Abstract Long Valley Caldera and the Mono–Inyo Domes volcanic field in eastern California lie in a left-stepping offset along the eastern escarpment of the Sierra Nevada, at the northern end of the Owens Valley and the western margin of the Basin and Range Province. Over the last 4 Ma, this volcanic field has produced multiple volcanic eruptions, including the caldera-forming eruption at 760 000 a BP and the recent Mono–Inyo Domes eruptions 500–660 a BP and 250 a BP . Beginning in the late 1970s, the caldera entered a sustained period of unrest that persisted through the end of the century without culminating in an eruption. The unrest has included recurring earthquake swarms; tumescence of the resurgent dome by nearly 80 cm; the onset of diffuse magmatic carbon dioxide emissions around the flanks of Mammoth Mountain on the southwest margin of the caldera; and other indicators of magma transport at mid- to upper-crustal depths. Although we have made substantial progress in understanding the processes driving this unrest, many key questions remain, including the distribution, size, and relation between magma bodies within the mid-to-upper crust beneath the caldera, Mammoth Mountain, and the Inyo Mono volcanic chain, and how these magma bodies are connected to the roots of the magmatic system in the lower crust or upper mantle.

Abstract In the last four decades, Campi Flegrei caldera has been the world’s most active caldera characterized by intense unrest episodes involving huge ground deformation and seismicity, but, at the time of writing, has not culminated in an eruption. We present a careful review, with new analyses and interpretation, of all the data and recent research results. We deal with three main problems: the tentative reconstruction of the substructure; the modelling of unrest episodes to shed light on possible pre-eruptive scenarios; and the probabilistic estimation of the hazards from explosive pyroclastic products. The results show, for the first time at a volcano, that a very peculiar mechanism is generating episodes of unrest, involving mainly activation of the geothermal system from deeper magma reservoirs. The character and evolution of unrest episodes is strongly controlled by structural features, like the ring-fault system at the borders of the caldera collapse. The use of detailed volcanological, mathematical and statistical procedures also make it possible to obtain a detailed picture of eruptive hazards in the whole Neapolitan area. The complex behaviour of this caldera, involving interaction between magmatic and geothermal phenomena, sheds light on the dynamics of the most dangerous types of volcanoes in the world.

Abstract The Neogene ignimbrite flare-up of the Altiplano Puna Volcanic Complex (APVC) of the Central Andes produced one of the best-preserved large silicic volcanic fields on Earth. At least 15 000 km 3 of magma erupted as regional-scale ignimbrites between 10 and 1 Ma, from large complex calderas that are typical volcano-tectonic depressions (VTD). Simple Valles-type calderas are absent. Integration of field, geochronological, petrological, geochemical and geophysical data from the APVC within the geodynamic context of the Central Andes suggests a scenario where elevated mantle power input, subsequent crustal melting and assimilation, and development of a crustal-scale intrusive complex lead to the development of APVC. These processes lead to thermal softening of the sub-APVC crust and eventual mechanical failure of the roofs above batholith-scale magma chambers to trigger the massive eruptions. The APVC ignimbrite flare-up and the resulting VTDs are thus the result of the time-integrated impact of intrusion on the mechanical strength of the crust, and should be considered tectonomagmatic phenomena, rather than purely volcanic features. This model requires a change in paradigm about how the largest explosive eruptions may operate.

Abstract Here we review the studies of pluton emplacement and caldera collapse, proposing a model linking the two processes. The shallow rise of magma occurs by dyking, and its emplacement occurs along major anisotropies. Many plutons are emplaced at subhorizontal discontinuities, forming sills. These will eventually grow, forming laccoliths, the most common mechanism to store shallow magma. Calderas are a surface expression of these reservoirs, and can be approximated by a piston cylinder, which sinks due to: (1) underpressure in the reservoir, producing outward-dipping reverse faults; (2) overpressure within a silllike chamber undergoing doming; and (3) overpressure within a laccolith generating apical tensile stresses; in the latter two cases, inward-dipping normal faults form. We suggest that collapse calderas are the surface expression of pressure variations within laccoliths or tabular intrusions, Their geometric relationships depend on the shape and aspect ratio of the intrusion and its pressure conditions. Tabular intrusions, generating ring-faults mainly at their tips, form calderas with a similar width to the intrusion; laccoliths may generate ring-faults along the intrusion roof, forming calderas narrower than the intrusion. The outward or inward dip of the caldera faults results from the underpressure or overpressure conditions within the reservoir.

