Intermittent explosions, ash plumes, seismicity, and a growing lava dome during March-April 2021

Soufrière St. Vincent (also referred to as “La Soufrière”) is the northernmost stratovolcano on St. Vincent Island in the southern part of the Lesser Antilles. The NE rim of the 1.6-km-wide summit crater is cut by a crater (500 m wide and 60 m deep) that formed in 1812. Recorded eruptions date back to 1718, with notable eruptions occurring in 1812, 1902, and 1979. An eruption that began on 27 December 2020 was characterized by increased seismicity, the growth of a new lava dome on the SW edge of the main crater, crater incandescence, and gas-and-steam emissions (BGVN 46:03). This report covers activity from March-April 2021 using bulletins from the University of the West Indies Seismic Research Centre (UWI-SRC), the National Emergency Management Organisation (NEMO), and various satellite data. Soufrière St. Vincent is monitored by the SRC assisted by the Soufrière Monitoring Unit (SMU) from the Ministry of Agriculture in Kingstown.

As of 2004, the monitoring network had consisted of five seismic stations, eight GPS stations, and several dry tilt sites. Seismic data are transmitted from field sites to the Belmont Observatory (9 km SSW), which is operated by the SMU. On 1 March a new seismic station installed at the National Parks Interpretation Center at Bamboo Range on the E side of the volcano became operational. Additionally, a new GPS monitoring station was placed in Fancy. Two more GPS stations were installed at Jacob’s Well and Table Rock on 19 March.

Activity during March 2021. The lava dome in the main crater continued to grow during March, expanding to the N, NW, S, and SE. Gas emissions persisted in the summit area from the top of the dome and along the contact area between the pre-existing 1979 dome and the 2020-21 dome, damaging vegetation on the SW side of the volcano. A team visited the dome on 19 and 23 March to make observations, take measurements, and maintain monitoring equipment. On 19 March the dome was 105 m tall, 921 m long, and 243 m wide, with an estimated volume of 13.13 million cubic meters, based on drone surveys (figure 18). Compared to the previous drone survey on 12 February, the dome had grown roughly 6.3 million cubic meters. On 23 March at 1030 the seismic network recorded a swarm of small low-frequency events that lasted about 45 minutes, possibly associated with magma movement beneath the dome. Starting at 1653 the network also detected volcano-tectonic (VT) earthquakes, some of which had been felt in communities, including Fancy, Owia, and Sandy Bay. By 26 March the period of elevated VT earthquakes had ended; only small, low-frequency events associated with lava dome growth were recorded.

Figure 18. Map (left) showing the growth of the new lava dome at Soufrière St. Vincent from 27 December 2020 to 19 March 2021. The dome is located on the SW edge of the crater rim and WSW of the 1979 lava dome that is covered in vegetation. Photo (right) of the laterally growing lava dome (dark mass at left) taken on 19 March 2021. White gas-and-steam emissions are visible along the contact area with the 1979 dome (right), as well as damaged vegetation. Courtesy of SMU, UWI-SRC, and NEMO SVG.

Activity during April 2021. The lava dome continued to grow to the N and S during April, accompanied by small earthquakes and gas-and-steam emissions rising both from the top of the dome and along the contact with the pre-existing 1979 dome. A new swarm of VT earthquakes was detected on 5 April at 0638, the number and magnitude continued to increase until 0830, after which they stabilized. These earthquakes were more intense and averaged 50/hour; the 23-25 March swarm averaged 1.5/hour. The largest event was registered as a Mw 3.5 at 0920 and was felt by local residents. At 1400 the rate of the earthquakes started to decline and then significantly dropped at 1600, though small VT events continued through 1800. Small earthquakes associated with dome growth were recorded on 7 April.

On 8 April UWI-SRC reported an increase in the volume of white gas-and-steam emissions rising from the summit crater and at 0300 scientists at the Belmont Observatory reported episodes of tremor that indicated the movement of magma and fluids near the surface. Subsequent tremor episodes at 0530, 0800, 1015, and 1300 showed increasing magnitude. Five long-period (LP) earthquakes and two brief swarms of VT events were detected between the tremor episodes; ash venting was also observed during the last episode. Sulfur dioxide emissions were also identified in satellite data. Visual observations from the Belmont Observatory during the early evening showed that the dome height had increased significantly during the day, and incandescent material over the vent area was visible in webcam images (figure 19). As a result, the Volcanic Alert Level (VAL) was raised to Red (the highest level on a four-color scale) at 1830 and the Prime Minister issued an evacuation order for communities in the Red Zone at the N end of the island, affecting 16,000-20,000 people.

Figure 19. Webcam image of incandescent material in the summit crater of Soufrière St. Vincent on 8 April 2021, accompanied by white gas-and-steam emissions. Courtesy of UWI-SRC.

On 9 April at 0840 scientists at the Belmont Observatory reported an explosion that generated an ash plume up to 8 km altitude that drifted mainly ENE (figure 20). Ash was deposited across the island, affecting communities including Chateaubelair, Petite Bordel, the Belmont Observatory, and areas at the S end of the island; it also forced the Argyle International Airport (20 km S) to close (figure 21). NEMO stated that evacuations in the Orange and Red Zones were affected by significant ashfall and poor visibility, but by the evening most residents had been evacuated. Seismicity increased at 1130, including an earthquake swarm lasting until 1440, though tremor continued to be reported. A second explosion at 1445 produced a vigorous ash plume that initially rose to 4 km altitude with lightning visible in the plume (figure 22). Successive explosions and ash pulses fed the plume for hours, rising to 16 km altitude. A third explosive series began at 1835 (figure 23). Ash venting continued through the night and into 10 April. Periods of banded tremor associated with explosive activity and stronger pulses of ash emissions to higher altitudes began at 0330 on 10 April, lasting for periods of 20-30 minutes with 1-3 hour gaps. The resulting ash plumes rose to 10.6-16 km altitude throughout the day. The Washington VAAC stated that ash plumes during 9-10 April had drifted as far as 1,200 km ESE and about 3,000 km ENE. The Grantley Adams International Airport (GAIA) in Barbados also closed.

Figure 20. Photo of the first ash plume from Soufrière St. Vincent on 9 April 2021 that rose to 8 km altitude and drifted ENE at 0840. Courtesy of UWI-SRC.

Figure 21. Photo showing ashfall in Chateaubelair during the eruption of Soufrière St. Vincent on 9 April 2021. Courtesy of Thomas Christopher, MVO, UWI-SRC.

Figure 22. Photo of the second ash plume from Soufrière St. Vincent on 9 April 2021 that rose to 4 km altitude at 1445. Courtesy of UWI-SRC.

Figure 23. Photo of the third explosive eruption and its resulting ash plume from Soufrière St. Vincent on 9 April 2021 at 1835. Photo has been color corrected. Courtesy of UWI-SRC.

Ashfall continued to be widespread on 10 April; deposit thickness varied from less than 1 mm in Colonaire (12.5 km SSE) to 10-15 mm in Rabacca (7.4 km SSE). Analysis of satellite imagery by UWI-SRC indicated that the explosions that began on 9 April destroyed the pre-existing 1979 and 2020-21 lava domes and created a crater 800 m in diameter. During the night of 10-11 April tremor lasting up to 20 minutes continued at intervals of 1.5-3 hours, associated with periods of explosions and ash emissions. Ashfall was observed in most areas of the St. Vincent Island, as well as on neighboring islands of the Grenadines (SSW), Barbados (165 km E), and Saint Lucia (50 km NNE). Explosions early on 11 April were followed by widespread power and water outages on the island, as reported by NEMO, and some houses had collapsed due to significant ashfall (figure 24). Around midday, the time between episodes of high-amplitude tremor lengthened from 1.5-4 hours to 5-8 hours, which continued to coincide with explosions and ash emissions. Small VT earthquakes were recorded around 1800. The VAAC reported that ash plumes rose to 12.2-16 km altitude throughout the day and continued to drift long distances to the ENE, E, and SE. On 12 April at 0415 a large explosion produced an ash plume that rose to 12.8 km altitude. The plume collapsed, generating pyroclastic flows that descended several valleys on S and W flanks, reaching the coast at Morne Ronde (4.3 km W), Larikai (3.5 km WNW), and Trois Loupes Bay (3.5 km NW) (figure 25). Extensive damage to vegetation was noted along the W coast, extending from Larikai Bay to Turner Bay. High-amplitude tremor episodes had stopped, but two low-amplitude and one high-amplitude episodes were recorded during 0600-1700. Explosivity or notable ash venting coincided with the episodes.

Figure 24. Photo of collapsed structures due to the thick ash deposits from the explosions in early April 2021 at Soufrière St. Vincent. Courtesy of UWI-SRC.

Figure 25. Image of Soufrière St. Vincent taken from the Sentinel-1 satellite on 11 April 2021 showing the changes of the summit crater with a possible smaller vent in the new crater. The darker black areas to the W of the summit are pyroclastic flow deposits from earlier explosive events. Courtesy of Adam Stinton, MVO, UWI-SRC.

A series of Vulcanian explosions that began at 0630 on 13 April, and lasted about 30 minutes, was accompanied by large seismic tremors followed by more than three hours of smaller continuous tremors (figure 26). The VAAC stated that a dense ash plume rose to 11 km altitude and drifted E, SE, N, and NE (figure 27). Resulting pyroclastic flows traveled about 6 km WSW. Observations made during the afternoon indicated that pyroclastic flows had reached the ocean in every valley extending from Larikai to Wallibou, a length of about 5 km. The Washington VAAC reported a new ash emission rising to 7.6 km altitude at 1850 that was associated with a thermal anomaly. Another explosion at 2030 pulsed for more than 40 minutes and produced pyroclastic flows that traveled toward the Rabacca River on the E coast of the island. Lahar deposits were reported in the Sandy Bay area. Another large explosion was detected at 2300, according to UWI-SRC. The Montserrat Volcano Observatory stated that a sulfur odor and minor ashfall from Soufrière St. Vincent was observed on their island (390 km NNW).

Figure 26. Plot summarizing the seismic data at Soufrière St. Vincent during 8-21 April 2021. The top panel shows the Real-Time Seismic Amplitude Measurement (RSAM) data from seismic stations SSVA and SVV. The SSVA data shows the banded tremor that started on 8 April and the continuous tremor prior to the first explosion. The red triangles represent the time of explosions, based on satellite imagery that was analyzed by scientists at the University of East Anglia. Data transmission from SSVA stopped when it was likely destroyed by an explosion. The SVV data shows the high-amplitude tremor generated during a series of explosions. The bottom two panels show the number of volcano-tectonic (VT) and long-period (LP) earthquakes detected by the seismic network. The number and intensity of LP earthquakes have increased since late 11 April. Courtesy of Roderick Stewart, MVO, UWI-SRC.

Figure 27. Photo of a strong ash plume rising above Soufrière St. Vincent on 13 April 2021. Courtesy of Thomas Christopher, MVO, UWI-SRC.

During 14 April seismicity continued to be characterized by episodes of tremor about 13-15 hours apart, separated by swarms of small LP earthquakes. At 1135 a band of seismic tremor, associated with an explosion, was recorded, though the emissions were gas-rich and less energetic. Another episode of tremor began at 0230 on 15 April, as well as increased ash venting. The seismic network recorded a near-constant swarm of LP and hybrid earthquakes throughout the day with three brief episodes of low-level tremor each lasting less than 30 minutes. Tremor starting at 2100 lasted about 40 minutes and may have been associated with a minor increase in venting. Sulfur dioxide emissions were successfully measured by ground-based instruments; scientists on a Coast Guard boat along the W coast recorded 809 tons/day.

Periods of tremor and near-constant LP and hybrid earthquake swarms continued on 16 April. One tremor episode on 16 April at 0615 that lasted about three minutes was accompanied by an explosion, followed by more than two hours of low-level tremor. The resulting ash plumes rose to 8 km altitude and drifted W. Sulfur dioxide measurements showed an average of 460 tons/day; some SO₂ plumes were reported to have reached India.

