Montserrat Volcano Observatory, Montserrat, West Indies

Scientific Report 37
05 October 1996

This report covers the 1-week period to 24:00 Saturday 05 October. Future Scientific Reports will follow this convention. This is a slight change from previous reports, which covered the period to 16:00 Saturday.


A new dome is growing within the crater of Soufriere Hills volcano. The dome first appeared on 1 October, at the base of the large scar caused by the explosion of 17/18 September. The rate of dome growth is similar to that observed before the explosion. Dome growth has been virtually aseismic, particularly in the latter part of the week. A volcano-tectonic earthquake swarm which started during the last reporting period died out at the beginning of the week, with some deeper earthquakes.

Visual observations

There was a period of clear weather from 1 to 4 October, which allowed excellent views and measurements of the dome complex. Limited views of the dome were also possible on 29 September and 5 October.

The major new development has been the appearance of a new lava dome in the scar left by the 17/18 September explosion. Deformation caused by this dome was first noticed on 29 September, and the dome itself was sighted on 1 October. It is called the "October 1" dome.

The new dome was initially light grey in colour and appeared to be more fluid than previous domes, when it was first sighted. It has a blocky texture, but no spines have appeared. During the early part of the week, the dome developed a flat-topped, steep-sided solid cap, which was apparently raised up intact by later growth (FIGURE 1). As the dome grew, it's base expanded southwards to fill the bottom of the scar. The dome volume increased by a factor of 2.4 from 2 to 3 October, giving an effusion rate of 3.75 m3/s or 324,000 m3/day. This is similar to the rate of about 300,000 m3/day measured before the explosion. Surveys of the dome using GPS and laser-sighting binoculars show that it grew in height from 782 m to 793 m between 1-2 October.

Small rockfalls were observed from the dome for the first time on 2 October, and several small collapses of the eastern face were seen over the following days. An initial talus slope was formed below the cap. None of the resulting talus material escaped from the scar area. Rockfalls also occurred from the back wall of the scar (the western side), and glowing cracks were observed in the west and north side of the scar. Several areas of vigorous steaming have been present at different times, particularly from the base of the new dome.

There has been no evidence of fresh growth in other parts of the dome complex.


The seismicity has remained at a generally low level, a pattern established following the explosion of 17/18 September. The volcano-tectonic (VT) earthquake swarm that began during the last reporting period peaked on 28 September, and VT and hybrid earthquakes had become scarce by the end of the week. The number of rockfall signals has increased during the week.

Earthquake types: 29 September to 05 October 1996
These earthquake counts are of events which triggered the seismic network event recording system. Future reports will report earthquake counts made in the same way. This is the same data set used for the daily reports, except that the daily reports describe the 24-hr period to 1600 each day. The figures here are for 0000 to 2400 each day.

Date      VT        LP        Hybrid    RF        Tremor
29 SEP 96 38        4         51        5         Moderate
30 SEP 96 24        5         10        8         Moderate
01 OCT 96 24        0         2         6         Moderate
02 OCT 96 2         0         0         17        Moderate
03 OCT 96 1         0         0         20        Low
04 OCT 96 1         0         0         20        Moderate
05 OCT 96 0         0         1         19        Low

The VT earthquake swarm ended on 01 October, with a gradual decline in earthquake numbers from 28 September. After 1900 on 29 September, there were some deeper earthquakes, with hypocentral depths of 3-4 km BSL. The shallower earthquakes, with depths 1-3 km (mostly shallower than 2 km), continued at the same time. The deeper earthquakes had higher frequency signals, and some S picks were possible, giving better depth constraint. The number of small hybrid earthquakes decreased throughout the week and, like the VTs, had ceased by the end of the reporting period. These earthquakes were probably caused by a fresh batch of magma starting to rise within the conduit.

The number of rockfall signals has increased, following the appearance of new dome growth on October 1. The cause of these long-period signals is still uncertain. The signals are often associated with observed rockfalls, but the possibility remains that they are due to some internal dome process (gas expansion?) which often results in collapses from the surface of the dome. The signals are similar to "isolated tremor" episodes observed during dome growth at Unzen Volcano, Japan.

The level of tremor during this period has been quite low. The Gages station continues to experience intermittent tremor, thought to be caused by water and steam movement within the volcano.

Several small regional earthquakes have been recorded by the seismic network, indicating an increase in tectonic earthquake activity, probably to the north of Montserrat.

Ground Deformation

EDM measurements have been restricted by low cloud. Measurements were made of the triangle between Chances Peak, Galways and O'Garros on 30 September, 2 and 4 October. The Chances Peak reflector could not be seen from O'Garros, and was cleaned and reoriented on 1 October. The results for this triangle are consistent with continued slow outward movement of the reflector (line shortening). The EDM reflector on Castle Peak was destroyed during the 17/18 September explosion. On 2 October a new reflector was installed at the same site, using the hole that was drilled last October. Measurements to this new reflector were attempted on 5 October, but were not possible due to low cloud.

No GPS surveys were undertaken during this period.

The GPS network operated by the University of Puerto Rico was reoccupied during the reporting period using stations at Hermitage, Chances Peak, Reids Hill and St Georges Hill. The results of these measurements are not yet available.

Gas and ash measurements

The COSPEC instrument is still out of operation, and no recent results from SO2 tube analysis are available. The air and ash sampling programme started recently has continued. Samples are sent to the UK for analysis, and the results will be available shortly.

Other work

A new network of broadband seismic stations is in the process of being installed. The network consists of 5 three-component, broadband and three single-component, short-period sensors, and will initially complement the existing short-period network. The new network stations are intended to be permanent. The signals are telemetered to the Observatory as 24-bit digital data.

MVO scientists have undertaken a complete review of the volcanic hazard and risk associated with the volcano. A revised set of alert procedures and a detailed volcanic risk map have been drafted and presented to the civilian authorities. The new Montserrat Alert System defines a set of alert stages and recommends actions which would be required in the event of a further escalation in volcanic activity.

Staff changes
Dr Willy Aspinall, BGS
Dr Paul Cole, BGS
Dr Richard Herd, BGS

Dr Simon Young, BGS
Dr Bill McGuire, BGS
Mr Wilkie Balgobin, SRU

Montserrat Volcano Observatory