Abstract Most ring-faults of collapse calderas are either circular or slightly elliptical in plan view, vertical or steeply dipping, and have vertical displacements from several hundred metres to a few kilometres. Most ring-faults are dip slip. Many, however, are partly dip-slip faults and thus shear fractures, and partly ring-dykes and thus extension fractures. Ring-faults have been modelled using analytical, analogue and numerical methods. While analytical models throw light on some aspects of the conditions for ring-fault initiation, they are too general for detailed analysis of ring-fault development. By contrast, analogue and numerical models, combined with rigorous testing on field data, have in recent years provided many new ideas and improved understanding of ring-fault initiation and development. Here we present numerical models for ring-fault formation in isotropic (non-layered) and anisotropic (layered) host rocks. Magmatic underpressure (below lithostatic) or excess pressure (above lithostatic) as the only loading normally favours dyke injections rather than ring-fault formation. A spherical (circular) chamber in a volcanic field subject to (centimetre-scale) doming, tension, or both, is unlikely to trigger ring-fault formation unless the chamber is located in a very soft (low Young’s modulus) layer or has recently injected dykes. The numerical models indicate that a ring-fault (and a ring-dyke) is most likely to form, in layered as well as non-layered host rocks, in a local stress field generated by a shallow sill-like chamber in a volcanic field subject to doming, tension, or both. The diameter of the chamber must be much smaller than the diameter of the volcanic field subject to doming; in many of the numerical models the volcanic field diameter is three to five times that of the chamber diameter. For a 20-km-wide layered volcanic field, either tension or tension combined with doming may result in ring-fault formation. For a 40-km-wide layered volcanic field, tension is not necessary; doming alone is sufficient to trigger ring-fault formation.

Abstract: Subsurface volume and pressure increases triggering surface inflation at active calderas are generally deduced by inverting ground-deformation time-series using isotropic and homogeneous half-space models (IHM). These models represent simplified mathematical analogues of the mechanical behaviour of the Earth’s crust. Using three-dimensional numerical modelling, we show that lateral discontinuities such as intracaldera- or caldera-ring-faults can significantly amplify and distort the ground deformation pattern during unrest. As a consequence, data inversions using IHMs, which do not consider lateral discontinuities, can provide erroneous results on causative source parameters. We also find that the degree of amplification and distortion in the form of abrupt changes in displacement/distance gradients in proximity to faults is dependent on source geometry. Prolate bodies represent a particularly critical geometry for which pressure increases may be overestimated by a factor of up to three. Our 3D analysis suggests that amplification effects can be much larger than predicted by earlier 2D models. We validate theoretical results by applying our model to investigate the effect of boundary faults and source geometries on the displacement field during ground uplift at the restless calderas of Campi Flegrei (Italy) and Sierra Negra (Galapagos Islands, Ecuador). Based on the discrepancy in results from IHMs and our numerical model, we argue that employing IHMs for inversion of ground displacement and gravity time-series may in some cases lead to a biased assessment of hazards associated with ground uplift.

Abstract: The predominantly constructive life cycles of large, long-lived, stratovolcanoes and basaltic shields are punctuated by transitory episodes during which large volumes of material are divested from the flanks. Such shedding typically takes place catastrophically in the form of a lateral collapse, generating a debris avalanche and leaving a scar that may attain caldera dimensions. Collapse may follow instability development arising from a single, discrete, event, such as a crypto-dome intrusion, or may be the end-product of progressive destabilization over a long period of time. Lateral collapses may also occur at persistent slumps, which may have been active over periods as long as 10 4 – 10 5 a prior to catastrophic failure. Collapse velocities may exceed 40 m s –1 , leading to completion of the process within a few hundreds of seconds. Collapse volumes span several orders of magnitude, ranging from less than 1 km 3 to more than 10 km 3 at many continental and subduction-zone volcanoes, to 1000 km 3 or more at the great basalt shields of Hawaii. Lateral collapse may be accompanied by a wide range of associated hazards, including atmospheric shock wave, pyroclastic flows and surges, extensive tephra fall, and secondary lahars. For volcanoes in the marine environment, potentially destructive and lethal tsunamis can be added to the inventory. Here, the potential for a lateral collapse at an ocean island volcano to generate a ‘mega-tsunami’ (more than 100 m high at source and destructive at oceanic distances) is discussed and evaluated.