Satellite images showed a series of changes to the newly formed crater in the SW part of the crater (figure 28). Updated measurements were taken of the new crater, by 17 April according to an Institut de Physique du Globe de Paris scientist. It was approximately 900 m N to S, 750 m E to W, and at least 100 m deep, based on RADAR data (figure 29). Several vents were identified or inferred by emissions points of gas-and-steam and ash plumes. During 17 and 18 April the average SO₂ flux was 232 and 391 tons/day, respectively.

Figure 28. New crater features were visible at the Soufrière St. Vincent summit in a 15 April 2021 satellite image from Planet Labs. Continued explosions created a new crater (red line) in the SW part of the larger summit crater that was approximately 900 m N-S, at least 750 m E-W, and 100 m deep. The orange patterned section represents tephra deposits. The yellow line outlines the 2020-21 lava dome. Several vents were established, based on gas-and-steam and ash plumes; only one could be seen clearly (black outline). Courtesy of UWI-SRC.

Figure 29. Satellite radar image from Capella Space acquired on 14 April 2021 showed the changes of the Soufrière St. Vincent summit crater following numerous explosions. The large depression in the middle of the image used to contain the 1979 and 2020-21 lava domes; explosions that began on 9 April destroyed both, leaving a smaller crater about 800 m in diameter. The base of the crater is distorted due to the presence of ash in the atmosphere and the ongoing eruption. North in the image is to the upper left corner. Courtesy of Adam Stinton, MVO, UWI-SRC.

A period of tremor at 1649 on 18 April was accompanied by an explosion that sent an ash plume to 8 km altitude and drifted S and SW (figure 30). This was the 30th distinct explosion since the explosive phase began. The period of tremor lasted until about 2100 and was followed by ongoing small VT, LP, and hybrid earthquakes. The rate of these events decreased by 0100 on 19 April. A single rockfall event was detected at 0139, possibly due to the growing lava dome, though there was no visual confirmation. Few, but intermittent, rockfall events, VT earthquakes, and an increasing number of small LP and hybrid earthquakes were detected through 21 April (see figure 26). The seismic station at Bamboo Range recorded a lahar on 20 April at 0400 which lasted about 30 minutes and may have descended a SE-flank valley. SO₂ measurements taken near the W coast showed an average flux of 350 tons/day. NEMO reported that 12,775 people were displaced by 19 April, with 6,208 people in 85 public shelters and 6,567 people (1,800 families) in private shelters.

Figure 30. Photo of ash plumes rising above Soufrière St. Vincent and drifted S and SW on 18 April 2021. Photo has been color corrected. Courtesy of Thomas Christopher, MVO, UWI-SRC.

On 22 April at 1108 explosions generated an ash plume that rose to 8 km altitude and drifted mainly N over the ocean (figure 31). A high-level seismic tremor was also detected at 1109 that lasted for 20 minutes. During the initial stages of the explosion a pyroclastic flow was observed traveling down the W flank, reaching an average speed of 33 km/hour. Around 2100 the seismic network recorded multiple lahars, though their location could not be determined. According to data from NEMO, 13,154 people had been displaced by 21 April, with 6,208 people in 85 public shelters and 6,790 people (1,618 families) in private shelters.

Figure 31. Photo of an ash plume rising to 8 km altitude above Soufrière St. Vincent at 1108 on 22 April 2021. Courtesy of Richard Robertson, UWI-SRC.

Seismicity dropped to low levels after the explosion on 22 April and remained low through 27 April; only a few VT, LP, and hybrid earthquakes were recorded. The average SO₂ flux was 992 tons/day on 23 April, again recorded from the W coast. Satellite radar imagery acquired on 24 April indicated possible growth at the lava dome following the explosions on 18 and 22 April. During an overflight on 26 April scientists observed white gas-and-steam emissions from several locations along the crater floor, though visibility was poor. No lava dome was visible, but a possible spine-like protrusion was seen through the clouds. Lahars in the Red and Orange zones were recorded by the seismic network at 0900 and 1000 on 27 April, during and after rainfall. Multiple lahars continued to be recorded by the seismic network within all drainages during 28-30 April due to heavy rainfall (figure 32). Trees brought down by lahars reached the ocean, where the floating logs created a hazard for small boats along the coastline. SO₂ measurements taken from a boat on the W coast yielded a flux of 1,036 tons/day. On 6 May the Volcanic Alert Level was lowered to Orange (the second highest on a four-color scale) because no explosions had been recorded since 22 April and the number of volcanic earthquakes had decreased significantly.

Figure 32. Photo of lahars in the Wallibou River near Soufrière St. Vincent on the morning of 29 April 2021 due to heavy rainfall. Courtesy of UWI-SRC.

Additional satellite data. Daily satellite-based sulfur dioxide measurements were notable beginning on 9 April. The initial explosion at 0840 showed relatively minor SO₂ emissions, though as the day progressed stronger plumes drifted E over the Atlantic Ocean (figure 33). As strong explosive activity continued at Soufrière St. Vincent, strong SO₂ plumes exceeding 2 Dobson Units (DUs) were detected during 9-14 April, moving generally E, based on data from the TROPOMI instrument on the Sentinel-5P satellite.

Figure 33. Strong and distinct SO2 plumes from Soufrière St. Vincent were detected by the TROPOMI instrument on the Sentinel-5P satellite during 9-14 April 2021 on days with strong explosive activity and ash venting. The plumes generally drifted E over the Atlantic Ocean. Courtesy of NASA Global Sulfur Dioxide Monitoring Page.

MIROVA (Middle InfraRed Observation of Volcanic Activity) analysis of MODIS satellite data detected a single thermal anomaly during mid-March 2021, followed by a brief break in activity (figure 34). Thermal activity resumed in April, occurring in three distinct clusters of moderate-power anomalies throughout the month. Two thermal hotspots were detected using the MODVOLC thermal algorithm on 24 April. Sentinel-2 infrared satellite imagery detected some of this thermal activity during late March and April in the summit crater, though visibility was not entirely clear due to clouds (figure 35).

Figure 34. A single thermal anomaly at Soufrière St. Vincent was detected during mid-March 2021. Thermal activity resumed in early April, occurring in three distinct clusters of moderate-power anomalies, as reflected in the MIROVA data (Log Radiative Power). Courtesy of MIROVA.

Figure 35. Sentinel-2 infrared satellite imagery showing a faint thermal anomaly (orange) in the summit crater of Soufrière St. Vincent on 24 March (top left), 3 April (top right), 8 April (bottom left), and 23 April (bottom right) 2021, though clouds covered most of the crater. Images using “Atmospheric penetration” rendering (bands 12, 11, 8a). Courtesy of Sentinel Hub Playground.

Information Contacts: University of the West Indies Seismic Research Centre (UWI-SRC), University of the West Indies, St. Augustine, Trinidad & Tobago, West Indies (URL: http://www.uwiseismic.com/); National Emergency Management Organisation (NEMO), Government of Saint Vincent and the Grenadines, Biseé, PO. Box 1517, Castries, Saint Lucia, West Indies (URL: http://nemo.gov.lc/); Washington Volcanic Ash Advisory Center (VAAC), Satellite Analysis Branch (SAB), NOAA/NESDIS OSPO, NOAA Science Center Room 401, 5200 Auth Rd, Camp Springs, MD 20746, USA (URL: www.ospo.noaa.gov/Products/atmosphere/vaac, archive at: http://www.ssd.noaa.gov/VAAC/archive.html); Hawai'i Institute of Geophysics and Planetology (HIGP) - MODVOLC Thermal Alerts System, School of Ocean and Earth Science and Technology (SOEST), Univ. of Hawai'i, 2525 Correa Road, Honolulu, HI 96822, USA (URL: http://modis.higp.hawaii.edu/); NASA Global Sulfur Dioxide Monitoring Page, Atmospheric Chemistry and Dynamics Laboratory, NASA Goddard Space Flight Center (NASA/GSFC), 8800 Greenbelt Road, Goddard, Maryland, USA (URL: https://so2.gsfc.nasa.gov/); MIROVA (Middle InfraRed Observation of Volcanic Activity), a collaborative project between the Universities of Turin and Florence (Italy) supported by the Centre for Volcanic Risk of the Italian Civil Protection Department (URL: http://www.mirovaweb.it/); Sentinel Hub Playground (URL: https://www.sentinel-hub.com/explore/sentinel-playground); Jenni Barclay, University of East Anglia, Norwich Research Park, Norwich, Norfolk, NR4 7TJ, UK (URL: https://research-portal.uea.ac.uk/en/persons/jenni-barclay); Ian Renfrew, University of East Anglia, Norwich Research Park, Norwich, Norfolk, NR4 7TJ, UK (URL: https://research-portal.uea.ac.uk/en/persons/ian-renfrew).

Information is preliminary and subject to change. All times are local (unless otherwise noted)

November 1971 (CSLP 92-71)

Eruption causes growth of island in crater lake

Card 1313 (08 November 1971) Steam and sulfurous smells from crater lake

"Disturbed conditions" have been noted from the air in the crater of . . . Soufrière in St. Vincent. Rising steam and a strong sulfurous smell have been reported coming from the water in the crater lake . . . . The condition was first noted by a pilot who flew over Soufrière on 31 October. His report was confirmed 1 November by a government official who flew over the crater in a small plane."

William C. Buell IV (Washington, D.C.) reports that Soufrière is a "very dangerous" volcano and that the heat level has recently risen from 20 to 84°C, a very significant change.

Card 1319 (23 November 1971) Eruption confirmed; island appears in crater lake

An island 100 m in diameter and 2 m high appeared in the lake center on 20 November. This island consists of angular blocks 1-10 m across and appears to be the top of a lava dome standing 200 m above the former lake floor. The water level is rising faster again with a 38-cm increase in the last 48 hours, but the temperature at the lake edge remains steady at 82°C. A 10-second series of explosions like thunder was heard on 20 November within the crater but the source was not seen because of hot vapor concealing most of the lake. No clear volcanic earthquakes have been recorded, although a seismograph 6 km SW of the summit has recently recorded indeterminate signals of 10-60 second duration. More seismographs are being installed and close watch continues.

Information Contacts:
Card 1313 (08 November 1971) John F. Tomblin, Seismic Research Unit, University of the West Indies, St. Augustine, Trinidad; Bridgetown Radio, Barbados.
Card 1319 (23 November 1971) John F. Tomblin, Seismic Research Unit, University of the West Indies, St. Augustine, Trinidad.

December 1971 (CSLP 92-71)

Dome growth continues; lava compositions

Card 1321 (01 December 1971) Lava dome growth in crater lake continues

"The islands which appeared on 20 November have grown in height and in numbers and they have coalesced to form a continuous land mass in the southeastern part of the lake. This land mass is 50-feet in height at the highest point as of 28 November, and the area covered by this larger island is over 100 m in diameter. This island is composed largely of three E-W trending ridges of a black massive rock. In addition, a number of pinnacles and small islands are sticking out of the water NW of this larger island. The lake temperature remains between 79-80°C at the margin of the lake. The lake level has been dropping; it dropped a couple of feet in the last three days.

"I have a report that there has been some leakage or seepage at the flank of the cone of the crater lake waters. The crater lake waters have found an outlet through the flank of the cone and are seeping out. This may account for some of the drop in water level. Earthquakes have been recorded on our seismograms for the last five days, between 10 and 18 events per day. These are not felt; they are instrumental, that is, recorded. There are four seismometers on St. Vincent now; one on the rim of the crater, two on the crater flank, one in Kingstown. There is one tiltmeter site being set up on the eastern flank of the volcano.

