Source of large pumice rafts traced to Havre seamount eruption

Large pumice rafts observed floating in the SW Pacific Ocean in the central Kemadec Islands, midway between North Island (New Zealand) and Tonga, have been traced by various investigators and monitoring systems to a mid-July 2012 eruption of the Havre submarine seamount (figure 1). The eruption was strong enough to result in thermal alerts and produce an ash plume that breached the ocean surface from a depth of at least 700 m.

Early observations. Maggie de Grauw, resident of Paeroa, New Zealand, took photographs (one of which is shown in figure 2) on 31 July 2012, from a commercial airplane (Virgin Pacific flight ##DJ94 from Apia, Samoa to Auckland, New Zealand) of a "peculiar large mass floating on the ocean between Tonga and Auckland."

She noted that there was another larger mass nearby, but was unable to photograph it because of the difficult angle. Believing it to be a pumice raft, she emailed the photos to Scott Bryan, author of an article on pumice rafts (Bryan and others, 2012), who forwarded it to Bulletin staff.

de Grauw noted that "the date stamp on the photograph says 1441 NZST [New Zealand standard time], which meant we had another 1 hour and 15 minutes flying time to Auckland. Total flight time was estimated at 3 and a half to 4 hours. Looking at the map, that would have put us some where near the Kermadec trench/ridge or islands. (Though I did not see any islands nearby)."

On 4 August, Bryan commented that the source of the pumice, if in the Kermadecs, may be Raoul, Macauley, Giggenbach, or Volcano W. He suggested that Monowai seamount seemed too far away. Brad Scott, New Zealand's GNS Science, noted that around the same time North Island of New Zealand had two volcanoes erupting (Tongariro and White Island) and seismic signals indicated that Monowai seamount (in the Kermadec Islands) was erupting as well.

On 10 August 2012, the New Zealand Defence Force (NZDF) reported an area of floating pumice in the open ocean. The area where the pumice was abundant was 463 km (250 nmi) in length and 56 km (30 nmi) wide, for a total area of 25,700 km² (7,500 nmi²). A photograph taken on 9 August 2012 shows an example of the pumice as seen from a Royal New Zealand Air Force (RNZAF) Orion patrol plane flying between Samoa and New Zealand (figure 3); video of the raft from this aircraft was shown in a press release from NZDF (2012).

The pilots relayed the information to the Royal New Zealand Navy vessel HMNZS Canterbury, which later that day encountered floating pumice ~160 km (85 nmi) WSW of Raoul island (29.27°S, 177.92°W). The Canterbury crew found that the pumice raft was ~0.6 m thick, 1 km wide, and extended to their right and left as far as the eye could see. The crew retrieved some pieces of pumice from the ship's water filters for later analysis and documentation. According to a news article by Priestley (2012) other samples of pumice were collected using buckets. These samples ranged from golf ball to soccer ball sized (figures 4 and 5). The pumice samples were "rough around the edges and irregular shapes." At that time, the origin of the pumice was still unknown.

In addition, Alain Bernard of the Laboratoire de Volcanologie, Université Libre de Bruxelles, Belgium, and Olivier Hyvernaud of the Laboratoire de Géophysique, Tahiti, observed the pumice raft in MODIS/Terra satellite images taken 3 August 2012.

Search for the pumice source. The search for the pumice source involved a number of investigators and their institutions, and several monitoring systems. Table 1 gives a summary of locations of the pumice source based on various observed phenomena associated with the July 2012 eruption as determined from various investigators.

Table 1. Summary of reported locations for events and features associated with tracking the source of the pumice rafts from the July 2012 eruption of Havre seamount. These locations may be compared with the location of Havre seamount (table 2) from Wright, Worthington and Gamble (2006). Compiled from listed references.

Source - Feature	Coordinates (as reported)	Coordinates (decimal degrees)	Dates and comments	Reference(s)
Seismic - source of pumice raft	31.13°S, 178.96°W	31.13°S, 178.96°W	17-18 July 2012, short seismic swarm	Hyvernaud (2012), Laboratoire de Géophysique, Papeete, Polynesian Network (Scott, 2012)
MODIS satellite - hot spot	31°7'S, 179°12'W	31.1°S, 179.2°W	1050 on 18 July 2012 UTC; band 22, 3.959 µm	Bernard (2012)
MODIS satellite - point of vapor plume	31°5'S, 179°1'W	31.1°S, 179.0°W	2150 on 18 July 2012 UTC; band 20, 3.75 µm; plume image "pointing to the source of the eruption"	Bernard (2012)
Satellite - source of pumice	30.95°S, 179.13°W	30.95°S, 179.13°W	19 July 2012 UTC, satellite data, raft becomes visible after 0205	Laboratoire de Géophysique, Papeete, Polynesian Network (Scott, 2012)

Table 2. Locations of Havre seamount and other nearby features, for comparison with early locations of pumice rafts' source vent (table 1). Compiled from listed references.

