lack of noise,
and their unpredictability. Another evil fact is that at night many
blocks ejected by such explosions are virtually invisible because they
are not incandescent. It is for these reasons that all of the 13 deaths
directly attributable to eruptive activity of Etna during the past century
were caused by phreatic or phreatomagmatic explosions.
This type of activity thus represents one of the main hazards to human
lives at this volcano, but fortunately it is limited to the immediate
surroundings of the summit craters. Large clasts ejected by phreatic
or phreatomagmatic eruptions as those seen during the past few decades
fall in a range of a few tens to a few hundreds of meters from the source
vent and are therefore only dangerous for those who are too close.
If phreatic or phreatomagmatic summit activity generates significant
amounts of ash, this may disrupt car and air traffic and cause drainage
problems in inhabited areas.
I.2.
High-altitude subterminal or flank eruptions
Eruptive
activity from vents (normally fissures) away from the summit craters
and their immediate surroundings is called subterminal
if it is very closely related to the activity of one or more of the
summit craters. Subterminal can also be termed "quasi-summit"
activity, and in recent years a trend toward considering it part of
the summit activity is evident in the volcanological literature. This
is mostly based on the fact that in the case of the opening of subterminal
vents the activity may shift back and forth between these vents and
the summit craters. If new eruptive fissures open at a certain distance
from the summit craters and their activity fully replaces that of those
craters, it becomes a flank eruption.
Although there is no clearly fixed transition, in terms of distance
from the summit craters, between subterminal and flank eruptions, the
latter commonly are followed by a period of repose at the summit craters,
while the earlier may be immediately followed by renewed activity at
the summit craters.
The activity of these types of eruption normally consists of quiet lava
emission with minor Strombolian activity or spattering (weak explosive
activity that ejects clots of fluid lava to a distance of a few meters
from the vents) and does not endanger human lives except in cases when
people get extremely close to the vents or move on top of lava tubes.
Effusion rates may be higher than during persistent summit activity
and allow the emplacement of longer lava flows. These, as in the case
of persistent summit activity, might bury the dirt roads in the summit
area and threaten, damage or destroy one of the few manmade structurers
in the upper parts of the volcano, such as the "Torre del Filosofo"
mountain hut which stands only 1 km south of the SE Crater. Since little
explosive activity is associated with these eruptions they do not have
any far-reaching effects. If the eruptive vents lie at elevations of
about 2900 m or less, the effusion rates are high enough and the activity
lasts for a certain period, lava flows might advance as far as several
kilometers and affect the tourist facilities on the southern or northern
flanks. In fact, during the July-August 2001 eruption lavas fed by an
eruptive fissure at about 2700 m elevation came close to the "Rifugio
Sapienza", a large mountain hut at about 1900 m altitude, and the
nearby departure station of the cable car. Much of the hazard potential
of such eruptions depends on where they take place - if they occur on
the W or E side of the volcano, they would affect areas without any
man-made structures whereas the S side is highly vulnerable.
II.
Medium-frequency, medium to high-hazard events
II.1.
Mainly effusive flank eruptions
Flank
eruptions occur at irregular intervals, which may last from less than
one year to several decades. The frequency of flank eruptions since
1971 has been unusually high - between 1971 and 1993 the average interval
between the beginning of one flank eruption and that of the next
one was about 1.5 years. The July-August 2001
eruption occurred more than 8 years after the end of the preceding
flank eruption, which is probably due to a cyclic eruptive behavior
of Mt. Etna that leads to flank eruptions being clustered in time rather
than occurring at more or less random intervals. However, the 2001 eruption
was atypical for some reasons and should not be taken as strictly representative
of the eruptive behavior of the volcano.
The hazards from eruptions of this kind are definitively higher than
those associated with summit eruptions but depend on the location of
the eruptive vents - both in the sense of elevation and geographical
location - and from the mass eruption rates and the eruption's duration.
The lower the eruptive vents, the greater the probability that lava
flows arrive in populated and agricultural areas. The higher the mass
eruption rates, the more rapid and longer the resulting lava flows and
thus the risk of invasion of populated and cultivated areas. While a
distinction according to the physical eruption parameters like elevation
of the eruptive vents, eruption duration and mass eruption rates is
often applied for the evaluation of the volcanic hazard at Etna, it
seems more useful to make this distinction relative to the vulnerability
of the areas on the various sides of the mountain and consider the various
eruptive parameters in this framework.