"There have been no volcanic explosions, or any emission of steam or gases, and all that has happened since 31 October is the extrusion of a very viscous dome-forming lava onto the floor of the crater. This lava dome is gradually emerging above the water level, higher and higher at approximately 4-6 feet per day. It seems to rise as a continuous mass and there are no disturbances of any sort except for the minor tremors which we have started to record."

Card 1330 (15 December 1971) Eruption continues during 8-13 December with dome growth

The eruption continues as before with emission of lava into the crater. The level of the water in the crater lake has dropped 3.1 m since 3 November. This drop can be accounted for entirely by evaporation and there is no evidence of leakage through the crater wall. The lava island in the crater lake was visited on 13 December by Haraldur Sigurdsson and Peter Baker in a dinghy and it was found that the water temperature increased from 74°C at the lake margin to 90°C at a distance of 10 m from the island. Water in contact with the island was boiling vigorously. The island is 30 m high above lake level and is approximately 600 m in length along its longest axis NW-SE. The rate of rise of the island has slowed down, and the major growth is now lateral by very slow extension of blocky lava lobes and ridges radiating from the central region. Five kilograms of rocks were collected from the southern end of the island. These are samples of plagioclase-phyric dark gray basaltic andesite with sparse phenocrysts of olivine and pyroxene.

In the period from 8 to 14 December only three events have been positively identified as local earthquakes. In addition, 40-60 minor seismic events are being reported every 24 hours by high-gain seismographs close to the crater rim. These events have a fairly constant frequency of 2.5 Hz and are propagated outward from the crater at a surface velocity of 2-3 km/s. Visual observations and timing of events and processes on the island have permitted correlation of some of the larger tremors with rockfall and lava creep.

Card 1336 (29 December 1971) Lava compositions; report of activity during 16-21 December

Information gained since our last report gives grounds for optimism with regards to events in the Soufrière and we consider that the probability of a violent eruption is decreasing. A chemical analysis of the lava collected on 13 December is now available and supports our former identification of a basaltic andesite, with a silica content of 55.3% and an overall composition virtually identical to the numerous basaltic andesite lavas of the Soufrière, which are the products of a quiet outpouring of lava from the crater. No local earthquakes have been recorded by tremors associated with rockfalls on the island continue to occur. The lack of significant seismic activity supports our interpretation of the present events as a relatively quiet eruption of a gas-poor lava.

Observations. Table 1 summarizes observations at the crater.

Table 1. Summary of Soufriere St. Vincent crater observations, 15-31 December 1971. Courtesy of the Seismic Research Unit.

The island has now reached a height of 120 feet above water level and continues to grow in area as new lava spreads laterally in the lake. The main areas of growth are the radial ridges to the N and SE. New observations of the process of growth indicate that the sequence of events is as follows: hot molten lava is extruded below water level at the base of the ridges and, as this lava comes into contact with the lake water, minor steam explosions are generated. Some of these explosions breach the surface of the lake near the island as areas of boiling water, and are followed by the spread of a red-brown stain which probably consists of stirred-up sediments from the lake bottom. Many of these minor explosions dislodge rocks from the island which have been made unstable by general ridge expansion. These minor explosions accompanied by rockfalls are the most likely source of the numerous minor tremors which are recorded by the seismographs close to the summit.

Water temperature at the observation point on the S edge of the lake has dropped from 74°C to a nearly constant value of 69°C. This may be a local phenomenon, due to relatively small output of lava into the southern part of the lake in the last few days, and we believe that other inaccessible regions of the lake are hotter. The drop in lake level has accelerated slightly in this period and is now approximately 1 foot 6 inches every 24 hours. Sigurdsson and Baker went boating on the crater lake on 21 December and carried out a series of bathymetric soundings along a profile from the S shoreline towards the island (table 2). Unfortunately, this survey had to be abandoned due to unfavorable conditions on the lake. The distance from the S shore of the crater lake to the island is approximately 12 m.

Table 2. Summary of depth and temperature measurements in the Soufriere St. Vincent crater lake, 21 December 1971. Courtesy of the Seismic Research Unit.

The survey indicates a maximum depth of 33 fathoms in this profile, at a distance of 38 fathoms from the S shore. Closer to the island the water depth starts decreasing at a rapid rate. Rumors of warm water in the stream S of Windsor Forest village were checked by ground observations on 17 December and found to be untrue. This stream is at a temperature of 25°C and the Falls of Baleine at 23°C. Five seismograph stations continue to operate on a 24-hour basis but three of these stations are within 2 km of the crater. These three stations continue to record the tremors discussed above, and attributed to growth processes in the lava island. No sub-surface earthquakes have been recorded in this period nor have any earthquakes been reported felt.

Composition of the lava. A chemical analysis of the rock samples collected from the island on 13 December is now at hand and the following table is an average of three determinations.

Table 3. Chemical analysis of Soufriere St. Vincent samples from the island in the crater lake collected on 13 December 1971. Data is the average of three determinations; analysis was done by the U.S. Geological Survey. Courtesy of the Seismic Research Unit.

The composition of the lava is that of a basaltic andesite, virtually identical with older Soufrière lavas such as those at Warbishy River, in the cliffs around the Larikai River, Morne Ronde Point, Campobello Bay, and the numerous lava flows which form the cliffs surrounding the Soufrière crater. The new lava is significantly more basaltic, i.e., lower in silica (55.3%) than the products of the most violent phases of the 1902 eruption, which were more andesitic in composition, with silica content of 56-58%.

Interpretation. In volcanic eruptions in general the most silica-rich products appear first, when the risk of violent eruptions and explosive activity is greatest. Products of later stages of an eruption have a lower silica content and are less explosive. We therefore expect that as the present eruption proceeds, lavas with progressively lower silica content will be erupted. All evidence therefore leads us to the conclusion that the likelihood of explosive activity is very small at this time and will become progressively smaller. Were the Soufrière eruption to enter an explosive phase, this would be preceded by major changes in seismic activity and other observable phenomena for at least one or two days.

Information Contacts:
Card 1321 (01 December 1971) Haraldur Sigurdsson, Seismic Research Unit, University of the West Indies.
Card 1330 (15 December 1971) Haraldur Sigurdsson and J.P. Shepherd, Seismic Research Unit, University of the West Indies.
Card 1336 (29 December 1971) Haraldur Sigurdsson and J.P. Shepherd, Seismic Research Unit, University of the West Indies.

January 1972 (CSLP 92-71)

Island now 145 feet above water level; lake temperature drops

Card 1341 (11 January 1972) Island now 145 feet above water level; lake temperature drops

Table 1 summarizes daily measurements [during 23-31 December] at the Soufrière. The island is now 145 feet in height above water level, but its growth has decelerated in the last few days. The cooling of the lake to 66°C is an indication that at present very little hot lava is in contact with the water. The lake level has dropped about one foot per day in the last six days. Minor rockfalls continue to take place off the edges of the lava mass, but these are less frequent than in the past weeks as is also indicated by a slight decrease in the number of tremors recorded by the summit seismograph stations. Most of the evidence tends to suggest that there is a slight new lava being extruded at a slower rate into the crater, but whether this decrease in activity is temporary or not is too early to comment on at this stage.

Information Contacts: Haraldur Sigurdsson and J.P. Shepherd, Seismic Research Unit, University of the West Indies.

February 1972 (CSLP 92-71)

Island growth continues; historical comparisons

Card 1350 (03 February 1972) Analyses of volcanic products

J. Tomblin reports the following. "In Card 1336 it was stated by Sigurdsson and Shepherd that 'the new hot lava is significantly more basaltic, i.e., lower in silica (55.3%) than the products of the most violent phases of the 1902 eruption, which were more andesitic in composition, with silica content of 56-58%.' These statements are incorrect. Table 4 contains mean values for all available wet chemical analyses of Soufrière rocks, summarized from UWI Seismic Research Special Publication No. 15 (Tomblin, 1968).

Table 4. Mean values for all available wet chemical analyses of Soufrière St. Vincent rocks, summarized from UWI Seismic Research Special Publication No. 15 (Tomblin, 1968). Courtesy of John Tomblin, UWI.

"From the above it is clear that there is no significant difference in silica content between lava flow and pyroclastic material erupted from the Soufrière Volcano. This writer concludes that the explosiveness of Soufrière magmas is a function of gas, not silica content, there being no direct relationship between these two components. The only noteworthy features of the above figures are that 1) aeolian differentiation of the 7-8 May 1902 ash appears to have been negligible at 100 miles but considerable at the 275-mile range; and 2) pyroclasts erupted at the end of the 1902-03 eruption were about 2% lower in silica than at the beginning of this eruption."

Card 1358 (23 February 1972) Crater observations 12-16 February; island growth continues

Table 5 is a list of observations made at the crater.

Table 5. Summary of Soufriere St. Vincent daily crater observations, 12 February-18 April 1972. Courtesy of the Seismic Research Unit.

The increase in the rate of growth of the island which occurred on 9 February is continuing at present. The current rate of increase is 3 ft/day as compared with a rate of increase of 1 ft/day prior to 9 February, representing a 200% increase in the growth rate. While the N of the island continues to rise, the S end shows little or no increase in height. This indicates that the activity has now shifted almost entirely to the N end of the lava island. Another survey of the island is planned and would probably be carried out within the next two weeks, in order to determine the rate of increase in area of the island. The water temperature has maintained a level of about 58°C for the past week, while the water level continues to decrease at the rate of 6 in/day. Tremors in the vicinity of the crater continue to be recorded, at a rate of about 10/day for the past week. There was a temporary failure of all three seismograph stations in the vicinity of the Soufrière, between 14-15 February. However, the summit and Rabacca station are back in operation, and repairs are continuing on the Wallibou station.

Card 1359 (29 February 1972) Comparison of current eruption with historical activity

The current eruption is a mild one and involves the growth of a volcanic dome. It is completely different in character from the three earlier historic eruptions of the Soufrière, all of which were violently explosive throughout. It is, however, similar to the second half of the two large historic eruptions of Mt. Pelée, Martinique, in 1902-05 and 1929-32. In each of these eruptions there was an initial, highly explosive phase which lasted several months, and this was followed by a second phase in which fresh lava welled up quietly into the crater to build a dome. In both Martinique eruptions, dome growth continued for more than two years. Comparable periods of between six months and three years have been reported for dome building at volcanoes in other parts of the world. Thus the present eruption in St. Vincent is likely to continue for one year and could possibly continue for two. This conclusion is supported by the constancy of the rate of lava emission (~0.5 x 10⁶ m³/day) from the beginning of November 1971 to the present, and by the small total volume so far erupted (80 x 10⁶ m³) in comparison with the total fresh lava produced in 1902-03 (2,600 x 10⁶ m³).

The likelihood of dangerously explosive activity during the present eruption is now remote. In at least 95% of historic eruptions at similar volcanoes where there has been any explosive activity, this has begun within the first two months of the onset of eruption. The Soufrière has now been in eruption for over four months and has comfortably exceeded this initial period of high uncertainty. At the present rate of lava emission, the summit of the new dome should reach the level of the lowest parts of the crater rim by May 1972, and the crater will be almost completely filled by early 1973. The viscosity of the new lava is high so that even if the dome grows to the point at which its flanks overlap the present crater rim, it is unlikely that molten lava will escape down the mountainside.

Information Contacts:
Card 1350 (03 February 1972) John F. Tomblin, Seismic Research Unit, University of the West Indies.
Card 1358 (23 February 1972) W. Aspinall and H.J. Almorales, Seismic Research Unit, University of the West Indies.
Card 1359 (29 February 1972) John F. Tomblin, Seismic Research Unit, University of the West Indies.

March 1972 (CSLP 92-71)

Dome survey shows expansion; growth rate slows in March

Card 1364 (07 March 1972) Crater observations, 17-28 February

Observations made at the crater [are included in Table 5].

