Abstract The Montserrat Volcano Observatory (MVO) is a statutory body of the Government of Montserrat and is the organization responsible for volcano monitoring operations on the island. It was formed shortly after the first phreatic explosions from Soufrière Hills Volcano occurred on 18 July 1995, and evolved from a hastily created, interim entity to a fully established volcano monitoring operation. Participating scientific teams have been drawn mainly from the Seismic Research Unit of the University of the West Indies, the US Geological Survey, the British Geological Survey and universities from various countries including the USA, UK, France and Puerto Rico. Despite its hurried inception, the MVO has been able to provide timely, high quality hazard advice to the civil authorities and has maintained an exceptional documentary record of all scientific aspects of the eruption. Its public education and information efforts have been extensive and there have been unusually high levels of interaction between scientists and the civil authorities, and between scientists and the public, both within Montserrat and outside in the wider world. The experience of setting up and running the MVO, under difficult and stressful conditions, has exemplified the advantages of teamwork and flexibility within monitoring operations and the benefits of openness and clarity in public interactions. Novel techniques have been applied to the appraisal of hazards and advances in scientific understanding have proved invaluable for risk assessment and management.

Abstract Volcanic hazard and risk at Soufrière Hills Volcano (SHV), Montserrat has been assessed in a consistent and quantitative manner for 14 years (1997–2011) during highly variable eruptive activity involving andesitic lava-dome growth, which has placed serious constraints on Montserratian society. This work has been carried out by the Scientific Advisory Committee (and predecessors) in collaboration with the Montserrat Volcano Observatory. We describe the organizational context of these assessments, the types of hazards and the methods used to analyse them. Knowledge elicitation using hazard scenarios and analysis by the Classical Model method were employed to formulate probabilistic forecasts of future hazardous events over the next year, and to quantify risks to individuals and Montserrat society generally. We devised a scheme for assessing the likelihood that the volcanic system had stopped receiving basalt magma, considered to be the main driver of the eruption. The accuracy of forecasts was tested using Brier Skill Scores: 83% of forecasts for events that were critical to life had positive skill, as measured by this method. We also discuss how government responded to our assessments. The continuous series of quantitative volcanic hazard and risk assessments described here is the only one of its kind.

Abstract The eruption on Montserrat during 1995-1999 was the most destructive in the Caribbean volcanic arc since that of Mont Pelee (Martinique) in 1902. It began on 18 July 1995 at the site of the most recent previous activity, on the flank of a c. 350-year-old lava dome within a sector-collapse scar. Phreatic explosivity occurred for 18 weeks before the onset of extrusion of an andesitic lava dome. Dome collapses produced pyroclastic flows that initially were confined by the sector-collapse scar. After 60 weeks of unsteadily accelerating dome growth and one episode of sub-Plinian explosivity, the dome eventually overtopped the confining scar. During 1997 almost two-thirds of the island was devastated following major dome collapses, two episodes of Vulcanian explosivity with fountain-collapse pyroclastic flows, and a flank failure with associated debris avalanche and explosive disruption of the lava dome. Nineteen people were killed directly by the volcanic activity and several were injured. From March 1998 until November 1999 there was a pause in magma ascent accompanied by reduced seismic activity, substantial degradation of the dome, and considerable degassing with venting of ash. The slow progress and long duration of the volcanic escalation, coupled with the small size of the island and the vulnerability of homes, key installations and infrastructure, resulted in a style of emergency management that was dominantly reactive. In order to minimize the disruption to life for those remaining on the island, following large-scale evacuations, scientists at the Montserrat Volcano Observatory had to anticipate hazards and their potential extents of impact with considerable precision. Based on frequent hazards assessments, a series of risk management zone maps was issued by administrative authorities to control access as the eruption escalated. These were used in conjunction with an alert-level system. The unpreparedness of the Montserrat authorities and the responsible UK government departments resulted in hardship, ill feeling and at times acrimony as the situation deteriorated and needs for aid mounted. Losses and stress could have been less if an existing hazards assessment had registered with appropriate authorities before the eruption.

