Intermittent activity most months, October 2018-June 2019; photographs during February and May 2019

Historical observations of eruptive activity from the glacier-covered Mount Michael stratovolcano on Saunders Island in the South Sandwich Islands were not recorded until the early 19th century at this remote site in the southernmost Atlantic Ocean, and remain extremely rare. With the advent of satellite observation technology, indications of more frequent eruptive activity have become apparent. Vapor emission is frequently reported from the summit crater, and AVHRR and MODIS satellite imagery has revealed evidence for lava lake activity in the summit crater (Lachlan-Cope and others, 2001). Limited thermal anomaly data and satellite imagery indicated at least intermittent activity during May 2000-November 2013, and from November 2014 through April 2018 (Gray and others, 2019). Ongoing observations, including photographs from two site visits in February and May 2019 suggest continued activity at the summit during most months through May 2019, the period covered in this report. Information, in addition to on-site photographs, comes from MIROVA thermal anomaly data, NASA SO₂ instruments, and Sentinel-2 and Landsat satellite imagery.

Near-constant cloud coverage for much of the year makes satellite data intermittent and creates difficulty in interpreting the ongoing nature of the activity. Gray and others (2019) concluded recently after a detailed study of shortwave and infrared satellite images that there was continued evidence for the previously identified lava lake on Mount Michael since January 1989. MIROVA thermal anomaly data suggest intermittent pulses of thermal energy in September, November, and December 2018, and April 2019 (figure 17). Satellite imagery confirmed some type of activity, either a dense steam plume, evidence of ash, or a thermal anomaly, each month during December 2018-March 2019. Sulfur dioxide anomalies were recorded in January, February, and March 2019. Photographic evidence of fresh ash was captured in February 2019, and images from May 2019 showed dense steam rising from the summit crater.

After satellite imagery and thermal anomaly data in late September 2018 showed evidence for eruptive activity (BGVN 43:10, figure 16), a single thermal anomaly in MIROVA data was recorded in mid-November 2018 (figure 17). A rare, clear Sentinel-2 image on 2 December revealed a dense steam plume over the active summit crater; the steam obscured the presence of any possible thermal anomalies beneath (figure 18).

Clear evidence of recent activity appeared on 1 January 2019 with both a thermal anomaly at the summit crater and a streak of ash on the snow (figure 19). Steam was also present within the summit crater. A distinct SO₂ anomaly appeared in data from the TROPOMI instrument on 14 January (figure 20).

Multiple sources of satellite data and sea-based visual observation confirmed activity during February 2019. SO₂ emissions were recorded with the TROPOMI instrument on 10, 11, 15, and 16 February (figure 21). A Landsat image from 10 February showed a dense steam plume drifting NW from the summit crater, with the dark rim of the summit crater well exposed (figure 22). Sentinel-2 images in natural color and atmospheric penetration renderings identified a dense steam plume drifting S and a thermal anomaly within the summit crater on 15 February (figure 23). An expedition to the South Sandwich Islands between 15 February and 8 March 2019 sponsored by the UK government sailed by Saunders in late February and observed a stream of ash on the NNE flank beneath the cloud cover (figure 24).

Faint SO₂ emissions were recorded twice during March 2019 (figure 25), and a dense steam plume near the summit crater was visible in Landsat imagery on 23 March (figure 26). Two thermal anomalies were captured in the MIROVA data during April 2019 (figure 17).

A volcano-related research project "SSIVOLC" explored the South Sandwich Islands volcanoes during 15 April-31 May 2019. A major aim of SSIVOLC was to collect photogrammetric data of the glacier-covered Mount Michael (Derrien and others, 2019). A number of still images were acquired on 17 and 22 May 2019 showing various features of the island (figures 27-30). The researchers visually observed brief, recurrent, and very weak glow at the summit of Mount Michael after dark on 17 May, which they interpreted as reflecting light from an active lava lake within the summit crater.

References: Lachlan-Cope T, Smellie J L, Ladkin R, 2001. Discovery of a recurrent lava lake on Saunders Island (South Sandwich Islands) using AVHRR imagery. J. Volcanol. Geotherm. Res., 112: 105-116.

Gray D M, Burton-Johnson A, Fretwell P T, 2019. Evidence for a lava lake on Mt. Michael volcano, Saunders Island (South Sandwich Islands) from Landsat, Sentinel-2 and ASTER satellite imagery. J. Volcanol. Geotherm. Res., 379:60-71. https://doi.org/10.1016/j.volgeores.2019.05.002.