Feature	Coordinates (as reported)	Coordinates (decimal degrees)	Comments	Reference(s)
Havre Seamount	31°6.500'S, 179°2.450'W	31.11°S, 179.04°W	Summit depth 720 m, basal depth 1,750 m; used in this report and by Klemetti (2012a).	Wright, Worthington, and Gamble (2006)
Havre Rock	31°17.3'S, 178°54.7'W	31.29°S, 178.9°W	Summit elevation ~70 m.	New Zealand Land Information (2008)
L'Esperance Rock	31°21.4'S, 178°48.4'W	31.36°S, 178.82°W	Summit slightly above sea level.	New Zealand Land Information (2008)

GNS Science issued a news bulletin on 11 August 2012 (Scott, 2012) noting that a report from the Laboratoire de Géophysique, Tahiti, confirmed two indications of eruptive activity in the Kermadec Islands, one from satellite tracking and another from seismic monitoring.

An examination of satellite data by the Laboratoire de Géophysique traced the pumice back to a source at 30.95°S, 179.13°W, 72 km SW of Curtis Island at 0205 on 19 July 2012 UTC. According to Olivier Hyvernaud (2012), between 0733 on 17 July 2012 UTC and 0300 on 18 July 2012 UTC, 157 hydroacoustic events from Kermadec ridge were measured. He noted that "the waveforms are all very similar, with a short length and a steep rise. For some events, seismic Rayleigh and Pn phases from regional seismic stations were associated." Among the 157 events of magnitudes between 3.0 and 4.8, 68 events were located (figures 6 and 7).

According to Hyvernaud, the "mean location is at 31.13°S, 178.96°W, a position in the vicinity of Havre seamount. The best locations are obtained with a mix of hydroacoustic and seismic phases. The focal depths are impossible to constrain, but we assume that they are shallow. Usually, we record several types of hydroacoustic events during volcanic submarine activity: submarine explosions, tremors and small earthquakes. Submarine explosions and tremors are never recorded in seismic [data] (unless you have a very close seismic station). For Havre, the strongest events have both seismic and hydroacoustic [signals], that's why I interpret them as small earthquakes. The weakest have only hydroacoustic phase[s], because seismic phases are below the detection threshold. Tremors and explosions have not been recorded for Havre: why?, I don't know...Perhaps the explosive sources are on the opposite side of the volcano and couldn't propagate towards French Polynesia?..."

Alain Bernard sent an email to the Bulletin reporting that he had analyzed nighttime imagery from a MODIS satellite and found a thermal hot spot from the eruption at 1050 on 18 July 2012 UTC, the earliest evidence of a hot spot from the Havre Seamount eruption reaching the ocean surface (figure 8a). He noted that "apparently, the first appearance of pumice rafting is on MODIS/Terra [satellite images] of July 18 [2012]...There is an intriguing feature associated with the raft, it looks like a plume of vapour(?) with a clear thermal contrast as seen in band 20 at 3.75µm [figure 8b]. I really don't know what this could be and if this feature is pointing to the source of the eruption. Anyway, the geographic location is close to 31°5'S and 179°1'W but as far as I know there is no identified submarine volcano there."

Havre identified as pumice source. According to a report by Erik Klemetti (2012), he and Robert Simmon, both working independently of GNS Science and using NASA's Moderate Resolution Imaging Spectroradiometer (MODIS) Terra and Aqua images, discovered the first signs of the eruption (discolored water, gray pumice, and a volcanic plume) in imagery from 0950 and 1410 on 19 July 2012 (local time) between Macauley Island and Volcano W. NASA satellite images acquired during 18-21 July 2012 show an obvious plume on both 18 and 19 July, then only the pumice raft on 20 July, suggesting the eruption may have only lasted a couple of days (figure 9). The eruption was strong enough to generate a thermal pulse from a depth of at least 700 m that could be measured at the ocean surface by satellite (figure 8a).

To identify known features on the sea floor that might correspond to the source vent for the pumice, Klemetti overlaid the MODIS/Terra image on Google Earth to find the location relative to Macauley volcano and Volcano W. Bathymetric maps indicated that the source of the plume appeared to be a U-shaped edifice that had no label in the Smithsonian's Global Volcanism Program Google Earth layer. However, from a map of the Kermadec Islands (Smith and Price, 2006), it appeared that the edifice was Havre seamount (near Havre Rock), a relatively unknown seamount volcano without documented eruptive history.

Klemetti concluded his report saying that this event "shows how easily an eruption can happen in the middle of the ocean and not be noticed for 3 weeks - even in the 21st century!... Most eruptions will be noticed either as heat spots or sources of sulfur dioxide emissions if not visually on these satellite images. However, sometimes we get the evidence of an eruption well after it happened and have to backtrack through remote sensing data to find the source and in the case of Havre, this was the only way that the source could be found so quickly."

By 21 July 2012, the eruption appeared to have waned, leaving behind rafts of pumice. Winds and currents spread the pumice into a series of twisted filaments, spread over an area of ~450 by 250 km as of 13 August. A 31 August 2012 issue of New Zealand Notices to Mariners (New Zealand Land Information, 2012) announced a recent insertion, "Volcanic Activity," to Chart NZ 222 (SW Pacific - Kermadec Islands; New Zealand Land Information, 2008) at a position 30°57.00'S, 179°07.80'W (figure 10). This location is identical to the location for the source of the ash plume as identified from satellite images by the Laboratoire de Géophysique.

Fate of pumice rafts. According to Bernard, Hyvernaud, Klemetti, and Simmon, satellite images revealed that Havre seamount erupted a tightly-packed raft of floating pumice on 19 and 20 July 2012. Over several weeks, wind and waves dispersed the pumice to the W, NW, N, and then E. A 28 July image showed one pumice raft, twisted by ocean currents, appearing as a well-defined strand (figure 11). By 6 August, the pumice was largely dispersed, spread over an area at least 450 km wide (figure 12a). Filaments of pumice remained in the area on 13 August, and the pumice was spread over an area of ~450 by 258 km (figure 12b). None appeared to reach Raoul Island, site of a permanently staffed meteorological station.