Activity at the crater has continued along the lines established three months ago. The height of the lava dome has increased by an average of about 1.5 ft/day and the lake water has continued to drop at a mean of 9 in/day. The water temperature at the lake edge, which remained constant between 56-60°C during the previous four weeks, dropped to a new low of 53.5°C on February 27.

An excursion by rowboat was made on 27 February by R. Arculus and K. Willis, geology research students from the University of Durham, England. Conditions were very unfavorable, with high winds, poor visibility, and frequent rockfalls from the flanks of the dome which are now relatively steep. Consequently, it was impossible to carry out a complete sounding survey. The maximum water depth recorded was 99.5 ft, at a point approximately halfway between the island and the lake edge. This depth is 48.5 ft less than the maximum obtained during the previous sounding survey on 27 January, and confirms that the lava dome is still expanding horizontally across the former lake floor.

Measurements have been made on near-vertical air photos of the crater taken on 11 February and kindly placed at our disposal by Mrs. M. Barnard. These indicate that the dome is approximately circular in plan at water level with dimensions on 11 February of 590 m (1,970 ft) from N to S, and 540 m (1,800 ft) from E to W.

On 28 February, a re-triangulation by theodolite was carried out by R. Waters of the Directorate of Overseas Surveys. This showed that the dome had expanded to 640 m (2,130 ft) in N-S diameter at water level and 590 m (1,970 ft) from E to W. This represents an increase in both horizontal dimensions at water level by 50 m in the last 17 days. Approximately 7 m of this increase is apparent and not real, i.e., due to the drop in water level between 11-27 February. The true amount of horizontal expansion of the dome is therefore 44 m.

The W half of the dome is now 10-20 feet higher than the E half, and the two halves are still separated by a N-S valley. The N part of this valley is the most actively steaming area of the dome. The dome summit was 295 ft. above current lake level on 27 February.

The only local seismic activity consists of small tremors which continue to originate close to the crater. Between 10-20 such events are currently occurring per day. There have been no deeper or larger earthquakes of the kind which we would expect to see prior to a major increase in the scale of the eruption. On the other hand, the shallow tremors have decreased to about one-quarter of their rate of occurrence in early December, without any decline having taken place in the rate of injection of new lava into the dome.

All available evidence indicates that the eruption will neither end nor become violent in the early future. The risk of a dangerous eruption is now remote and will continue to decrease with time. However, so long as this continues, the present monitoring system should be maintained.

Card 1366 (20 March 1972) Crater observations, 29 February-12 March

The eruption continues as before. The rate of rise of the rocks has decreased slightly to about one foot per day but the rate at which the water level is dropping is still about nine inches per day. The water temperature shows a downward trend and reached a new low of 51°C on 8 March, but this was followed by rise to 57°C by 10 March. The most noticeable feature of this period has been the very small number of crater tremors, and no tremors have been recorded since 7 March. There is still no sign of deeper seismic activity. Rockfalls continue to be frequent.

No comprehensive survey of the island has been carried out during this period but from visual observation the general shape is still as previously described.

Card 1370 (28 March 1972) Activity and growth rate slow in March

It is evident from the data of the past three weeks that activity in the Soufriere crater is very slight. The rise of the island mass slowed down on 1 March and in the period 1-22 March the total increase in island height has been 2 feet 4 inches or only 1.3 inches/day on the average. On 22 March the western summit of the island was the highest point, 226 feet. The northern peak was at 216 feet and the eastern summit at 201 feet. Water level is dropping at a slower rate than before 7 March, when the daily drop was 9 inches on the average. The present rate of drop in water level is 4 inches per day, but the lake has finally reached its pre-eruption level, 85 feet below 3 November datum. When the lake was at its highest level on November 20th, it was thus some 92.5 feet above the normal pre-eruption level. Seismic activity is low, and limited to one crater tremor per day on the average.

Sigurdsson and Almorales carried out a bottom survey of the crater lake on 2 March (table 6) along a profile from the south point of the island to the near shore (282 feet). The maximum depth in this area is 111 feet. Surface water temperature near the southern shore of the island is 62°C. A boat trip was made around the island and water temperature recorded at different points in the lake. Maximum temperature of 81°C was recorded in bubbling hot springs on the northern shore of the lava mass, but minimum temperature of 49°C was encountered in center of the wide bay east of the island.

Table 6. Depth soundings in the southern part of the Soufriere St. Vincent crater lake, 2 March 1972. Courtesy of the Seismic Research Unit.

Information Contacts:
Card 1364 (07 March 1972) John F. Tomblin, Seismic Research Unit, University of the West Indies.
Card 1366 (20 March 1972) J.B. Shepherd, Seismic Research Unit, University of the West Indies.
Card 1370 (28 March 1972) H.J. Almorales and H. Sigurdsson, Seismic Research Unit, University of the West Indies.

April 1972 (CSLP 92-71)

Lava island stopped growth on 20 March; eruption over

Card 1380 (17 April 1972) Lava island stopped growth on 20 March; eruption over

Since 20 March the island in the middle of the crater lake has maintained a maximum height of 226 feet on he western summit. The greatest heights on the eastern and northern summits are 196 and 211 feet, respectively. These two heights which are now considered reliable are not in agreement with the heights quoted in the previous report when the measuring poles were interfered with by other parties. Loose rocks continue to crumble off the slopes, thus narrowing the peaks and forming more lateral spread towards the base of the island.

The temperature of the water at the crater edge has been decreasing by 0.5°C/day on the average since 22-31 March and an increased drop of 3° was detected three days after. The water level is now dropping at the rate of 3.5 inches per day leaving on the slopes of the crater a thin scum of muddy substance which on hardening becomes flaky and is eventually washed back into the lake by the rain.

The number of crater tremors have gradually declined to one about every two days. These tremors are of shallow depths, very low magnitude and can only be detected by the summit station. This decline in seismic activity is probably due to the island not growing or the absence of new lava emitting from the crater.

On 11 April 1972, H. Sigurdsson reported that "the Soufriere eruption has now ceased. Height of the lava island has remained constant at 226 feet above 3 November datum since 20 March, or some 316 feet above present water level. Crater tremors were of the order one a day since 7 March, but none have been recorded since 1 April."

Information Contacts: H. Sigurdsson, Seismic Research Unit, University of the West Indies.

May 1972 (CSLP 92-71)

Water temperature and level decrease

Card 1388 (02 May 1972) Crater situation unchanged in over a month

The situation in the crater remains the same as it has been for more than a month. The island height is unchanging; loose material on the island is being washed into the lake by the rains, revealing the structure of the lave mass. Water continues to evaporate from the lake thereby reducing the lake level fractionally; the color of the water is now clearer and more translucent indicating that the sediments disturbed by the initial eruptive activity are resettling. The number of crater tremors is practically zero. The phase of lava emission appears to have come to a halt, but crater observations will continue.

Card 1394 (22 May 1972) Water temperature and level decrease

"There has been no further increase in the height of the island of lava since 20 March, indicating that no more lava has been introduced into the crater lake. The highest point on the island is still 226 feet relative to the 3 November datum.

The water temperature continues to decrease with minor fluctuations. The temperature was 46°C on 7 May and 41°C on 11 May. The water level is decreasing at an average rate of 1.5 inches/day, as evaporation over the crater lake continues. The water level was 94 ft on 23 April, 95 ft 6 inches on 7 May and 96 ft on 11 May.

The frequency of crater tremors continued on a greatly diminished scale and only six tremors were recorded during the past three weeks.

The situation at the crater has maintained the current pattern over the past seven weeks. Observations at the crater will continue until a state of normalcy has been established."

Information Contacts:
Card 1388 (02 May 1972) J.B. Shepherd and W.P. Aspinall, Seismic Research Unit, University of the West Indies.
Card 1394 (22 May 1972) H.J. Almorales, Seismic Research Unit, University of the West Indies.

June 1972 (CSLP 92-71)

Water temperatures and level stabilize

Card 1400 (09 June 1972) Water temperatures and level stabilize

During the period 13-26 May 1972, the situation at the crater continued along the lines stated in the last report. The highest point on the island maintained its height of 226 feet relative to the 3 November datum. This point is now 324.5 feet above the present water level.

The water temperature varied between 40-41°C. The temperature was 40°C on 15 May, 41°C on 18 May, 40°C on 22 May, and 41°C on 25 May.

The water level continued to decrease, but at a slightly higher average rate, from 1.5 inches/day to 2.0 inches/day. The water level was -97 feet on 15 May, -98 feet on 22 May, and -98.5 feet on 25 May, and is now 13.5 feet below the original level of the lake.

The occurrence of crater tremors was practically zero, with one tremor recorded during the last two weeks.

Work on the construction of a tiltmeter station at Orange Hill is now completed. The installation of instruments, initial adjustments, and readings to establish reference levels is scheduled for the next two weeks.

Information Contacts: H.J. Almorales, Seismic Research Unit, University of the West Indies.

April 1979 (SEAN 04:04) Cite this Report

Major explosive eruption; 17,000 evacuated

A series of powerful explosions from Soufrière produced large ash clouds and several pyroclastic avalanches, forcing the evacuation of more than 17,000 persons from the N end of St. Vincent. This eruption is particularly noteworthy because of the wide variety of observations made by various scientific teams (from land, low-flying aircraft, a high-altitude research plane, and from satellites). When these data are analyzed and integrated, the geophysical community can look forward to an unusually well-documented account of an episode of explosive island-arc volcanism.

The first pre-eruption seismic event, telemetered to the Seismic Research Unit of the University of the West Indies by seismometers 3 and 9 km from Soufrière's summit, was a strong local earthquake at 1106 on 12 April, within 1 hour of the fortnightly earth tide maximum (calculated by F.J. Mauk). Seismic activity gradually increased through the day, and by 1900 about 15 clearly identifiable earthquakes, apparently B-type, were occurring per hour. Continuous harmonic tremor began to build an hour later, and within 2 hours was saturating the seismometers.

A team of volcanologists and seismologists from the Seismic Research Unit arrived on St. Vincent 13 April, and were later joined by researchers from several other institutions. The first explosive activity was observed at dawn (about 0500) on 13 April. Subsequent explosive events were reported at 1115, 1700, and 2050 on the 13th, 0300 and 1200 on 14 April, 1705 on 17 April, 0635 on 22 April, and 2355 on 25 April. Analysis of infrared imagery from NOAA's SMS-1 weather satellite indicates that most of these high eruption clouds were fed briefly (less than 1/2 hour) by the volcano. The largest of the clouds, from the 17 April explosion, reached an estimated height of 18 km and ultimately grew to a diameter of 140 km. The noon explosion on 14 April produced a 100 km-diameter cloud, and two explosions on 13 April also produced sizeable clouds, 60-km-diameter at 1700 and 40 km at 2050. The first two explosion clouds on 13 April were smaller, and the explosion reported at 0300 on 14 April produced no infrared signature. Most of the explosions occurred close to diurnal earth tide maxima, and the 17 April explosion also fell on the fortnightly earth tide minimum (calculations from F.J. Mauk).

A NASA P-3 Electra aircraft, equipped with lidar, a 10-stage quartz-crystal microbalance, and an NO₂ remote-sensing instrument, flew through the periphery of the 17 April cloud minutes after it was ejected. Ash was collected and photographs were taken. The next evening, the NASA aircraft's lidar detected significant quantities of ejecta in the stratosphere, in patches 0.5-3 km thick with a base altitude of about 18.7 km. Scientists at NCAR, Kyushu University, and other institutions, will search at higher latitudes for stratospheric aerosols from this event.

Hot pyroclastic avalanches have accompanied at least three of the explosions. The largest avalanche flowed down the Larikai River valley (figure 1) at noon on 14 April and continued beyond the mouth of the river (3 km W of the crater) several kilometers out to sea. Its deposit at the coast was 1.5 m thick, ~300 m wide, and contained scoria blocks up to 60 cm in diameter. The surface temperatureof this deposit was well in excess of 100°C when it was inspected 28 hours after emplacement. A hot avalanche from one of the earlier explosions traveled 2.5 km SE from the summit, down the Rabacca Valley, and the 17 April explosion produced numerous hot avalanches that moved down several valleys on the flanks of the cone.