It has to be noted that while material damage resulting from flank eruptions
can be overwhelmingly high, such events do not threaten human lives
because lava flows move quite slowly once they have reached a certain
distance from their source vents. All sources attributing human deaths
to lava flows at Etna (such as the striking number of up to 20,000 deaths
in the 1669 eruption) are pure inventions or misconceptions based on
confusing eruptions with earthquakes - the latter in fact have caused
more than 200,000 deaths in Sicily during the past 1000 years, while
Etna's volcanism can only be blamed for less than 100 deaths during
the past 2000 years (see "Etna and Man").
II.1.1.
Eruptions on the E flank
12 out of the 23 flank eruptions during the 20th century occurred in
the eastern sector of Mt. Etna and were at least partially confined
to the Valle del Bove, a large collapse
structure on the E flank of the volcano. If the eruptive fissure(s)
are located within that depression there is a good chance that the lava
flows they generate will accumulate on its floor, a vast, desert-like
area covered by numerous lava flows of previous eruptions. Usually flank
eruptions occurring in the Valle del Bove are greeted by the local population
with a sense of relief, since it acts like a giant natural catchment
basin. In the Valle del Bove there is little to destroy; only a few
patches of forest have survived the numerous invasions by lava flows
especially since 1950, which have led to the resurfacing of about 80%
of the Valle del Bove floor in this period.
However, some eruptions that occurred in this area did cause damage
and seriously threatened population centers lying below the lower (eastern)
end of the Valle del Bove, such as in 1950-1951, 1979 and 1991-1993.
In 1950 the lava flows from the initial phase of the 1950-1951 eruption,
fed by high effusion rates, advanced close to the village of Fornazzo
and damaged nearby fruit gardens. In 1979 high effusion rates generated
a fast-moving lava flow that advanced on top of the 1950 flow and again
seriously threatened Fornazzo, which was evacuated for a few days but
fortunately was not touched by the lava flow; however, fruit gardens
and a road were buried by the lava. A different situation developed
during the first months of the 1991-1993 eruption. In that case a constant,
moderately high effusion rate led to the establishment of an efficient
system of lava tubes, which allowed the transport of lava without significant
heat loss (lava flowing through tubes is protected from the chilling
effect of the outside air) over many kilometers. As a result the lava
flow field attained a length of more than 8 km and extended very close
to the town of Zafferana on the SE flank of Etna (Calvari and Pinkerton,
1998). This eruption is a striking example of how the various factors
determining the hazard potential of an eruption may interact with each
other and produce unexpected effects. Eruptions like those of 1979 and
1991-1993 will inevitably lead to a repetition of the threat to the
villages below the Valle del Bove, and since this sector of the volcano
has been particularly active in recent decades, many of the eruptions
in the foreseeable future can be expected to take place there.
II.1.2.
Eruptions on the S flank
Although less frequent than E flank eruptions, eruptions that occur
on the S flank have a high hazard potential since this is the side of
the mountain where man-made structures extend to the highest altitudes.
A large complex of tourist facilites such as hotels, restaurants, souvenir
shops and the Etna cable car (partially destroyed during the 1983, 1985
and July-August 2001 eruptions) lie at an altitude of about 1900 m,
so that even small eruptions occurring in this area might cause significant
damage. Both the 1983 and 2001 eruptions were highly destructive, although
in both cases the threat to towns lying further downslope was vastly
exaggerated by the press and local administrations. Any future eruption
in this area will threaten or damage the tourist facilities, but rebuilding
and maintenance of these structures is of paramount interest since this
is the main access route for tourists to the higher regions of Etna,
in spite of the presence of a beautiful alternative route on the N flank
(Linguaglossa-Piano Provenzana).
II.1.3.