The NASA Earth Observatory continued tracking the spread of the pumice from the Havre eruption. By 19 August 2012 the pumice was spread over an area of 270,000 km² of the Pacific Ocean and was continuing to spread. This pumice will likely stay afloat for months if not longer and eventually make landfall wherever the currents dictate - potentially as far away as South America.

According to GNS, the crew on a flight between Auckland and Apia on 1 October 2012 reported "floating pumice in the Kermadec Islands NE of New Zealand. The GeoNet duty volcanologist received this from the MetService Aviation Forecaster as part of the routine exchange of volcano data and observations between the organisations and airline pilots." "It is most likely this pumice raft is the same one [generated in mid-July 2012 and attributed to Havre as a source], just more spread out now. We have no direct evidence that Havre has erupted again."

The fate of the Havre pumice is unknown at this time, but a recent study by Scott Bryan and others (2012) details what happened to pumice from the 2006 Home Reef eruption in Tonga (see BGVN 31:09, 31:10, 31:12, 32:04, 33:05, and 33:12). That 2006 eruption (VEI 2 where the main vent was likely tens of meters below the ocean surface) was strong enough to create an ash plume that likely reached as high as 15 km altitude at its maximum, and did produce a small island that might have been as high as 75 m above sea level (wave action quickly removed the tephra forming the island). For the Home Reef eruption, the drifting pumice quickly hosted upwards of 80 different species of marine life over the course of its journey. Pumice rafts might be one of the ways that the ocean can redistribute organisms throughout the world oceans. Within eight months of the eruption, some of the pumice clasts had traveled over 5,000 km. Many clasts stayed afloat for ~2 years (Bryan and others 2012).

October 2012 cruise confirms Havre as pumice source. On 26 October 2012 the New Zealand National Institute of Water and Atmospheric Research's (NIWA) Research Vessel Tangaroa mapped Havre submarine volcano. NIWA ocean geology scientist Joshu Mountjoy announced finding a new volcanic cone which has formed on the edge of the volcano, towering 240 m above the crater rim that was first mapped in 2002 (Wright and others, 2006). The 2012 Havre eruption was strong enough to breach the ocean surface from a depth of more than 700 m by producing an ash plume, thermal alert, and a pumice raft that covered an area of 22,000 km², all visible by satellite.

According to a press release from NIWA (2012), the voyage leader, NIWA's volcanologist Richard Wysoczanski, said that "we know the shape of the volcano from previous research. Using the multibeam echosounder, we made a before and after comparison of the volcano to determine the size of the eruption and the change it has made to the seafloor." NIWA previously mapped Havre volcano in 2002 (Wright and others, 2006), showing a 1-km-high undersea mountain with a 5-km-wide, 800-m-deep central crater. This central steep-walled crater is a caldera, which is a collapse feature of volcanoes, like Lake Taupo, often known to produce large and violent eruptions.

Mountjoy noted that "One side of the caldera wall is bulging in towards the volcano's centre. The bulging may indicate where an eruption may occur in the future, or it might lead to an undersea avalanche." Several cubic kilometers of new material had been added to the volcano. Large volumes of freshly erupted pumice have accumulated on the caldera floor, raising the floor by up to 10 m. Glassy volcanic rocks were sampled from the fresh crater wall, typical of newly erupted material. Wysoczanski noted that there were new volcanic cones in one area. Volcanic rocks were collected, up to beach ball size, that vary in color and texture from black glassy material to white pumice. Round pebbles of pure sulphur were also retrieved.

Havre Seamount background. Smith and Price (2006) published one of the first bathymetric maps showing the main features of the Tonga-Kermadec arc/back-arc system and the location of Havre seamount (figure 13). Wright and others (2006) reported on the first full-scale mapping of Havre seamount in 2002 and some of its geology (figure 14). Table 2 lists locations for Havre seamount and other nearby features.

References. Anonymous, 2012. First sighting responsible for undersea eruption, Bay of Plenty Times, URL: http://www.bayofplentytimes.co.nz/news/first-sighting-responsible-undersea-eruption/1598061/, updated 27 October 2012, accessed 1 November 2012.

Bernard, A., 2012. Hot Spots from the July 18 Eruption in Kermadec volcanic arc, International Association of Volcanology and Chemistry of the Earth's Interior (IAVCEI) Commission of Volcanic Lakes (CVL), URL: http://www.ulb.ac.be/sciences/cvl/havre/pumice_raft_Havre_eruption.html, updated 13 August 2012, accessed 13 August 2012.

Bryan, S.E., Cook, A.G., Evans, J.P., Hebden, K., Hurrey, L., Colla, P., Jell, J.S., Weatherley, D., and Firn, J., 2012. Rapid, Long-Distance Dispersal by Pumice Rafting. PLoS ONE, v. 7, no. 7: e40583 (DOI: 10.1371/journal.pone.0040583).

