The following condenses Scientific Reports of the Montserrat Volcano Observatory (MVO) from 23 November 1997 to 4 January 1998.

Overview. Activity during 23 November-21 December was at a relatively low level, although the 22 October dome (BGVN 22:11) continued to grow on the S side in the Galway's area while the N flanks remained quiet. Seismicity was dominated by rockfall and long-period events, with two periods of pyroclastic flow on 27 November and 1 December. During 21 December-4 January, volcanic activity increased to a high level with a large hybrid earthquake swarm on 25 December followed by a debris avalanche and dome collapse down the White River early on 26 December. A series of very large pyroclastic flows destroyed a wide area SW of the volcano and created a surge cloud that may have been associated with a lateral blast.

Visual observations. Good observations of the dome were possible during 23 November-21 December. A small dome collapse occurred on 27 November at 1445 and continued until after 1800. Although some of the flows reached the sea, most of the collapse was comprised of continuous but relatively small flows that traveled less than 2 km. Further pyroclastic flow activity occurred on 1 December starting around 2000 preceded by 2 hours of elevated rockfall activity. Only the largest of these pulses reached as far as the delta. While the flows occurred, glow was observed in the sky above the Galway's area. Neither period of activity produced ash clouds above 3 km or involved significant volumes of material.

On 4 December observers noticed that the dome's talus apron in the Galway's area had significantly increased in height and width. Continued growth of the 22 October dome had begun to encroach on Chances Peak and the Tar River to the E. The height of the active growth center was estimated at ~980 m, well above the height of the explosion crater rim. Around this time, theodolite surveys confirmed that the N lobe of the 22 October dome had not grown recently. Occasional minor rockfalls occurred on the N and E faces in the Tar River valley as old surfaces of the dome gradually degraded.

Near-continuous rockfalls started to cut into the S margin of Chance's Peak during 7-21 December. The talus slope above the Galway's Soufriere area extended significantly SE and the summit area of Galways Mountain degraded slightly. The growing talus apron extended over the remnants of the Galways wall, depositing a large volume of material outside the old crater margins. The remnants of the pre-September 1997 explosion crater above Gages and Tar River eroded, spilling a small amount of material down the Tar River. Theodolite measurements from Jack Boy Hill indicated no movement on the N flanks. A few small rockfalls occurred down Tuitts Ghaut on 10 December.

The low-level activity that had prevailed for 6 weeks was broken at about 0300 on 26 December. Reports from the police checkpoint in Salem indicated explosions at 0315 and 0325. At 0400, observers on Garibaldi Hill reported light ashfall and burning buildings S of Plymouth. A pilot saw an ash plume heading S at ~11 km altitude. Helicopter flights later in the morning revealed extensive damage in the area of Aymer's Ghaut, S of Kinsale, to the N flanks of South Soufriere Hills. Deposits from a large debris avalanche traveled down the White River to the sea and added considerable material to the pyroclastic fan.

Pyroclastic flows and surges covered from Kinsale almost to the summit of South Soufriere Hills. The area closest to the White River, including St. Patrick's and Morris's, was completely destroyed. A small pyroclastic flow traveled down Dry Ghaut towards Sweeney's Well to within 300 m of the SE coast. Brief views of a void at the top of White River indicated the loss of a large amount of material from the area around the Galway's Soufriere and of a portion of Galway's wall. The new talus apron and S lobe above Galways wall had been excavated. Later flights and visits allowed measurement of the deposit's thickness and temperature (table 24).

Table 24. Thicknesses and temperatures of deposits from the 26 December explosion at Soufriere Hills. Courtesy of MVO.

Observations from Old Road Bay indicated that a large wave came ashore there immediately after the eruption. At about 0300, an observer in the vicinity of the bay's N end reported that the sea was "sucked backwards" before coming onto land near the jetty. Measurements of detritus on the shore revealed that the wave must have been ~1 m higher than the road, but there was little evidence of substantial waves elsewhere along the coast. This wave probably resulted from a debris avalanche entering the sea; the wave was most likely focused in the Old Road Bay area because of the shape of the bay.

