Etna Week (Part 2) - The current dynamics and activity of Etna

This is Part 2 of 3 from guest blogger Dr. Boris Behncke. Check out Part 1 here.

The current dynamics and activity of Etna
by guest blogger Dr. Boris Behncke

The recent behavior of Etna is characterized by nearly continuous eruptive activity from the summit craters and eruptions from new vents on the flanks at intervals of a few years to decades. Summit eruptions vary from quiet lava emission to mild Strombolian explosions to high-discharge-rate Hawaiian to sub-Plinian style lava and fire fountaining accompanied by the emplacement of fast-moving lava flows; usually the strongest activity concentrates in episodes lasting from a few tens of minutes to a few hours. Most flank eruptions are predominantly effusive - that is, characterized by the emission of lava flows, and explosive activity during these events is often limited to Strombolian explosions or mild spattering. This leads to the emplacement of extensive lava flow-fields and only minor cones are built at the eruptive vents, including the smallest constructive volcanic features, called hornitos (Spanish: "small furnaces").

Pyroclastic constructs on Etna come in all forms and sizes. The photograph at top shows a cluster of steep, narrow spires a few meters tall built up around small vents by the ejection of liquid blobs of lava (spattering activity), seen against the backdrop of the huge composite cone of the Southeast Crater, one of the summit craters of Etna. The conspicuous yellow hue is from sulfur deposits. The bottom photograph shows the largest pyroclastic flank cone formed during the historical period, Monti Rossi at about 700 m elevation near the village of Nicolosi, on the south flank of Etna. The name, literally, means "the red mountains" - the plural stands for the two summit peaks, the crater actually lying between them; but the original name - Monte della Ruina, "mountain of devastation" - more accurately refers to the catastrophic impact of this eruption. The cone is approximately 250 m tall from base to top. Photos taken in 1999 and 2000 by Boris Behncke

Some flank eruptions, however, show much more intense explosive activity, like the 2001 and 2002-2003 eruptions, and a number of previous eruptions as in 1852-1853, 1879, 1886, and 1892. Conspicuous pyroclastic cones (also called cinder cones or scoria cones) are formed during such explosive activity, which can be up to a few hundred meters tall, like the prominent double-peaked Monti Rossi cone formed during the unusually voluminous and explosive 1669 eruption on the south flank of Etna. A typical feature of the flank vents is that each erupts only once, like the eruptive centers in monogenetic cone fields worldwide (e.g., the famous "new volcano" ParÃcutin in Mexico, 1943-1952). As a matter of fact, the numerous pyroclastic cones of Etna could be considered a monogenetic cone field, were it not for the huge central volcano on whose flanks they are sitting.

Flank eruptions represent a considerable hazard for the populated areas on the lower flanks of the mountain, which are home to approximately one million people. During the historical period, new flank vents have occasionally opened within those areas that are now densely urbanized, especially on the southern and southeastern flanks, most recently in 1669 near the village of Nicolosi. During the past 1000 years, lava flows have reached the coast of the Ionian sea on three occasions, in ~1030, 1224, and 1669. The map below shows the extent of historical lava flows, distinguishing those of flank eruptions (in different shades of pink, yellow and red) from those emitted during summit eruptions (in green). It is evident that summit lava flows have never come anywhere close to the populated areas and therefore summit activity poses no immediate threat to those areas.

Map of historical lava flows of Etna, distinguishing summit from flank eruptions. Modified from Crisci et al. (2010)

Evolution of the summit area
The summit area of Etna has undergone profound changes in the past century. Until 1911, there was a single large crater at the summit, about half a kilometer wide, and truncating a broad cone about 300 m tall, which had grown since a major summit collapse accompanying the large 1669 flank eruption. This crater was known as the Central Crater. At the beginning of the 20th century, it was a funnel-shaped pit about 200 m deep, but intermittent eruptive activity on its floor led to its gradual filling, and in the mid-1950s, lava flows for the first time overflowed from the Central Crater onto the upper flanks of the volcano. Vigorous activity from several vents within the crater in the early 1960s led to the complete filling and obliteration of the Central Crater, and two large cones built up around the two main vents, the Voragine ("Big Mouth") that had been present since 1945, and a smaller vent known as "the 1964 crater". In 1968, a third vent opened, which became known as Bocca Nuova ("New Mouth"), and which progressively enlarged in diameter mostly due to the caving in of its unstable rims.