Bryner, J., 2012. Pumice 'raft' floating off New Zealand coast created by undersea volcano eruption, researchers say, Huffington Post Science, URL: http://www.huffingtonpost.com/2012/10/27/pumice-raft-volcano_n_2028058.html, updated 26 October 2012, accessed 1 November 2012.

de Grauw, M. and Stradling, S., 2012. Personal communication (email to GVP), 10 Septebmer 2012.

Gardner, J.P.A., Curwen, M.J., Long, J., Williamson, R.J., and Wood, A.R., 2006. Benthic community structure and water column characteristics at two sites in the Kermadec Islands Marine Reserve, New Zealand, New Zealand Journal of Marine and Freshwater Research, v. 40, pp. 179-194.

Hyvernaud, O., 2012. Personal communication, Havre seamount volcanic eruption (email to GVP), 10 October 2012.

Klemetti, E., 2012a. Havre Seamount: The source of Kermadec Island pumice raft?, Wired: Eruptions Blog, URL:http://www.wired.com/wiredscience/2012/08/source-of-kermadec-island-pumice-raft-eruption-identified, updated 13 August 2012, accessed 2 October 2012.

Klemetti, E., 2012b. What Is the Fate of Volcanic Pumice Rafts?, Wired: Eruptions Blog, URL: http://www.wired.com/wiredscience/2012/08/the-biology-of-volcanic-pumice-rafts/, updated 22 August 2012, accessed 2 October 2012.

Memmott, M., 2012. 7,500 square miles of pumice floating in the Pacific is 'weirdest thing I've seen', National Public Radio, URL: http://m.npr.org/story/158577099?url=/blogs/thetwo-way/2012/08/10/158577099/7-500-square-miles-of-pumice-floating-in-pacific-is-weirdest-thing-ive-seen, updated 10 August 2012, accessed 13 September 2012.

New Zealand Land Information (LINZ), 2012. New Zealand Notices to Mariners Notices NZ 151-154, Edition 18, pp. 6-9, New Zealand Hydrographic Authority, Wellington, NZ, URL: http://www.linz.govt.nz/docs/hydro/ntm/pdf12/nz18-3108-151-154.pdf, updated 31 August 2012, accessed 13 September 2012.

New Zealand Land Information (LINZ), 2008, Kermadec Islands, South Pacific Ocean, New Zealand, map NZ222, scale 1:300,000, Sourced from Land Information New Zealand data. Crown Copyright Reserved. URL: http://data.linz.govt.nz/layer/1267-chart-nz-222-kermadec-islands/##, updated 27 August 2012, accessed 13 September 2012.

New Zealand National Institute of Water and Atmospheric Research (NIWA), 2012. First sighting of volcano responsible for undersea eruption, Press Release, NIWA, URL: http://www.scoop.co.nz/stories/SC1210/S00054/first-sighting-of-volcano-responsible-for-undersea-eruption.htm, updated 27 October 2012, accessed 1 November 2012.

Priestley, R., 2012. The mystery of the pumice raft, Listener, issue 3774, URL: http://www.listener.co.nz/current-affairs/science/the-mystery-of-the-pumice-raft/, updated 8 September 2012, accessed 25 September 2012 (see also http://blogs.scientificamerican.com/expeditions/2012/08/10/kermadecs-islands-a-serendipitous-event/; http://rebeccapriestley.com/2012/08/12/kermadecs-voyage-2-the-mystery-of-the-floating-pumice).

Scott, B., 2012. Volcanic activity: Kermadec Islands, media release, Institute of Geological and Nuclear Sciences Limited, Wairakei Research Centre, Taupo, NZ.

Smith, I.E.M., and Price, R.C., 2006. The Tonga-Kermadec arc and Havre-Lau back-arc system: Their role in the development of tectonic and magmatic models for the western Pacific, Journal of Volcanology and Geothermal Research, v. 156 (3-4), p. 315-331.

Wright, I.C., Worthington, T.J., and Gamble, J.A., 2006. New multibeam mapping and geochemistry of the 30°-35° S sector, and overview, of southern Kermadec arc volcanism, Journal of Volcanology and Geothermal Research, v. 149 (3-4), p. 263-296.

Information Contacts: Alain Bernard, Laboratoire de Volcanologie, Dept. Earth and Environmental Sciences CP160/02, Université Libre de Bruxelles 50, Ave. Roosevelt 1050 Brussels, Belgium; Bryan Scott, Queensland University of Technology, Brisbane, AU; Maggie de Grauw, Paeroa, New Zealand; Olivier Hyvernaud, Laboratorie de Géophysique, BP 640 Papeete, Tahiti, French Polynesia; Bradley J. Scott, Institute of Geological and Nuclear Sciences Limited (GNS) (URL: http://www.gns.cri.nz/); Eric Klemetti, Denison University (URL: https://www.wired.com/category/eruptions/); Roger Matthews, Unitec Institute of Technology, Auckland, NZ; NASA Earth Data Near Real Time (Orbit Swath) Images (URL: https://earthdata.nasa.gov/earth-observation-data/near-real-time); New Zealand Listener magazine (URL: http://www.noted.co.nz/the-listener/); New Zealand Defense Force (NZDF) (URL: http://www.nzdf.mil.nz/); Pew Environment Group (URL: http://www.pewenvironment.org); Rebecca Priestley, Victoria University of Wellington, New Zealand (URL: https://rebeccapriestley.com/); Robert Simmon and Jeff Schmaltz, NASA Earth Observatory (URL: http://earthobservatory.nasa.gov).