On 1 January increased activity included at least one large pyroclastic flow that traveled down the White River to within 1 km of the sea and much rockfall activity from above Galway's Wall. A new talus apron started to accumulate at the base of the remains of Galway's Wall and some large wall-parallel cracks were observed in the Chance's Peak side. Until 4 January, rockfalls created diffuse ash clouds that generally drifted W over Plymouth.

Seismicity. During 23 November-7 December, the decrease in seismic activity that began around 14 November continued (figure 37). However, the lack of hybrid earthquakes in the second week despite a constant number of rockfalls was unusual. The frequency content of the rockfall signals changed slightly towards longer periods probably due to the attenuation of higher frequencies. This indicated a possible change of ground coupling between rockfalls or pyroclastic flows and unconsolidated deposits.

Figure 37. Daily hybrid earthquake counts at Soufriere Hills during 23 November 1997-3 January 1998. Courtesy of MVO.

During 7-21 December seismicity was dominated by rockfall signals and long-period earthquakes. In many cases it was difficult to distinguish between the two event types. A number of the events classified as long-period earthquakes resembled short bursts of harmonic tremor that lasted up to a few tens of seconds and were nearly monochromatic at ~2 Hz, although there were often a few wavelengths at 1 Hz present at the start of an event. Almost all the signals classified as rockfall had a dominant frequency of ~2 Hz, even those clearly correlated visually with rockfalls at the dome. Events were thus classified by the relative importance of the dominant peak, creating a gray area between the types. The rockfall signals may also have included ash venting or degassing; this was supported by a signal recorded during ash venting after an explosion with a corresponding monochromatic seismic signal near 2 Hz.

During 21-24 December seismicity remained relatively quiet. Slightly more hybrids and noticeably fewer rockfall signals occurred than in preceding weeks but it was not indicative of any great changes. The quiet period ended on 24 December when a hybrid swarm began, leading to the large collapse on 26 December (figure 37).

The swarm, which began at 1420 on 24 December, started out sparsely with events every 20 minutes and slowly increased in intensity until about 2000 on 25 December. Individual events also generally increased in amplitude as the swarm progressed, but even the largest were relatively small in amplitude, an order of magnitude smaller than those recorded in early November, for example. At 2000 the hybrids occurred too frequently to trigger the networks and the signal was effectively a tremor signal. The tremor's amplitude increased until about 2300 and then declined until the collapse at 0300 the next day.

At about 0300 a continuous high-amplitude signal lasted ~16 minutes and included several pulses. The signal then continued at reduced amplitude for 9 minutes, during which a roaring sound similar to but louder than that heard after explosions and associated with ash venting. Intervals of monochromatic seismicity at 1.9 Hz were recorded during this time, as was also the case after explosions. The combination of the roaring noise and monochromatic seismicity indicated vigorous degassing after the main pyroclastic flows had finished, suggesting that the conduit was exposed.

After the collapse, seismicity settled into a cyclical pattern with peaks in RSAM every 6-8 hours. Hybrid earthquakes occurred during the periods of raised background amplitude but in most cases there were not enough in any given cycle to constitute a swarm. Two days after the collapse the number of hybrids decreased to previous levels but the cycles in RSAM continued until 4 January.

Ground deformation. GPS surveys of the BIGNET (Harris, Whites, Windy Hill and Broderick's) and LEESNET (Old Towne, WaterWorks, St George's Hill and Lees Yard) networks were taken during 23 November-28 December. During 23 November-7 December the Harris-Whites baseline had recovered half of the shortening that occurred over the last five months; the line was ~2 cm shorter than its pre-June mean. The data suggested a slow rise of the Broderick's site. No clear trends were identified in the data collected on LEESNET.