Figure 1. Sketch map of Soufrière and the N portion of St. Vincent, based on the map in the Catalog of Active Volcanoes of the World. Arrows show routes taken by pyroclastic avalanches, which traveled down valleys on Soufrière's flanks during at least three of the explosions in April 1979. Lengths are approximately proportional to distances traveled by the hot avalanches. Arrows labeled 1 and R indicate avalanches in the Larikai and Rabacca valleys. The lake and island (a lava dome extruded during the 1971 eruption) shown in Soufrière's summit crater were both destroyed during the present eruption. Rocks older than the Soufrière edifice are shown by a stippled pattern.

Ashfalls from most of the explosions were limited to a few centimeters in the N portion of St. Vincent, and small amounts on Barbados, ~150 km E. However, during the 22 April eruption (figure 2) abnormal winds dropped 4 mm of ash on Kingstown (the capital of St. Vincent), on the S end of the island.

Figure 2. Photograph taken shortly after the onset of the 22 April 1979 explosion of Soufriere of St. Vincent. Height of the eruption cloud is estimated to be ~6 km. Photo taken by R. Fiske.

The character of the seismic activity varied considerably during the eruption. Harmonic tremor continued to saturate the instruments from 2200 on 12 April through 1600 on 14 April. Beginning at about 0900 on the 15th, a decline in tremor amplitude made it possible to identify more than 100 individual events per hour at the station 9 km from the summit. The number of individual events declined irregularly through the morning of 16 April. At about 2000 on the 16th, the seismicity changed entirely to rock fracture events, which continued until the explosion of 17 April. After this explosion, the seismicity consisted primarily of 30-50 small explosion events per hour, which could be correlated with the ejection of steam puffs that sometimes contained a little ash. Most of the later explosions were preceded by brief periods (up to 3 hours) of seismic quiet. Seismicity ended almost completely at about 0300 on 29 April. With the exception of a 20-minute period of tremor during the night of 29-30 April, the volcano has remained seismically quiet through early 3 May.

. . . . The amount of tephra produced by the current eruption is about two orders of magnitude less than in 1902. . . .

Information Contacts: J. Tomblin, W. Aspinall, K. Rowley, and J. Shepherd, UWI; A. Kreuger and R. LaPorte, NOAA; M. McCormick, NASA LaRC; F. Mauk, Univ. of Michigan; R. Fiske, SI.

May 1979 (SEAN 04:05) Cite this Report

Explosions and seismicity end; new lava dome in summit crater

A new phase of activity from Soufrière began in late April. At 0300 on 29 April, the seismicity that had accompanied the eruption since 12 April ended almost completely, and there have been virtually no local earthquakes since then. Katia Krafft climbed to the crater rim on 3 May and observed a small new lava dome growing in the base of the crater, the same area that had been occupied by the center of the now-destroyed 1971 dome. A sample of the new dome collected by John Tomblin in mid-May was described in hand specimen as a basaltic andesite, similar to the 1971 dome.

By the end of May, the new dome had grown to ~500 m in diameter and 60 m in height. The dome's growth rate was difficult to estimate, but was probably <0.5 x 10⁶ m³/day in late May. Virtually all of the recent expansion of the relatively fluid dome has been lateral rather than vertical.

A few small explosions have taken place from the dome, but none were large enough to project material above the rim (about 300 m above the crater floor). Fumarolic activity from the top of the dome was moderately strong during the early stages of its growth, but had stopped by 20 May. In late May, fuming was still fairly vigorous around the dome, on the floor of the crater. A team from Guadeloupe's volcanological observatory measured a temperature of 239°C from the fumaroles in late May, and collected samples from the dome.

The April explosions deposited a total of about 30 cm of ash on the zone within 1 km of the crater rim. Large numbers of massive dense remnants of the 1971 dome and scoria blocks up to 60 cm in diameter were also found in this zone.

H.H. Lamb's preliminary estimate of Dust Veil Index (Lamb, 1970) for this eruption is from 3 to 9, probably nearer to 3. The Dust Veil Index for the 1902 eruption of Soufrière was calculated at 300; the Krakatau 1883 eruption = 1,000.

Much of the crater floor is presently mantled with rubble. The new dome had spread over the remnants of the old crater lake by about 20 May, but a new lake had begun to form at the beginning of June because of the onset of the rainy season.

Most of the more than 17,000 evacuees were allowed to return to their homes on 14 May, including residents of the two largest towns near the volcano, Georgetown and Chateaubelair. Restrictions remain in effect for the 4,000 persons who live in the zone devastated by the 1902 eruption, N of the Rabacca and Wallibou Rivers.

Reference. Lamb, H.H., 1970, Volcanic dust in the atmosphere: Philosophical Transactions of the Royal Society of London, series A, v. 266, no. 1178, p. 425-533.

Information Contacts: W. Aspinall, K. Rowley, J. Shepherd, and J. Tomblin, UWI; K. Krafft, Cernay, France; H. Lamb, Univ. of East Anglia.

June 1979 (SEAN 04:06) Cite this Report

Aseismic lava extrusion persists; heavy rains generate large secondary mudflows

A period of summit crater lava extrusion, accompanied by little or no seismicity, began in late April and was continuing at the end of June.

Richard Fiske and Haraldur Sigurdsson descended into Soufrière's crater on 18 June. Lava extrusion continued from the same vent that produced the 1971 dome and the 1979 explosions. Blocky lava had flowed to the N wall of the crater, reaching maximum dimensions of 725 m across and 110 m high according to their tape and compass survey. The small number of flow front rock avalanches observed during their 4-hour stay may indicate a rate of extrusion substantially lower than in late April. Heavy steaming in the crater limited visibility.

Since the recent beginning of the rainy season, large quantities of tephra have been eroded from Soufrière's flanks. Major mud flows that traveled down the larger valleys (mostly on the W side of the volcano) carved deep, narrow canyons, dramatically exposing pre-1979 valley fill deposits. The mudflows disrupted road crossings in the Rabacca Valley (E flank) but otherwise did little property damage. Revegetation of the areas devastated by the 1979 eruption had begun by mid-June.

The zone of destruction from the 1902 eruption (N of the Wallibou and Rabacca Rivers) remains partially evacuated. Some of the several thousand evacuees work in the area during the day, but most leave at night.

Information Contacts: R. Fiske, SI; H. Sigurdsson, Univ. of Rhode Island.

July 1979 (SEAN 04:07) Cite this Report

Lava extrusion continues

Lava extrusion at Soufrière continued through July. Between surveys on 2 July and 4 August, the lava had expanded 30-50 m horizontally (except on the N side where it had reached the crater wall, 4:6) and about 6 m vertically, to a mean diameter of 820 m and a mean height of 85 m. Assuming 45° sides, its volume was 36.5 x 10⁶ m³ on 4 August, having increased an average of 0.36 x 10⁶ m³/day in July (similar to the June rate).

During the August survey, some pulsing steam emission took place, mostly at the S edge of the lava, but no explosions were observed. The summit seismic station recorded 50-200 small earthquakes per day during July, but none of these were detected by instruments on the flanks.

Information Contacts: J. Tomblin, UWI.

August 1979 (SEAN 04:08) Cite this Report

Lava extrusion slows

Lava extrusion continued through August. However the extrusion rate has decreased during the past 3 months, as shown in table 7. The mean diameter of the lava extrusion increased from 820 m on 4 August to 832 m on 21 August, but the highest point has remained at 130 m above the crater floor since 11 August, after 13 m of vertical growth from 10 July to 11 August.

Table 7. Rate of lava extrusion in the summit crater of Soufrière St. Vincent, 25 May-25 October 1979, calculated by John Tomblin.

Between 50 and 200 very small local earthquakes continued to be recorded daily by the summit seismic station, but very few were detected by seismometers on the flanks.

When the summit was not obscured by clouds, gentle but continuous steam emission was visible. Steam frequently rose slightly above the crater rim. Water was sometimes present in the crater, especially after heavy rains, but the volume of water remained small and water depths did not exceed ~1 m.

Information Contacts: J. Tomblin, UWI.

September 1979 (SEAN 04:09) Cite this Report

Lava extrusion continues

Extrusion of lava into Soufrière's summit crater continued through September. However, the rate of extrusion continued to decrease; the late August-late September rate was an order of magnitude less than peak values observed in mid-May.

Between 21 August and 23 September, the mean diameter of the lava body grew from 832 m to 840 m, covering about 60% of the crater floor, and its maximum height increased 1 m to 131 m. Its volume was calculated at 37.6 x 10⁶ m³.

Small seismic events continued to be recorded by the summit seismograph. There has been no significant change in seismicity since a several-day increase in the number of events in late June and early July.

Information Contacts: J. Tomblin, UWI.

October 1979 (SEAN 04:10) Cite this Report

Lava extrusion virtually stopped

By late October, extrusion of lava into Soufrière's central crater had virtually stopped. The October extrusion rate was an order of magnitude less than that of late September and two orders of magnitude less than the May rate.

The mean diameter of the lava extrusion increased by only 1.5 m between 2 and 25 October, to 870 m, and the maximum height remained ~130 m. Vigorous steaming from the lava was continuing in late October, but the number of small local earthquakes recorded by the summit seismograph had declined markedly since early October.

Information Contacts: J. Tomblin, UWI.

November 1979 (SEAN 04:11) Cite this Report

Lava extrusion stopped

No lava has been extruded into Soufrière's central crater since the survey of 25 October. However, monitoring of the volcano by the Seismic Research Unit, University of the West Indies, continues. Data from four seismometers are telemetered continuously to Trinidad, and scientists visit the volcano every weekend.

Further References. Fiske, R., and Sigurdsson, H. (eds)., 1982, Soufrière Volcano, St. Vincent: Observations of its 1979 eruption from the ground, aircraft, and satellites: Science, v. 216, no. 4550, p. 1105-1126 (11 papers).

Shepherd, J.B., Aspinall, W.P., Rowley, K.C., and others, 1979, The eruption of Soufrière Volcano, St. Vincent April-June 1979: Nature, v. 282, p. 24-28.

Shepherd, J.B., and Sigurdsson, H., 1982, Mechanism of the 1979 Explosive Eruption of Soufrière Volcano, St. Vincent: JVGR, v. 13, p. 119-130.

Sparks, R.S.J., and Wilson, L., 1982, Explosive volcanic eruptions - V. Observations of plume dynamics during the Soufrière eruption, St. Vincent: Geophysical Journal of the Royal Astronomical Society, v. 69, p. 551-570.

Information Contacts: J. Tomblin, UWI.

March 2005 (BGVN 30:03) Cite this Report

Anomalous winds spread sulfurous odors, causing unwarranted fears

Widespread sulfurous odors and haze during mid-February 2005 on the island of St. Vincent and as far as the Grenadines (50-75 km S) led some people to conclude that the smells reflected increased output of volcanic gases from the Soufrière volcano, St. Vincent, a possible harbinger of an eruption. Sulfurous odors are common on the volcano's W flank, but less frequent on other parts of the island. Scientists determined that typical winds diminish the sulfurous odors over much of the island, and the onset of the odors resulted from changes in wind patterns rather than increased gas output or other demonstrable changes.

The Seismic Research Unit (SRU) collaborates with a small local unit called the Soufrière Monitoring Unit (which operates from the Ministry of Agriculture in Kingstown). The following report on the subject comes from SRU's Richard Robertson.

