Bayesian event tree for eruption forecasting (BET_EF) at Vesuvius,Italy: a retrospective forward application to the 1631 eruption

Reliable forecasting of the next eruption at Vesuvius is the main scientific factor in defining effective strategies to reduce volcanic risk in one of the most dangerous volcanic areas of the world. In this paper, we apply a recently developed probabilistic code for eruption forecasting to new and independent historical data related to the pre-eruptive phase of the 1631 eruption. The results obtained point out three main issues: (1) the importance of “cold” historical data (according to Guidoboni 2008) related to pre-eruptive phases for evaluating forecasting tools and possibly refining them; (2) the BET_EF code implemented for Vesuvius would have forecasted the 1631 eruption satisfactorily, marking different stages of the pre-eruptive phase; (3) the code shows that pre-eruptive signals that significantly increase the probability of eruption were likely detected more than 2 months before the event.     

Volcanic hazard at Vesuvius: An analysis for the revision of the current emergency plan

Mt Somma-Vesuvius is a composite volcano on the southern margin of the Campanian Plain which has been active since 39 ka BP and which poses a hazard and risk for the people living around its base. The volcano last erupted in 1944, and since this date has been in repose. As the level of volcanic risk perception is very high in the scientific community, in 1995 a hazard and risk evaluation, and evacuation plan, was published by the Italian Department of Civil Protection (Dipartimento della Protezione Civile). The plan considered the response to a worst-case scenario, taken to be a subplinian eruption on the scale of the 1631 AD eruption, and based on a volcanological reconstruction of this eruption, assumes that a future eruption will be preceded by about two weeks of ground uplift at the volcano's summit, and about one week of locally perceptible seismic activity. Moreover, by analogy with the 1631 events, the plan assumes that ash fall and pyroclastic flow should be recognized as the primary volcanic hazard. To design the response to this subplinian eruption, the emergency plan divided the Somma-Vesuvius region into three hazard zones affected by pyroclastic flows (Red Zone), tephra fall (Yellow and Green Zone), and floods (Blue Zone). The plan at present is the subject of much controversy, and, in our opinion, several assumptions need to be modified according to the following arguments: a) For the precursory unrest problem, recent scientific studies show that at present neither forecast capability is realistic, so that the assumption that a future eruption will be preceded by about two weeks of forecasts need to be modified; b) Regarding the exposure of the Vesuvius region to flow phenomena, the Red Zone presents much inconsistency near the outer border as it has been defined by the administrative limits of the eighteen municipality area lying on the volcano. As this outer limit shows no uniformity, a pressing need exists to define appropriately the flow hazard zone, since there are some important public structures not considered in the current Red Zone that could be exposed to flow risk; c) Modern wind records clearly indicate that at the time of a future eruption winds could blow not only from the west, but also from the east, so that the Yellow Zone (the area with the potential to be affected by significant tephra fall deposits) must be redefined. As a result the relationship between the Yellow Zone and Green Zone (the area within and beyond which the impact of tephra fall is expected to be insignificant) must be reconsidered mainly in the Naples area; d) The May 1998 landslide, caused in the Apennine region east of the volcano by continuous rain fall, led to the definition of a zone affected by re-mobilisation of tephra (Blue Zone), confined in the Nola valley. However, as described in the 1631 chronicles of the eruption, if generation of debris flows occurs during and after a future eruption, a much wider region east of the Somma-Vesuvius must be affected by events of this type.     

