Three caldera-shaped accidents: Volcanic calderas,meteoric scars and lunar cirques

A caldera is a large volcanic depression, more or less circular, the diameter of which is many times greater than those of the included volcanic vents. Calderas must be separated from feetono-volcanic depressions, which have an irregular shape. Volcanic calderas are produced by engulfment. The scars, produced by the impact of meteorites on the earth, are circular or elliptical depressions. Lunar cirques are nearly all circular; some of them have a polygonal, and then sometimes an hexagonal shape. On the surface of the moon elliptical depressions are wholly absent. Moreover, on the moon it is a rigid law that when intersections of ring plains do occur, the smaller cirque is entire, the rim of the middlemost being interrupted by the smallest, whereas the biggest is interrupted by the middle one. This phenomenon would be in accordance with a volcanic origin and a decreasing volcanic activity whereas it is incompatible with an impact of meteorites.     

Erosion calderas: origins, processes, structural and climatic control

 The origin and development of erosion-modified, erosion-transformed, and erosion-induced depressions in volcanic terrains are reviewed and systematized. A proposed classification, addressing terminology issues, considers structural, geomorphic, and climatic factors that contribute to the topographic modification of summit or flank depressions on volcanoes. Breaching of a closed crater or caldera generated by volcanic or non-volcanic processes results in an outlet valley. Under climates with up to ∼2000–2500 mm annual rainfall, craters, and calderas are commonly drained by a single outlet. The outlet valley can maintain its dominant downcutting position because it quickly enlarges its drainage basin by capturing the area of the primary depression. Multi-drained volcanic depressions can form if special factors, e.g., high-rate geological processes, such as faulting or glaciation, suppress fluvial erosion. Normal (fluvial) erosion-modified volcanic depressions the circular rim of which is derived from the original rim are termed erosion craters or erosion calderas, depending on the pre-existing depression. The resulting landform should be classed as an erosion-induced volcanic depression if the degradation of a cluster of craters produces a single-drained, irregular-shaped basin, or if flank erosion results in a quasi-closed depression. Under humid climates, craters and calderas degrade at a faster rate. Mostly at subtropical and tropical ocean-island and island-arc volcanoes, their erosion results in so-called amphitheater valleys that develop under heavy rainfall (>∼2500 mm/year), rainstorms, and high-elevation differences. Structural and lithological control, and groundwater in ocean islands, may in turn preform and guide development of high-energy valleys through rockfalls, landsliding, mudflows, and mass wasting. Given the intense erosion, amphitheater valleys are able to breach a primary depression from several directions and degrade the summit region at a high rate. Occasionally, amphitheater valleys may create summit depressions without a pre-existing crater or caldera. The resulting, negative landforms, which may drain in several directions and the primary origin of which is commonly unrecognizable, should be included in erosion-transformed volcanic depressions. Received: 4 January 1998 / Accepted: 18 January 1999     

Crustal structures that produce eruptions of welded tuff and formation of calderas

Accumulation of a huge amount of welded tuff and subsequent formation of calderas are typical examples of volcanisms on a large scale. Crustal structures which are related to the phenomena may be a key for the study of volcanisms on the earth. In Japan there are some 10 calderas of low gravity anomaly type, large and small in diameter, and accordingly in volume of their ejecta, welded tuff. The low gravity anomalies are very characteristic inside the calderas and just concentric with their rims, and are due to mass deficiency caused by eruptions of a huge amount of ejecta. The author (1957), hitherto, has clarified the subterranean structures resulting from the caldera formations by a gravimetric method. In this paper, the author describes the crustal structures beneath Sikotu Caldera in Hokkaido, Japan, deduced from the gravimetric, geomagnetic, seismic and geologic surveys. And referring several examples of the subterranean structures of the other calderas in Japan, the author wishes to discuss the crustal structures that would produce the caldera formations.     

