Italy's Volcanoes: The Cradle of Volcanology

Etna index

   
Geology Geological history Cones and craters
Eruptive characteristics Eruptions before 1971 Eruptions since 1971
Etna and Man References Web sites
Weather forecasts FAQ Latest news

 

The storage and transport of magma: The search for the hidden magma chamber
Part 2: Eruptive cycles reveal insights into the dynamics of magma storage and volcano instability


Going back in the chronology of documented eruptions of Mount Etna, it is seen that the record is reasonably complete and accurate for the past 400 years or so (back to the year A.D. 1600). Before that date, it becomes more and more sketchy, and acribic research carried out by a French team of scientists led by J.-C. Tanguy since the late 1970s has shown that many of the eruptions described for the pre-1600 period did either not take place, or at least the lava flows attributed to them do not correspond with the alleged eruption dates. Nearly all of the major lava flows shown on the geological map of Etna (Romano et al. 1979) with ages older than A.D. 1600 are older than indicated on the map, in most cases by 200 to 500 years (Tanguy et al. 1999). So any analysis of the long-term behavior of the volcano is limited to the about 400 years since A.D. 1600, which covers at least 51 flank eruptions and about as many significant episodes of summit activity. This might seem much, but actually it isn't. Statistical analysis works with amazing amounts of data, but risks to fail with this relatively limited number of eruptions (yet studies are under way to see whether they can be used for some limited statistical methods).

A few important things, though, can be read from the post-1600 eruptive record. The first, and most important, is that the activity of Etna is characterized by very strong variations in time. During the first 70 years of the post-1600 period (1600-1669), the activity was at amazingly high levels, with a series of important flank eruptions, most of which were quite voluminous (up to 1 cubic kilometer and more), and some lasted over many years. Consequently, some had devastating consequences, such as those of 1651-1653 (partial destruction of the town of Bronte) and 1669 (destruction of 15 villages and damage to the city of Catania), because their lava flows invaded inhabited areas. That period of intense activity closed with the enormous 1669 eruption, when one cubic kilometer of magma was released from the plumbing system of the volcano in only four months. The 1669 eruption brought the total volume of eruptive products since A.D. 1600 to 3 cubic kilometers, which means that during those 70 years, magma was erupted at an average rate of 1.4 cubic meters per second. And then, something changed.

For nearly 100 years after the 1669 eruption, Etna was much less active. There were remarkably fewer flank eruptions (only three clearly defined flank eruptions are documented, in 1689, 1702, and 1755), and these were of much smaller volume than the eruptions of the 1600-1669 period. There was intense activity at the summit, and a new summit cone was built up (following the collapse of a previous summit cone during the 1669 eruption). Yet the rate of magma emission was probably only a small fraction of that observed between 1600 and 1669, maybe it was lower by up to two orders of magnitude (that is, in the range of 0.01 cubic meters per second).

In 1763, activity at Etna lived a fresh surge, with two flank eruptions in a single year (February-March 1763 on the western flank, and June-September 1763 on the southern flank, forming the conspicuous cone of the Montagnola), then there were two more flank eruptions in 1764-1765 and 1766 before the activity settled into a more regular pattern, with flank eruptions occurring about once every 10 years. For the next 100 years, flank eruptions occurred in many areas of the volcano, with volumes varying from a few million cubic meters to more than 100 million cubic meters: 1780, 1792-1793, 1802, 1809, 1819, 1832, 1843, 1852-1853 and 1865. The mean eruption rate during the full period from 1763 to 1865 was 0.2 cubic meters per second, the whole amounting to 0.75 cubic kilometers - still quite a small quantity as compared to the 3 cubic kilometers erupted between 1600 and 1669.

After the 1865 eruption, Etna's behavior became more complex. This was the time of the well-defined eruptive cycles, which are still continuing to the present day. At the same time, the productivity of the volcano has risen sharply, especially since 1950. The total volume of eruptive products since 1865 is about 1.7 cubic kilometers, corresponding to an average output rate of nearly 0.4 cubic meters per second, twice that of the 1763-1865 period. But if the period since 1950 alone is considered, the jump in productivity is still more evident, because more than one cubic kilometer of lavas and an ever more significant quantity of pyroclastics were erupted since then at a mean rate of more than 0.6 cubic meters per second. Focusing still closer, on the period from 1971 to 2003, there is yet another increase: in the arc of 22 years, the volume of erupted products amounted to slightly less than 0.7 cubic kilometers, and the average eruption rate for this period is 1 cubic meter per second. This is not all that far from the output rate of the 17th century (through 1669), and the message is loud and clear: Etna is getting more and more active. If this trend continues, the levels of activity will get closer and closer to those seen between 1600 and 1669.