Towards real-time eruption forecasting in the Auckland Volcanic Field: application of BET_EF during the New Zealand National Disaster Exercise ‘Ruaumoko’

The Auckland Volcanic Field (AVF) is a young basaltic field that lies beneath the urban area of Auckland, New Zealand’s largest city. Over the past 250,000 years the AVF has produced at least 49 basaltic centers; the last eruption was only 600 years ago. In recognition of the high risk associated with a possible future eruption in Auckland, the New Zealand government ran Exercise Ruaumoko in March 2008, a test of New Zealand’s nation-wide preparedness for responding to a major disaster resulting from a volcanic eruption in Auckland City. The exercise scenario was developed in secret, and covered the period of precursory activity up until the eruption. During Exercise Ruaumoko we adapted a recently developed statistical code for eruption forecasting, namely BET_EF (Bayesian Event Tree for Eruption Forecasting), to independently track the unrest evolution and to forecast the most likely onset time, location and style of the initial phase of the simulated eruption. The code was set up before the start of the exercise by entering reliable information on the past history of the AVF as well as the monitoring signals expected in the event of magmatic unrest and an impending eruption. The average probabilities calculated by BET_EF during Exercise Ruaumoko corresponded well to the probabilities subjectively (and independently) estimated by the advising scientists (differences of few percentage units), and provided a sound forecast of the timing (before the event, the eruption probability reached 90%) and location of the eruption. This application of BET_EF to a volcanic field that has experienced no historical activity and for which otherwise limited prior information is available shows its versatility and potential usefulness as a tool to aid decision-making for a wide range of volcano types. Our near real-time application of BET_EF during Exercise Ruaumoko highlighted its potential to clarify and possibly optimize decision-making procedures in a future AVF eruption crisis, and as a rational starting point for discussions in a scientific advisory group. It also stimulated valuable scientific discussion around how a future AVF eruption might progress, and highlighted areas of future volcanological research that would reduce epistemic uncertainties through the development of better input models.     

Chronology of Vesuvius’ activity from A.D. 79 to 1631 based on archeomagnetism of lavas and historical sources

The activity of Vesuvius between A.D. 79 and 1631 has been investigated by means of precise archaeomagnetic dating of primary volcanic deposits and taking into account the stratigraphy of lavas and tephra, historical written accounts, archaeological evidence related to the developing urbanisation, and radiocarbon ages. We found that the historical records are highly useful in constraining the timing of the main events, even if the data are often too scarce and imprecise for ascertaining the details of all phases of activity, especially their magnitude and emplacement of all the deposit types. In addition, some eruptions that took place in the 9^th and 10^th centuries appear to be unnoticed by historians. The archaeomagnetic study involved 26 sites of different lavas and 2 pyroclastic deposits. It shows that within the 15 centuries which elapsed between A.D. 79 and 1631, the effusive activity of Vesuvius clustered in the relatively short period of time between A.D. 787 and 1139 and was followed by a 5-century-long repose period. During this time Vesuvius prepared itself for the violent explosive eruption of 1631. The huge lavas shaping the morphology of the coast occurred largely through parasitic vents located outside the Mount Somma caldera. One of these parasitic vents is located at low elevation, very close to the densely inhabited town of Torre Annunziata. Among the various investigated lavas, a number of which were previously attributed to the 1631 eruption, none is actually younger than the 12^th century. Therefore it is definitively concluded that the destructive 1631 event was exclusively explosive.Editorial handling: J. McPhie     

Geochemical evidence of the renewal of volcanic activity inferred from CO<Subscript>2</Subscript> soil and SO<Subscript>2</Subscript> plume fluxes: the 2007 Stromboli eruption (Italy)

On 27 February 2007, a new eruption occurred on Stromboli which lasted until 2 April. It was characterized by effusive activity on the Sciara del Fuoco and by a paroxysmal event (15 March). This crisis represented an opportunity for us to refine the model that had been developed previously (2002–2003 eruption) and to improve our understanding of the relationship between the magmatic dynamics of the volcano and the geochemical variations in the fluids. In particular, the evaluation of the dynamic equilibrium between the volatiles (CO₂ and SO₂) released from the magma and the corresponding fluids discharged from the summit area allowed us to evaluate the level of criticality of the volcanic activity. One of the major accomplishments of this study is a 4-year database of summit soil CO₂ flux on the basis of which we define the thresholds (low–medium–high) for this parameter that are empirically based on the natural volcanological evolution of Stromboli. The SO₂ fluxes of the degassing plume and the CO₂ fluxes emitted from the soil at Pizzo Sopra la Fossa are also presented. It is noteworthy that geochemical signals of volcanic unrest have been clearly identified before, during and after the effusive activity. These signals were found almost simultaneously in the degassing plume (SO₂ flux) and in soil degassing (CO₂ flux) at the summit, although the two degassing processes are shown to be clearly different. The interpretation of the results will be useful for future volcanic surveillance at Stromboli.     

