Introduction to special section on the Eruption of Soufrière Hills Volcano, Montserrat, the CALIPSO Project, and the SEA-CALIPSO Arc-Crust Imaging Experiment


1. Introduction

[1] Since 1995 the eruption of the andesitic Soufrière Hills volcano (SHV), Montserrat, has wreaked destruction on the small Caribbean Island of Montserrat and disrupted its economy and manner of living. But the eruption has been studied in impressive detail, and the volcano has become an important natural laboratory for investigations of volcanic processes and mitigation measures. Here we present the results of monitoring, observations and associated research on the eruption of SHV from the onset of the eruption in July 1995 until January 2009, but with emphasis on work since 2000. This is the most recent of several volumes to provide comprehensive repositories of data and interpretations concerning this noteworthy and scientifically-valuable eruption, but the need is timely. A much-cited series of 25 papers was published on the Montserrat eruption in Geophysical Research Letters in 1998, representing research carried out through August 1997 [Young et al., 1998]. This was followed by Geological Society Memoir 21, which contained 30 papers and a large photographic record, documenting research from July 1995 until November 1999 [Druitt and Kokelaar, 2002]. The August 2003 issue of the Journal of Petrology presented eight papers focused on petrological and geochemical aspects of the eruption, and additional papers on SHV have appeared from time to time in assorted journals, but it has been a decade since a comprehensive review has appeared. Thus we believe our Special Section fills an important niche. This Special Section contains a series of ten articles on the eruption of SHV, seven papers on data obtained from the CALIPSO borehole observatory network and allied surface monitoring stations, and eight papers on results from the SEA-CALIPSO experiment conducted in 2007 to image the arc crust and lithosphere about Montserrat using tomography, reflection seismology, and profiling.

2. Eruption of Soufrière Hills Volcano

[2] The SHV resumed its activity in 1995 after a respite of several hundred years including three aborted attempts to erupt in the last Century. The current eruption has since been thoroughly studied [Druitt and Kokelaar, 2002; Sparks and Young, 2002], but knowledge of deep structure is scarce, and the magmatic system underneath SHV has hitherto been imprecisely characterized. Some of these deficiencies are remedied in this issue by the observations made from the CALIPSO borehole geophysical observatory and from the results of the SEA-CALIPSO seismic experiment.

[3] In the opening paper, Wadge et al. provide a comprehensive examination of the lava production at SHV from 1995 to 2009, a task more difficult than it might first appear because of the separation of eruption products into tephra and talus as well as lava, and because more than half of the eruption products are deposited under the sea (see LeFriant et al.). The valuable database is included in the supplementary material. This effort is complimented by that of Christopher et al., who take a similarly ranging 1995–2009 overview of volatile production, with implications on mafic magma supply and degassing. The SHV eruption straddles a major improvement in satellite monitoring of SO2, and this topic is explored by Carn and Prata in their paper on explosive degassing.

[4] The next series of papers focuses on lava domes, which follow on from a previous study of the gigantic 2003 collapse by Herd et al. [2005]. Ryan et al. compare growth of the lava dome and extrusion rates for the period 2005–2008, Loughlin et al. examine lava dome evolution, dome collapse, and cyclicity over 2005–2007, and Baptie examines the detection of dome collapse using passive seismic interferometry. Insights into the Vulcanian explosions in 2008–09 at SHV, and associated pyroclastic flows and tephra deposition, are reported by Komorowski et al. A series of three papers on petrological issues follow. Barclay et al. examine the implications of increased abundance of mafic enclaves in the SHV lava, Genareau and Clarke use clast heterogeneity to reveal details of magma mingling, and magma hybridization and diffusive exchange is studied by Humphreys et al.


[5] Data reported here were taken from the collaborative Caribbean Andesite Lava Island Precision Seismo-geodetic Observatory, CALIPSO, a volcano monitoring system installed late 2002 and early 2003 for investigations of the dynamics of SHV magmatic activity [Mattioli et al., 2004]. The system consists of an integrated array of specialized instruments in four strategically located ∼200-m-deep boreholes in concert with several shallower holes and surface sites. CALIPSO was initiated because the volcanic activity was continuing and provided an opportunity to take advantage of the high-sensitivity of borehole measurements to record data critical for investigating properties of the activity. To the extent that SHV is typical of andesitic dome-building volcanoes, results from this research can be expected to apply more generally.