Derrien A, Richter N, Meschede M, Walter T, 2019. Optical DSLR camera- and UAV footage of the remote Mount Michael Volcano, Saunders Island (South Sandwich Islands), acquired in May 2019. GFZ Data Services. http://doi.org/10.5880/GFZ.2.1..2019.003

Information Contacts: 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/); 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/); Sentinel Hub Playground (URL: https://www.sentinel-hub.com/explore/sentinel-playground); Chris Darby (URL: https://twitter.com/ChrisDDarby, image at https://twitter.com/ChrisDDarby/status/1100686838568812544).

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

February 2003 (BGVN 28:02) Cite this Report

Lava lake detected in satellite imagery during 1995-2002

Although previous eruptions have been recorded in the South Sandwich Islands (Coombs and Landis, 1966), ongoing volcanic activity has only recently been detected and studied. These islands (figure 1) are all volcanic in origin, but sufficiently distant from population centers and shipping lanes that eruptions, if and when they do occur, currently go unnoticed. Visual observations of the islands probably do not occur on more than a few days each year (LeMasurier and Thomson, 1990). Satellite data have recently provided observations of volcanic activity in the group, and offer the only practical means to regularly characterize activity in these islands.

Using Advanced Very High Resolution Radiometer (AVHRR) data, Lachlan-Cope and others (2001) discovered and analyzed an active lava lake on the summit of Saunders Island (figure 2) that was continuously present for intervals of several months between March 1995 and February 1998; plumes originating from the island were observed on 77 images during April 1995-February 1998. J.L. Smellie noted that during helicopter overflights on 23 January 1997 (Lachlan-Cope and others, 2001) "dense and abundant white steam was emitted from the crater in large conspicuous puffs at intervals of a few seconds alternating with episodes of less voluminous, more transparent vapour." Smellie also observed that the plume commonly extended ~8-10 km downwind.

The MODIS Thermal Alert system also detected repeated thermal anomalies throughout 2000-2002 in the summit area (figure 3), indicating that activity at the lava lake has continued. Anomalous pixels (1 km pixel size) were detected intermittently and were all 1-2 pixels in size, consistent with the relatively small confines of the crater. The timing of anomalous images in this study likely has more to do with the viewing limitations imposed by weather (persistent cloud cover masks any emitted surface radiance in the majority of images) than it has to do with fluctuations in activity levels, so this plot of radiance (figure 4) should not be used as a proxy for lava lake vigor.

References. Coombs, D.S., and Landis, C.A., 1966, Pumice from the South Sandwich eruption of March 1962 reaches New Zealand: Nature, v. 209, p. 289-290.

Holdgate, M.W., and Baker, P.E., 1979, The South Sandwich Islands, I, General description: British Antarctic Survey Science Report, v. 91, 76 p.

Lachlan-Cope, T., Smellie, J.L., and Ladkin, R., 2001, Discovery of a recurrent lava lake on Saunders Island (South Sandwich Islands) using AVHRR imagery: Journal of Volcanology and Geothermal Research, v. 112, p. 105-116.

LeMasurier, W.E., and Thomson, J.W. (eds), 1990, Volcanoes of the Antarctic Plate and Southern Oceans: American Geophysical Union, Washington, D.C., AGU Monograph, Antarctic Research Series, v. 48.

Wright, R., Flynn, L.P., Garbeil, H., Harris, A.J.L., and Pilger, E, 2002, Automated volcanic eruption detection using MODIS: Remote Sensing of Environment, v. 82, p. 135-155.

Information Contacts: Matt Patrick, Luke Flynn, Harold Garbeil, Andy Harris, Eric Pilger, Glyn Williams-Jones, and Rob Wright, HIGP Thermal Alerts Team, Hawai'i Institute of Geophysics and Planetology (HIGP) / School of Ocean and Earth Science and Technology (SOEST), University of Hawai'i, 2525 Correa Road, Honolulu, HI 96822, USA (URL: http://modis.higp.hawaii.edu/); John Smellie, British Antarctic Survey, Natural Environment Research Council, High Cross, Madingly Road, Cambridge CB3 0ET, United Kingdom (URL: https://www.bas.ac.uk/).