Eruptions on the W and N flanks
During the 20th century, the W and N flanks of Etna have been affected
by fewer eruptions than the E and S flanks, but two dramatic eruptions
on the NE flank in 1911 and 1923 caused significant damage and threatened
the town of Linguaglossa, and the 1981 eruption on the NNW flank led
to the destruction of hundreds of buildings, roads and railway lines
and nearly consumed the scenic town of Randazzo. The complex of tourist
facilities (hotels, restaurants, souvenir shops, and ski lifts) of Piano
Provenzana, located at about 1800 m elevation on the NE flank, was virtually
wiped out and buried by lava flows at the beginning of the 2002-2003
eruption. On the other hand, the somewhat atypical 1974 eruption on
the W flank was very small and affected the least densely populated
and developed sector of the volcano. As a rule, eruptions with high
effusion rates are more likely to produce lava flows long enough to
extend into vulnerable areas, while low-effusion rate events may even
take place at relatively low elevations without representing a significant
threat. The 1974 vents opened at 1500-1600 m elevation, which is fairly
low, and in 1981 the most destructive lava flows were emitted at exceptionally
high effusion rates from vents at 1300-1700 m elevation while vents
at only 1150 m altitude generated small, slow flows because the effusion
rates there were low. The areas where flank eruptions are most likely
to occur in the future are the so-called Northeast Rift (a zone where
eruptive fissures and pyroclastic cones are densely clustered along
a line of structural weakness) and the much less distinct so-called
West Rift Zone, in which numerous sizeable pyroclastic cones occur.
Many Northeast Rift eruptions in the past were characterized by high
effusion rates and similar events in the future might endanger the town
of Linguaglossa or one of its neighboring villages, and the extensive
areas of wine and hazelnut production on the N and NE sides of the mountain
as well as the largest contiguous natural forest on Etna, the "Pineta
Ragabo", which was indeed severely damaged by lava flows at the
beginning of the 2002-2003 eruption.
II.2.
Explosive activity during flank eruptions
The
first two flank eruptions of the new millennium, in July-August 2001
and October 2002-January 2003, have revealed a hazard that had until
then been largely ignored: tephra falls over large areas around the
volcano, generated by strongly explosive activity. Until 2001, explosive
activity was substantially believed to be a privilege of the summit
craters, but large amounts of airborne ash were produced from vents
on the southern flank in 2001 and, even more significantly, in 2002-2003.
Ash fell heavily in downwind sectors, amounting to several centimeters
up to 30 km away, interrupting air traffic, and causing enormous logistical
and economic problems to the Catania region. Fine ash fell up to several
hundred kilometers distant, in Greece to the east, and northern Africa
to the south. Near the eruptive vents, pyroclastic deposits reached
thicknesses of more than 50 m, and sizeable pyroclastic cones were built
around the vents, reaching heights of 100 m in 2001 and more than 200
m in 2002-2003.
The explosive eruptions of 2001 and 2002-2003 marked the end of a fairly
peaceful period of essentially non-explosive flank eruptions that had
lasted for more than a century. The common belief that Etna was not
capable of producing strongly explosive eruptions was largely a result
of this period, even though flank eruptions had become more frequent
during the last 30 years of the 20th century, and brief episodes of
highly explosive activity had characterized much of the summit activity
since the 1960s. In fact, the most recent strongly explosive flank eruption
occurred in 1892, in a time when the Etna region was strongly underdeveloped,
and much less densely populated than it is now. However, a simple look
at all the large pyroclastic cones (many of them being of historical
age) that dot large areas on the flanks of the volcano should have served
as an indicator of potentially explosive flank eruptions. One of the
main reasons why explosive flank eruptions have been largely neglected
until recently lies in the fact that such events in the past (that is,
before the end of the 19th century) have caused much less consternation
and problems to the population at that time, because the society and
its infrastructures were considerably less vulnerable. A fall of ash
surely meant a nuisance for those living in the fallout sector, but
it could also be used to fill the holes in the dirt roads that made
up much of the lifeline network then. Airplanes had yet to be invented,
and sophisticated techniques now occupying broad space in everybody's
everyday life were not even dreamt of. This is why the 2001 and 2002-2003
eruptions came as a very bad surprise, affecting a socienty that had
advanced close to mid-European standards.