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

September 2012 (BGVN 37:09) Cite this Report

Source of large pumice rafts traced to Havre seamount eruption

Large pumice rafts observed floating in the SW Pacific Ocean in the central Kemadec Islands, midway between North Island (New Zealand) and Tonga, have been traced by various investigators and monitoring systems to a mid-July 2012 eruption of the Havre submarine seamount (figure 1). The eruption was strong enough to result in thermal alerts and produce an ash plume that breached the ocean surface from a depth of at least 700 m.

Early observations. Maggie de Grauw, resident of Paeroa, New Zealand, took photographs (one of which is shown in figure 2) on 31 July 2012, from a commercial airplane (Virgin Pacific flight ##DJ94 from Apia, Samoa to Auckland, New Zealand) of a "peculiar large mass floating on the ocean between Tonga and Auckland."

She noted that there was another larger mass nearby, but was unable to photograph it because of the difficult angle. Believing it to be a pumice raft, she emailed the photos to Scott Bryan, author of an article on pumice rafts (Bryan and others, 2012), who forwarded it to Bulletin staff.

de Grauw noted that "the date stamp on the photograph says 1441 NZST [New Zealand standard time], which meant we had another 1 hour and 15 minutes flying time to Auckland. Total flight time was estimated at 3 and a half to 4 hours. Looking at the map, that would have put us some where near the Kermadec trench/ridge or islands. (Though I did not see any islands nearby)."

On 4 August, Bryan commented that the source of the pumice, if in the Kermadecs, may be Raoul, Macauley, Giggenbach, or Volcano W. He suggested that Monowai seamount seemed too far away. Brad Scott, New Zealand's GNS Science, noted that around the same time North Island of New Zealand had two volcanoes erupting (Tongariro and White Island) and seismic signals indicated that Monowai seamount (in the Kermadec Islands) was erupting as well.

On 10 August 2012, the New Zealand Defence Force (NZDF) reported an area of floating pumice in the open ocean. The area where the pumice was abundant was 463 km (250 nmi) in length and 56 km (30 nmi) wide, for a total area of 25,700 km² (7,500 nmi²). A photograph taken on 9 August 2012 shows an example of the pumice as seen from a Royal New Zealand Air Force (RNZAF) Orion patrol plane flying between Samoa and New Zealand (figure 3); video of the raft from this aircraft was shown in a press release from NZDF (2012).

The pilots relayed the information to the Royal New Zealand Navy vessel HMNZS Canterbury, which later that day encountered floating pumice ~160 km (85 nmi) WSW of Raoul island (29.27°S, 177.92°W). The Canterbury crew found that the pumice raft was ~0.6 m thick, 1 km wide, and extended to their right and left as far as the eye could see. The crew retrieved some pieces of pumice from the ship's water filters for later analysis and documentation. According to a news article by Priestley (2012) other samples of pumice were collected using buckets. These samples ranged from golf ball to soccer ball sized (figures 4 and 5). The pumice samples were "rough around the edges and irregular shapes." At that time, the origin of the pumice was still unknown.

In addition, Alain Bernard of the Laboratoire de Volcanologie, Université Libre de Bruxelles, Belgium, and Olivier Hyvernaud of the Laboratoire de Géophysique, Tahiti, observed the pumice raft in MODIS/Terra satellite images taken 3 August 2012.

Search for the pumice source. The search for the pumice source involved a number of investigators and their institutions, and several monitoring systems. Table 1 gives a summary of locations of the pumice source based on various observed phenomena associated with the July 2012 eruption as determined from various investigators.

Table 1. Summary of reported locations for events and features associated with tracking the source of the pumice rafts from the July 2012 eruption of Havre seamount. These locations may be compared with the location of Havre seamount (table 2) from Wright, Worthington and Gamble (2006). Compiled from listed references.

Source - Feature	Coordinates (as reported)	Coordinates (decimal degrees)	Dates and comments	Reference(s)
Seismic - source of pumice raft	31.13°S, 178.96°W	31.13°S, 178.96°W	17-18 July 2012, short seismic swarm	Hyvernaud (2012), Laboratoire de Géophysique, Papeete, Polynesian Network (Scott, 2012)
MODIS satellite - hot spot	31°7'S, 179°12'W	31.1°S, 179.2°W	1050 on 18 July 2012 UTC; band 22, 3.959 µm	Bernard (2012)
MODIS satellite - point of vapor plume	31°5'S, 179°1'W	31.1°S, 179.0°W	2150 on 18 July 2012 UTC; band 20, 3.75 µm; plume image "pointing to the source of the eruption"	Bernard (2012)
Satellite - source of pumice	30.95°S, 179.13°W	30.95°S, 179.13°W	19 July 2012 UTC, satellite data, raft becomes visible after 0205	Laboratoire de Géophysique, Papeete, Polynesian Network (Scott, 2012)

Table 2. Locations of Havre seamount and other nearby features, for comparison with early locations of pumice rafts' source vent (table 1). Compiled from listed references.

