Eruptive activity . . . began with a phreatic explosion on 18 July that caused ashfall around the island (20:6). Formation of a second vent in the summit crater on 28 July and increased steam-and-ash emissions prompted evacuations from communities near the summit (20:7). Activity was variable but generally low in early August, with small mudflows and continued steaming. Vent 1 reactivated on 11 August, and some earthquakes were centered beneath St. George's Hill, 3 km WNW of the summit. Relatively heavy cloud cover and bad weather prevented observations on many days in late August.

Moderate emissions continued from both vents through 19 August (20:7). An ash-and-steam eruption from Vent 1 at 1220 on 19 August was similar in size to the two previous events that caused ashfall in Plymouth. The eruption lasted for ~35 minutes, during which a 10-minute-long phase of more vigorous activity deposited ~1 mm of ash in areas NW of the vent. Ashfall along the road between Lees and Gages has caused reduced traction; police were advising motorists to drive slower on the slippery surface.

Another small phreatic episode at 1657 on 20 August produced minor ashfall. Since the disappearance of continuous tremor on 19 August, seismicity consisted of low-intensity spasmodic tremor and occasional small volcano-tectonic (VT) earthquakes attributed to very shallow activities under the summit area.

Formation of Vent 3 on 22 August. Seismicity generally increased in frequency and amplitude until the largest single phreatic eruption episode to date at 0803 on 21 August produced a column to a height of ~2 km. Ash blown down the W flank engulfed Plymouth within eight minutes, caused darkness for ~25 minutes, and deposited an estimated 2 mm of ash. An analysis of the new ash showed that it consisted only of old altered material. Other reported phenomenon were projectiles in the Long Ground area and a density current in Gages. Over 70 locatable VT earthquakes were recorded between noon on 20 August through noon the next day; most of these events occurred before the 21 August eruption and were very shallow.

Following a contingency plan, the volcano observatory was relocated to the Vue Pointe Hotel in Old Towne (~ 4.5 km N of Plymouth) at approximately 1600 on the 21st; by 1730 all of the seismic instruments were back on-line. Seismicity was less frequent and vigorous through noon on 22 August, with only four eruption signals of smaller intensity and duration than the previous episodes. Over 5,000 residents were evacuated from the capital city of Plymouth to camps on the N part of the island. Day use was permitted, but restricted. Government offices were also relocated.

Six phreatic eruptions occurred between noon on 22 August and noon the next day. The largest, at 1551 on 22 August, produced ashfall ~3-3.5 km to the SW. This event was followed by three long-period (LP) earthquakes (events associated with the movement of pressurized fluids). When the crater area was visited on the afternoon of 22 August it was discovered that a new vent (Vent 3) had opened along the inside of English's Crater rim. Only 24 earthquakes were located during 21-23 August, all very shallow. Seismicity was comparable or slightly lower during the next twenty-four hours. There were six episodes of increased gas venting, some of which were followed by small VT events. Fifteen earthquakes, including two LP events, were located beneath Soufriere Hills at depths ranging from near-surface to ~5 km.

Aerial reconnaissance on the afternoon of 24 August showed a NNW-SSE trending line of several small explosion craters in the summit crater, with Vent 1 at the N end and Vent 3 at the S; Vent 2 was offset to the SE. The rate of steam emission from Vent 2 was very low, while slightly more steam was being emitted from Vent 3. No gas emission was observed from Vent 1. Seismicity continued to be relatively low, with seven earthquakes distributed beneath St. George's Hill and Soufriere Hills at depths of 0-4 km. Three episodes of increased gas venting and 30 minutes of broadband tremor also occurred. A small mudflow originating from Gages Upper Soufriere during heavy rains on the afternoon of 24 August flowed through Fort Ghaut. Seismicity remained low until a swarm of VT earthquakes that lasted from 2157 on 25 August until 0230 the next day. Located earthquakes consisted of 22 VT events at depths of 0-5 km beneath the N flank. Five episodes of gas venting during 25-26 August had repose intervals of ~4 hours.

Formation of Vent 4 on 27 August. On 27 August there was one episode of broadband tremor that lasted ~20 minutes and a small mudflow in Fort Ghaut that began around 0820. From 26 to 28 August, fourteen small VT earthquakes were located beneath Soufriere Hills and St. Georges Hill at depths of <5 km. During this same period there were twelve episodes of increased gas venting. One episode at 1443 on 26 August ejected ash that could be seen from the Vue Pointe Hotel; other emissions caused light ashfall in the Tar River area. Observations on the morning of 28 August confirmed the presence of a fourth vent that had probably opened the day before. Located on the NNE flank of Castle Peak dome, it was vigorously emitting steam and ash through mid-day on 28 August; emissions from the other vents were low. Eight VT earthquakes were located beneath Soufriere Hills at depths of 0-4 km. Five episodes of increased gas venting occurred.

Vent 4 was still emitting mainly steam at a reduced rate on 29-30 August. Another nine episodes of increased gas venting were detected, and five small shallow earthquakes were located beneath and to the N of the Soufriere Hills during 29-30 August. Unusually good visibility allowed Castle Peak dome to be inspected at around 0900 on 30 August. Steam emissions from all vents were low and there was no ash. The main vent system (a linear chain of vents extending from Vent 1 on the NW margin of the dome SE to the S margin of the dome) had enlarged since 24 August. Mud or muddy water was locally present in the bottom of the main vent system. Several pools of standing water were located atop Castle Peak, and the moat pond on the NW side of the dome still existed. A recent mudflow from the W side of the dome southward down the Tar River had buried Vent 2, on the S side of the dome.

In terms of earthquake activity, the 24-hour period beginning at 1400 on 30 August was probably the most active since the 21 August phreatic eruption. Thirty-four shallow earthquakes were located WNW of Soufriere Hills. A few earthquakes were also located beneath Windy Hill (3.5 km NNW) and in the area between Windy Hill and Soufriere Hills. Seismicity decreased the next day, when only ten shallow earthquakes were located WNW of Soufriere Hills; two were also located beneath Windy Hill. In addition, four LP earthquakes occurred at shallow depths beneath the NW edge of Soufriere Hills. During these two days, thirteen episodes of increased gas venting were detected, but steam and ash emissions from all vents remained low.

Deformation and SO₂ measurements. A review of the Brodrick's dry-tilt data completed on 23 August indicated that some deformation of the volcano may have occurred between January and 9 August, confirming that magma may be at a shallow depth (as suggested by the earthquake data). Tiltmeter readings in late August were generally within background noise levels; no tilt related to volcanism was observed. EDM reflectors were deployed on 30 August in Gages Upper Soufriere and on Castle Peak dome.

COSPEC gas measurements taken on the afternoon of 20 August indicated that the rate of SO₂ emission was just above the detection level, ~50 metric tons/day (t/d). Additional measurements taken during favorable conditions on the next afternoon and morning of 22 August did not detect any SO₂. This lack of SO₂ was thought to be a result either of the system running out of gases or a sealing off of the fluid access path to the surface. A COSPEC flight on the afternoon of 23 August detected a slight trace of SO₂ (~ 40 t/d) while a flight the next morning showed none. The flux rate on the morning of 26 August was ~50 t/d, and on 28 August was ~85 t/d. Further COSPEC measurements on 29, 30, and 31 August showed no detectable SO₂.

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: VDAP, USGS; Seismic Research Unit, UWI; Montserrat EOC.