[6] The CALIPSO section opens with the paper by Mattioli et al. on the surface deformation of SHV from GPS geodesy. A long time in gestation, this paper is an important addition to the scientific knowledge of SHV, and a rare, very long GPS record of deformation at an erupting andesitic volcano. As noted by one reviewer, it is a data-heavy iceberg of a paper with much residing as supplementary material, but none the worse for that. The data and methods are transparent, and this paper will clearly serve as the primary GPS data reference for the SHV eruption for some time to come.

[7] The following paper by Voight et al. on the “magma-sponge hypothesis for stratovolcanoes” addresses an enigmatic issue– why are the pressure centers as derived from GPS data so deep, in comparison to the petrological indicators that implied a depth around 5 km? Resolution is sought in a vertically-elongate magma reservoir system, and a related conundrum, the imbalance between the dense-rock volume of lava erupted and the crustal volume change associated with deformation, is explained by the response of compressible bubbly reservoir magma. An alternative view of magma system structures assumes the existence of several discrete reservoirs, and this idea is explored for the circumstance of layered elastic media by Foroozan et al.

[8] Next is a series of four papers focused on Vulcanian explosion dynamics as revealed by data from the CALIPSO borehole strainmeters. The first of these, by Voight et al., investigates three explosions that occurred in July 2003, triggered by the collapse of a prodigious lava dome. The investigation constrains eruption volumes, mass fluxes of order 107 kg s−1, and conduit pressures, and each explosion generated remarkable very-long-period gravity waves that propagated at 30 m s−1. The second paper by Linde et al. examines, for a Vulcanian explosion in 2004 (Figure 1) large, roughly equal but opposite polarity strain changes at two relatively “near” sites, and small changes at a distal site, and uses these data to constrain the geometry and attitude of a shallow feeder dike source. A similar constraint regarding a dike source is developed in the third paper by Chardot et al., for one of three explosions investigated in 2008–2009; the other explosions were broadly similar in strain and barometric signatures to those in 2003 and yielded similar interpretations on dynamics. The final paper in the series by Ripepe et al. considers observations of infrasonic and gravity waves at SHV in 2008, and complements the CALIPSO data with records from an acoustic array set up by the Università di Firenze.

Figure 1.

Rapidly ascending eruption column of the Soufrière Hills volcano on March 3, 2004, viewed from 6 km northwest. The explosion removed a small lava dome that had grown at the center of the crater formed by the July 12–13, 2003, dome collapse. The convecting plume rose to about 7 km and small pyroclastic flows and surges formed at the base of the eruption column. Strainmeter data for this explosion suggest the influence of a pressurized dike source of WNW orientation (Photo by B. Voight).


[9] The internal structure of an active volcano remains one of the most enigmatic issues in geosciences. Several passive seismic tomography experiments have been carried out at active volcanoes (Kilauea, Mt. St. Helens, Etna, Unzen), but an inhomogeneous distribution of earthquakes can compromise resolution. At Montserrat, volcanic earthquakes are <4 km deep and limit passive methods to study of shallow features.

[10] Thus an active-source seismic experiment to study the structure of arc crust under Montserrat and Soufrière Hills volcano was carried out in December 2007, and data are reported by Paulatto et al. [2010a, 2010b] and in this issue. SEA-CALIPSO (Seismic Experiment with Airgun-source) is conducted under the umbrella of the CALIPSO consortium project. The aim is to obtain three-dimensional structural images of the island and its volcanic centers, using seismic tomography, seismic reflection/refraction imaging techniques, and offshore seismic profiling. Before this experiment, knowledge of the deeper structure of Montserrat and SHV was limited, with proposed models based on restricted and sometimes conflicting geophysical, geological and petrological data. This multi-national experiment—involving institutes from the USA, United Kingdom, New Zealand, Trinidad, and Montserrat – is generating high resolution images of this island, its volcanic edifices, and adjacent crust [Paulatto et al., 2010a, 2010b] and should advance our understanding of crustal evolution in arc systems, magma storage and transport systems, and volcanic processes.