Feature	Coordinates (as reported)	Coordinates (decimal degrees)	Comments	Reference(s)
Havre Seamount	31°6.500'S, 179°2.450'W	31.11°S, 179.04°W	Summit depth 720 m, basal depth 1,750 m; used in this report and by Klemetti (2012a).	Wright, Worthington, and Gamble (2006)
Havre Rock	31°17.3'S, 178°54.7'W	31.29°S, 178.9°W	Summit elevation ~70 m.	New Zealand Land Information (2008)
L'Esperance Rock	31°21.4'S, 178°48.4'W	31.36°S, 178.82°W	Summit slightly above sea level.	New Zealand Land Information (2008)

GNS Science issued a news bulletin on 11 August 2012 (Scott, 2012) noting that a report from the Laboratoire de Géophysique, Tahiti, confirmed two indications of eruptive activity in the Kermadec Islands, one from satellite tracking and another from seismic monitoring.

An examination of satellite data by the Laboratoire de Géophysique traced the pumice back to a source at 30.95°S, 179.13°W, 72 km SW of Curtis Island at 0205 on 19 July 2012 UTC. According to Olivier Hyvernaud (2012), between 0733 on 17 July 2012 UTC and 0300 on 18 July 2012 UTC, 157 hydroacoustic events from Kermadec ridge were measured. He noted that "the waveforms are all very similar, with a short length and a steep rise. For some events, seismic Rayleigh and Pn phases from regional seismic stations were associated." Among the 157 events of magnitudes between 3.0 and 4.8, 68 events were located (figures 6 and 7).

According to Hyvernaud, the "mean location is at 31.13°S, 178.96°W, a position in the vicinity of Havre seamount. The best locations are obtained with a mix of hydroacoustic and seismic phases. The focal depths are impossible to constrain, but we assume that they are shallow. Usually, we record several types of hydroacoustic events during volcanic submarine activity: submarine explosions, tremors and small earthquakes. Submarine explosions and tremors are never recorded in seismic [data] (unless you have a very close seismic station). For Havre, the strongest events have both seismic and hydroacoustic [signals], that's why I interpret them as small earthquakes. The weakest have only hydroacoustic phase[s], because seismic phases are below the detection threshold. Tremors and explosions have not been recorded for Havre: why?, I don't know...Perhaps the explosive sources are on the opposite side of the volcano and couldn't propagate towards French Polynesia?..."

Alain Bernard sent an email to the Bulletin reporting that he had analyzed nighttime imagery from a MODIS satellite and found a thermal hot spot from the eruption at 1050 on 18 July 2012 UTC, the earliest evidence of a hot spot from the Havre Seamount eruption reaching the ocean surface (figure 8a). He noted that "apparently, the first appearance of pumice rafting is on MODIS/Terra [satellite images] of July 18 [2012]...There is an intriguing feature associated with the raft, it looks like a plume of vapour(?) with a clear thermal contrast as seen in band 20 at 3.75µm [figure 8b]. I really don't know what this could be and if this feature is pointing to the source of the eruption. Anyway, the geographic location is close to 31°5'S and 179°1'W but as far as I know there is no identified submarine volcano there."

Havre identified as pumice source. According to a report by Erik Klemetti (2012), he and Robert Simmon, both working independently of GNS Science and using NASA's Moderate Resolution Imaging Spectroradiometer (MODIS) Terra and Aqua images, discovered the first signs of the eruption (discolored water, gray pumice, and a volcanic plume) in imagery from 0950 and 1410 on 19 July 2012 (local time) between Macauley Island and Volcano W. NASA satellite images acquired during 18-21 July 2012 show an obvious plume on both 18 and 19 July, then only the pumice raft on 20 July, suggesting the eruption may have only lasted a couple of days (figure 9). The eruption was strong enough to generate a thermal pulse from a depth of at least 700 m that could be measured at the ocean surface by satellite (figure 8a).

To identify known features on the sea floor that might correspond to the source vent for the pumice, Klemetti overlaid the MODIS/Terra image on Google Earth to find the location relative to Macauley volcano and Volcano W. Bathymetric maps indicated that the source of the plume appeared to be a U-shaped edifice that had no label in the Smithsonian's Global Volcanism Program Google Earth layer. However, from a map of the Kermadec Islands (Smith and Price, 2006), it appeared that the edifice was Havre seamount (near Havre Rock), a relatively unknown seamount volcano without documented eruptive history.

Klemetti concluded his report saying that this event "shows how easily an eruption can happen in the middle of the ocean and not be noticed for 3 weeks - even in the 21st century!... Most eruptions will be noticed either as heat spots or sources of sulfur dioxide emissions if not visually on these satellite images. However, sometimes we get the evidence of an eruption well after it happened and have to backtrack through remote sensing data to find the source and in the case of Havre, this was the only way that the source could be found so quickly."

By 21 July 2012, the eruption appeared to have waned, leaving behind rafts of pumice. Winds and currents spread the pumice into a series of twisted filaments, spread over an area of ~450 by 250 km as of 13 August. A 31 August 2012 issue of New Zealand Notices to Mariners (New Zealand Land Information, 2012) announced a recent insertion, "Volcanic Activity," to Chart NZ 222 (SW Pacific - Kermadec Islands; New Zealand Land Information, 2008) at a position 30°57.00'S, 179°07.80'W (figure 10). This location is identical to the location for the source of the ash plume as identified from satellite images by the Laboratoire de Géophysique.