Subsidence of ash-flow calderas: relation to caldera size and magma-chamber geometry

 Diverse subsidence geometries and collapse processes for ash-flow calderas are inferred to reflect varying sizes, roof geometries, and depths of the source magma chambers, in combination with prior volcanic and regional tectonic influences. Based largely on a review of features at eroded pre-Quaternary calderas, a continuum of geometries and subsidence styles is inferred to exist, in both island-arc and continental settings, between small funnel calderas and larger plate (piston) subsidences bounded by arcuate faults. Within most ring-fault calderas, the subsided block is variably disrupted, due to differential movement during ash-flow eruptions and postcollapse magmatism, but highly chaotic piecemeal subsidence appears to be uncommon for large-diameter calderas. Small-scale downsag structures and accompanying extensional fractures develop along margins of most calderas during early stages of subsidence, but downsag is dominant only at calderas that have not subsided deeply. Calderas that are loci for multicyclic ash-flow eruption and subsidence cycles have the most complex internal structures. Large calderas have flared inner topographic walls due to landsliding of unstable slopes, and the resulting slide debris can constitute large proportions of caldera fill. Because the slide debris is concentrated near caldera walls, models from geophysical data can suggest a funnel geometry, even for large plate-subsidence calderas bounded by ring faults. Simple geometric models indicate that many large calderas have subsided 3–5 km, greater than the depth of most naturally exposed sections of intracaldera deposits. Many ring-fault plate-subsidence calderas and intrusive ring complexes have been recognized in the western U.S., Japan, and elsewhere, but no well-documented examples of exposed eroded calderas have large-scale funnel geometry or chaotically disrupted caldera floors. Reported ignimbrite "shields" in the central Andes, where large-volume ash-flows are inferred to have erupted without caldera collapse, seem alternatively interpretable as more conventional calderas that were filled to overflow by younger lavas and tuffs. Some exposed subcaldera intrusions provide insights concerning subsidence processes, but such intrusions may continue to evolve in volume, roof geometry, depth, and composition after formation of associated calderas. Received: 13 February 1997 / Accepted: 9 August 1997     

Formation and development of normal-fault calderas and the initiation of large explosive eruptions

The ring fractures that form most collapse calderas are steeply inward-dipping shear fractures, i.e., normal faults. At the surface of the volcano within which the caldera fault forms, the tensile and shear stresses that generate the normal-fault caldera must peak at a certain radial distance from the surface point above the center of the source magma chamber of the volcano. Numerical results indicate that normal-fault calderas may initiate as a result of doming of an area containing a shallow sill-like magma chamber, provided that the area of doming is much larger than the cross-sectional area of the chamber and that the internal excess pressure in the chamber is smaller than that responsible for doming. This model is supported by the observation that many caldera collapses are preceded by a long period of doming over an area much larger than that of the subsequently formed caldera. When the caldera fault does not slip, eruptions from calderas are normally small. Nearly all large explosive eruptions, however, are associated with slip on caldera faults. During dip slip on, and doming of, a normal-fault caldera, the vertical stress on part of the underlying chamber suddenly decreases. This may lead to explosive bubble growth in this part of the magma chamber, provided its magma is gas rich. This bubble growth can generate an excess fluid pressure that is sufficiently high to drive a large fraction of the magma out of the chamber during an explosive eruption. Received: 2 January 1997 / Accepted: 22 April 1998     

Caldera resurgence during magma replenishment and rejuvenation at Valles and Lake City calderas

A key question in volcanology is the driving mechanisms of resurgence at active, recently active, and ancient calderas. Valles caldera in New Mexico and Lake City caldera in Colorado are well-studied resurgent structures which provide three crucial clues for understanding the resurgence process. (1) Within the limits of ⁴⁰Ar/³⁹Ar dating techniques, resurgence and hydrothermal alteration at both calderas occurred very quickly after the caldera-forming eruptions (tens of thousands of years or less). (2) Immediately before and during resurgence, dacite magma was intruded and/or erupted into each system; this magma is chemically distinct from rhyolite magma which was resident in each system. (3) At least 1?km of structural uplift occurred along regional and subsidence faults which were closely associated with shallow intrusions or lava domes of dacite magma. These observations demonstrate that resurgence at these two volcanoes is temporally linked to caldera subsidence, with the upward migration of dacite magma as the driver of resurgence. Recharge of dacite magma occurs as a response to loss of lithostatic load during the caldera-forming eruption. Flow of dacite into the shallow magmatic system is facilitated by regional fault systems which provide pathways for magma ascent. Once the dacite enters the system, it is able to heat, remobilize, and mingle with residual crystal-rich rhyolite remaining in the shallow magma chamber. Dacite and remobilized rhyolite rise buoyantly to form laccoliths by lifting the chamber roof and producing surface resurgent uplift. The resurgent deformation caused by magma ascent fractures the chamber roof, increasing its structural permeability and allowing both rhyolite and dacite magmas to intrude and/or erupt together. This sequence of events also promotes the development of magmatic–hydrothermal systems and ore deposits. Injection of dacite magma into the shallow rhyolite magma chamber provides a source of heat and magmatic volatiles, while resurgent deformation and fracturing increase the permeability of the system. These changes allow magmatic volatiles to rise and meteoric fluids to percolate downward, favouring the development of hydrothermal convection cells which are driven by hot magma. The end result is a vigorous hydrothermal system which is driven by magma recharge.     