The long break in strongly explosive flank eruption between 1892 and
2001 was very probably an exceptional period in Etna's recent history,
and it is unlikely that another such period of considerable length will
be seen in the near future. Given the current dynamics of the volcano
(with a certain proportion of magma being supplied into a new, "eccentric"
reservoir that has formed next to the central conduit system), it must
be rather expected that there will be more flank eruptions with similar
degrees of explosivity as those of 2001 and 2002-2003 in the near future.
As seen during these events, widespread and voluminous tephra falls
create a variety of problems in the affected sectors. Closer to the
volcano, thick tephra deposits on roads represent a significant obstacle
to road traffic and render car driving dangerous especially if mixed
with rainfall water, which renders pavements quite slippery. Crops may
temporarily suffer damage, although in the long term the ash is a powerful
fertilizer (one of the main reasons why the Etna region is so densely
populated). The ash enters every home, and thus affects the quality
of life of those living in the tephra fall sectors. A hazard to health
(mostly from inhaling the ash) has not been recognized, because Etna's
ash is generally to heavy and coarse-grained to remain suspended in
the air - it simply falls to the ground and thus cannot be breathed.
In areas of intense road traffic (foremost the city of Catania with
its dense traffic, which is unbearable already when there is no ash),
however, ash grains are likely to be crushed and pulverized by frequently
passing vehicles and thus transform into fine dust that may be suspended
in the air every time there is wind. In that case it cannot be excluded
that it is inhaled by people and affect their health. No detailed study
has been made public so far in the wake of the recent eruptions, although
the subject has been intensely discussed in local mass media, and local
administrations have categorically excluded any risk to public health
(and done little to remove the enormous quantities of black ash still
persisting all over the Catania area, although much of it has been removed
by heavy rainfalls between January and May 2003).
The sector that is most sensible to tephra falls is aviation, and with
it tourism, and thus a large business sector in the Catania area. During
the 2001 and 2002-2003 eruptions, the international airport of Catania,
ominously named "Fontanarossa" (red spring), had to be closed
and flights were rerouted to Palermo (200 km distant) or Reggio Calabria
(90 km distant, and on the opposite side of the Strait of Messina),
or simply cancelled. In the case of the latter eruption the problems
came essentially after the end of the tourist season (which had strongly
suffered from the worldwide effects of the terrorism attacks of 11 September
2001) but persisted over a period of several months, compared to about
two weeks in 2001.
Following the 2002-2003 experience, Italian civil aviation experts have
shown eager activity to mitigate the effects from future explosive volcanism
at Etna, including a recent meeting with representatives from the Japanese
city of Kagoshima, which has a decades-long experience in dealing with
frequent ash falls from nearby Sakurajima volcano, and seems to deal
with them pretty well. Catania will have much to learn from that example.
And this is warranted. There is a good possibility that future eruptions
of Etna will show similar degrees of explosivity as the most recent
ones in 2001 and 2002-2003.
III.
Low-frequency, high-hazard events
III.1.
Low-altitude flank eruptions
Throughout
the historical period, flank eruptions at Etna have been most frequent
at medium to high elevations - that is, between 1500 and 2500 m altitude.
However, flank eruptions may also occur at much lower elevations, and
these represent the most hazardous type of activity seen during the
past 2000 years. Two historical eruptions from the S flank may serve
as the possible extremes of what may occur in the case of a flank eruption
- that of March-July 1985 whose vents lay above 2500 m elevation, and
an eruption generally ascribed to the year 1381 (evidence presented
by Tanguy, 1981 and Tanguy and Patanè, 1996, indicates that this
eruption occurred about 200 years earlier, i.e. during the 12th century)
when an eruptive fissure opened at about 400 m elevation. In the earlier
case - an eruption characterized by low effusion rates - the eruptive
fissure cut through the Piccolo Rifugio which had already been damaged
by the 1983 eruption and abandoned; lava flows extended below 2000 m
but did not threaten the tourist facilities around the Rifugio Sapienza
which had suffered extensive destruction in the 1983 eruption. Both
the low mass eruption rates and the high elevation of the eruptive vents
contributed to a fairly tranquil course of events (from a human standpoint).