For much of the 1970s to 1990s, the evolution of the Voragine and the Bocca Nuova was characterized by periodic intracrater activity and rim collapse, leading to their growth in diameter, until the two pits began to coalesce with only a thin septum remaining between the two, known as the "diaframma" (diaphragm). During a period of exceptionally intense summit eruptions in 1997-1999, both craters were filled to overflowing before subsidence of magma in their conduits led to the formation of new collapse pits, which gradually enlarged and coalesced into a single large depression, Etna's new Central Crater.

Aerial views of Etna's summit area showing the evolution from the single Central Crater in the early 20th century to the currently four summit craters. Top photograph was taken in the 1920s, when the Northeast Crater was already present (but is barely discernible in this view); the size and depth of the Central Crater is well recognizable here. The view is from the west. Center photo is of 1961 and shows the Central Crater filled-to-overflow with pyroclastic cones and lava; the much smaller Northeast Crater is seen behind the Central Crater to the left. The view is from the south. Photograph at bottom was taken in May 2008, the Bocca Nuova and Voragine are in the upper center, nearly coalescing into a new Central Crater, whereas the Northeast Crater is emitting a dense white vapor plume at right, and the Southeast Crater is at center left, showing conspicuous light-colored sulfur deposits lining its rim. The view is from the east. Photographers for top and center photographs unknown, bottom photo by Stefano Branca (INGV-Catania)

In the spring of 1911, a collapse pit opened at the northeastern base of the central summit cone, from which issued a vapor plume but which showed no eruptive activity until 1917. This pit became known as "the Northeast subterminal Crater" (the term subterminal is applied to eruptive vents lying close to Etna's summit craters and showing a eruptive behavior different from the vents of flank eruptions); it is now called Northeast Crater. The new crater remained a pit until 1923, when a small cone grew within and filled the pit, leading to the first lava overflows from the Northeast Crater. In the 1950s, cone growth intensified, as the crater became the site of virtually continuous, mild Strombolian activity accompanied by slow lava emission; this type of activity was termed "persistent" and for a long time was believed to represent the most common type of Etnean eruptive manifestation. In 1977, however, the Northeast Crater switched to a more dramatic form of volcanism, which proved highly efficient in making it become the highest point on Etna - brief but violent episodes of high lava fountaining with voluminous, fast-moving lava flows and tall tephra columns.

The Northeast Crater showing different types of eruptive activity. Top photo, taken in 1969, shows the cone of the Northeast Crater nearly as tall as the rim of the former Central Crater (in the foreground), and displaying weak Strombolian activity from its summit, while lava quietly issues from a small crack on the left side of the cone. This activity lasted with few interruptions from 1955 until 1971, and again from 1974 until 1977. Photographer T. Micek (?). The bottom photograph shows one of about twenty episodes of violent fire fountaining and tall tephra plumes that occurred between July 1977 and March 1978; this was one of the latest episodes of that series. View is from the village of Monterosso on the southeast flank of Etna, photo by Carmelo Sturiale.

By 1978, the Northeast Crater had grown to about 3340 m elevation and thus become the highest point ever measured on Etna. It produced a few more episodes of lava fountaining in late-1980 and early-1981, which brought its height to 3350 m. On 24 September 1986, an unprecedentedly violent eruptive episode caused a reduction in height by 10 m of its cone, and further collapse occurred throughout the following decade. Although the Northeast Crater went through another phase of intense activity in 1995-1996, its height continued to decrease, and in 2007 was 3329.6 m (Neri et al., 2008).

The Southeast Crater seen from the air immediately after its formation in spring 1971 (top) and in May 2008 (bottom). Note that the field of view in the latter photo is much wider than in the earlier. Photos taken by Carmelo Sturiale and Boris Behncke

The latest addition to Etna's summit crater family is the Southeast Crater, which formed during a flank eruption in May 1971 at the southeastern base of the central summit cone as a sort of pressure valve - while lava was emitted a few kilometers further downslope to the northeast, it emitted vapor-rich ash clouds for a couple of weeks. It then remained quiet until spring 1978 and then sprang to life with high lava fountains accompanying a series of flank eruptions in rapid succession - April-June, August, and November 1978, and August 1979. Since then, it has been the most persistently active vent on Etna, and its appearance on the stage was accompanied by a marked change in the eruptive behavior of the volcano. As a matter of fact, since the birth of the Southeast Crater, Etna has practically doubled its average output rate (Behncke and Neri, 2003a).

