Aeolian Islands, Italy (38.79 N, 15.21 E)

(This report covers the period 16.05.1996 - 31.1.1997)

As reported in last contribution to the GVN (Bulletin, v.21, no. 4), Stromboli volcano started an intense eruptive phase on 16 April 1996, during which the activity was marked, from the seismik point of view, by events that followed each other very rapidly. The number of events suddenly passed from a "normal" value of 286 on 15 April to a value of 540 on the following day, and reached a striking 100% of minutes in which the trigger procedure signalled an event on 22 April. The average number of events per day between 16 April and 1 June 1996 was about 800. This means, as the trigger is done on a "per minute" basis, that the volcano was erupting for more than 50% of the time. Visual observations made by Stromboli volcano guide N. Zerilli confirmed that the NE Crater (i.e. Crater 1) was responsible for most of these events, which were generated by a moderate but almost continuous strombolian activity sometimes resembling real lava fountaining (an example of this unusual continuous spattering activity can be seen in the video made by R. Carniel on 28 April 1996 and made available by J. Alean and R. Carniel on STROMBOLI ON-LINE website at the URL http://www.ezinfo.ethz.ch/volcano/verschiedenes/movies/str7.mp4). Noteworthy during this period was also the number of more energetic events, i.e. the ones that saturate our acquisition system. The average number of such events in the period 16 April - 1 June was about 90 per day, exceeding 10% of the total number of recorded events.

The period of strong strombolian activity was interrupted on 1 June 1996, when a violent explosion was observed at 21.50 GMT. N. Zerilli reports that the explosion was clearly heard and seen from the village of Stromboli, and that it threw very large blocks to a considerable distance from the craters, especially into the zone of the "canneto", uphill from the village of Stromboli-San Bartolo. A bush fire was lit by the incandescent pyroclastic material. On STROMBOLI ON-LINE, at the special events page (http://www.ezinfo.ethz.ch/volcano/verschiedenes/besonderee.html), a photo can be seen, taken from Punta Labronzo only few minutes after the eruption (courtesy of Wolfgang Müller). Although information is quite contradictory, it seems that some tourists were surprised by the explosion while sleeping near the summit, and got injured. Reports suggest that the explosion was more violent than the other ones recorded in recent years (10 February and 16 October 1993, 5 March 1995, 16 February 1996). A sudden decrease of the seismic activity followed the explosion like in many similar cases in the past, with the number of events going down to 276 on 2 June (On 31 May the number was 1097), the major events to 10 (119 on 31 May) and the tremor intensity to 5.6 Vs. For a comparison, on 31 May the tremor intensity was estimated at 13 Vs but this value is somehow questionable as in this period it is very difficult to apply the "usual" definition of tremor as "the ground movement in the absence of events", as the events followed each other very rapidly.

The relative calm in the activity did not last for long: on 7 June the number of events already exceeded again the value of 500 and, after another short drop during the last days of June, it reached another climax between 29 July and 3 August (1440 triggering minutes over 1440!). The number of major events recorded during this climax is particularly impressive (250 per day on the average, with a peak of 334 on 2 August). For comparison, the average total number of events per day between 1 January and 15 April 1996 was 238.

During the first half of August the activity increased also at SW crater (Crater 3), in particular at vent 3/1, where the magma level became more and more shallow until it produced a very small lava flow on 16-17 August; the flow proceeded for only a few metres inside the crater. In the evening of 18 August N. Zerilli observed that it had already stopped. Only four red incandescent holes in a small cone were visible, but no more strombolian activity at this vent. During the days of the small lava flow in SW crater the level of the magma was very high also in NE Crater.

On 22 August 1996 a tourist was hit by a bomb at 2:30 AM local time, while sleeping in his bag just 80 m from the craters. He was part of a group and, after the accident, had to be transported by helicopter to Messina for head surgery. Apparently this was not a particularly violent eruption but simply the result of the person being too close to the craters. N. Zerilli reported that during the preceding days ejecta from SW Crater (Crater 3) had already reached the lowest of the series of observing sites between the Pizzo and SW Crater.