Abstract Extrusion during Phase 5 (8 October 2009–11 February 2010) produced significant volumetric and geomorphic changes to the lava dome and surrounding valleys at the Soufrière Hills Volcano, Montserrat. Approximately 74×10 6 m 3 of lava was extruded at an average rate of 7 m 3 s −1 during the short period of activity. Addition of lava to the pre-existing dome resulted in a net volumetric increase of up to 38×10 6 m 3 . Pyroclastic density current (PDC) and ashfall deposits accounted for the remaining 36×10 6 m 3 . A series of thick, blocky lobes were extruded from a central vent. In addition, several short-lived spines and two large shear lobes were also extruded. Significant PDC activity resulted in substantial valley filling of up to 108 m. The large pre-existing dome significantly influenced the growth of lobes, such that many block-and-ash flows were generated from viscous lobes draped over the summit and upper slopes. Geomorphic changes caused by rapid filling of the surrounding valleys aided in both flow avulsion and the emplacement of deposits up to 6 km from the dome. These geomorphic changes have important consequences for hazards from PDCs.

Abstract The Soufrière Hills Volcano (SHV) crystallizes cristobalite (crystalline silica) in its lava domes, and inhalation of cristobalite-rich ash may pose a chronic respiratory hazard. We investigate the causes of variation in cristobalite abundance (measured by X-ray diffraction) in ash from dome collapses, explosions and ash venting from 1997 to 2010. Cristobalite abundance in bulk dome-collapse ash varies between 4 and 23 wt%. During periods of slow lava extrusion (<5 m 3 s −1 ), cristobalite is abundant (7–23 wt%), which we attribute to extensive devitrification in slow-cooling lava; it can also form rapidly (15 wt% in 2 months), but we find no correlation between cristobalite abundance and dome residence time (DRT). By contrast, during rapid extrusion (>5 m 3 s −1 ), cristobalite abundance is low (4–7 wt%, similar to that associated with Vulcanian explosions), and correlates strongly with DRT. We attribute this correlation to progressive vapour-phase mineralization or devitrification, and the lack of contamination by older lava. Cristobalite abundance is expected to be >7 wt% for collapse of slowly extruded lava, for ash venting through a dome or for incorporation of hydrothermally altered edifice during explosions; cristobalite abundance is expected to be <7 wt% for collapse of rapidly extruded lava, for ash venting without dome incorporation and from Vulcanian explosions at SHV.

Abstract Sulphur dioxide (SO 2 ) diffusion tube monitoring has been undertaken on Montserrat since 1995, providing a unique and insightful long-term dataset of ground-level SO 2 concentrations during the eruption of the Soufrière Hills Volcano (SHV). The monitoring of ground-level SO 2 is important to assess the potential of human exposure to high levels of SO 2 that may impact on health. Air-quality objectives for SO 2 are present in some countries to prevent potential health impacts. Here we summarise diffusion tube monitoring in Montserrat and analyse concentrations with respect to the potential for exposure to levels above recommended levels. We explore relationships that may exist with SO 2 flux measurements and volcanic events. Concentrations have been higher during pauses in lava extrusion. Diffusion tube concentrations are highest within 5 km of the volcano and at locations downwind of the plume. Areas where concentrations have exceeded relevant limits have been uninhabited since 1996. The potential for human exposure above recommended limits is, therefore, currently considered low, as the population would not have been exposed to high concentrations for extended periods of time. Full-time occupation and/or long-term exposure in the areas where concentrations exceed the relevant limits would not be advised.