Ages of calderas,large explosive craters and active volcanoes in the Kuril-Kamchatka region,Russia

The ages of most of calderas, large explosive craters and active volcanoes in the Kuril-Kamchatka region have been determined by extensive geological, geomorphological, tephrochronological and isotopic geochronological studies, including more than 600 ¹⁴C dates. Eight Krakatoa-type and three Hawaiian-type calderas and no less than three large explosive craters formed here during the Holocene. Most of the Late Pleistocene Krakatoa-type calderas were established around 30 000–40 000 years ago. The active volcanoes are geologically very young, with maximum ages of about 40 000–50 000 years. The overwhelming majority of recently active volcanic cones originated at the very end of the Late Pleistocene or in the Holocene. These studies show that all Holocene stratovolcanoes in Kamchatka were emplaced in the Holocene only in the Eastern volcanic belt. Periods of synchronous, intensified Holocene volcanic activity occurred within the time intervals of 7500–7800 and 1300–1800 ¹⁴C years BP.     

Bedding in Scottish (Fifeshire) Tuff-pipes and its relevance to maars and calderas

Along the coast between Lower Largo and St. Monance, in Fife, Scotland, Carboniferous sediments are piecerd by 13 exceptionally well-exposed basaltic tuff-pipes. The pyroclastic rocks, which are bedded in either centroclinal or collapsed form, were originally formed subaerially. They are separated by ring faults from the surrounding sediments which are turned down against the pipe margins. The tuffs in the pipes have undergone cauldron subsidence of at least 500 m. Clastic and magmatic minor intrusions, particularly at the margins of the pipes, accompanied the subsidence. As comparable amounts of subsidence are recorded in many basaltic tuffpipes (some with, others without ring fractures) in other parts of the world, it is suggested that subsidence may have contributed to the formation of maars. Comparison is made between cauldron-subsidence features in tuff-pipes and those of calderas.     

Hidden calderas evidenced by multisource geophysical data; example of Cappadocian Calderas, Central Anatolia

The Cappadocian volcanic field in central Anatolia (Turkey) is characterised by a sequence of 10 Neogene ignimbrites. The associated calderas have been partly dismantled and buried by subsequent tectonic and sedimentary processes and, therefore, cannot be readily recognized in the field. Recent progress in the understanding of the stratigraphic correlations and flow patterns has identified two main probable source areas for the ignimbrites. Detailed study of these areas, based on gravity surveys, remote sensing data (SPOT and ERS1 images) and digital elevation models (DEM), has provided evidence for two major caldera complexes and their relationship to old stratovolcanoes and Neogene tectonics. The older Nevsehir–Acigöl caldera complex, located between the towns of Acigöl, Nevsehir and Cardak, is inferred to be the source of the Kavak and Zelve ignimbrites. The Nevsehir–Acigöl caldera complex is defined mainly by a −35 mGal circular gravimetry anomaly about 15 km in diameter. The boundaries of this, now buried, caldera complex are shown by high gradients on the Bouguer gravity anomaly map. The younger Derinkuyu caldera complex, located between the Erdas stratovolcano and the Ciftlik basin, is inferred to be the source of the Sarimaden, Cemilköy, Gördeles and Kizilkaya ignimbrites. It is well-defined by a rectangular (35×23 km) gravity low (−30 mGal) with a positive high (+20 mGal) in the center. Gravity, remote sensing data and the DEM provide evidence that the Erdas stratovolcano, on the northern margin of the Derinkuyu caldera complex, represents the remnants of a large stratovolcano partly cut by one or more caldera collapses. The positive anomaly within the Derinkuyu caldera complex is centered on the 15-km-wide Sahin Kalesi volcanic massif. Field evidence and structural features inferred from the DEM and remote sensing data strongly suggest that this massif is a resurgent doming associated with the Gördeles ignimbrite eruption. High-resolution ERS1, SPOT and DEM images reveal that the transtensive regime, active at least since the Miocene, influenced the location of eruptive centers and caldera complexes in Cappadocia. The two caldera complexes are located in transtensive grabens. The subsidence of these grabens, continuing after the caldera collapse events, most likely resulted in the burying of the calderas and could explain the difficulties in identifying them in the field.     