March 2004 (BGVN 29:03) Cite this Report

A MODIS thermal alert on 7 May 2003 (no secondary confirmation)

The only previous report on the remote Michael volcano was in BGVN 28:02, which commented on a lava lake detected by satellite imagery over the period 1995-2002. A review of MODIS data for the period from that report (end 2002) to 16 March 2004 (UTC) reveals one thermal alert, on 7 May 2003 (UTC). No corroborative report is available, although previous alerts were interpreted as possibly representing lava lake activity.

Information Contacts: Rob Wright, Luke Flynn, and Eric Pilger; MODIS Thermal Alert System, Hawaii Institute of Geophysics and Planetology (HIGP), School of Ocean and Earth Science and Technology, University of Hawaii at Manoa (URL: http://modis.higp.hawaii.edu/).

April 2006 (BGVN 31:04) Cite this Report

Lack of new thermal signals suggesting any eruption since October 2005

The last reported activity of Mount Michael was noted in the SI/USGS Weekly Report of 12-18 October 2005. At that time the first MODVOLC alerts for the volcano since May 2003 indicated an increased level of activity in the island's summit crater and a presumed semi-permanent lava lake that appeared confined to the summit crater. Those alerts occurred on 3, 5, and 6 October 2005. Since that time there has been no additional information concerning Mount Michael and presumably little to no activity.

Information Contacts: Matt Patrick, Luke Flynn, Harold Garbeil, Andy Harris, Eric Pilger, Glyn Williams-Jones, and Rob Wright, HIGP Thermal Alerts Team, Hawai'i Institute of Geophysics and Planetology (HIGP) / School of Ocean and Earth Science and Technology (SOEST), University of Hawai'i, 2525 Correa Road, Honolulu, HI 96822, USA (URL: http://modis.higp.hawaii.edu/); John Smellie, British Antarctic Survey, Natural Environment Research Council, High Cross, Madingly Road, Cambridge CB3 0ET, United Kingdom (URL: https://www.bas.ac.uk/).

October 2006 (BGVN 31:10) Cite this Report

Clear IR satellite view on 28 October 2006 suggests lava inside the crater

Matt Patrick sent a new Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) image, collected 28 October 2006 over Saunders Island . In his opinion this is the best image collected to date owing to the lack of a plume obscuring the summit crater, which was a problem in all previous images. The improved image provides a clear view of the crater (figures 5 and 6).

Analyzing the VNIR, SWIR, and Thermal Infrared (TIR) (not shown in figures 5 or 6) images together shows that the outer crater is 500-600 m wide, with a 180m high-temperature crater interior. The latter shows up as an SWIR anomaly and may indicate the rough extent of active lava flow being ~ 180 m wide. Matt Patrick chose Villarrica volcano in Chile for comparison to Mt. Michael (figure 7) since it presents a potentially good analogue in terms of morphology and activity style. Maximum radiant heat flux values were similar (up to ~ 150 MW), suggesting that the maximum intensity of activity may be similar. Mt. Michael shows a much lower frequency of thermal alerts, which may be the result of more frequent cloud cover in the South Sandwich Islands or a greater depth to molten lava in the Mt. Michael crater.

Table 1 shows a summary of thermal anomalies and possible eruptions from Moderate Resolution Imagine Spectroradiometer (MODIS) satellites since November 2005. The last reported activity of Mount Michael was noted in the SI/USGS (Smithsonian Institution/U.S. Geological Survey) Weekly Volcanic Activity Report of 12-18 October 2005 (see BGVN 31:04). At that time the first MODVOLC alerts for the volcano since May 2003 indicated an increased level of activity in the island's summit crater and a presumed semi-permanent lava lake that appeared confined to the summit crater. Those alerts occurred on 3, 5, and 6 October 2005.

Table 1. Thermal anomalies measured by MODIS satellites for Mount Michael for the period 3 October 2005 to 1 November 2006. All of the anomalies appeared on the SW side of the volcano. Courtesy of Hawai'i Institute of Geophysics and Planetology (HIGP) Thermal Alerts Team.