During the days after the accident (23-28 August) our seismic station highlighted another period with a noteworthy number of major events, with a peak of 255 on 26 August (more than 70% of the total number of events!). These days also marked a significant change in the activity at the craters, as reported by N. Zerilli. Vent 3/1, which was the source of the small lava flow, became completely inactive. The activity at vents 3/2 and 3/3, almost joined, consisted of very powerful strombolian explosions, which sometimes managed to send small fragments on the Pizzo. So, the SW Crater (Crater 3) was the source of the high number of saturating events. However, the most obvious change in the activity was seen in NE Crater (Crater 1), where vent 1/4 showed only rare explosions, but with strong gas jets, and vent 1/3 stopped the continuous activity which had begun on 16 April 1996. This explains why the total number of events decreased to more "normal" values (less than 400 events per day).

29 August marked the beginning of a new period, with more usual values for the number of saturating events (now less than 15 per day); the total number of events showed values between 150 and 300 per day, and the tremor intensity dropped to moderate values, below 2 Vs. On 4 September 1996, at 13.45 GMT, another violent explosion was recorded at the craters. The blast threw incandescent pyroclastic material in almost every direction. Some of it again reached vegetation on the volcano slopes, starting several fires. As N. Zerilli reported, the explosion was clearly heard and observed from the village of Stromboli. Once again, the reports about the presence of injured people are quite contradictory. On the special events page of STROMBOLI ON-LINE a photo of the explosion, taken from the village of S. Vincenzo by Antonio Di Sarno, is presented courtesy of the author and of Boris Behncke. On 5 September, as a consequence of this new powerful explosion, the Mayor of Lipari ordered the closure of the path to the craters for safety reasons. The access to the summit area is still forbidden at the time of writing of this report (13 February 1997).

After the explosion the strombolian activity continued to decrease. Both the number of major events and the total number of events decreased to very low values, which are still maintained at the end of January 1997.

A visit of the crater area was made by R. Carniel at the end of September. The activity was extremely low; moreover, smoke often covered most of the crater area. Only few eruptions could be seen (or heard), and of very moderate intensity. Remarkable were some very long rumbles, which could be heard even from the village of Stromboli. These rumbles were probably associated with strong degassing from Crater 2, although this interpretation cannot be confirmed by direct observation because of poor visibility. Fig. 2 illustrates the change in crater morphology between April 1996 (Fig. 2a) and September 1996 (Fig. 2b). These sketches were drawn by J. Alean on the base of simple terrestrial stereo photographs taken by J. Alean (a) and R. Carniel (b). Unfortunately the morphological changes in Crater 1 (which usually sees the continuous construction and destructions of several small cones) are not inferrable from the photographs taken in September, because smoke always covered the northeasternmost section of the crater terrace. However, changes in the other two craters are easily seen in the sketches. The most striking ones affected Crater 3. The front vent 3/3, which was a sort of a pit in April, had the appearance of a small cone in September, surrounded by a lot of pyroclastic material ejected by the same vent during these 5 months. In the westernmost part of Crater 3 vents 3/1 and 3/2 merged into a single great chasm, characterized by fairly steep edges and well separated from the rest of the crater.

The "record" of low activity goes to the day 10 November, when the seismic station recorded only 17 events within 24 hours (none of them saturating the acquisition system). Several other days also showed exceptionally low values. Remarkable were also some long continuous periods without a single recorded event (7 hours on 25 October, 8 hours on 10 November, 10 hours during the night between 12 and 13 November). The tremor intensity slowly increased from September 1996 (1-2 V.s) to January 1997 (3-5 V.s). No significant increase of the number of recorded events or of the saturating events was observed during this period (the maximum was recorded on 6 January, with 160 events). At the end of January the activity can therefore still be defined as "low to moderate".

Fig 1. caption:

Seismicity recorded at Stromboli, 15.05.1996 - 31.01.1997.

open bars = number of recorded events per day

solid bars = saturating events, i.e. with ground velocity exceeding 100 µm/s

line = daily average of tremor intensity on hourly 60-seconds samples.

Arrows highlight the two powerful explosions of 1 June and 4 September 1996.

Courtesy of Roberto Carniel.

Fig 2. caption:

Sketches drawn on the base of terrestrial stereo photographs indicating morphological changes of the crater area between April 1996 and September 1996.

a) 25.4.96. 3/2? and 3/3? vents are indirectly inferred from trajectory of ejecta

b) 19.9.96. ? indicate areas impossible to see because of fumaroles

Courtesy of Jürg Alean.