First historical evidence of a significant Mt. Etna eruption in 1224

The 1224 Mt. Etna eruption is a significant event both in terms of the mass of erupted materials and because it involved the lower eastern slope of the volcano, reaching down to the sea. Nevertheless, it is unknown to current historical catalogues. According to the historical sources, only two other lava flows actually reached as far as the sea: in 396 BC, just north of the present-day inhabited area of Acireale, according to the geological data alone, and in 1669, when the lava covered the south-eastern flank of Mt. Etna and damaged Catania. We present and discuss the two medieval sources that attest to the eruption of 1224 and make available the original texts. Furthermore, through the close analysis of the historical and topographic context of the Etna area, taking account of the roads and ports in the early 13th century, we have tried to single out the possible area of the lava's outlet into the sea in 1224 on historical grounds. A repeat of an eruption similar to that of 1224 would have a serious impact today as the coast is densely populated.     

Soil CO2 degassing on Mt Etna (Sicily) during the period 1989–1993: discrimination between climatic and volcanic influences

Wide variations were measured in the diffuse CO₂ flux through the soils in three selected areas of Mt Etna between August 1989 and March 1993. Degassing of CO₂ from the area of Zafferana Etnea-S. Venerina, on the eastern slope of the volcano, has been determined to be more strongly influenced by meteorological parameters than the other areas. The seasonal component found in the data from this area has been excluded using a filtering algorithm based on the best fitting equation calculated from the correlation between CO₂ flux values and those of air temperature. The filtered data appear to have variations temporally coincident with those from the other areas, thus suggesting a common and probably deep source of gas. The highest fluxes measured in the two most peripheral areas may correlate well with other geophysical and volcanological anomalous signals that preceded the strong eruption of 1991–1993 and that were interpreted as deep pressure increases. Anomalous decreases in CO₂ fluxes accompanied the onset and the evolution of that eruption and have been interpreted as a sign of upward migration of the gas source. The variations of CO₂ flux at the 1989 SE fracture have also given interesting information on the timing of the magmatic intrusion that has then fed the 1991–1993 eruption.     

Reconstruction of the eruptive history of Usu volcano,Hokkaido, Japan,inferred from petrological correlation between tephras and dome lavas

Usu volcano has erupted nine times since 1663. Most eruptive events started with an explosive eruption, which was followed by the formation of lava domes. However, the ages of several summit lava domes and craters remain uncertain. The petrological features of tephra deposits erupted from 1663 to 1853 are known to change systematically. In this study, we correlated lavas with tephras under the assumption that lava and tephra samples from the same event would have similar petrological features. Although the initial explosive eruption in 1663 was not accompanied by lava effusion, lava dome or cryptodome formation was associated with subsequent explosive eruptions. We inferred the location of the vent associated with each event from the location of the associated lava dome and the pyroclastic flow deposit distribution and found that the position of the active vent within the summit caldera differed for each eruption from the late 17th through the 19th century. Moreover, we identified a previously unrecognized lava dome produced by a late 17th century eruption; this dome was largely destroyed by an explosive eruption in 1822 and was replaced by a new lava dome during a later stage of the 1822 event at nearly the same place as the destroyed dome. This new interpretation of the sequence of events is consistent with historical sketches and documents. Our results show that petrological correlation, together with geological evidence, is useful not only for reconstructing volcanic eruption sequences but also for gaining insight into future potential disasters.     