Fate of pumice rafts. According to Bernard, Hyvernaud, Klemetti, and Simmon, satellite images revealed that Havre seamount erupted a tightly-packed raft of floating pumice on 19 and 20 July 2012. Over several weeks, wind and waves dispersed the pumice to the W, NW, N, and then E. A 28 July image showed one pumice raft, twisted by ocean currents, appearing as a well-defined strand (figure 11). By 6 August, the pumice was largely dispersed, spread over an area at least 450 km wide (figure 12a). Filaments of pumice remained in the area on 13 August, and the pumice was spread over an area of ~450 by 258 km (figure 12b). None appeared to reach Raoul Island, site of a permanently staffed meteorological station.

The NASA Earth Observatory continued tracking the spread of the pumice from the Havre eruption. By 19 August 2012 the pumice was spread over an area of 270,000 km² of the Pacific Ocean and was continuing to spread. This pumice will likely stay afloat for months if not longer and eventually make landfall wherever the currents dictate - potentially as far away as South America.

According to GNS, the crew on a flight between Auckland and Apia on 1 October 2012 reported "floating pumice in the Kermadec Islands NE of New Zealand. The GeoNet duty volcanologist received this from the MetService Aviation Forecaster as part of the routine exchange of volcano data and observations between the organisations and airline pilots." "It is most likely this pumice raft is the same one [generated in mid-July 2012 and attributed to Havre as a source], just more spread out now. We have no direct evidence that Havre has erupted again."

The fate of the Havre pumice is unknown at this time, but a recent study by Scott Bryan and others (2012) details what happened to pumice from the 2006 Home Reef eruption in Tonga (see BGVN 31:09, 31:10, 31:12, 32:04, 33:05, and 33:12). That 2006 eruption (VEI 2 where the main vent was likely tens of meters below the ocean surface) was strong enough to create an ash plume that likely reached as high as 15 km altitude at its maximum, and did produce a small island that might have been as high as 75 m above sea level (wave action quickly removed the tephra forming the island). For the Home Reef eruption, the drifting pumice quickly hosted upwards of 80 different species of marine life over the course of its journey. Pumice rafts might be one of the ways that the ocean can redistribute organisms throughout the world oceans. Within eight months of the eruption, some of the pumice clasts had traveled over 5,000 km. Many clasts stayed afloat for ~2 years (Bryan and others 2012).

October 2012 cruise confirms Havre as pumice source. On 26 October 2012 the New Zealand National Institute of Water and Atmospheric Research's (NIWA) Research Vessel Tangaroa mapped Havre submarine volcano. NIWA ocean geology scientist Joshu Mountjoy announced finding a new volcanic cone which has formed on the edge of the volcano, towering 240 m above the crater rim that was first mapped in 2002 (Wright and others, 2006). The 2012 Havre eruption was strong enough to breach the ocean surface from a depth of more than 700 m by producing an ash plume, thermal alert, and a pumice raft that covered an area of 22,000 km², all visible by satellite.

According to a press release from NIWA (2012), the voyage leader, NIWA's volcanologist Richard Wysoczanski, said that "we know the shape of the volcano from previous research. Using the multibeam echosounder, we made a before and after comparison of the volcano to determine the size of the eruption and the change it has made to the seafloor." NIWA previously mapped Havre volcano in 2002 (Wright and others, 2006), showing a 1-km-high undersea mountain with a 5-km-wide, 800-m-deep central crater. This central steep-walled crater is a caldera, which is a collapse feature of volcanoes, like Lake Taupo, often known to produce large and violent eruptions.

Mountjoy noted that "One side of the caldera wall is bulging in towards the volcano's centre. The bulging may indicate where an eruption may occur in the future, or it might lead to an undersea avalanche." Several cubic kilometers of new material had been added to the volcano. Large volumes of freshly erupted pumice have accumulated on the caldera floor, raising the floor by up to 10 m. Glassy volcanic rocks were sampled from the fresh crater wall, typical of newly erupted material. Wysoczanski noted that there were new volcanic cones in one area. Volcanic rocks were collected, up to beach ball size, that vary in color and texture from black glassy material to white pumice. Round pebbles of pure sulphur were also retrieved.

Havre Seamount background. Smith and Price (2006) published one of the first bathymetric maps showing the main features of the Tonga-Kermadec arc/back-arc system and the location of Havre seamount (figure 13). Wright and others (2006) reported on the first full-scale mapping of Havre seamount in 2002 and some of its geology (figure 14). Table 2 lists locations for Havre seamount and other nearby features.

References. Anonymous, 2012. First sighting responsible for undersea eruption, Bay of Plenty Times, URL: http://www.bayofplentytimes.co.nz/news/first-sighting-responsible-undersea-eruption/1598061/, updated 27 October 2012, accessed 1 November 2012.

Bernard, A., 2012. Hot Spots from the July 18 Eruption in Kermadec volcanic arc, International Association of Volcanology and Chemistry of the Earth's Interior (IAVCEI) Commission of Volcanic Lakes (CVL), URL: http://www.ulb.ac.be/sciences/cvl/havre/pumice_raft_Havre_eruption.html, updated 13 August 2012, accessed 13 August 2012.

Bryan, S.E., Cook, A.G., Evans, J.P., Hebden, K., Hurrey, L., Colla, P., Jell, J.S., Weatherley, D., and Firn, J., 2012. Rapid, Long-Distance Dispersal by Pumice Rafting. PLoS ONE, v. 7, no. 7: e40583 (DOI: 10.1371/journal.pone.0040583).