Mobility and depositional controls of radioelements in hydrothermal systems at the Long Valley and Valles calderas

The loci and abundance of U and Th were examined in tuffaceous rocks encompassing hydrothermal systems at the Long Valley caldera, California and the Valles caldera, New Mexico. Aspects of these systems may be analogous to conditions expected in a potential site for a high-level waste repository in welded tuff. Examination of radioelements in core from scientific drill holes at these sites was accomplished by gamma-ray spectrometry and fission-track radiography. In the lateral-flowing hydrothermal system at the Long Valley caldera, where temperatures range from 140 to 200 °C, U is concentrated to 20 ppm in Fe-rich zones of varved tuff and to 50 ppm with Fe-rich mineral phases in tuff fragments of a calcite-cemented breccia. U-series disequilibrium in some of these samples suggests mobilization/deposition of parent U and/or its daughters. In the vapor zone of the Valles caldera's hydrothermal system (temperature ˜ 100 °C), the concordance of high U, low Th/U and decreasing whole-rock O-isotope ratios suggests that U was concentrated in response to hydrothermal circulation when the system was formerly liquid-dominated. In the underlying present-day liquid-dominated zone (temperature to 210 °C), U, up to several tens of parts per million, occurs with pyrite and Fe-oxide minerals, and in concentrations to several percents with a Ti-Nb-Y-rare earth mineral. In the Valles system's outflow zone, U is also concentrated in Fe-rich zones as well as in carbonaceous-rich zones in the Paleozoic sedimentary rocks that underlie the Quaternary tuff. Th, associated with accessory minerals, predominates in breccia zones and in a mineralized fault zone near the base of the Paleozoic sedimentary sequence. Relatively high concentrations of U occur in springs representative of water recharging the Valles caldera's hydrothermal system. In contrast, considerably lower U concentrations occur in hot waters (> 220 °C) and in the system's outflow plume, suggesting that U is concentrating in the hotter part of the system. The Long Valley and Valles observations indicate that U and Ra are locally mobile under hydrothermal conditions, and that reducing conditions associated with Fe-rich minerals and carbonaceous material are important factors in the adsorption of U, and thus can retard its transport in water at elevated temperature.     

Post-collapse volcanic history of calderas on a composite volcano: an example from Roccamonfina,southern Italy

Roccamonfina, part of the Roman Potassic Volcanic Province, is an example of a composite volcano with a complex history of caldera development. The main caldera truncates a cone constructed predominantly of this caldera may have been associated with one of the ignimbritic eruptions of the Brown Leucitic Tuff (BLT) around 385 000 yr BP. The Campagnola Tuff, the youngest ignimbrite of the BLT, however, drapes the caldera margin and must postdate at least the initial stages of collapse. During the subsequent history of the caldera there were several major explosive eruptions. The largest of these was that of the Galluccio Tuff at about 300 000 yr BP. It is likely that there was further collapse within the main caldera associated with these eruptions. It is of note that despite these subsequent major explosive eruptions later collapse occurred within the confines of the main caldera. Between eruptions caldera lakes developed producing numerous lacustrine beds within the caldera fill. Extensive phases of phreatomagmatic activity generated thick sequences of pyroclastic surge and fall deposits. Activity within the main caldera ended with the growth of a large complex of basaltic trachyandestite lava domes around 150 000 yr BP. Early in the history of Roccamonfina sector collapse on the northern flank of the volcano formed the northern caldera. One of the youngest major events on Roccamonfina occurred at the head of this northern caldera with explosive activity producing the Conca Ignimbrite and associated caldera. There is no evidence that there was any linkage in the plumbing systems that fed eruptions in the main and northern calderas.     