The Southeast Crater has grown much more rapidly than the Northeast Crater, and nearly 40 years after its birth its cone stands approximately 300 m above the site where it came to life in 1971, reaching a height of 3290 m as of 2007. This rapid growth is the result of numerous periods of frantic eruptive activity which are unparalleled in the documented history not only of Etna but of all volcanoes on Earth. The culmination was a series of 64 episodes of violent lava or fire fountaining between January and June 2000, followed by two more in August and 16 more in May-July 2001 (Behncke et al., 2006). The Southeast Crater has erupted more recently in 2006 and 2007-2008, again producing numerous episodes of strong Strombolian activity and lava fountaining, the latest - and possibly most violent - on 10 May 2008, when lava flows advanced 6.4 km in 4 hours, an unprecedented value for Etnean summit eruptions.

What are the reasons for such variable and, for a basaltic volcano, often unusually violent explosive behavior?

Eruption types and styles
It seems that much of the explosivity of Etna is driven by magmatic gases, foremost water vapor (H₂O) and carbon dioxide (CO₂). Etna is emitting significant quantities of these gas species, up to 200,000 metric tons of water vapor and about 20,000 tons of carbon dioxide per day. Eruptions tend to be more explosive when magma rises fast, which is the case when batches of new primitive magma enter into the plumbing system of the volcano, so that the most explosive eruptions of Etna in the past few thousand years have also produced the most mafic magmas (Coltelli et al., 2005; Kamenetsky et al., 2007). In particular, a powerful sub-Plinian eruption about 3930 years before present produced picritic magma, which was also extremely enriched in CO₂. In contrast, the Plinian eruption of 122 BC was apparently triggered by the sudden decompression of the magmatic system, which led to the catastrophic exsolution of gas although the pre-eruptive water content of the magma was found to be only about 1 weight-% (Del Carlo and Pompilio, 2004).

Hypothetical and simplified scheme of the magmatic plumbing system of Etna, illustrating magma transport feeding summit activity and the two different types (lateral vs. eccentric) flank eruptions, from Behncke and Neri (2003b)

Most magma ascends to the surface through the central conduit system of Etna, which leads to the frequent summit activity. Unless magma ascent is very rapid, much gas is lost from the magma during its ascent to the surface, and significant volumes of relatively gas-poor magma are stored in the shallow plumbing system of the volcano. During many flank eruptions of Etna, such gas-poor magma exits laterally from the central conduits, resulting in relatively weak or almost no explosive activity but copious lava outflow. Most flank eruptions during the 20th century were of this type; they are commonly called "lateral" flank eruptions. Typically such eruptions are accompanied by the cessation of summit activity and some collapse at the summit craters, as the central conduit system is drained of magma.

Extremes in eruptive styles at Etna: totally non-explosive extrusion of gas-poor lava near the Southeast Crater in 1999 (top), and the 10-km-high eruption column formed during a sub-Plinian eruption from the Voragine on 22 July 1998 as seen from Catania. Photos taken by Boris Behncke and Sandro Privitera

Another type of Etnean flank eruption is characterized by much more pronounced explosive activity, resulting in the emission of significant volumes of ash even for prolonged periods of up to several months, as in 1892, 2001, and 2002-2003. These eruptions occur when magma, rather than rising through the central conduits, pushes its way forcefully through the flank of the volcano to form new conduits called "eccentric" or "peripheral" (Rittmann, 1964; Neri et al., 2005). Being in a closed system until eruption, the magma does not lose significant amounts of its gas during ascent, and therefore the ensuing activity is considerably more explosive. The 1974 and 2002-2003 eccentric eruptions did in fact produce more tephra than lava (Andronico et al., 2004; Corsaro et al., 2009), belying the widespread notion of Etna being a rather non-explosive volcano!

Eruptions and flank instability
The question why Etna makes flank eruptions at all is not easy to answer. Certainly the fact that the volcano lies above the intersection of several main regional fault systems helps in rendering its flanks unstable and subject to fracturing. Mazzarini and Armienti (2001) demonstrated that the distribution of Etna's flank cones is largely controlled by intersections between tectonic lines of weakness. It has also been suggested (e.g., Chester et al., 1985) that the hydrostatic (or rather "magmastatic") pressure exerted on the conduit walls by the rising magma column within the conduit might lead to the opening of lateral cracks through which the magma could escape to feed flank eruptions. Bousquet and Lanzafame (2001) specified that magma transfer from the central conduits into the flank occurred in a more or less horizontal manner, rather than rising upward vertically. All of these scenarios concerned exclusively lateral flank eruptions, not eccentric ones, which had effectively been nearly forgotten prior to the 2001 and 2002-2003 eruptions.