Uranium and thorium in the volcanic processes

Distribution of radioactive elements in the Quaternary alkaline volcanites of Northern Latium has been studied and conclusions of volcanological interest, both as to differentiation of magma in the more superficial levels of the crust and as to the relationship between volcanic eruption and concentration of particular elements, have been drawn. The following principal results are emphasized:

1. i)
   There are two well defined types of distribution of U and Th corresponding to fractional crystallization differentiation and to pneumatolytic (gaseous transfer) differentiation. From the volcanological viewpoint, this double distribution mirrors the hypomagma or pyromagma conditions of the melt, thus allowing the physico-chemical characters of the magma to be defined, in respect to the different magma chambers considered and to the different levels of the same magma chamber, before an eruption.     
   
   

A long-term volcanic hazard event tree for Teide-Pico Viejo stratovolcanoes (Tenerife,Canary Islands)

We propose a long-term volcanic hazards event tree for Teide-Pico Viejo stratovolcanoes, two complex alkaline composite volcanoes that have erupted 1.8–3 km³ of mafic and felsic magmas from different vent sites during the last 35 ka. This is the maximum period that can be investigated from surface geology and also represents an upper time limit for the appearance of the first phonolites on that volcano. The whole process of the event tree construction was divided into three stages. The first stage included the determination of the spatial probability of vent opening for basaltic and phonolitic eruptions, based on the available geological and geophysical data. The second, involved the analysis of the different eruption types that have characterised the volcanic activity from Teide during this period. The third stage focussed on the generation of the event tree from the information obtained in the two previous steps and from the application of a probabilistic analysis on the occurrence of each possible eruption type. As for other volcanoes, the structure of the Teide-Pico Viejo Event Tree was subdivided into several steps of eruptive progression from general to more specific events. The precursory phase was assumed as an unrest episode of any geologic origin (magmatic, hydrothermal or tectonic), which could be responsible for a clear increase of volcanic activity revealed by geophysical and geochemical monitoring. According to the present characteristics of Teide-Pico Viejo and their past history, we started by considering whether the unrest episode would lead to a sector collapse or not. If the sector collapse does not occur but an eruption is expected, this could be either from the central vents or from any of the volcanoes' flanks. In any of these cases, there are several possibilities according to what has been observed in the period considered in our study. In the case that a sector collapse occurs and is followed by an eruption we considered it as a flank eruption. We conducted an experts elicitation judgement to assign probabilities to the different possibilities indicated in the event tree. We assumed long term estimations based on existing geological and historical data for the last 35 Ka, which gave us a minimum estimate as the geological record for such a long period is incomplete. However, to estimate probabilities for a short term forecast, for example during an unrest episode, we would need to include in the event tree additional information from the monitoring networks, as any possible precursors that may be identified could tell us in which direction the system will evolve. Therefore, we propose to develop future versions of the event tree to include also the precursors that might be expected on each path during the initial stages of a new eruptive event.     