The "1381" eruption, on the other hand, was characterized
by higher effusion rates and occurred in an area which is now practically
a part of Catania's suburbs. Lava flows from that event reached the
sea to the south of Porto d'Ulisse, after overrunning an area now occupied
by Ognina and Picanello, parts of present-day Catania.
The historical record shows that eruptions from low-lying vents on the
S and SE flank were clustered in time, such as various eruptions during
the Roman age and again between the 10th and 15th centuries. The large
1669 eruption, though, appears to have been an isolated event, but it
entered the historical record as the most devastating and violent known
flank eruption of Mt. Etna. All these eruptions were characterized by
high eruption rates and their lava flows frequently reached the sea;
in 1669 nearly one cubic kilometer of lava (about 40 times the volume
of the July-August 2001 lava) was emitted within four months, which
corresponds to an average effusion rate of 80-100 cubic meters per second
(compared to an average rate of 12-13 cubic meters during the 2001 eruption).
While no eruption at Etna since 1669 has occurred from vents lying below
1000 m elevation, there is absolutely no reason to assume that low-altitude
eruptions will not occur in the future. Considering the long period
elapsed since 1669 one could actually reason that such an eruption is
overdue.
Low-altitude flank eruptions are relatively well documented for the
southern and southeastern flanks of Mt. Etna but only sketchy, confusing
or erroneous information exists for such eruptions in all other sectors
of the volcano, since these areas were much less densely populated and
virtually no people with a certain level of culture and education dwelled
there to witness and describe those events. Yet the abundance of numerous
large pyroclastic cones associated with extensive lava flow fields on
the W and N to NE flanks testifies to major eruptions from vents at
low elevations in the past.
A resumption of eruptions from vents below 1000 m elevation would cause
widespread destruction today, especially if they occurred on the S or
SE side of the volcano. The vent areas and lava flows of the eruptions
of Roman age, of the 10th to 15th century and of 1669 all lie in areas
that are now densely populated, including the city of Catania. While
lava flows may be (partially) diverted in the case of eruptions characterized
by low effusion rates and vents at high elevations (such as in the 1983,
1991-93 and 2001 eruptions), such measures will not be possible if the
vents open in populated areas and lava flows could only be diverted
from one town to another because there is practically no area left to
where a flow might be diverted to without causing damage.
The only possibility would be a protection at least of Catania or other
coastal towns, which in any case would be relatively far from an eruptive
vent. The 1669 eruption serves as a good example of how Catania might
be protected from being invaded by a lava flow. In 1669, the lava was
kept from reaching and destroying the center of Catania by simply constructing
thick barriers across the main streets of the city. Similarly, gaps
between building complexes in the marginal areas of the town could be
closed by concrete walls that could be erected after the onset of an
eruption at low elevations, once the probable course of the lava flows
is more or less predictable (e.g. with the help of computer simulations).
Furthermore it should be expected that the incredibly huge accumulation
of buildings, with many of them being contiguous, would exert some resistance
to a lava flow by itself. During modern history, no large city of Catania's
dimensions in the industrialized part of the world has been reached
or invaded by lava flows, and the effect of large building complexes
on a slowly approaching, but thick lava flow is not known. The only
precedent, which the world witnessed via the mass media as recently
as in January 2002, was the invasion by lava flows of the city of Goma
in the Democratic Republic of Congo. Even though that city cannot be
compared in the least sense with Catania, it was fascinating to see
how the lavas erupted from Nyiragongo volcano were channelized by roads
and buildings, of which many were burned but only few were crushed.
Even in the case of a low-altitude flank eruption in a more remote and
less densely populated area, devastating consequences are to be expected.
The next eruption of this type needs not necessaritly affect the highly
urbanized SE flank but might equally strike the N flank and the valley
of the Alcantara river. A major eruption in that area would lead to
the destruction of the town of Francavilla and several smaller villages
and cause the loss of the world-famous Alcantara gorge which attracts
numerous tourists and local residents. Furthermore, immediately before
reaching the sea the lava flow would cut a highway, various important
roads and the Messina-Catania railway. The damming of the Alcantara
river would cause flooding over a wide area. While much of the recent
hazard evaluations at Etna are concerned with the S and SE flanks and
the Catania area, the possibility of a disastrous low-altitude flank
eruption in other sectors of the volcano has been mostly neglected so
far.