An 18 December survey of EASTNET (Harris, Whites, Long Ground, Windy Hill, and Hermitage) indicated a 4.5 cm shortening of the Harris-Whites baseline during April-September 1997. Since September the line has returned to within 2 cm of its pre-April mean. It appeared to be stable and the last three baseline measurements were within 4 mm of one another. Hermitage continued to move NE. It has moved 10.7 cm since mid-January 1997; 6.2 cm of this in the last three months, a marked rate increase.

Volume measurements. On 28 November surveys between Mosquito Ghaut in the N and White's Ghaut in the E found that the total volume of material had increased by 24 x 10⁶ m³ since 13 August 1997. The increase included materials from the 21 September collapse, column-collapse flows during explosive activity, and pyroclastic flow activity on the N flank from mid-August to 21 September.

As of 7 December, the volume of erupted material was then estimated at 203 x 10⁶ m³ including the dome, pyroclastic flow material, and estimates of the eruption column material. A previous dome survey measured an extrusion rate of 5 m³/s, down from summer 1997, but the overall trend for the average extrusion rate was still increasing.

A volume survey of the dome on 8 December revealed that the W, N, and E flanks remained mostly unchanged. The S lobe continued its active growth over the whole face. Growth appeared to be caused by the extrusion of large slabs and blocks from the central cleft between the N and S lobes. The talus apron at the base of the S lobe had extended significantly since the last survey, increasing by 26 x 10⁶ m³. Most of the new material accumulated in the Galway's Soufriere region on the S side of the remnants of Galway's wall. Deposits in this area were over 140 m thick. The dome volume was 102 x 10⁶ m³ with the total erupted volume for the entire eruption was estimated at 232 x 10⁶ m³.

A volume survey of the White River valley fan taken on 4 January showed that it had not yet extended into the sea significantly. On the other hand, the steep underwater shelf in this area caused most material that reached the sea to slump off the edge of the fan into deep water where it could not be surveyed. A survey of the material in White River valley was hampered due to unsafe, ashy conditions; however, rough estimates of the amount of material ponded in the valley were 20-30 x 10⁶ m³. Due to poor visibility, the size of the scars in the dome were hard to determine. Estimates for the amount of material lost during the 26 December event were 30-60 x 10⁶ m³, including old Galway's Wall material.

Environmental monitoring. Ambient dust sampling was conducted using a Dust Trak (PM10) instrument at four fixed sites. Each value is an average of the concentration measured over approximately 24 hours. The values at the fixed sites were low from 23 November 1997 to 4 January 1998 except for values at the Catholic school, which sometimes recorded raised levels. The dust concentration was judged to be higher due to proximity of the main road and the large amount of human activity there.

Sulfur dioxide diffusion tubes were re-sited as a result of volcanic activity but showed that in the N of the island there were no significant concentrations of SO₂. On 17 December, scorching of leaves and grass in the Woodlands area resulted from light ashfall.

Geological Summary. The complex, dominantly andesitic Soufrière Hills volcano occupies the southern half of the island of Montserrat. The summit area consists primarily of a series of lava domes emplaced along an ESE-trending zone. The volcano is flanked by Pleistocene complexes to the north and south. English's Crater, a 1-km-wide crater breached widely to the east by edifice collapse, was formed about 2000 years ago as a result of the youngest of several collapse events producing submarine debris-avalanche deposits. Block-and-ash flow and surge deposits associated with dome growth predominate in flank deposits, including those from an eruption that likely preceded the 1632 CE settlement of the island, allowing cultivation on recently devegetated land to near the summit. Non-eruptive seismic swarms occurred at 30-year intervals in the 20th century, but no historical eruptions were recorded until 1995. Long-term small-to-moderate ash eruptions beginning in that year were later accompanied by lava-dome growth and pyroclastic flows that forced evacuation of the southern half of the island and ultimately destroyed the capital city of Plymouth, causing major social and economic disruption.

Information Contacts: Montserrat Volcano Observatory (MVO), c/o Chief Minister's Office, P. O. Box 292, Plymouth, Montserrat (URL: http://www.mvo.ms/).