THE HISTORICAL RECORDS OF VOLCANIC ERUPTION IN THE KOREAN PENINSULA

The historical document record is of vital significance to determine the volcanic eruption history age in the volcanology research and it cannot be replaced by ¹⁴C dating and other methods. The volcanoes are widely distributed in the northeast area of China, but there is lack of relevant historical records. However, there are the records of the volcanic eruption in the historical documents of Goryeo Dynasty(AD918-1392)and Joseon Dynasty(AD1391-1910)in the Korean Peninsula which is separated by a river with China only. Some of the records have been widely used as important information to the research of Changbaishan Tianchi volcano eruption history by researchers both at home and abroad, but they have different opinions. On the basis of the historical documents in the Korean Peninsula, that is, the History of Goryeo Dynasty and the Annals of the Joseon Dynasty so on, the phenomena of volcanic eruptions, including the intuitive eruptive events and the doubtful volcanic eruption phenomenon such as "the ash fall", "the white hair fall", "the sky fire", "the dust fall" are investigated and put in order systematically in this paper. The results are as follows:1)The intuitive eruptive events are the 1002AD eruption of Mt. Halla volcano on Jeju Island, Korea Peninsula, and the 1007AD volcanic eruption offshore to the west of Jeju Island, Korea Peninsula, as well as the 1597AD eruption of Mt. Wangtian'e volcano in Changbai County, Jilin Province, China; 2)"The ash fall" is airborne volcanic ash, and those "ash falls" happening in 1265, 1401-1405, 1668, 1673 and 1702AD are possibly the tephra of Changbaishan Tianchi volcano; 3)"The white hair fall" is Pele's hair and it is speculated that the "white hair fall "happening in 1737AD is related to Changbaishan Tianchi volcanic eruption; 4)If regarding "the sky fire" as the volcanic eruption phenomenon, "the sky fire" happening in 1533AD is possibly the Changbaishan volcanic eruption event, and "the sky fire" in 1601-1609AD may be the eruptive event of the Longgang volcano in Jilin Province, China or Changbaishan Tianchi volcano; 5)"The dust fall" is recorded in many historical documents. However, "the dust fall" is not the volcanic ash fall but the phenomenon of loess fall. So, it is improper to determine the eruptive events of Changbaishan Tianchi volcano on the basis of "the dust fall".     

Improved prediction and tracking of volcanic ash clouds

During the past 30 years, more than 100 airplanes have inadvertently flown through clouds of volcanic ash from erupting volcanoes. Such encounters have caused millions of dollars in damage to the aircraft and have endangered the lives of tens of thousands of passengers. In a few severe cases, total engine failure resulted when ash was ingested into turbines and coating turbine blades. These incidents have prompted the establishment of cooperative efforts by the International Civil Aviation Organization and the volcanological community to provide rapid notification of eruptive activity, and to monitor and forecast the trajectories of ash clouds so that they can be avoided by air traffic. Ash-cloud properties such as plume height, ash concentration, and three-dimensional ash distribution have been monitored through non-conventional remote sensing techniques that are under active development. Forecasting the trajectories of ash clouds has required the development of volcanic ash transport and dispersion models that can calculate the path of an ash cloud over the scale of a continent or a hemisphere. Volcanological inputs to these models, such as plume height, mass eruption rate, eruption duration, ash distribution with altitude, and grain-size distribution, must be assigned in real time during an event, often with limited observations. Databases and protocols are currently being developed that allow for rapid assignment of such source parameters. In this paper, we summarize how an interdisciplinary working group on eruption source parameters has been instigating research to improve upon the current understanding of volcanic ash cloud characterization and predictions. Improved predictions of ash cloud movement and air fall will aid in making better hazard assessments for aviation and for public health and air quality.     

Geochemistry of Vesuvius volcanics during 1631–1944 period

Major- and trace-element systematics of the magmas erupted during the recent activity of Vesuvius (1631–1944) show that composition ranges are narrow and identical in each activity cycle except for the 1631 eruption which emitted also more differentiated products. These rocks are the most basic magmas erupted in Vesuvius. The high K and incompatible-element contents of these products are characteristic of the whole Roman-Campanian Province.Lavas and tephra of the 1906 and 1944 eruptions are highly porphyric. Geochemical modelling shows that these products are mechanical mixing of a slightly differentiated melt and various proportions of phenocrysts. Except ratios of highly incompatible trace elements (Th/La, Th/Ta, Th/Ba), no chemical characteristics of primary and differentiated melts can be simply inferred from bulk erupted products. An inverse method for calculating melt and mineral compositions is presented. Calculated weigth fractions are consistent with measured modal compositions. A model of crystal differential accumulation controlled by relative phase densities and eruption rates is proposed.     