The Pernicana fault cuts through the northeastern flank of Etna, from an elevation of about 2000 m at the Northeast Rift, down to sea level near the village of Fondachello. From Neri et al. (2004)

Displacement along the Pernicana fault during the massive flank movement of 2002, along the Fornazzo-Linguaglossa road (top) and the Catania-Messina highway (bottom). From Neri et al. (2004)

Since the early 1990s scientists (Borgia et al, 1992; Lo Giudice and Rasà, 1992; Rust and Neri, 1996; Bousquet and Lanzafame, 2001) proposed that a large portion of the volcano, encompassing its eastern and southern flank sectors, was subject to lateral sliding, much in the same manner as the southern flank of Kīlauea on Hawai'i. There was some debate as for the cause of the sliding - was it caused by gravitational pull, the push of accumulating magma below the volcano, or by more shallow intrusion of magma into the flanks? Also the extent of the mobile sector was not unanimously defined; whereas there was agreement that the northern boundary of this sector was defined by the transcurrent (mostly horizontally moving) Pernicana fault, the southern or southwestern boundary was variously attributed to different fault systems cutting the southeastern and southwestern flanks of Etna. It is now known that the extreme southwestern boundary is the Ragalna fault system (Rust and Neri, 1996; Rust et al., 2005; Neri et al., 2007).

Speculation became truth in the fall of 2002, when a large sector of the eastern and southeastern flank of Etna underwent a massive move toward the Ionian Sea. During a powerful and complex flank eruption in the summer of 2001, the southern flank and summit area of the volcano were violently ripped open, and the eastern flank started to move away from the remainder of the mountain at accelerating speed. Though this was recognized only in hindsight (Bonforte et al., 2008, 2009; Puglisi et al., 2008), many of us were convinced that the 2001 eruption had significantly destabilized the volcanic edifice, and that further flank eruptions would occur from now on in rapid succession.

On 24 September 2002, a shallow earthquake occurred on the northeast flank of Etna, along the upper portion of the Pernicana fault system, which had been very active during the 1980s but had not shown any seismic activity or significant displacement since 1988. The earthquake was accompanied by conspicuous ground rupturing along the fault, similar to numerous events between 1980 and 1988. A few weeks later, on 27 October 2002, a more pronounced movement along the fault heralded the onset of a new flank eruption, which affected both the south and northeast sides of Etna and destroyed numerous tourist facilities as well as forested areas. During a few days, a part of the northeastern flank moved by more than 2 m eastward; then the movement extended over an ever larger area to the southeast side of Etna, where earthquakes accompanying the displacement caused severe damage in several villages, such as Santa Venerina and Milo.

Top: The sector of Etna's eastern to southern flanks affected by flank instability and displacement is shown in pink. PFS = Pernicana fault system; VB = Valle del Bove; RN = Ripe della Naca; ZE = Zafferana Etnea; SV = Santa Venerina; TFS = Timpe fault system; AC = Acireale; TF = Trecastagni fault; R = Ragalna fault system. From Neri et al. (2004). Bottom: Distribution of earthquake epicenters accompanying the 2002 eruption and flank movement helped to distinguish several blocks (Blocks 1, 2 and 3) within the unstable sector, moving at different times and speeds. From Neri et al. (2005)

This immense mass movement, which was later revealed to have involved about 2000 cubic kilometers of rock (Walter et al., 2005), both of the volcanic pile and of the underlying sedimentary basement, was documented in extreme detail, thanks to improved monitoring equipment placed on the volcano a few years before. It could thus be established that the movement started at the Pernicana fault in the northwestern portion of the moving sector, and then extended both eastward - to the Ionian coast - and southward, affecting numerous fault systems cutting through the eastern and southeastern portions of the volcano. In the entire area, earthquakes were strongly felt and often caused damage, and cracks ripped through buildings and other constructions as well as roads.

Since the fall of 2002, the movement of Etna's eastern flank has continued, most of the time at somewhat reduced speed, but often with new accelerations accompanied by shallow earthquakes. Since 2004, the southern block in the unstable sector has started moving slowly southward. At the Pernicana fault, dramatic slip accompanied by earthquakes and rupturing of the ground surface has occurred several times in 2003 and 2004, and again in early April 2010. This all indicates that the volcano has not yet returned to a state of relative stability and equilibrium as before 2002 (or 2001, if we consider the eruption of that year a significant factor in destabilizing the volcano). As a matter of fact, the behavior of Etna has changed profoundly since 2001.