III.2.
Major explosive summit eruptions
Until
a few years ago, explosive
activity at Etna was considered a less common and thus less significant
type of activity than lava emission, and generally it was believed to
be quite modest. This view has changed since a group of researchers
began to publish the results of their search for evidence for large-scale
explosive events at the volcano, an evolution that has received further
stimulus from the frequent, vigorously explosive events at the summit
craters in the past decade. Documented explosive summit eruptions during
the past 2000 years were exclusively paroxysmal episodes as described
above, but as a matter of fact the historical records of the past 400
years describe no period of similar explosive activity as that between
1995 and 2001.
Detailed field work by Coltelli et al. (1998, 2000) has led to the discovery
and study of numerous large-scale explosive events over the past 100,000
years. The most recent - and apparently one of the most violent - of
these occurred in 122 BC. That eruption had until recently been attributed
to a flank vent (M. Trigona), located at 425 m elevation on the SE flank
near the village of Trecastagni, which is associated with a lava flow.
Historical records report strong tephra falls that caused most of the
houses in Catania to collapse. Damage was in fact so widespread that
the inhabitants of Catania were exempted from paying taxes to the government
of Rome for ten years. Coltelli et al.'s research indicates that the
eruption occurred from the summit area rather than from a flank vent,
and therefore must have been of exceptional violence. Furthermore the
study shows that pyroclastic flows occurred in the summit area, but
surges rushed much farther downslope to reach areas that are now densely
populated. A similar event today would certainly cause much damage and,
due to its suddenness, there would be a direct risk to human lives.
It is of paramount interest to understand the dynamics that could lead
to an event of such proportions, and Coltelli et al. (1998) propose
that the 122 BC eruption was caused by a sudden decompression in the
central conduit system, which in turn was a result of major fracturing
and extension in the upper part of the volcano. The magic word in this
context is volcano instability (see the next section, III.3.), which
opens the stage to a spectrum of worrisome scenarios, none of us would
desire to see become reality.
Evidence of still larger, and more devastating, eruptions in the past
is readily accessible in spectacular outcrops near the town of Biancavilla
on the lower SW flank of Mt. Etna. This evidence consists of thick deposits
of pyroclastic flows, also known as "nuées ardentes"
or glowing avalanches, which rushed down the flanks of an ancestral
Etna volcano (named "Ellittico" by geologists) some 13,000
to 15,000 years ago. The composition of the magma from which these pyroclastic
flows were derived is different from that of the magmas erupted since
then and indicates that before that eruption a large quantity of magma
accumulated in a shallow reservoir below the volcano. This process is
assumed to be the main reason for the exceptionally explosive and voluminous
activity at the end of the "Ellittico" volcanism at Etna,
and nothing indicates that a similar process is occurring today. For
an eruption of this magnitude and character to occur thousands of years
of magma accumulation and chemical evolution would be needed; at present
it seems that Etna is erupting most of the magma it receives from a
mysterious source at depth (see "The
storage and transport of magma: The search for the hidden magma chamber")
and there is no evidence of magma accumulation in a large shallow reservoir.
III.3.
Volcano instability and sector collapse
One
of the most hazardous processes that can occur at a volcano, active
or not, is the collapse of one of its flanks leading to a huge avalanche
of volcanic debris, a process that is generally known as sector collapse.
The famous eruption of Mount St. Helens in Washington (U.S.A.) on 18
May 1980 was triggered by such a sector collapse, which generated a
series of devastating events, starting with a debris avalanche immediately
followed by a laterally directed explosion (or blast) and a Plinian
eruption. During the years following that eruption it was discovered
that a surprising quantity among the volcanoes on Earth had experienced
similar events, but many of these had been of far greater magnitude
than the 1980 collapse, debris avalanche and eruption at Mount St. Helens.
Sector collapse soon became known to leave peculiar morphological features
in the sides of the affected volcanoes, so-called collapse amphitheaters,
and distinct deposits showing a hummocky surface were interpreted to
be the remainders of the collapsed volcano flanks, which had advanced
at awesome speed for up to tens of kilometers from their original positions.