"During the night of 16 February and most of the day of 17 February there were widespread reports of sulfurous smells throughout southern St. Vincent and as far as the Grenadines. The day of the 17th was hazy; people put these two things together and came up with the conclusion that the volcano was acting up. The sulfur smell is unusual since the wind direction is such that most of the smell from the fumaroles at the summit of the volcano gets blown out to sea and is usually only smelt by a few residents on the eastern flank of the volcano.

"[SRU] . . . worked with Ms. Aisha Samuels, the head of the local volcano monitoring unit, to first investigate the report and later to quell fears that the volcano was doing anything unusual. We determined very early on that nothing serious was happening, since we have seismic stations both on the volcano and throughout the island [figure 3], none of which had recorded any increased seismicity. Further, we had just completed a GPS campaign on the island during January 2005, which revealed nothing unusual. It also involved two days of measurements on the summit of the volcano during which scientists were in very close proximity to the vent from which future eruptions will [likely] originate.

Figure 3. A sketch map showing the island of St. Vincent, including Soufrière volcano, other volcanic centers, geographic features, and Seismic Research Unit monitoring instrumentation (as of February 2004). January 2005 discussion of the instrumentation noted that it then included five seismic stations, eight GPS stations, and several dry-tilt sites. Courtesy of SRU.

"We quickly determined that the reported 'activity' was due to an unusual southerly wind combined with the phenomena of Sahara dust which is common around this time of the year in St. Vincent and which results in very hazy conditions. However, to completely rule out the possibility of anything unusual happening in the crater that may not have been possibly detected by our various measurements, we advised the local Unit that they should visit the crater summit the next day (18 February)."

That visit found nothing out of the ordinary. Accordingly, SRU did not think it necessary to update their website since it was so insignificant—"'a 10 day wonder' as they say in the West Indies, or a 'pseudo-crisis.'" Such reports are common for St. Vincent and the entire region.

Information Contacts: Richard Robertson, Seismic Research Unit, The University of the West Indies, St. Augustine, Trinidad (URL: http://www.uwiseismic.com/); Aisha Samuel, Soufriere Monitoring Unit, Ministry of Agriculture, St. Vincent.

March 2021 (BGVN 46:03) Cite this Report

New lava dome on the SW edge of the main crater in December 2020

Soufrière St. Vincent is the northernmost stratovolcano on St. Vincent Island in the southern part of the Lesser Antilles. The NE rim of the 1.6-km-wide summit crater is cut by a crater (500 m wide and 60 m depth) that formed in 1812. Recorded eruptions date back to 1718, with notable eruptions occurring in 1812, 1902, and 1979. The eruption of 1979 was characterized by ashfall, pyroclastic flows, and lahars, in addition to a series of Vulcanian explosions during 13-26 April 1979 that destroyed the lava dome in the summit crater, which had formed during a 1971 effusive eruption (SEAN 04:04). As a result, more than 20,000 people were evacuated. Beginning around 3 May 1979 another lava dome began to form in the main crater (SEAN 04:05; Shepherd et al., 1979) that continued to grow until the end of October 1979, expanding to 850 m in diameter and 120 m high (SEAN 04:11; Cole et al., 2019).

No further eruptive activity took place until December 2020, when a new lava dome began to grow SW of the pre-existing 1979 lava dome, accompanied by increased seismicity, crater incandescence, and gas-and-steam emissions. This report reviews information through February 2021 using bulletins from the University of the West Indies Seismic Research Centre (UWI-SRC), the National Emergency Management Organisation (NEMO), and various satellite data. Soufrière St. Vincent is monitored by the SRC assisted by the Soufrière Monitoring Unit (SMU) from the Ministry of Agriculture in Kingstown. As of 2004, the monitoring network had consisted of five seismic stations, eight GPS stations, and several dry tilt sites. Seismic data are transmitted from field sites to the Belmont Observatory (9 km SSW), which is operated by the SMU (figure 4). On 1 January 2021 a new seismic station was installed at Georgetown, on 10 January one was installed in Owia, followed on 15 January by another on the upper S flank, station SSVA at the summit on 18 January, and in Fancy on 21 January. In February 2021 the USGS-USAID (US Geological Survey-US Agency for International Development), through the Volcano Disaster Assistance Program (VDAP), donated equipment to build four more seismic stations.

Figure 4. Location map of the Belmont Observatory (yellow star) located in Rosehall, St. Vincent, 9 km SSW from the Soufrière St. Vincent summit crater (red triangle). Base map satellite imagery courtesy of Google Earth.

A spike in seismicity was recorded during June-July 2019 (figure 5), though no cause was reported. The number of events sharply declined after July but continued intermittently through November 2020. Seismicity began to increase in early November through 23 December 2020, which included 126 earthquakes described as volcano-tectonic events and rockfall signals that were captured on one reliable seismic station (SVB) located 9 km from the volcano. The maximum daily count was 11 events on 16 November. After 23 December a total of eight events were detected before seismicity briefly subsided.

Figure 5. Daily count of volcanic earthquakes recorded at Soufrière St. Vincent during 1 January 2019 through February 2021. Increased seismicity was detected during June-July 2019 and mid-October 2020 through February 2021. An installation of station SVV on 6 January 2021 at Wallibou is annotated on this graph. Data courtesy of UWI-SRC.

Activity during December 2020. Staff members of the Soufrière Monitoring Unit (SMU) made visual observations of the crater on 16 December and reported minor changes in fumarolic activity and a small lake on the E side of the crater floor. On 27 December UWI-SRC and NEMO reported that an effusive eruption had begun, which was characterized by a new lava dome in the main crater on the SW perimeter of the 1979 dome (figures 6 and 7). A thermal hotspot in the crater was also detected that day using satellite data by NASA FIRMS. As a result, the Volcanic Alert Level (VAL) was raised to Orange (the second highest level on a four-color scale) on 29 December (figure 8). The Volcano Ready Communities Project, a collaboration between NEMO SVG and UWI Seismic Research Centre, distributed their volcano hazard map for the surrounding communities, in preparation for a potential evacuation (figure 9).

Figure 6. Photo of the first documented observation of the new lava dome at Soufrière St. Vincent on 27 December 2020 taken from the E side of the summit. Courtesy of Melanie Grant, IG, UWI-SRC.

Figure 7. Photo of an early observation of the new lava dome at Soufrière St. Vincent on 29 December 2020 growing WSW of the 1979 lava dome on the SW edge of the summit crater, accompanied by gas-and-steam emissions. The dome was estimated to be 60 m high on 30 December. Courtesy of Kemron Alexander (color corrected), SMU, UWI-SRC.

Figure 8. Volcanic Hazard Alert Level System for Soufriere St. Vincent. Courtesy of UWI-SRC.

Figure 9. Volcanic hazard map for Soufrière St. Vincent, showing different areas that are likely to experience hazardous volcanic events which would require evacuations. The hazard map is divided into four zones: Zone 1 (Red), which is a very high hazard location; Zone 2 (Orange), which is a high hazard location; Zone 3 (Yellow), which is a moderate hazard location; and Zone 4 (Green), which is a low hazard location. This poster was created prior to the current eruption as part of the Volcano Ready Communities Project, a collaboration between NEMO SVG and UWI Seismic Research Centre. Courtesy of UWI-SRC and NEMO.

Activity during January-February 2021. Observations made during a field visit on 5 January, during a helicopter overflight on 6 January, and based on 9 January drone video noted that the new dome was expanding to the W on the WSW edge of the 1979 lava dome and continued to gradually grow through February 2021 (figure 10). Growth of the 2020/21 lava dome produced small, hot rockfalls and gas-and-steam emissions that were visible from the Belmont Observatory. The gas emissions were most notable from a small depression at the top of the dome. Two seismic stations were installed on the flank of the volcano at Wallibou (SVV) and at the summit (SSVA) on 6 and 18 January, respectively.

Figure 10. Map showing the growth of the new 2020/21 lava dome at Soufrière St. Vincent from 27 December 2020 to 12 February 2021. The dome is located on the SW edge of the crater rim and WSW of the 1979 lava dome that is covered in vegetation. Courtesy of UWI-SRC.

Seismic stations recorded 573 events through 0730 on 30 January; this number continued to grow into February (up to 703 events by 0830 on 4 February) (figure 5). Observations on 14 January showed that the dome was growing taller and expanding to the E and W. An overflight on 15 January showed extensive vegetation damage on the E, S, and W inner crater walls; damage previously noted on the upper SW crater rim had expanded downslope (figure 11). Scientists visited on 16 January and recorded temperatures of 590°C at the dome surface (figure 12). During 15-17 January residents to the W of the volcano reported nighttime crater incandescence. Persistent gas-and-steam emissions were observed rising above the dome, as well as from the contact between the 2020/21 and 1979 domes during the rest of the month and through February.

Figure 11. Oblique aerial view of the lava dome at Soufrière St. Vincent between the 1979 dome and the SW crater rim on 15 January 2021, accompanied by gas-and-steam emissions. On this day, the dome was 340 m long, 160 m wide, and 80 m high. Courtesy of Adam Stinton, MVO, UWI-SRC.

Figure 12. Thermal measurements were taken at the base of the freshly extruded lava dome at Soufrière St. Vincent on 16 January 2021. Top: Photo (color corrected) of the base of the new lava dome. Bottom: Thermal FLIR (Forward-Looking InfraRed) image of the base of the new lava dome showing a maximum temperature of 590.8°C. Courtesy of Adam Stinton, MVO, UWI-SRC.

Sulfur dioxide emissions were first detected on 1 February using a Multi-Gas Instrument and a filter pack; the dome had reached an estimated volume of 5.93 million cubic meters. Vegetation on the NW part of the crater (N of the dome) was damaged, likely due to fire. The dome continued to expand laterally to the N and S, according to reports issued on 6 and 8 February. After that it grew about 15 m to the NW and SE, according to 11 and 15 February reports (figure 13). NEMO reported that the growth rate of the lava dome ranged from 1.9 to 2.13 m³/s (figure 14). Active gas-and-steam emissions originated dominantly at contact areas between the pre-existing 1979 dome and the 2020/21 dome, as well as at the top of the new dome.

Figure 13. Photo of the 2020/21 lava dome (dark mass at left) at Soufrière St. Vincent on 12 February 2021 showing continuous gas-and-steam emissions and damaged vegetation on the 1979 lava dome (right). On this day, the dome was 618 m long, 232 m wide, 90 m high, and an estimated volume of 6.83 million cubic meters. Courtesy of Kemron Alexander, SMU, UWI-SRC.

Figure 14. Estimated lava extrusion rates and added volume of material at Soufrière St. Vincent’s 2020/21 lava dome during 27 December 2020 through 3 February 2021. Calculations were based on UAV photography and photogrammetry. Data courtesy of UWI-SRC.

Thermal satellite data. MIROVA (Middle InfraRed Observation of Volcanic Activity) analysis of MODIS satellite data shows the beginning of thermal activity in late December 2020 and continuing at a lower power into early February (figure 15). A single MODVOLC thermal alert was detected on 29 December. This activity marks the beginning of the effusive eruption and the formation of the new lava dome. Sentinel-2 thermal satellite imagery detected a thermal anomaly on the SW side of the main crater during clear weather days in January 2021, which represents the active 2020/21 lava dome (figure 16). Fresh, hot material is also visible surrounding the thermal anomaly, which demonstrates the growth of the lava dome over time.

Figure 15. Thermal activity at Soufrière St. Vincent was detected beginning in late December 2020 and continued through early February 2021, as reflected in the MIROVA data (Log Radiative Power). The power of the thermal anomalies had slightly decreased after December. Courtesy of MIROVA.

Figure 16. Sentinel-2 thermal satellite imagery showing a persistent thermal anomaly (bright yellow-orange) in Soufrière St. Vincent’s growing lava dome on the WSW edge of the main crater during 3 January through 28 January 2021. The dark black color is the freshly cooled material from the effusive activity, which also demonstrates the increasing size of the lava dome. Images using “Atmospheric penetration” rendering (bands 12, 11, 8a). Courtesy of Sentinel Hub Playground.