Fluctuations in Etna's eruptive behavior since 1600 AD, with marked variations in the frequency, style, and size (volume) of eruptions. The output rate was exceptionally high from about 1607 until 1669, when ten - most of them very large - flank eruptions occurred (see black vertical bars in the graph at the bottom of the figure) and up to 3 cubic kilometers of magma was erupted. Very low output and few flank eruptions are seen during the following ~100 years, until the 1760s when flank eruptions pick up in frequency and size. A marked acceleration in the activity of Etna is evident starting in the second half of the 20th century. Unpublished figure by Boris Behncke and Marco Neri

Eruptive cycles
If one looks at the historical record of Etna's eruptions, it becomes evident that the intervals between these events, as well as their characteristics (duration, location, volume, eruptive style) vary strongly. Unfortunately the record is complete only since the beginning of the 17th century, yet these past little more than 400 years show remarkable fluctuations in Etna's activity. The first 70 years of the 17th century showed unusually high levels of activity, with frequent summit activity and ten flank eruptions. Some of these flank eruptions lasted for years - the one of 1614-1624 being the longest flank eruption in the historical record of Etna - and produced large volumes of lava (1614-1624: about 1 km3, 1634-1638: about 200 million m3, 1646-1647: about 160 million m3, 1651-1653: about 450 million m3, 1669: about 650 million m3). A few of the flank eruptions were quite explosive and built large pyroclastic cones, like Monte Nero during the 1646-1647 eruption and Monti Rossi in 1669.

The last eruption in this series, in 1669, apparently emptied a shallow magma reservoir that had existed throughout the previous decades - evidence for such a reservoir lies in the presence of abundant up to centimeter-sized plagioclase feldspar crystals in the lavas of all eruptions from 1600 until 1669. The rounded shape and pale yellow color of these crystals has led the locals to call the lavas of this period "cicirara", which means something like "chickpea lava", because the crystals resemble chick peas! Due to a prolonged presence in a reservoir relatively close to the surface, the magma could cool and crystallize to the degree that plagioclase grew to the "chick pea" size crystals seen in the 17th century lavas. Further evidence for wholesale magma extraction from a shallow reservoir and a dramatic withdrawal of the magma column in the central conduit is the collapse of Etna's summit cone during the 1669 eruption (Corsaro et al., 1996).

After the 1669 eruption, Etna has never again produced "cicirara". Furthermore, the frequency and size of flank eruptions dropped sharply for about 100 years, with only three minor flank eruptions being recorded in 1689, 1702, and 1755. It seems that the magma reservoir that had fed the intense activity of the 17th century had disappeared, the feeding system of the volcano had been disrupted, and the mountain had become structurally stable. Much of the time, all magma that made it to the surface rose to the summit, where a new cone was constructed. Flank eruptions became frequent again from 1763 on, and for the next 100 years occurred about once per decade, with volumes of a few tens to rarely more than 100 million cubic meters per eruption.

Interestingly, throughout the 18th and the first half of the 19th centuries, there are no records of significant earthquakes in the unstable eastern sector of Etna as those of the 1980s and of 2002 and the following years. A powerful and destructive earthquake in 1818 near Acireale was probably caused by movement along a regional tectonic fault, not by movement of Etna's unstable flank.

Etna's unstable eastern to southern flank sector, and a selection of earthquakes presumably caused by movement of this unstable sector. Note that there have been many more earthquakes in this area during the period since 1865, when the first of these events took place. Unpublished figure by Boris Behncke

Then came the year 1865, which brought a large eruption on the northeast flank - the Monti Sartorius eruption - and soon after its end, a very localized, extremely shallow (close to the surface) earthquake on Etna's eastern flank, which devastated the village of Macchia di Giarre and killed about 70 people. Similar earthquakes have since then occurred at a recurrence rate of a few years, luckily rarely resulting in as many fatalities, but often causing significant damage and a few human deaths. Most, if not all, of these earthquakes are now known to be related to slippage, or movement, of Etna's unstable eastern to southern flank sectors.