Abstract Vulcanian explosions generated at Soufrière Hills Volcano between 2008 and 2010 varied from simple events involving minimal pyroclastic density currents (PDCs) to complex events involving more than one explosion. Calculated volumes for the deposits of the PDCs formed by these explosions ranged up to 2.7×10 6 m 3 , with more than half the explosions having volumes greater than 1×10 6 m 3 . The deposits formed by the explosions varied in lithology, with some explosions generating pumice-rich PDCs (e.g. 29 July 2008 and 11 February 2010) showing development of sinuous lobes. These explosions are similar to those formed in 1997, with gas-rich, conduit-derived magma being the dominant driving mechanism. Other explosions were pumice-poor ( c. 5 wt% pumice) and generated morphologically distinct PDC deposits. Many of the pumice-poor explosions were associated with lower tephra plumes of <8 km, but were some of the largest volume events in terms of PDC production and suggest a generation mechanism involving destruction of significant quantities of the lava dome. Analysis of video footage shows that PDC formation was pulsatory, probably related to destabilization of portions of the lava dome during the initial phases of the explosion.

Abstract This paper describes ash-venting activity at Soufrière Hills Volcano, Montserrat that was precursory to the onset of three phases of lava extrusion in 2005, 2008 and 2009, and similar ash venting that occurred during the fifth phase of lava extrusion. We describe in detail a style of mild, tephra-generating activity termed ash venting and its associated tephra products. The nature of the seismicity associated with ash venting is compared with that of explosive activity. All explosive events, from small explosions to large Vulcanian explosions, have impulsive, low-frequency onsets. These are absent in ash-venting events, which have subtle, emergent onsets. Microscope and grain-size analyses show that ash-venting events and large Vulcanian explosions generate tephra that is similar in grain size (in medial and distal regions), although phreatic events in 2005 were finer grained. Ash-venting products are either composed of fine-grained, variably altered pre-existing material or juvenile material. There is a general correlation between the length of the pause and the length of the period of precursory activity prior to lava extrusion following it. Syn-extrusive ash venting is frequently associated with short-term increases in extrusion rate and is considered to be related to shear-induced fragmentation at the conduit margin.

Abstract On 11 February 2010, a partial dome collapse, the largest since 20 May 2006, occurred at Soufrière Hills Volcano (SHV), Montserrat. The collapse is also the largest generated on the northern flank of SHV since the eruption began in 1995. Approximately 50×10 6 m 3 was removed from the dome, resulting in widespread pyroclastic density currents (PDCs). Mapping revealed a complex stratigraphy that varied widely across the northern and NE flanks, and reflected the complex evolution of the collapse. The deposits included a range of fine-grained ash-rich and pumice-rich units deposited by dilute PDCs, and several types of coarse-grained, blocky deposits from dense PDCs. Several previously unaffected areas, including Bugby Hole, Farm River Valley, the village of Harris and Trants, suffered significant damage to the natural and built environments. The collapse lasted 107 min but the bulk of the activity occurred in a 15 min period that included five of the six peaks in PDC generation and two Vulcanian explosions. Although powerful, the PDCs generated were not associated with a lateral blast. The likely cause was the piecemeal collapse of a series of large, unstable lobes that had been extruded on the northern flank of the pre-existing dome.

Abstract Dome growth at Soufrière Hills Volcano halted in early March 1998. After dome growth ceased, seismicity reduced significantly, but activity related to dome disintegration and degassing of magma at depth continued. A sustained episode of pyroclastic flows on 3 July 1998 marked the single largest collapse from March 1998 to November 1999. This led to a remarkable episode of dome collapses, low-energy explosions and ash-venting that resulted in the regular production of ash plumes, commonly reaching 1.5–6 km above sea level (a.s.l), but sometimes up to 11 km a.s.l., and the development of a small block-and-ash cone around the explosion crater. During the period of this residual activity, higher levels of activity occurred approximately every five to six weeks. This periodicity was similar to the cycles observed during active dome growth during 1995 to 1998, and probably had a similar cause. The relatively high level of observed activity caused continued concern regarding volcanic hazards and their potential to impact upon the resident population. Vigorous magma extrusion resumed in November 1999. The activity of the intervening period is attributed to the continued cooling and degassing of the dome, conduit and deep magma body, the impact of rising volcanic gases in the volcanic edifice, and limited magma flow in the conduit.