It was logical that the concept of sector collapse would eventually
be tested in the case of the Valle del Bove on the E flank of Etna.
Since the mid 1980s and especially since the early 1990s various groups
of researchers have proposed an origin of the Valle del Bove by one
or more sector collapse events (Guest et al., 1984; Borgia et al., 1992),
the later hypotheses associated such events with the gravitational spreading
of the E flank of the volcano which is open to the sea. Gravitational
spreading was seen as a possible cause - or a possible result - of repeated
intrusions of magma in a set of fracture systems extending from the
summit area to ENE and SE, actually these fracture systems have been
particularly active since 1971. A major problem in painting the picture
was the apparent lack of a deposit that could be clearly interpreted
as a debris avalanche deposit, but in 1998 Calvari et al. presented
evidence for such a deposit which they dated at no less than 8400 years
before present. More recently, a group of researchers from the U.K.
has begun a study of rocks on the S rim of the Valle del Bove, which
led them to the preliminary conclusion that the latest collapse event
in the Valle del Bove occurred only some 3500 years ago. The data presented
by Calvari et al. (1998) and by Deeming et al. (2001) indicate that
catastrophic sector collapse occurred at least twice during the Holocene
(that is, during the past 10,000 years) at Etna. This is not a happy
bit of news. Geological events that happened in the past are very likely
to be repeated in the future. Sector collapse is the most dramatic scenario
to be envisaged at Etna, and various processes observed at the volcano
in the past few decades are interpreted by some geologists to point
to continued volcano instability which might eventually lead to further
collapse in the upper parts of the Valle del Bove.
One of these is the frequent intrusion of magma into the fracture systems
that bound the W wall of the Valle del Bove. Each time such an intrusion
take place in the area (this has happened five times since 1980), the
area to the E of the intruding dike is displaced eastward, and this
area is the steep W face of the Valle del Bove. In other words, the
upper part of the Valle del Bove headwall is pushed away from the rest
of the mountain. This is a steep slope about 1000 m high, and if a large
part of it were to transform into a major landslide, it would fall from
a height of may hundreds of meters toward the valley floor, which would
give the falling mass an incredible acceleration and momentum. The resulting
debris avalanche would speed eastward across the floor of the Valle
del Bove and almost surely extend far beyond its lower end into the
densely populated area below, possibly down to the Ionian Sea. All this
would occur within a few minutes. Devastation would be beyond imagination,
and, worst of all, the present state-of-the-art of volcanology has few
means to predict events of this kind in time. It can simply be hoped
that those who believe that Etna is prone to further collapse in the
near future are wrong, and that the volcano will rather behave in the
same manner it has done during the past 2000 years.
But even if a sector collapse of catastrophic dimensions is probably
a remote possibility, the eyes of many scientists are fixed on the W
rim of the Valle del Bove as this area is moved eastward by one intrusion
after the other, at intervals of a few years, and on the SE Crater,
whose cone sits immediately on that mobile rim and which has grown at
an unbelievable speed since the late 1990s. Can the rapidly increasing
weight of that cone, plus the weight of several voluminous lava flow
fields emplaced since 1999 on the W rim of the Valle del Bove, destabilize
that rim and trigger its collapse? What if such a collapse entrains
the SE Crater cone, exposing its conduit and all the magma it contains
to the fresh air and instantaneous decompression? Wouldn't this be a
perfect scenario uniting the hypotheses of sector collapses and major
explosive eruptions in the past? The truth is, no one knows. Most scientists
studying Etna prefer to assume that the volcano will do the same things
they have seen personally during their many years of observations of
the volcano and its activity, and it is very likely that things will
exactly go that way. But at times, like a nightmare, the vision of that
area plunging into the Valle del Bove and beyond, leaving behind it
a mass of decompressing magma that transforms into a huge explosion,
haunts the minds of some of us. This is where volcanologists can only
wish that the more catastrophism oriented colleagues possess a vivid
phantasy. Most of the hazards described on this page can be mitigated
if the involved people and institutions are willing to reason and to
collaborate. A sector collapse and all that it might unleash simply
must not take place, not here and not now.
|