Abstract Campi Flegrei caldera, west of Naples in southern Italy, has an exceptional documented record of ground deformation from Roman times onwards. Systematic recording began in the nineteenth century. For earlier dates, information has been obtained from archaeological studies and from contemporary descriptions of the locations of buildings, usually Roman, with respect to sea-level. Especially important have been accounts related to the Serapis, a Roman market-place built in the second century bc and now incorporated within the modern town of Pozzuoli. The long-term patterns of ground deformation have conventionally been investigated on the premise that Campi Flegrei naturally tends to a state of static equilibrium. This study argues that, instead, the area naturally tends to a steady rate of subsidence, at about 17 mm a -1 . After this background rate has been removed, the data indicate that a permanent uplift of some 33 m has occurred from Roman times (up until the present day: 2005 at the time of writing), attributable to the intrusion of 1.85 km 3 of magma, of which only 1% has been erupted. Uplift has occurred in three episodes, the third of which is still in progress. The behaviour can be interpreted in terms of the intermittent ascent of magma between a reservoir of c . 10 2 —10 3 km 3 at depths of 8-15 km or greater, to a much smaller, shallower system at depths of about 3 – 4 km. Should the current pattern of deformation follow previous trends, uplift is expected to continue for another 80–90 years, during which time Campi Flegrei will be characterized by an elevated possibility of eruption.

Abstract Previous and new results from probabilistic approaches based on available volcanological data from real eruptions of Campi Flegrei, are assembled in a comprehensive assessment of volcanic hazards at the Campi Flegrei caldera, in order to compare the volcanic hazards related to the different types of events. Hazard maps based on a very wide set of numerical simulations, produced using field and laboratory data as input parameters relative to the whole range of fallout and pyroclastic-flow events and their relative occurrence, are presented. The results allow us to quantitatively evaluate and compare the hazard related to pyroclastic fallout and density currents (PDCs) in the Campi Flegrei area and its surroundings, including the city of Naples. Due to the dominant wind directions, the hazard from fallout mostly affects the area east of the caldera, and the caldera itself, with the level of probability and expected thickness decreasing with distance from the caldera and outside the eastern sectors. The hazard from PDCs decrease roughly radially with distance from the caldera centre and is strongly controlled by the topographic relief, which produces an effective barrier to propagation of PDCs to the east and northeast, areas which include metropolitan Naples. The main result is that the metropolitan area of Naples would be directly exposed to both fallout and PDCs. Moreover, the level of probability for critical tephra accumulation by fallout is relatively high, even for moderate-scale events, while, due to the presence of topographic barriers, the hazard from PDCs is only moderate and mostly associated with the largest events.

Abstract Long Valley Caldera and the Mono-Inyo Craters chain form a large volcanic complex in eastern California that has experienced persistent earthquake activity and ground uplift over the past 25 years. The central part of Long Valley Caldera (an area of more than 100 square km) has been slowly rising since 1980 at an average rate of 3 cm a Inversion of micro-gravimetry and deformation data using a single vertical prolate ellipsoid source has helped to define the existence of a relatively shallow (5–8 km) silicic magma intrusion of 0.11–0.19 km 3 beneath the caldera’s resurgent dome. We use the information from the singlesource inversion to constrain a more general three-dimensional distribution of volume changes in the subsurface. The distributed inversion identifies two main inflation areas beneath the resurgent dome: one following the regional trend of north–south faults, and another in the dome’s southern section, parallel to a strike-slip fault that is responsible for most of the seismic activity in the caldera’s south moat.

Abstract A model of caldera resurgence was applied to the Island of Ischia to explain uplift, volcanic activity and tectonics on Mount Epomeo, as well as historical seismicity and slow ground movements recorded for the past 2000 years. A two-dimensional mechanical model was utilized for the crust, which was considered to be an elastic plate overlying a laccolith. Geometric dimensions and mechanical parameters were constrained using geological, geophysical and geochemical data. We propose that a laccolith, with a diameter L of c . 10 km, and a depth of up to 1 km in the centre of the island, triggered the caldera resurgence after the Mount Epomeo Green Tuff eruption forming the caldera (55 000 a bp ). A bending phase and a punched laccolith phase are thought to have caused the observed deformations in the caldera. These processes control the tectonics at the boundary of the Mount Epomeo resurgent structure, volcanic activity and dynamics of the island.

Abstract Large caldera collapses represent catastrophic natural events, second only to large meteoritic impacts. In addition, some calderas are densely populated, making the risk extreme, even for moderate eruptions. Understanding caldera mechanisms, unrest and the danger of eruption is therefore a crucial challenge for Earth sciences. Several key features of caldera behaviour have yet to be fully understood. Through a combination of case studies and theoretical modelling, the following topics are addressed in this volume: the conditions required to produce and to release large volumes of magma erupted during caldera formation; how magmatic feeding systems evolve before and after a caldera has formed; the processes that limit the behaviour of precursors to eruptions; how pre-emptive precursors can be distinguished from those that drive unrest without an eruption; and given that post-collapse eruptions may occur across a wide area, the optimum procedures for designing hazard maps and mitigation strategies.