Dynamics of Mount Etna in the past 400 years
Returning to the period of awesome activity in the 17th century (1600-1669), there is an additional detail to be considered. The lavas and pyroclastics erupted in that period are compositionally different from anything that has come out of the volcano ever since. One characteristic can be readily seen with the naked eye: those products are extremely rich in relatively large (up to 1 cm) crystals of plagioclase, and are known locally as "cicirara", "pea lavas". The amount and size of these plagioclase crystals shows some variations from eruption to eruption, but all flows of that period, without exception, are "cicirara". None of the lavas erupted from 1669 until today are "cicirara". Thus it is not only the behavior of the volcano that has changed, something has also changed in the dynamics of magma storage.

In an interesting paper published more than 10 years ago, Hughes et al. (1990) speculate that the ubiquitous presence of "cicirara" lavas in the 17th century (through 1669) is a result of the existence of a major, relatively shallow (the authors give no further indications about the depth) reservoir, where magma was stored for some time before erupting, and thus there was enough time to permit the growth of the plagioclase crystals that constitute the main characteristic of those magmas. With the large eruptions of the 17th century, most notably that of 1669, this presumed magma reservoir was emptied more or less completely. Evidence for magma withdrawal from a relatively shallow reservoir is seen by Hughes et al. (1990) also in the fact that during the 1669 eruption, the summit cone of Etna collapsed, forming a small caldera. After that event, magma rose from a much deeper source directly through the central conduit system without resting somewhere in a shallower reservoir, and thus growth of large plagioclase crystals was no longer possible, and no "cicirara" lavas could form.

Another important detail lies in the fact that the lavas of the 1600-1669 period are slightly more mafic (that is, less rich in silica) than nearly all of the post-1669 lavas. However, in recent years, some eruptions have again produced slightly more mafic lavas and pyroclastics, such as those of 1974 and 1989, and the eccentric activity in 2001 and 2002-2003. Furthermore, the mineralogy of the eccentric products of the latest two eruptions show a quite different mineralogy when compared to nearly all other products of the past 350 years or so. The most important observation is that they contain certain amounts of a mineral called amphibole, which points to the existence of higher quantities of water in these magmas than in most other recent products of Etna. Only one eruption in the past 350 years is known to have produced similar amounts of amphibole, that of 1892 on the southern flank, which generated the famous cones of the Monti Silvestri, near the tourist complex of the Rifugio Sapienza (J.-C. Tanguy and R. Clocchiatti, personal communication). That was also very probably an eccentric eruption, which was characterized furthermore by a high degree of explosivity, like the eccentric activity in 2001 and 2002-2003.

The return of eccentric activity in the latest eruptions coincides with the discovery of distinct, relatively shallow magma reservoirs below the volcano, one of which might be a fairly recent feature, and this poses a number of important questions. Does the appearance and growth of these reservoirs mark a return to a similar situation as possibly existed between 1600 and 1669? Will these reservoirs continue to grow, will they eventually merge to form one larger reservoir? Will "cicirara" lavas reappear? Is Mount Etna returning to a style and magnitude of activity as in the period between 1600 and 1669?

The sharp increase in the productivity seen at the volcano in the past 50 years seems to be a confirmation of this hypothesis. Etna is presently more active than at any other time since 1669, and its activity is growing more complex, with compositionally distinct magmas erupted at the same time, a dramatic increase in the explosivity of the eruptions (both summit and flanks), and with increasing evidence for relatively shallow magma storage below the volcano. There are no indicators that the current trend might invert in the foreseeable future, and it seems that the stage is set for impressive eruptions in the near future, which might become a serious problem for the population and authorities around the volcano.

But how can this recent upsurge in the activity be placed in the long-term dynamic framework of Etna's behavior? If the volcano will really return to a similar level of activity as in the 17th century, does this mean that there is a cyclic behavior also in the long term, on a scale of centuries? One might feel tempted to say, yes, this is so. But to get the correct answer, it would be necessary to know exactly what the volcano has done before A.D. 1600, and this is virtually impossible, for the reasons named above. It would be necessary to prove that there was an increase in the activity prior to that year like the increase observed today. From the sketchy record available this is not very evident. The last well-documented flank eruptions prior to A.D. 1600 occurred in the 1530s, and for 70 years there is no unequivocal evidence for flank eruptions. A large flank eruption reported for 1595 on the southwestern flank is seriously doubted by Tanguy et al. (1999), for the lava flow dated "1595" on the geological map of Etna actually consists of two distinct lava flows, which were erupted in about A.D. 1060 and A.D. 1200, respectively. There is no evidence for a consistent buildup in the activity before 1600. And at the same time it seems that "cicirara" lavas were erupted for quite some time before the 17th century, including the period of apparently relatively low-level activity prior to A.D. 1600. Establishing a model of the long-term behavior of Mount Etna is thus not all that easy as it might seem from looking only at the last 400 years.