Hazard assessment of far-range volcanic ash dispersal from a violent Strombolian eruption at Somma-Vesuvius volcano, Naples, Italy: implications on civil aviation

Long-range dispersal of volcanic ash can disrupt civil aviation over large areas, as occurred during the 2010 eruption of Eyjafjallaj?kull volcano in Iceland. Here we assess the hazard for civil aviation posed by volcanic ash from a potential violent Strombolian eruption of Somma-Vesuvius, the most likely scenario if eruptive activity resumed at this volcano. A Somma-Vesuvius eruption is of concern for two main reasons: (1) there is a high probability (38?%) that the eruption will be violent Strombolian, as this activity has been common in the most recent period of activity (between AD 1631 and 1944); and (2) violent Strombolian eruptions typically last longer than higher-magnitude events (from 3 to 7?days for the climactic phases) and, consequently, are likely to cause prolonged air traffic disruption (even at large distances if a substantial amount of fine ash is produced such as is typical during Vesuvius eruptions). We compute probabilistic hazard maps for airborne ash concentration at relevant flight levels using the FALL3D ash dispersal model and a statistically representative set of meteorological conditions. Probabilistic hazard maps are computed for two different ash concentration thresholds, 2 and 0.2?mg/m³, which correspond, respectively, to the no-fly and enhanced procedure conditions defined in Europe during the Eyjafjallaj?kull eruption. The seasonal influence of ash dispersal is also analysed by computing seasonal maps. We define the persistence of ash in the atmosphere as the time that a concentration threshold is exceeded divided by the total duration of the eruption (here the eruption phase producing a sustained eruption column). The maps of averaged persistence give additional information on the expected duration of the conditions leading to flight disruption at a given location. We assess the impact that a violent Strombolian eruption would have on the main airports and aerial corridors of the Central Mediterranean area, and this assessment can help those who devise procedures to minimise the impact of these long-lasting low-intensity volcanic events on civil aviation.     

Lava fountains during the episodic eruption of South–East Crater (Mt. Etna), 2000: insights into magma-gas dynamics within the shallow volcano plumbing system

Mt. Etna, in Sicily (Italy) is well known for frequent effusive and explosive eruptions from both its summit and flanks. South-East Crater (SE Crater), one of the four summit craters, has been the most active in the last 20 years and often produces episodic lava fountains over periods lasting from a few weeks to months. The most striking of such eruptive phases was in 2000. Sixty four lava fountains, separated by quiescent intervals and sometimes associated with lava overflows, occurred that year between January and June, a time period during which we consider the volcano to have been in episodic eruption. This paper presents mainly results of petrochemical investigations carried out on both tephra and lavas collected during a number of the lava fountain episodes in 2000. The new data have been integrated with volcanological and seismic information in order to correlate the features of the eruptive activity with magma-gas dynamics in the plumbing system of SE Crater. The main findings allow us to characterise the 2000 episodic eruption in the framework of the recent SE Crater activity. In particular, we infer that the onset of the 2000 eruption was triggered by the ascent of new, more primitive and volatile-rich magma that progressively intruded into the SE Crater reservoir, where it mixed with the resident, more evolved magma. Furthermore, we argue that the 2000 SE Crater lava fountains largely resulted from the instability of a foam layer accumulated at the top of the underlying reservoir and rebuilt prior to each episode, in agreement with the collapsing foam model for lava fountains.     

The 1631 Vesuvius eruption. A reconstruction based on historical and stratigraphical data