Field work during mid-January 2021. SRC collected rock samples from the new lava dome and sent them to scientists from the University of East Anglia, University of Plymouth, and University of Oxford on 16 January 2021 as a collaborative project to analyze their composition and compare them with the composition of rocks erupted in 1902, 1971, and 1979. Analyses showed that the new 2020/21 lava dome was basaltic andesite, similar in composition to the earlier domes (figure 17).

Figure 17. Backscattered electron image of a sample from the 2020/21 lava dome showing groundmass texture. Low-contrast dark gray crystals are feldspar microlites in glass (darkest gray). Some of the larger feldspar crystals have Ca-rich cores (paler gray). Clinopyroxenes also make up the groundmass (brighter gray) and some are breaking down to Fe-oxides (small oxides at edges of clinopyroxene bottom center and bottom right). In some areas dark glass is devitrifying (paler gray irregular shapes within dark gray glassy patches). Fe-Ti oxides are also common (bright white crystals). Total image width is about 0.3 mm. Image and description courtesy of Bridie Davies, UEA.

References: Cole P D, Robertson R E A, Fedele L, Scarpati C, 2019. Explosive activity of the last 1000 years at La Soufrière, St Vincent, Lesser Antilles. J. Volcanol. Geotherm. Res., 371:86-100.

Shepherd, J. B., Aspinall, W. P., Rowley, K. C., Pereira, J., Sigurdsson, H., Fiske, R. S., Tomblin, J. F., 1979. The eruption of Soufrière volcano, St Vincent April–June 1979. Nature, 282 (5734), 24–28. doi:10.1038/282024a0.

Information Contacts: University of the West Indies Seismic Research Centre (UWI-SRC), University of the West Indies, St. Augustine, Trinidad & Tobago, West Indies (URL: http://www.uwiseismic.com/); National Emergency Management Organisation (NEMO), Government of Saint Vincent and the Grenadines, Biseé, PO. Box 1517, Castries, Saint Lucia, West Indies (URL: http://nemo.gov.lc/); MIROVA (Middle InfraRed Observation of Volcanic Activity), a collaborative project between the Universities of Turin and Florence (Italy) supported by the Centre for Volcanic Risk of the Italian Civil Protection Department (URL: http://www.mirovaweb.it/); Hawai'i Institute of Geophysics and Planetology (HIGP) - MODVOLC Thermal Alerts System, School of Ocean and Earth Science and Technology (SOEST), Univ. of Hawai'i, 2525 Correa Road, Honolulu, HI 96822, USA (URL: http://modis.higp.hawaii.edu/); Sentinel Hub Playground (URL: https://www.sentinel-hub.com/explore/sentinel-playground); Google Earth (URL: https://www.google.com/earth/); Bridie Davies, University of East Anglia, Norwich Research Park, Norwich, Norfolk, NR4 7TJ, UK (URL: https://people.uea.ac.uk/bridie_davies).

May 2021 (BGVN 46:05) Cite this Report

Intermittent explosions, ash plumes, seismicity, and a growing lava dome during March-April 2021

Soufrière St. Vincent (also referred to as “La Soufrière”) is the northernmost stratovolcano on St. Vincent Island in the southern part of the Lesser Antilles. The NE rim of the 1.6-km-wide summit crater is cut by a crater (500 m wide and 60 m deep) that formed in 1812. Recorded eruptions date back to 1718, with notable eruptions occurring in 1812, 1902, and 1979. An eruption that began on 27 December 2020 was characterized by increased seismicity, the growth of a new lava dome on the SW edge of the main crater, crater incandescence, and gas-and-steam emissions (BGVN 46:03). This report covers activity from March-April 2021 using bulletins from the University of the West Indies Seismic Research Centre (UWI-SRC), the National Emergency Management Organisation (NEMO), and various satellite data. Soufrière St. Vincent is monitored by the SRC assisted by the Soufrière Monitoring Unit (SMU) from the Ministry of Agriculture in Kingstown.

As of 2004, the monitoring network had consisted of five seismic stations, eight GPS stations, and several dry tilt sites. Seismic data are transmitted from field sites to the Belmont Observatory (9 km SSW), which is operated by the SMU. On 1 March a new seismic station installed at the National Parks Interpretation Center at Bamboo Range on the E side of the volcano became operational. Additionally, a new GPS monitoring station was placed in Fancy. Two more GPS stations were installed at Jacob’s Well and Table Rock on 19 March.

Activity during March 2021. The lava dome in the main crater continued to grow during March, expanding to the N, NW, S, and SE. Gas emissions persisted in the summit area from the top of the dome and along the contact area between the pre-existing 1979 dome and the 2020-21 dome, damaging vegetation on the SW side of the volcano. A team visited the dome on 19 and 23 March to make observations, take measurements, and maintain monitoring equipment. On 19 March the dome was 105 m tall, 921 m long, and 243 m wide, with an estimated volume of 13.13 million cubic meters, based on drone surveys (figure 18). Compared to the previous drone survey on 12 February, the dome had grown roughly 6.3 million cubic meters. On 23 March at 1030 the seismic network recorded a swarm of small low-frequency events that lasted about 45 minutes, possibly associated with magma movement beneath the dome. Starting at 1653 the network also detected volcano-tectonic (VT) earthquakes, some of which had been felt in communities, including Fancy, Owia, and Sandy Bay. By 26 March the period of elevated VT earthquakes had ended; only small, low-frequency events associated with lava dome growth were recorded.

Figure 18. Map (left) showing the growth of the new lava dome at Soufrière St. Vincent from 27 December 2020 to 19 March 2021. The dome is located on the SW edge of the crater rim and WSW of the 1979 lava dome that is covered in vegetation. Photo (right) of the laterally growing lava dome (dark mass at left) taken on 19 March 2021. White gas-and-steam emissions are visible along the contact area with the 1979 dome (right), as well as damaged vegetation. Courtesy of SMU, UWI-SRC, and NEMO SVG.

Activity during April 2021. The lava dome continued to grow to the N and S during April, accompanied by small earthquakes and gas-and-steam emissions rising both from the top of the dome and along the contact with the pre-existing 1979 dome. A new swarm of VT earthquakes was detected on 5 April at 0638, the number and magnitude continued to increase until 0830, after which they stabilized. These earthquakes were more intense and averaged 50/hour; the 23-25 March swarm averaged 1.5/hour. The largest event was registered as a Mw 3.5 at 0920 and was felt by local residents. At 1400 the rate of the earthquakes started to decline and then significantly dropped at 1600, though small VT events continued through 1800. Small earthquakes associated with dome growth were recorded on 7 April.

On 8 April UWI-SRC reported an increase in the volume of white gas-and-steam emissions rising from the summit crater and at 0300 scientists at the Belmont Observatory reported episodes of tremor that indicated the movement of magma and fluids near the surface. Subsequent tremor episodes at 0530, 0800, 1015, and 1300 showed increasing magnitude. Five long-period (LP) earthquakes and two brief swarms of VT events were detected between the tremor episodes; ash venting was also observed during the last episode. Sulfur dioxide emissions were also identified in satellite data. Visual observations from the Belmont Observatory during the early evening showed that the dome height had increased significantly during the day, and incandescent material over the vent area was visible in webcam images (figure 19). As a result, the Volcanic Alert Level (VAL) was raised to Red (the highest level on a four-color scale) at 1830 and the Prime Minister issued an evacuation order for communities in the Red Zone at the N end of the island, affecting 16,000-20,000 people.

Figure 19. Webcam image of incandescent material in the summit crater of Soufrière St. Vincent on 8 April 2021, accompanied by white gas-and-steam emissions. Courtesy of UWI-SRC.

On 9 April at 0840 scientists at the Belmont Observatory reported an explosion that generated an ash plume up to 8 km altitude that drifted mainly ENE (figure 20). Ash was deposited across the island, affecting communities including Chateaubelair, Petite Bordel, the Belmont Observatory, and areas at the S end of the island; it also forced the Argyle International Airport (20 km S) to close (figure 21). NEMO stated that evacuations in the Orange and Red Zones were affected by significant ashfall and poor visibility, but by the evening most residents had been evacuated. Seismicity increased at 1130, including an earthquake swarm lasting until 1440, though tremor continued to be reported. A second explosion at 1445 produced a vigorous ash plume that initially rose to 4 km altitude with lightning visible in the plume (figure 22). Successive explosions and ash pulses fed the plume for hours, rising to 16 km altitude. A third explosive series began at 1835 (figure 23). Ash venting continued through the night and into 10 April. Periods of banded tremor associated with explosive activity and stronger pulses of ash emissions to higher altitudes began at 0330 on 10 April, lasting for periods of 20-30 minutes with 1-3 hour gaps. The resulting ash plumes rose to 10.6-16 km altitude throughout the day. The Washington VAAC stated that ash plumes during 9-10 April had drifted as far as 1,200 km ESE and about 3,000 km ENE. The Grantley Adams International Airport (GAIA) in Barbados also closed.

Figure 20. Photo of the first ash plume from Soufrière St. Vincent on 9 April 2021 that rose to 8 km altitude and drifted ENE at 0840. Courtesy of UWI-SRC.

Figure 21. Photo showing ashfall in Chateaubelair during the eruption of Soufrière St. Vincent on 9 April 2021. Courtesy of Thomas Christopher, MVO, UWI-SRC.

Figure 22. Photo of the second ash plume from Soufrière St. Vincent on 9 April 2021 that rose to 4 km altitude at 1445. Courtesy of UWI-SRC.

Figure 23. Photo of the third explosive eruption and its resulting ash plume from Soufrière St. Vincent on 9 April 2021 at 1835. Photo has been color corrected. Courtesy of UWI-SRC.

Ashfall continued to be widespread on 10 April; deposit thickness varied from less than 1 mm in Colonaire (12.5 km SSE) to 10-15 mm in Rabacca (7.4 km SSE). Analysis of satellite imagery by UWI-SRC indicated that the explosions that began on 9 April destroyed the pre-existing 1979 and 2020-21 lava domes and created a crater 800 m in diameter. During the night of 10-11 April tremor lasting up to 20 minutes continued at intervals of 1.5-3 hours, associated with periods of explosions and ash emissions. Ashfall was observed in most areas of the St. Vincent Island, as well as on neighboring islands of the Grenadines (SSW), Barbados (165 km E), and Saint Lucia (50 km NNE). Explosions early on 11 April were followed by widespread power and water outages on the island, as reported by NEMO, and some houses had collapsed due to significant ashfall (figure 24). Around midday, the time between episodes of high-amplitude tremor lengthened from 1.5-4 hours to 5-8 hours, which continued to coincide with explosions and ash emissions. Small VT earthquakes were recorded around 1800. The VAAC reported that ash plumes rose to 12.2-16 km altitude throughout the day and continued to drift long distances to the ENE, E, and SE. On 12 April at 0415 a large explosion produced an ash plume that rose to 12.8 km altitude. The plume collapsed, generating pyroclastic flows that descended several valleys on S and W flanks, reaching the coast at Morne Ronde (4.3 km W), Larikai (3.5 km WNW), and Trois Loupes Bay (3.5 km NW) (figure 25). Extensive damage to vegetation was noted along the W coast, extending from Larikai Bay to Turner Bay. High-amplitude tremor episodes had stopped, but two low-amplitude and one high-amplitude episodes were recorded during 0600-1700. Explosivity or notable ash venting coincided with the episodes.

Figure 24. Photo of collapsed structures due to the thick ash deposits from the explosions in early April 2021 at Soufrière St. Vincent. Courtesy of UWI-SRC.

Figure 25. Image of Soufrière St. Vincent taken from the Sentinel-1 satellite on 11 April 2021 showing the changes of the summit crater with a possible smaller vent in the new crater. The darker black areas to the W of the summit are pyroclastic flow deposits from earlier explosive events. Courtesy of Adam Stinton, MVO, UWI-SRC.