In any case, unless it will be possible to reconstruct the pre-1600 activity with appreciable precision, that period of slightly more than 400 years, starting with the year A.D. 1600, is the only source of reliable data available for analysis and interpretation. For this period at least, it can be established that the eruptive dynamics of Mount Etna were strongly controlled by the way magma was transferred from a deep source area to the surface. During the first 70 years of the 17th century, magma may have been stored in a relatively shallow reservoir lying below the base of the volcanic edifice - maybe at a depth of 5 km below the sea level, in the area where magma accumulation takes place also today - with the resulting emission of "cicirara" lavas. These conditions were abruptly altered by the 1669 eruption, which led to the draining of the reservoir, whereupon transport of magma went directly from the mysterious deep source area to the surface, without any significant temporary storage. Interestingly, this transition was shortly followed by the catastrophic earthquake of 1693 in southeastern Sicily, which caused terrific devastation in much of the area, including the Catania area, with some estimated 60,000 people killed. Some authors (e.g., Hirn et al. 1997) have speculated that there might be a correlation between the magmatic and seismic events in the 17th century, but much more data is needed to put this hypothesis on a solid foundation, though attractive it is.

Mount Etna is currently experiencing the most vigorous phase of activity since the 17th century, and there is no reason to assume that this trend will be interrupted in the near future. Whether the activity will become more similar to that of the 17th century or not, the volcano will continue to produce frequent eruptions, including numerous flank eruptions, both lateral and eccentric, and it is likely that the impressive eruptions of 2001 and 2002-2003 were just a wake-up call, a shape of things to come.

Volcano instability
As if things were not complicated enough, the dynamics of Mount Etna cannot be described in terms of magma storage and uprise alone. A crucial question related to the short-term eruptive cycles observed since 1865 has not yet been answered: why are there such cycles? It is only with geophysical data of the past 20-30 years and with the very limited observations of the preceding decades that one can attempt to find an answer. We now know that the edifice of this volcano (like those of many other volcanoes worldwide) shows a remarkable degree of instability, which is expressed in episodic displacements of a large sector of its flank. Known as flank "slip" or "spreading", such displacements occur where gravitational (or magma-induced) instability exceeds the buttressing (and thus stabilizing) effect exerted by surrounding relief (such as nearby volcanoes or mountain chains). Flank slip is very pronounced at Kilauea volcano on Hawaii and sometimes occurs in displacements of up to 8 m, as in 1975. These events, interpreted until recently to represent possible initial stages of catastrophic sector collapse, are now rather believed to occur at volcanoes where catastrophic collapse is impeded by various factors, and flank instability is expressed in numerous, relatively minor and non-catastrophic events.

At Etna, flank instability was for the first time proposed by two different groups of researchers in the early 1990s (Borgia et al. 1992; Lo Giudice and Rasà 1992). The earlier authors suggested that slippage affected an area extending from the east-west trending Pernicana Fault (now rather called Pernicana Fault System) to the north to the Ragalna fault system on the south-southwestern side of the volcano, encompassing the eastern, southeastern and southern flanks. In their model, the basal detachment plane is located at a depth of about 6 km below sea level, and instability resulted largely from lateral pressure exerted by magma accumulation in what they called a "plutonic complex" centered below the summit of Mount Etna. Lo Giudice and Rasà (1992) also saw evidence for displacement of the eastern flank but placed the basal detachment plane at much shallower depth, at the boundary between the volcanic pile of Etna and the sedimentary substratum, around 0-1 km above the sea level. Several years later, Rust and Neri (1996) essentially confirmed the views expressed by Borgia et al. (1992), but more recently, Bousquet and Lanzafame (2001) forwarded a quite different hypothesis, with a very shallow detachment plane located 1-2 km above the sea level, and flank instability being mainly a result of shallow magma intrusion during flank eruptions. In a model that aims at reconciling the earlier and strongly differing hypotheses, Tibaldi and Groppelli (2002) proposed that flank slip actually occurred above two different detachment planes located both at shallow and deep levels.

Things became abruptly much clearer when Mount Etna spoke itself, in late 2002. Continue


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

Page set up on 28 September 1999, last modified on 11 October 2003

 

Hosted by VolcanoDiscovery