Diffuse CO2 emission rate from Pululahua and the lake-filled Cuicocha calderas,Ecuador

We report herein the first results of two soil CO₂ efflux surveys carried out at Cuicocha lake-filled and Pululahua caldera volcanic systems, Ecuador. A total of 172 and 217 soil CO₂ efflux measurements were taken at the surface environment of Pululahua and Cuicocha calderas respectively, by means of the “accumulation chamber” method during the summer of 2006 to constrain the total CO₂ output from the studied area. Soil CO₂ efflux values ranged from non-detectable up to 48.5 and 141.7 g m^{− 2} d^{− 1} for Cuicocha and Pululahua calderas respectively. In addition, probability graphs were used to distinguish the existence of different geochemical populations. Sequential Gaussian Simulation was used to construct an average map for 100 simulations and to compute the total CO₂ emission at each studied area: 106 and 270 t d^{− 1} (metric tons per day) for Cuicocha (13.3 km²) and Pululahua (27.6 km²) volcanoes respectively.     

Eruptive versus non-eruptive behaviour of large calderas: the example of Campi Flegrei caldera (southern Italy)

Caldera eruptions are among the most hazardous of natural phenomena. Many calderas around the world are active and are characterised by recurrent uplift and subsidence periods due to the dynamics of their magma reservoirs. These periods of unrest are, in some cases, accompanied by eruptions. At Campi Flegrei caldera (CFc), which is an area characterised by very high volcanic risk, the recurrence of this behaviour has stimulated the study of the rock rheology around the magma chamber, in order to estimate the likelihood of an eruption. This study considers different scenarios of shallow crustal behaviour, taking into account the earlier models of CFc ground deformation and caldera eruptions, and including recent geophysical investigations of the area. A semi-quantitative evaluation of the different factors that lead to magma storage or to its eruption (such as magma chamber size, wall-rock viscosity, temperature, and regional tectonic strain rate) is reported here for elastic and viscoelastic conditions. Considering the large magmatic sources of the CFc ignimbrite eruptions (400–2,000 km³) and a wall-rock viscosity between 10¹⁸ and 10²⁰ Pa s, the conditions for eruptive failure are difficult to attain. Smaller source dimensions (a few cubic kilometres) promote the condition for fracture (eruption) rather than for the flow of wall rock. We also analyse the influence of the regional extensional stress regime on magma storage and eruptions, and the thermal stress as a possible source of caldera uplift. The present study also emphasises the difficulty of distinguishing eruption and non-eruption scenarios at CFc, since an unambiguous model that accounts for the rock rheology, magma-source dimensions and locations and regional stress field influences is still lacking.     

Downsag and extension at calderas: new perspectives on collapse geometries from ice-melt,mining, and volcanic subsidence

Structures at calderas may form as a result of precursory tumescence, subsidence due withdrawal of magmatic support, resurgence, and regional tectonism. Structural reactivation and overprinting are common. To explore which types of structures may derive directly from subsidence without other factors, evidence is reviewed from pits caused by the melting of buried ice blocks, mining subsidence, scaled subsidence models, and from over 50 calderas. This review suggests that complex patterns of peripheral deformation, with multiple ring and arcuate fractures both inside and outside caldera rims, topographic embayments, arcuate graben, and concentric zones of extension and compression may form as a direct result of subsidence and do not require a complex subsidence and inflation history. Downsag is a feature of many calderas and it does not indicate subsidence on an inward-dipping ring fault, as has been inferred previously. Where magmatic inflation is absent or slight, initial arcuate faults formed during collapse are likely to be multiple, and dip outwards to vertical. Associated downsag causes the peripheries of calderas undergo radial (centripetal) extension, and this accounts for some of the complex peripheral fractures, arcuate crevasses, graben, and some topographic moats. The structural boundary of a caldera, defined here as the outermost limits of subsidence and related deformation including downsag, commonly lies outside ring faults and outside the embayed topographic wall. It is likely to be funnel-shaped, i.e. inward-dipping, even though ring and arcuate fractures within it may dip outward. Inward-dipping arcuate normal faults at shallow levels and steep inward-dipping contacts between a caldera's fill and walls may both occur at a caldera that has initially subsided on outward-dipping ring faults. They arise due to peripheral surficial extension, gravitational spreading and scarp collapse. Topographic enlargement at some calderas and the formation of embayments may reflect general progressive downsag and localized downsag, respectively. These processes may occur in addition to surficial degradation of oversteep ring-fault scarps.     