Date	Time (UTC)	Number of pixels	Satellite
01 Nov 2006	0125	1	Terra
31 Oct 2006	1600	1	Aqua
21 Oct 2006	1120	1	Terra
20 Oct 2006	0250	2	Aqua
20 Oct 2006	0100	3	Terra
21 Jul 2006	0120	1	Terra
09 Jun 2006	0920	2	Aqua
21 Jan 2006	0100	1	Terra
20 Dec 2005	0100	1	Terra
06 Oct 2005	0115	1	Terra
05 Oct 2005	0220	1	Aqua
03 Oct 2005	0045	1	Terra

References. Lachlan-Cope, T., Smellie, J.L., and Ladkin, R., 2001, Discovery of a recurrent lava lake on Saunders island (South Sandwich Islands) using AVHRR imagery: Journal of Volcanology and Geothermal Research, vol. 112, no. 1-4, p. 105-116 (authors are members of the British Antarctic Survey).

LeMasurier, W.E., and Thomson, J.W. (eds), 1990, Volcanoes of the Antarctic Plate and Southern Oceans: American Geophysical Union, Washington, D.C., AGU Monograph, Antarctic Research Series, v. 48.

Information Contacts: Matt Patrick, Michigan Technological University, Houghton, MI; Thermal Alerts Team, Hawai'i Institute of Geophysics and Planetology (HIGP), School of Ocean and Earth Science and Technology (SOEST), University of Hawai'i, 2525 Correa Road, Honolulu, HI 96822, USA (URL: http://modis.higp.hawaii.edu/); John Smellie, British Antarctic Survey, Natural Environment Research Council, High Cross, Madingly Road, Cambridge CB3 0ET, United Kingdom (URL: https://www.bas.ac.uk/); ASTER Science Project Teams, United States and Japan (URL: https://asterweb.jpl.nasa.gov/).

April 2008 (BGVN 33:04) Cite this Report

No thermal anomalies detected since December 2006

The frigid, remote, and uninhabited region of Michael volcano is seldom visited. Thermal anomalies detected by satellite-based MODIS instruments, processed using the MODVOLC algorithm by the Thermal Alerts System of the Hawai'i Institute of Geophysics and Planetology, provide some data about possible eruptive activity (BGVN 28:02, 29:03, 31:04, and 31:10). During 3-6 October 2005 there were three days with thermal anomalies (BGVN 31:04). MODIS data indicates that anomalous pixels were also detected on 19 December 2005 (20 December UTC) and on 20 January 2006 (21 January UTC) (BGVN 31:10). The most recently reported MODIS thermal anomalies indicated activity during 19-21 October 2006 (20-21 October UTC) and again on 31 October-1 November 2006 (BGVN 31:10). The source of these anomalies was an inferred lava lake in a central vent as shown on an Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) image collected 28 October 2006 (BGVN 31:10). Additional anomalies occurred on 13 November and 6 December 2006 (7 December UTC). No anomalies were measured after that date through May 2008. Since August 2000 there have been six periods when thermal anomalies were detected in satellite imagery (table 2).

Table 2. Eruptive periods at Michael as inferred from MODIS thermal data from January 2000 through May 2008. Courtesy of the Hawai'i Institute of Geophysics and Planetology Thermal Alerts System.

Date (UTC)	Thermal pixel data	Bulletin reference
30 Aug 2000-03 Feb 2001	3 days with pixels	BGVN 28:02
05 Aug 2001-21 Nov 2001	10 days with pixels	BGVN 28:02
05 Jul 2002-01 Nov 2002	12 days with pixels	BGVN 28:02
07 May 2003	2 anomalous pixels	BGVN 29:03
03 Oct 2005-21 Jan 2006	5 days with pixels, three during 3-6 Oct	BGVN 31:04, 31:10
09 Jun 2006-07 Dec 2006	9 days with pixels	BGVN 31:10, 33:04

Information Contacts: Hawai'i Institute of Geophysics and Planetology (HIGP) 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/).

February 2013 (BGVN 38:02) Cite this Report

Eruption from ‘new’ vent

Matthew Patrick (USGS-HVO) notified Bulletin editors that in late 2012 images from thermal sensing satellites showed a 'new' active vent on Mount Michael on Saunders Island in the South Sandwich Islands (see location map, figure 1 in BGVN 28:02). This prompted scrutiny of the same vent in earlier images. Patrick noted that, although the vent was first identified in the 2012 images, it also appeared as activity in satellite images starting in 2006. The South Sandwich Islands are generally devoid of vegetation and habitants, and are largely ice-bound. Thus, satellite thermal alerts are strong evidence of volcanism.