The eruption of 1631 A.D. was the most violent and destructive event in the recent history of Vesuvius. More than fifty primary documents, written in either Italian or Latin, were critically examined, with preference given to the authors who eyewitnessed volcanic phenomena. The eruption started at 7 a.m. on December 16 with the formation of an eruptive column and was followed by block and lapilli fallout east and northeast of the volcano until 6 p.m. of the same day. At 10 a.m. on December 17, several nuées ardentes were observed to issue from the central crater, rapidly descending the flanks of the cone and devastating the villages at the foot of Vesuvius. In the night between the 16th and 17th and on the afternoon of the 17th, extensive lahars and floods, resulting from rainstorms, struck the radial valleys of the volcano as well as the plain north and northeast.Deposits of the eruption were identified in about 70 localities on top of an ubiquitous paleosol formed during a long preeruptive volcanic quiescence. The main tephra unit consists of a plinian fallout composed of moderately vesicular dark green lapilli, crystals and lithics. Isopachs of the fallout are elongated eastwards and permit a conservative volume calculation of 0.07 km³. The peak mass flux deduced from clast dispersal models is estimated in the range 3–6 × 10⁷ kg/s, corresponding to a column height of 17–21 km. East of the volcano the plinian fallout is overlain by ash-rich low-grade ignimbrite, surges, phreatomagmatic ashes and mud flows. Ash flows occur in paleovalleys around the cone of Vesuvius but are lacking on the Somma side, suggesting that pyroclastic flows had not enough energy to overpass the caldera wall of Mt. Somma. Deposits are generally unconsolidated, massive with virtually no ground layer and occasionally bearing sparse rests of charred vegetation. Past interpretations of the products emitted on the morning of December 17 as lava flows are inconsistent with both field observations and historical data. Features of the final phreatomagmatic ashes are suggestive of alternating episodes of wet ash fallout and rainfalls. Lahars interfingered with primary ash fallout confirm episodes of massive remobilization of loose tephra by heavy rainfalls during the final stage of the eruption.Chemical analyses of scoria clasts suggest tapping of magma from a compositionally zoned reservoir. Leucite-bearing, tephritic-phonolite (SiO₂ 51.17%) erupted in the early plinian phase was in fact followed by darker and slightly more mafic magma richer in crystals (SiO₂ 49.36%). During the nuées ardentes phase the composition returned to that of the early phase of the eruption.The reconstruction of the 1631 eruptive scenario supplies new perspectives on the hazards related to plinian eruptions of Vesuvius.     

The Kuwae (Vanuatu) eruption of<Emphasis Type="SmallCaps"> AD</Emphasis> 1452: potential magnitude and volatile release

Sometime during AD 1452, according to new evidence, a large-magnitude, initially phreatomagmatic eruption, destroyed the island of Kuwae (16.83°S, 168.54°E), located in the present-day Republic of Vanuatu. It created a 12×6-km submarine caldera composed of two adjacent basins. Based on estimates of caldera volume, between 30 and 60 km³ DRE of dacite magma was ejected as pyroclastic flow and fall deposits during this event. Annual layers of ice dating from the period AD 1450–1460 contain acidity peaks representing fallout of sulfuric acid onto both the Greenland and Antarctic ice caps. These acidity peaks have been attributed by others to the sedimentation of H₂SO₄ aerosols that originated from sulfur degassing during the Kuwae eruption. Improved dating techniques and new data from nineteen ice cores reveal a single acidity peak attributed to Kuwae lasting from 1453 to 1457. In this study, we present new electron microprobe analyses of the S, Cl, and F contents of matrix glasses and glass inclusions in phenocrysts from tephra ejected during the Kuwae eruption. We establish that the Kuwae event did indeed yield a large release of sulfur gases. From our glass inclusion data and analysis, we calculate that the total atmospheric aerosol loading from the 1452 Kuwae eruption was ≫100 Tg H₂SO₄. Much of the volatile mass released during the eruption was probably contained in a separate, volatile-rich, fluid phase within the pre-eruptive Kuwae magma body. Comparing the volatile release of the Kuwae eruption with other large-magnitude eruptions, places Kuwae as the greatest sulfuric acid aerosol producer in the last seven centuries, larger even than sulfur emissions from the eruption of Tambora (Indonesia) in 1815, and possibly Laki (Iceland) in 1783. The severe and unusual climatic effects reported in the mid- to late-1450s were likely caused by the Kuwae eruption.