Genesis of dacitic magmatism at batur volcano, Bali, Indonesia: Implications for the origins of stratovolcano calderas

Batur is an active stratovolcano on the island of Bali, Indonesia, with a large, well-formed caldera whose formation is correlated with the eruption about 23,700 years ago of a thick ignimbrite sheet. Our study of the volcanic stratigraphy and geochemistry of Batur shows the formation of the caldera was signalled by a change in the composition of the erupting material from basaltic and andesitic to dacitic. The dacitic rocks are glassy, possess equilibrium phenocryst assemblages, and display compositional characteristics consistent with an origin by crystal-liquid fractionation from more mafic parent magmas in a shallow chamber, possibly at 1.5 km depth and 1000–1070°C.However, although separated by a gap of 6 wt.% SiO₂, the dacitic rocks are clearly related in their minor- and trace-element geochemistry to those basalts and basaltic andesites erupted after the caldera was formed rather than to the andesites erupted immediately before the dacites first appeared. We infer from this and published experimental modelling of the possible crystallization behaviour of basaltic magma chambers that a magmatic cycle involving caldera formation began independently of the previous activity of Batur by formation of a new, closed-system magma chamber beneath the volcano. Fractional crystallization, possibly at the walls of the chamber, led to the early production of derivative siliceous magmas and, consequently, to caldera formation, while most of the magma retained its original composition. The postcaldera Batur basalts represent the largely undifferentiated core liquids of this chamber.This model contrasts with the traditional evolutionary model for stratovolcano calderas but may be applicable to the origins of calderas similar to that of Batur, particularly those in volcanic island arcs.     

现代海底热液活动的调查研究方法

介绍了国外在现代海底热液活动调查研究中使用的方法和技术手段。现代海底热液活动是普遍发育于大洋中活动板块边界及板内火山活动中心的一种在岩圈和大洋之间进行能量和物质交换的过程。其突出的表象是高温的热液从海底流出，并由此造成热液活动区和上限覆水体的物理和化学异常。海底热测量、海底岩石取样、水体的CH4、^3He和Mn异常观测、硅氧异常观测、多波束测量、OBS观测、深潜调查等都是进行现代海底热液活动调查研究的重要手段。其中水体异常的观测是最为快捷有效的，而深潜器直接观测是现代海底热液活动研究必不可少的。     

Volcanic systems and calderas in the Vatnajökull region,central Iceland: Constraints on crustal structure from gravity data

A Bouguer anomaly map is presented of southern central Iceland, including the western part of Vatnajökull and adjacent areas. A complete Bouguer reduction for both ice surface and bedrock topography is carried out for the glaciated regions. Parts of the volcanic systems of Vonarskarð-Hágöngur, Bárðarbunga-Veiðivötn, Grímsvötn-Laki, and to a lesser extent Kverkfjöll, show up as distinct features on the gravity map. The large central volcanoes with calderas: Vonarskarð, Bárðarbunga, Kverkfjöll and Grímsvötn, are associated with 15–20 mGal gravity highs caused by high density bodies in the uppermost 5 km of the crust. Each of these bodies is thought to be composed of several hundred km³ of gabbros that have probably accumulated over the lifetime of the volcano. The Skaftárkatlar subglacial geothermal areas are not associated with major anomalous bodies in the upper crust. The central volcanoes of Vonarskarð and Hágöngur belong to the same volcanic system; this also applies to Bárðarbunga and Hamarinn, and Grímsvötn and Þórðarhyrna. None of the smaller of the two volcanoes sharing a system (Hágöngur, Hamarinn and Þórðarhyrna) is associated with distinct gravity anomalies and clear caldera structures have not been identified. However, ridges in the gravity field extend between each pair of central volcanoes, indicating that they are connected by dense dyke swarms. This suggests that when two central volcanoes share the same system, one becomes the main pathway for magma, forming a long-lived crustal magma chamber, a caldera and large volume basic intrusive bodies in the upper crust. Short residence times of magma in the crust beneath these centres favour essentially basaltic volcanism. In the case of the second, auxillary central volcano, magma supply is limited and occurs only sporadically. This setting may lead to longer residence times of magma in the smaller central volcanoes, favouring evolution of the magma and occasional eruption of rhyolites. The eastern margin of the Eastern Volcanic Zone is marked by a NE–SW lineation in the gravity field, probably caused by accumulation of low density, subglacially erupted volcanics within the volcanic zone. This lineation lies 5–10 km to the east of Grímsvötn.