Patrick shared with us the following information from a paper by Patrick and Smellie (2013) about the vent, labeled as Old Crater (SE and outside of main crater, see figure 2 in BGVN 28:02). ASTER [Advance Spaceborne Thermal Emission and Reflection Radiometer] imagery provided "new information on the small subordinate crater, marked as 'Old Crater' by Holdgate and Baker (1979), presumably because it was inactive at the time of their observations." An ASTER image on 28 October 2006 showed an apparent SWIR [short-wave infrared] anomaly at Old Crater. The crater itself appeared to be snow-free and was approximately 150 m in diameter. An ASTER image from 5 January 2008, showed a steam plume coming from this vent, which appeared to be about 190 m wide, as well as a TIR [thermal infrared] anomaly. A very high resolution image from November 2009 available on Google Earth showed a small steam plume emanating from the crater, which is about 190 m wide (figure 8). An ASTER image from 17 November 2010, showed apparently recent eruptive activity in Old Crater, evidenced by tephra fallout emanating from the crater and a small TIR anomaly (at the time there was also a TIR anomaly in the main crater). According to Patrick and Smellie, the plume, tephra fall, SWIR anomalies, and crater enlargement (from 150 to 190 m) indicated that this vent had reactivated by late 2006.

MODVOLC satellite thermal alerts measured from the volcano since our last Bulletin report (BGVN 33:04, activity through May 2008) and to 4 April 2013 are shown in Table 3. A solitary alert appeared 25 October 2008, followed by a four year period of apparent inactivity. Then, another solitary alert was measured in late June 2012, followed by alerts for two days in October 2012 and two days in November 2012. Patrick noted that occasional and sporadic alerts are very typical for Michael.

Table 3. Satellite thermal alerts measured by MODVOLC over Michael from 2008-February 2013. Pixel sizes generally range from 1-1.5 km². Note that previous satellite thermal alerts for Michael were listed in BGVN 31:10 (October 2005-November 2006) and 33:04 (August 2000-May 2008). Courtesy of MODVOLC.

Date	Time (UTC)	Number of pixels	Satellite
25 Oct 2008	0100	1	Terra
30 Jun 2012	0100	1	Terra
02 Oct 2012	0110	1	Terra
28 Oct 2012	0200	2	Aqua
28 Oct 2012	1125	2	Terra
14 Nov 2012	0055	3	Terra
22 Nov 2012	1120	2	Terra

References. Patrick, M.R., and Smellie, J.L., 2013, A spaceborne inventory of volcanic activity in Antarctica and southern oceans, 2000-2010, Antarctic Science, v. 25, no. 4, p. 475-500.

Holdgate, M.W., and Baker, P.E., 1979. The South Sandwich Islands: I. General description, British Antarctic Survey Scientific Reports, No. 91, pp. 1-76.

Information Contacts: Matthew Patrick, Hawaiian Volcano Observatory (HVO), U.S. Geological Survey, PO Box 51, Hawai'i National Park, HI 96718, USA (URL: https://volcanoes.usgs.gov/observatories/hvo/); MODVOLC, Hawai'i Institute of Geophysics and Planetology (HIGP) 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/).

February 2016 (BGVN 41:02) Cite this Report

Thermal anomalies during August-October 2015

The Michael stratovolcano forms the emergent small, unpopulated, mostly ice-covered Saunders Island, part of the South Sandwich Islands. The volcano is primarily monitored using satellite imagery. Prior to August 2015, the most recent eruptive activity detected was in the second half of 2012 (BGVN 38:02).

Infrared MODIS satellite data processed by MODVOLC showed thermal alerts at the summit of Michael during 28 August to 8 October 2015 (figure 9, table 4). MODIS data presented on the MIROVA site shows increased radiative power (measured from satellite thermal data) from late August through October 2015, with a very strong maximum near the end of August (figure 10); lower-power gradually decreasing detections during late November 2015-January 2016 may represent a cooling lava flow. It should be noted that cloud cover over a volcano may prevent the measurement of infrared data.

Table 4. Satellite thermal alerts measured by MODVOLC over Michael during August-October 2015. Pixel sizes generally range from 1-1.5 km2. Courtesy of MODVOLC.