Bryner, J., 2012. Pumice 'raft' floating off New Zealand coast created by undersea volcano eruption, researchers say, Huffington Post Science, URL: http://www.huffingtonpost.com/2012/10/27/pumice-raft-volcano_n_2028058.html, updated 26 October 2012, accessed 1 November 2012.

de Grauw, M. and Stradling, S., 2012. Personal communication (email to GVP), 10 Septebmer 2012.

Gardner, J.P.A., Curwen, M.J., Long, J., Williamson, R.J., and Wood, A.R., 2006. Benthic community structure and water column characteristics at two sites in the Kermadec Islands Marine Reserve, New Zealand, New Zealand Journal of Marine and Freshwater Research, v. 40, pp. 179-194.

Hyvernaud, O., 2012. Personal communication, Havre seamount volcanic eruption (email to GVP), 10 October 2012.

Klemetti, E., 2012a. Havre Seamount: The source of Kermadec Island pumice raft?, Wired: Eruptions Blog, URL:http://www.wired.com/wiredscience/2012/08/source-of-kermadec-island-pumice-raft-eruption-identified, updated 13 August 2012, accessed 2 October 2012.

Klemetti, E., 2012b. What Is the Fate of Volcanic Pumice Rafts?, Wired: Eruptions Blog, URL: http://www.wired.com/wiredscience/2012/08/the-biology-of-volcanic-pumice-rafts/, updated 22 August 2012, accessed 2 October 2012.

Memmott, M., 2012. 7,500 square miles of pumice floating in the Pacific is 'weirdest thing I've seen', National Public Radio, URL: http://m.npr.org/story/158577099?url=/blogs/thetwo-way/2012/08/10/158577099/7-500-square-miles-of-pumice-floating-in-pacific-is-weirdest-thing-ive-seen, updated 10 August 2012, accessed 13 September 2012.

New Zealand Land Information (LINZ), 2012. New Zealand Notices to Mariners Notices NZ 151-154, Edition 18, pp. 6-9, New Zealand Hydrographic Authority, Wellington, NZ, URL: http://www.linz.govt.nz/docs/hydro/ntm/pdf12/nz18-3108-151-154.pdf, updated 31 August 2012, accessed 13 September 2012.

New Zealand Land Information (LINZ), 2008, Kermadec Islands, South Pacific Ocean, New Zealand, map NZ222, scale 1:300,000, Sourced from Land Information New Zealand data. Crown Copyright Reserved. URL: http://data.linz.govt.nz/layer/1267-chart-nz-222-kermadec-islands/##, updated 27 August 2012, accessed 13 September 2012.

New Zealand National Institute of Water and Atmospheric Research (NIWA), 2012. First sighting of volcano responsible for undersea eruption, Press Release, NIWA, URL: http://www.scoop.co.nz/stories/SC1210/S00054/first-sighting-of-volcano-responsible-for-undersea-eruption.htm, updated 27 October 2012, accessed 1 November 2012.

Priestley, R., 2012. The mystery of the pumice raft, Listener, issue 3774, URL: http://www.listener.co.nz/current-affairs/science/the-mystery-of-the-pumice-raft/, updated 8 September 2012, accessed 25 September 2012 (see also http://blogs.scientificamerican.com/expeditions/2012/08/10/kermadecs-islands-a-serendipitous-event/; http://rebeccapriestley.com/2012/08/12/kermadecs-voyage-2-the-mystery-of-the-floating-pumice).

Scott, B., 2012. Volcanic activity: Kermadec Islands, media release, Institute of Geological and Nuclear Sciences Limited, Wairakei Research Centre, Taupo, NZ.

Smith, I.E.M., and Price, R.C., 2006. The Tonga-Kermadec arc and Havre-Lau back-arc system: Their role in the development of tectonic and magmatic models for the western Pacific, Journal of Volcanology and Geothermal Research, v. 156 (3-4), p. 315-331.

Wright, I.C., Worthington, T.J., and Gamble, J.A., 2006. New multibeam mapping and geochemistry of the 30°-35° S sector, and overview, of southern Kermadec arc volcanism, Journal of Volcanology and Geothermal Research, v. 149 (3-4), p. 263-296.

Information Contacts: Alain Bernard, Laboratoire de Volcanologie, Dept. Earth and Environmental Sciences CP160/02, Université Libre de Bruxelles 50, Ave. Roosevelt 1050 Brussels, Belgium; Bryan Scott, Queensland University of Technology, Brisbane, AU; Maggie de Grauw, Paeroa, New Zealand; Olivier Hyvernaud, Laboratorie de Géophysique, BP 640 Papeete, Tahiti, French Polynesia; Bradley J. Scott, Institute of Geological and Nuclear Sciences Limited (GNS) (URL: http://www.gns.cri.nz/); Eric Klemetti, Denison University (URL: https://www.wired.com/category/eruptions/); Roger Matthews, Unitec Institute of Technology, Auckland, NZ; NASA Earth Data Near Real Time (Orbit Swath) Images (URL: https://earthdata.nasa.gov/earth-observation-data/near-real-time); New Zealand Listener magazine (URL: http://www.noted.co.nz/the-listener/); New Zealand Defense Force (NZDF) (URL: http://www.nzdf.mil.nz/); Pew Environment Group (URL: http://www.pewenvironment.org); Rebecca Priestley, Victoria University of Wellington, New Zealand (URL: https://rebeccapriestley.com/); Robert Simmon and Jeff Schmaltz, NASA Earth Observatory (URL: http://earthobservatory.nasa.gov).