A series of Vulcanian explosions that began at 0630 on 13 April, and lasted about 30 minutes, was accompanied by large seismic tremors followed by more than three hours of smaller continuous tremors (figure 26). The VAAC stated that a dense ash plume rose to 11 km altitude and drifted E, SE, N, and NE (figure 27). Resulting pyroclastic flows traveled about 6 km WSW. Observations made during the afternoon indicated that pyroclastic flows had reached the ocean in every valley extending from Larikai to Wallibou, a length of about 5 km. The Washington VAAC reported a new ash emission rising to 7.6 km altitude at 1850 that was associated with a thermal anomaly. Another explosion at 2030 pulsed for more than 40 minutes and produced pyroclastic flows that traveled toward the Rabacca River on the E coast of the island. Lahar deposits were reported in the Sandy Bay area. Another large explosion was detected at 2300, according to UWI-SRC. The Montserrat Volcano Observatory stated that a sulfur odor and minor ashfall from Soufrière St. Vincent was observed on their island (390 km NNW).

Figure 26. Plot summarizing the seismic data at Soufrière St. Vincent during 8-21 April 2021. The top panel shows the Real-Time Seismic Amplitude Measurement (RSAM) data from seismic stations SSVA and SVV. The SSVA data shows the banded tremor that started on 8 April and the continuous tremor prior to the first explosion. The red triangles represent the time of explosions, based on satellite imagery that was analyzed by scientists at the University of East Anglia. Data transmission from SSVA stopped when it was likely destroyed by an explosion. The SVV data shows the high-amplitude tremor generated during a series of explosions. The bottom two panels show the number of volcano-tectonic (VT) and long-period (LP) earthquakes detected by the seismic network. The number and intensity of LP earthquakes have increased since late 11 April. Courtesy of Roderick Stewart, MVO, UWI-SRC.

Figure 27. Photo of a strong ash plume rising above Soufrière St. Vincent on 13 April 2021. Courtesy of Thomas Christopher, MVO, UWI-SRC.

During 14 April seismicity continued to be characterized by episodes of tremor about 13-15 hours apart, separated by swarms of small LP earthquakes. At 1135 a band of seismic tremor, associated with an explosion, was recorded, though the emissions were gas-rich and less energetic. Another episode of tremor began at 0230 on 15 April, as well as increased ash venting. The seismic network recorded a near-constant swarm of LP and hybrid earthquakes throughout the day with three brief episodes of low-level tremor each lasting less than 30 minutes. Tremor starting at 2100 lasted about 40 minutes and may have been associated with a minor increase in venting. Sulfur dioxide emissions were successfully measured by ground-based instruments; scientists on a Coast Guard boat along the W coast recorded 809 tons/day.

Periods of tremor and near-constant LP and hybrid earthquake swarms continued on 16 April. One tremor episode on 16 April at 0615 that lasted about three minutes was accompanied by an explosion, followed by more than two hours of low-level tremor. The resulting ash plumes rose to 8 km altitude and drifted W. Sulfur dioxide measurements showed an average of 460 tons/day; some SO₂ plumes were reported to have reached India.

Satellite images showed a series of changes to the newly formed crater in the SW part of the crater (figure 28). Updated measurements were taken of the new crater, by 17 April according to an Institut de Physique du Globe de Paris scientist. It was approximately 900 m N to S, 750 m E to W, and at least 100 m deep, based on RADAR data (figure 29). Several vents were identified or inferred by emissions points of gas-and-steam and ash plumes. During 17 and 18 April the average SO₂ flux was 232 and 391 tons/day, respectively.

Figure 28. New crater features were visible at the Soufrière St. Vincent summit in a 15 April 2021 satellite image from Planet Labs. Continued explosions created a new crater (red line) in the SW part of the larger summit crater that was approximately 900 m N-S, at least 750 m E-W, and 100 m deep. The orange patterned section represents tephra deposits. The yellow line outlines the 2020-21 lava dome. Several vents were established, based on gas-and-steam and ash plumes; only one could be seen clearly (black outline). Courtesy of UWI-SRC.

Figure 29. Satellite radar image from Capella Space acquired on 14 April 2021 showed the changes of the Soufrière St. Vincent summit crater following numerous explosions. The large depression in the middle of the image used to contain the 1979 and 2020-21 lava domes; explosions that began on 9 April destroyed both, leaving a smaller crater about 800 m in diameter. The base of the crater is distorted due to the presence of ash in the atmosphere and the ongoing eruption. North in the image is to the upper left corner. Courtesy of Adam Stinton, MVO, UWI-SRC.

A period of tremor at 1649 on 18 April was accompanied by an explosion that sent an ash plume to 8 km altitude and drifted S and SW (figure 30). This was the 30th distinct explosion since the explosive phase began. The period of tremor lasted until about 2100 and was followed by ongoing small VT, LP, and hybrid earthquakes. The rate of these events decreased by 0100 on 19 April. A single rockfall event was detected at 0139, possibly due to the growing lava dome, though there was no visual confirmation. Few, but intermittent, rockfall events, VT earthquakes, and an increasing number of small LP and hybrid earthquakes were detected through 21 April (see figure 26). The seismic station at Bamboo Range recorded a lahar on 20 April at 0400 which lasted about 30 minutes and may have descended a SE-flank valley. SO₂ measurements taken near the W coast showed an average flux of 350 tons/day. NEMO reported that 12,775 people were displaced by 19 April, with 6,208 people in 85 public shelters and 6,567 people (1,800 families) in private shelters.

Figure 30. Photo of ash plumes rising above Soufrière St. Vincent and drifted S and SW on 18 April 2021. Photo has been color corrected. Courtesy of Thomas Christopher, MVO, UWI-SRC.

On 22 April at 1108 explosions generated an ash plume that rose to 8 km altitude and drifted mainly N over the ocean (figure 31). A high-level seismic tremor was also detected at 1109 that lasted for 20 minutes. During the initial stages of the explosion a pyroclastic flow was observed traveling down the W flank, reaching an average speed of 33 km/hour. Around 2100 the seismic network recorded multiple lahars, though their location could not be determined. According to data from NEMO, 13,154 people had been displaced by 21 April, with 6,208 people in 85 public shelters and 6,790 people (1,618 families) in private shelters.

Figure 31. Photo of an ash plume rising to 8 km altitude above Soufrière St. Vincent at 1108 on 22 April 2021. Courtesy of Richard Robertson, UWI-SRC.

Seismicity dropped to low levels after the explosion on 22 April and remained low through 27 April; only a few VT, LP, and hybrid earthquakes were recorded. The average SO₂ flux was 992 tons/day on 23 April, again recorded from the W coast. Satellite radar imagery acquired on 24 April indicated possible growth at the lava dome following the explosions on 18 and 22 April. During an overflight on 26 April scientists observed white gas-and-steam emissions from several locations along the crater floor, though visibility was poor. No lava dome was visible, but a possible spine-like protrusion was seen through the clouds. Lahars in the Red and Orange zones were recorded by the seismic network at 0900 and 1000 on 27 April, during and after rainfall. Multiple lahars continued to be recorded by the seismic network within all drainages during 28-30 April due to heavy rainfall (figure 32). Trees brought down by lahars reached the ocean, where the floating logs created a hazard for small boats along the coastline. SO₂ measurements taken from a boat on the W coast yielded a flux of 1,036 tons/day. On 6 May the Volcanic Alert Level was lowered to Orange (the second highest on a four-color scale) because no explosions had been recorded since 22 April and the number of volcanic earthquakes had decreased significantly.

Figure 32. Photo of lahars in the Wallibou River near Soufrière St. Vincent on the morning of 29 April 2021 due to heavy rainfall. Courtesy of UWI-SRC.

Additional satellite data. Daily satellite-based sulfur dioxide measurements were notable beginning on 9 April. The initial explosion at 0840 showed relatively minor SO₂ emissions, though as the day progressed stronger plumes drifted E over the Atlantic Ocean (figure 33). As strong explosive activity continued at Soufrière St. Vincent, strong SO₂ plumes exceeding 2 Dobson Units (DUs) were detected during 9-14 April, moving generally E, based on data from the TROPOMI instrument on the Sentinel-5P satellite.

Figure 33. Strong and distinct SO2 plumes from Soufrière St. Vincent were detected by the TROPOMI instrument on the Sentinel-5P satellite during 9-14 April 2021 on days with strong explosive activity and ash venting. The plumes generally drifted E over the Atlantic Ocean. Courtesy of NASA Global Sulfur Dioxide Monitoring Page.

MIROVA (Middle InfraRed Observation of Volcanic Activity) analysis of MODIS satellite data detected a single thermal anomaly during mid-March 2021, followed by a brief break in activity (figure 34). Thermal activity resumed in April, occurring in three distinct clusters of moderate-power anomalies throughout the month. Two thermal hotspots were detected using the MODVOLC thermal algorithm on 24 April. Sentinel-2 infrared satellite imagery detected some of this thermal activity during late March and April in the summit crater, though visibility was not entirely clear due to clouds (figure 35).

Figure 34. A single thermal anomaly at Soufrière St. Vincent was detected during mid-March 2021. Thermal activity resumed in early April, occurring in three distinct clusters of moderate-power anomalies, as reflected in the MIROVA data (Log Radiative Power). Courtesy of MIROVA.

Figure 35. Sentinel-2 infrared satellite imagery showing a faint thermal anomaly (orange) in the summit crater of Soufrière St. Vincent on 24 March (top left), 3 April (top right), 8 April (bottom left), and 23 April (bottom right) 2021, though clouds covered most of the crater. Images using “Atmospheric penetration” rendering (bands 12, 11, 8a). Courtesy of Sentinel Hub Playground.

Information Contacts: University of the West Indies Seismic Research Centre (UWI-SRC), University of the West Indies, St. Augustine, Trinidad & Tobago, West Indies (URL: http://www.uwiseismic.com/); National Emergency Management Organisation (NEMO), Government of Saint Vincent and the Grenadines, Biseé, PO. Box 1517, Castries, Saint Lucia, West Indies (URL: http://nemo.gov.lc/); Washington Volcanic Ash Advisory Center (VAAC), Satellite Analysis Branch (SAB), NOAA/NESDIS OSPO, NOAA Science Center Room 401, 5200 Auth Rd, Camp Springs, MD 20746, USA (URL: www.ospo.noaa.gov/Products/atmosphere/vaac, archive at: http://www.ssd.noaa.gov/VAAC/archive.html); Hawai'i Institute of Geophysics and Planetology (HIGP) - MODVOLC Thermal Alerts System, School of Ocean and Earth Science and Technology (SOEST), Univ. of Hawai'i, 2525 Correa Road, Honolulu, HI 96822, USA (URL: http://modis.higp.hawaii.edu/); NASA Global Sulfur Dioxide Monitoring Page, Atmospheric Chemistry and Dynamics Laboratory, NASA Goddard Space Flight Center (NASA/GSFC), 8800 Greenbelt Road, Goddard, Maryland, USA (URL: https://so2.gsfc.nasa.gov/); MIROVA (Middle InfraRed Observation of Volcanic Activity), a collaborative project between the Universities of Turin and Florence (Italy) supported by the Centre for Volcanic Risk of the Italian Civil Protection Department (URL: http://www.mirovaweb.it/); Sentinel Hub Playground (URL: https://www.sentinel-hub.com/explore/sentinel-playground); Jenni Barclay, University of East Anglia, Norwich Research Park, Norwich, Norfolk, NR4 7TJ, UK (URL: https://research-portal.uea.ac.uk/en/persons/jenni-barclay); Ian Renfrew, University of East Anglia, Norwich Research Park, Norwich, Norfolk, NR4 7TJ, UK (URL: https://research-portal.uea.ac.uk/en/persons/ian-renfrew).