Date	Time (UTC)	Number of pixels	Satellite
28 Aug 2015	0050	1	Terra
28 Aug 2015	0235	1	Aqua
30 Aug 2015	0225	1	Aqua
31 Aug 2015	0120	1	Terra
30 Sep 2015	0130	1	Terra
02 Oct 2015	0120	2	Terra
06 Oct 2015	1610	1	Aqua
08 Oct 2015	1660	1	Aqua

Information Contacts: MIROVA, 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/)

October 2018 (BGVN 43:10) Cite this Report

Intermittent thermal pulses and satellite imagery hotspots during September 2016-September 2018

Historical observations of eruptive activity on ice-covered Mount Michael stratovolcano on Saunders Island in the South Sandwich Islands were not recorded until the early 19th century at this remote site in the southernmost Atlantic Ocean. With the advent of satellite observation technology, indications of more frequent eruptive activity have become apparent. The last confirmed eruption evidenced by MODVOLC thermal alerts was during August-October 2015 (BGVN 41:02). Limited thermal anomaly data and satellite imagery since then have indicated intermittent activity through September 2018. Information for this report comes from MODVOLC and MIROVA thermal anomaly data and Sentinel-2, Landsat, and NASA Terra satellite imagery.

Evidence for thermal activity at Mount Michael tapered off in MIROVA data from October 2015 through January 2016. MODVOLC thermal alerts reappeared on 28 September 2016 and recurred intermittently through 6 January 2017. Low-level MIROVA thermal signals appeared in June and September-November 2017. During January-September 2018, evidence for some type of thermal or eruptive activity was recorded from either MODVOLC, MIROVA, or satellite imagery each month except for May and June.

Although MODVOLC thermal alerts at Mount Michael ended on 8 October 2015, the MIROVA radiative power data showed intermittent pulses of decreasing energy into early January 2016 (figure 10, BGVN 41:02). At a high-latitude, frequently cloud-covered site such as Saunders Island, this could be indicative of continued eruptive activity. A white plume in low resolution NASA's Terra satellite data was spotted drifting away from Saunders in April 2016, but no thermal activity was reported. The only high-confidence data available from April 2016 through May 2017 is from the MODVOLC thermal alert system, which recorded two thermal alerts on 28 September 2016, one the next day, one on 30 October, and eight alerts on four days in November. Activity continued into January 2017 with one alert on 17 December 2016, and six alerts on 2 and 6 January 2017 (figure 11).

A low-level log radiative power MIROVA signal appeared in early June 2017; two more signals appeared in September 2017, one in early October and one in late November (figure 12). Additional signals plotted as more than 5 km from the source may or may not reflect activity from the volcano. Steam plumes were visible in NASA Terra satellite images drifting away from the island in August, October, and December 2017, but no thermal signatures were captured.

More sources of evidence for activity became available in 2018 with the addition of the Sentinel-2 satellite data during the months of February-April and September. Multiple thermal signals appeared from MIROVA in January 2018 (figure 13), and the first Sentinel-2 satellite image showed a distinct hotspot at the summit on 10 February (figure 14).

A MODVOLC thermal alert appeared on 26 March 2019 followed by a significant hotspot signal in Sentinel-2 imagery on 29 March (figure 15). The hotspot was still present along with a substantial steam plume on 3 April 2018. Sentinel-2 imagery on 11 April revealed a large steam plume and cloud cover, but no hotspot.

MIROVA thermal signals appeared in mid-July and mid-August 2018 (figure 13) but little satellite imagery was available to confirm any thermal activity. The next clear signal of eruptive activity was evident in a Sentinel-2 image as a hotspot at the summit on 23 September. A small MIROVA signal was recorded the same day (figure 13, bottom). A few days later, on 28 September 2018, a Landsat 8 image showed a clear streak of dark-gray ash trending NW from the summit of Mount Michael (figure 16).

Information Contacts: 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/); 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); NASA Worldview (URL: https://worldview.earthdata.nasa.gov/).

August 2019 (BGVN 44:08) Cite this Report

Intermittent activity most months, October 2018-June 2019; photographs during February and May 2019

Historical observations of eruptive activity from the glacier-covered Mount Michael stratovolcano on Saunders Island in the South Sandwich Islands were not recorded until the early 19th century at this remote site in the southernmost Atlantic Ocean, and remain extremely rare. With the advent of satellite observation technology, indications of more frequent eruptive activity have become apparent. Vapor emission is frequently reported from the summit crater, and AVHRR and MODIS satellite imagery has revealed evidence for lava lake activity in the summit crater (Lachlan-Cope and others, 2001). Limited thermal anomaly data and satellite imagery indicated at least intermittent activity during May 2000-November 2013, and from November 2014 through April 2018 (Gray and others, 2019). Ongoing observations, including photographs from two site visits in February and May 2019 suggest continued activity at the summit during most months through May 2019, the period covered in this report. Information, in addition to on-site photographs, comes from MIROVA thermal anomaly data, NASA SO₂ instruments, and Sentinel-2 and Landsat satellite imagery.