[11] The SEA-CALIPSO sub-section contains eight papers, led by the paper by Shalev et al. on the 3D seismic velocity tomography of Montserrat. Striking features of the tomography include three relatively high-velocity zones below each of the main volcanic centers on Montserrat, and three low-velocity zones flanking Centre Hills. A low velocity zone is suggested under SHV below 5 km, about the limit of good resolution, and this is suggestive of the top of the magma reservoir [cf. Paulatto et al., 2010b]. The next paper in the series, by Paulatto et al., explores further constraints on the intrusive system beneath Montserrat by study of the attenuation of seismic signals. Byerly et al. then provide insights on the limitations of reflection imaging in an active source experiment for a small island, and use the onshore array to study deep crustal reflections using microearthquakes. The tomographic techniques yield information on the distribution of crustal velocities, but the next step requires interpretation of those velocities in terms of plausible rock types. This key step is taken by Kiddle et al., by merging velocity information with petrologic data from xenoliths and consideration of laboratory velocity data on selected rock types.

[12] The final four papers overlap in their consideration of tectonics both onshore and offshore, and/or offshore perspectives based on seismic reflection, swath bathymetry, and sediment cores. The first paper by Miller et al. explores volcano-tectonic earthquake activity during the early stages of volcanic activity (1995–96) at SHV, and considers seismicity in relation to regional tectonics and heterogeneity of crust as revealed by SEA-CALIPSO tomography. The clustered seismicity and relatively-aseismic zones are interpreted to reflect a broad weakened tectonic zone of ESE trend that crosses Montserrat, and the ascent of a magmatic dike which altered the stress distribution to promote localized fault movements and caused localized dilatation with changes in pore-fluid pressures. The next paper by Feuillet et al. places Montserrat in tectonic context with the broad Caribbean region, and provides new insights enabled by the French-supported 2009 GWADASEIS marine cruise. This is followed in turn by the paper by Kenedi et al., who examine tectonics on and near Montserrat using streamer data from the SEA-CALIPSO cruise, and provide images of volcaniclastic sediment packages offshore. The latter topic introduces the offshore sediment perspectives on the eruption of SHV, which are treated in further detail by LeFriant et al. in the final paper of the Special Section.

[13] All papers have been improved as a result of thorough technical reviews, and we thank our colleagues for their considerable and pain-staking efforts to ensure high standards in science and presentation. Our sincere gratitude goes especially to Fabio Florindo, whose devoted editorial guidance on the special section steered it through stormy seas, and whose wisdom regarding reviewer suggestions led to substantial enhancements of the manuscripts. Ruth Harris also edited several papers. We are very grateful also to Beverly Turner-Holmes and to the able and considerate publications staff at AGU.

[14] The scientific work at MVO has been supported financially by the UK Government and the Government of Montserrat, and the National Environmental Research Council (NERC), via the British Geological Survey and the Seismic Research Center of the University of West Indies, and to university scientists. CALIPSO is jointly funded by National Science Foundation (NSF), Continental Dynamics and Instrumentation and Facilities Programs (United States), and NERC (United Kingdom). SEA-CALIPSO funding was provided by NSF (Continental Dynamics, Geophysics, and Instrumentation and Facilities Programs), NERC, Discovery TV, and British Geological Survey and Foreign & Commonwealth Office (UK). The Incorporated Research Institutions for Seismology PASSCAL Instrument Center provided instrumentation and technical support to the project. We sincerely thank R. Reichlin, L. Johnson, R. Kelz, S. Esperanca, and A. Shor at NSF, H. Beadman at NERC, Marine Services staff under C. Day, the RRS Cook crew under P. Gauld, the GeoPro airgun team, the Scripps team, the OBS team, J. Fowler, B. Beaudoin, and M. Fort at PASSCAL, V. Hards, M. Strutt, Tappy Syers, P. Williams, Venus Bass and other staff at MVO, and the many people who worked hard aboard ship or on land to deploy seismic instruments and collect data during December 2007. We are grateful to C. McClintock at PSU for aid in funding efforts. Our thanks go also to the good people of Montserrat who allowed us use of their properties for instrument deployment, and who assisted our scientific work in innumerable helpful ways for over a decade.