At the same time, the intervals between flank eruptions have become systematically clustered into determined sequences, or parts of cycles. The first cycle started after the large 1865 flank eruption (and the first earthquake in modern time that can be with confidence attributed to flank displacement), initially with quiet emission of gas from the Central Crater and a few years later, mild activity within the Central Crater. From 1874 until 1892 there were five flank eruptions, which showed an overall increase in emitted volume in time, the latest - in 1892 - being the most voluminous (besides 120 million m3 this eruption also produced a significant amount of pyroclastics). This increase in the volume of flank eruptions was apparently the result of increasing structural instability of the volcano. Apparently the emission of a large volume of magma brought this cycle to a close, the volcano returned to relatively stable conditions, and a new cycle started, like the previous one, with a period of quiescence, followed by summit activity, which in turn was followed by a further series of flank eruptions. Four cycles of this type occurred between 1865 and 1993. The latest of these was longer than its predecessors - 42 years - and culminated in a series of no less than 13 flank eruptions, many of which were among the largest of the past 300 years. This cycle ended with the 472-days-long eruption that lasted from December 1991 until March 1993 and produced the greatest lava volume - about 250 million cubic meters - of any Etnean eruption since 1669.

Evolution of the 1952-1993 eruptive cycle at Etna, showing three main phases (Eruptive quiescence -> summit activity -> flank eruptions, ending with a particularly voluminous flank eruption). Unpublished figure by Boris Behncke and Marco Neri

Interestingly, there was very little seismic activity in Etna's unstable sector during the first two phases of this cycle, whereas they became more and more frequent during the third phase. Many episodes of accelerated flank displacement preceded flank eruptions by days to months, as in 1981, 1983, 1985, and 1989.

Etna's latest and ongoing cycle started after the end of the large 1991-1993 flank eruption. For two years, no eruptive activity occurred anywhere on the mountain. Then, in summer 1995, eruptive activity returned to the summit craters - first at the Bocca Nuova and then at the Northeast Crater; in 1996 and 1997 also the Southeast Crater and the Voragine joined the party (Allard et al., 2006). This period of summit eruptions continued until July 2001 and consisted of several long-lasting lava overflows and more than 150 episodes of violent Strombolian to sub-Plinian explosive activity, nearly always with copious lava emission. We called this exceptional period of activity "The Millennium Fireworks". Still more exciting fireworks came with the flank eruptions of 2001 and 2002-2003, and two more flank eruptions have occurred in 2004-2005 and 2008-2009, separated by a period of spectacular eruptions from the Southeast Crater in 2006-2008. These events are described in detail in the Bulletins of the Global Volcanism Network (scroll down the page to get to the more recent reports).

What should be noted is that since flank eruptions kicked in again in 2001, the unstable flank sector of Etna has moved at sometimes astonishing rates (up to several tens of centimeters in a few days in spring 2009), and seismic activity in this sector has been intense, including a series of rupturing events at the Pernicana fault as recently as April 2010. The volcano seems to be currently in the middle of an eruptive cycle, and it is likely that this will come to an end (and bring back the volcano to a state of temporary stability) only with a very large, voluminous flank eruption (Behncke and Neri, 2003a; Allard et al., 2006). From a scientific point of view this is rather exciting. From a human (and civil defense) point of view, these prospects are rather disconcerting and challenging.

So why does the flank of Etna move? It is now believed that much of the movement is caused by the pressure of magma accumulating within the volcano. As a matter of fact, much more magma enters into Etna's plumbing system than exits during eruptions. The quantity of this unerupted "excess" magma can be approximately calculated from the amounts of gas emitted from the volcano, in particular sulfur dioxide. It has thus been revealed (Spilliaert et al., 2005; Allard et al., 2006) that at least three-quarters of the magma that enter into the Etnean feeder system stay there, which leads to a constant volume increase. Where does all this magma go? There are certainly no empty spaces that can host this magma, so space must be created, and this is best done in pushing the volcano, both upwards (so that the volcano swells, or inflates), and sidewards, in whatever direction the side of the mountain gives way most easily. At Etna this is on the eastern, southeastern, and to a lesser degree, southern flanks, which are not buttressed by surrounding mountains as the northern and western flanks. It can be speculated that the more magma accumulates below the volcano, the more unstable it becomes, and this in turn facilitates the opening of fractures on the flanks, allowing magma to escape in flank eruptions. Possibly the presence of a large, relatively shallow magma reservoir during the 17th century led to a strong destabilization of the volcano, which thus had magma leaking through its open flanks on any given occasion, and in large volumes. A similar situation seems to be on the way to become established in recent decades - so no one would be really surprised to see Etna behave again like it did between 1600 and 1669, but once more, these are anything else than comforting prospects.