Near-constant cloud coverage for much of the year makes satellite data intermittent and creates difficulty in interpreting the ongoing nature of the activity. Gray and others (2019) concluded recently after a detailed study of shortwave and infrared satellite images that there was continued evidence for the previously identified lava lake on Mount Michael since January 1989. MIROVA thermal anomaly data suggest intermittent pulses of thermal energy in September, November, and December 2018, and April 2019 (figure 17). Satellite imagery confirmed some type of activity, either a dense steam plume, evidence of ash, or a thermal anomaly, each month during December 2018-March 2019. Sulfur dioxide anomalies were recorded in January, February, and March 2019. Photographic evidence of fresh ash was captured in February 2019, and images from May 2019 showed dense steam rising from the summit crater.

After satellite imagery and thermal anomaly data in late September 2018 showed evidence for eruptive activity (BGVN 43:10, figure 16), a single thermal anomaly in MIROVA data was recorded in mid-November 2018 (figure 17). A rare, clear Sentinel-2 image on 2 December revealed a dense steam plume over the active summit crater; the steam obscured the presence of any possible thermal anomalies beneath (figure 18).

Clear evidence of recent activity appeared on 1 January 2019 with both a thermal anomaly at the summit crater and a streak of ash on the snow (figure 19). Steam was also present within the summit crater. A distinct SO₂ anomaly appeared in data from the TROPOMI instrument on 14 January (figure 20).

Multiple sources of satellite data and sea-based visual observation confirmed activity during February 2019. SO₂ emissions were recorded with the TROPOMI instrument on 10, 11, 15, and 16 February (figure 21). A Landsat image from 10 February showed a dense steam plume drifting NW from the summit crater, with the dark rim of the summit crater well exposed (figure 22). Sentinel-2 images in natural color and atmospheric penetration renderings identified a dense steam plume drifting S and a thermal anomaly within the summit crater on 15 February (figure 23). An expedition to the South Sandwich Islands between 15 February and 8 March 2019 sponsored by the UK government sailed by Saunders in late February and observed a stream of ash on the NNE flank beneath the cloud cover (figure 24).

Faint SO₂ emissions were recorded twice during March 2019 (figure 25), and a dense steam plume near the summit crater was visible in Landsat imagery on 23 March (figure 26). Two thermal anomalies were captured in the MIROVA data during April 2019 (figure 17).

A volcano-related research project "SSIVOLC" explored the South Sandwich Islands volcanoes during 15 April-31 May 2019. A major aim of SSIVOLC was to collect photogrammetric data of the glacier-covered Mount Michael (Derrien and others, 2019). A number of still images were acquired on 17 and 22 May 2019 showing various features of the island (figures 27-30). The researchers visually observed brief, recurrent, and very weak glow at the summit of Mount Michael after dark on 17 May, which they interpreted as reflecting light from an active lava lake within the summit crater.

References: Lachlan-Cope T, Smellie J L, Ladkin R, 2001. Discovery of a recurrent lava lake on Saunders Island (South Sandwich Islands) using AVHRR imagery. J. Volcanol. Geotherm. Res., 112: 105-116.

Gray D M, Burton-Johnson A, Fretwell P T, 2019. Evidence for a lava lake on Mt. Michael volcano, Saunders Island (South Sandwich Islands) from Landsat, Sentinel-2 and ASTER satellite imagery. J. Volcanol. Geotherm. Res., 379:60-71. https://doi.org/10.1016/j.volgeores.2019.05.002.

Derrien A, Richter N, Meschede M, Walter T, 2019. Optical DSLR camera- and UAV footage of the remote Mount Michael Volcano, Saunders Island (South Sandwich Islands), acquired in May 2019. GFZ Data Services. http://doi.org/10.5880/GFZ.2.1..2019.003

Information Contacts: 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/); 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/); Sentinel Hub Playground (URL: https://www.sentinel-hub.com/explore/sentinel-playground); Chris Darby (URL: https://twitter.com/ChrisDDarby, image at https://twitter.com/ChrisDDarby/status/1100686838568812544).