Incursion of a large-volume,spatter-bearing pyroclastic density current into a caldera lake: Pavey Ark ignimbrite,Scafell caldera,England

The Scafell caldera-lake volcaniclastic succession is exceptionally well exposed. At the eastern margin of the caldera, a large andesitic explosive eruption (>5 km³) generated a high-mass-flux pyroclastic density current that flowed into the caldera lake for several hours and deposited the extensive Pavey Ark ignimbrite. The ignimbrite comprises a thick (≤125 m), proximal, spatter- and scoria-rich breccia that grades laterally and upwards into massive lapilli-tuff, which, in turn, is gradationally overlain by massive and normal-graded tuff showing evidence of soft-state disruption. The subaqueous pyroclastic current carried juvenile clasts ranging from fine ash to metre-scale blocks and from dense andesite through variably vesicular scoria to pumice (<10³ kg m⁻³). Extreme ignimbrite lithofacies diversity resulted via particle segregation and selective deposition from the current. The lacustrine proximal ignimbrite breccia mainly comprises clast- to matrix-supported blocks and lapilli of vesicular andesite, but includes several layers rich in spatter (≤1.7 m diameter) that was emplaced in a ductile, hot state. In proximal locations, rapid deposition of the large and dense clasts caused displacement of interstitial fluid with elutriation of low-density lapilli and ash upwards, so that these particles were retained in the current and thus overpassed to medial and distal reaches. Medially, the lithofacies architecture records partial blocking, channelling and reflection of the depletive current by substantial basin-floor topography that included a lava dome and developing fault scarps. Diffuse layers reflect surging of the sustained current, and the overall normal grading reflects gradually waning flow with, finally, a transition to suspension sedimentation from an ash-choked water column. Fine to extremely fine tuff overlying the ignimbrite forms ∼25% of the whole and is the water-settled equivalent of co-ignimbrite ash; its great thickness (≤55 m) formed because the suspended ash was trapped within an enclosed basin and could not drift away. The ignimbrite architecture records widespread caldera subsidence during the eruption, involving volcanotectonic faulting of the lake floor. The eruption was partly driven by explosive disruption of a groundwater-hydrothermal system adjacent to the magma reservoir.     

Crystallization and welding variations in a widespread ignimbrite sheet; the Rattlesnake Tuff,eastern Oregon,USA

The 7.05 Ma Rattlesnake Tuff covers ca. 9000 km², but the reconstructed original coverage was between 30000 and 40000 km². Thicknesses are remarkably uniform, ranging between 15 and 30 m for the most complete sections. Only 13% of the area is covered with tuff thicker than 30 m, to a maximum of 70 m. The present day estimated tuff volume is 130 km³ and the reconstructed magma volume of the outflow is 280 km³ DRE (dense rock equivalent). The source area of the tuff is inferred to be in the western Harney Basin, near the center of the tuff distribution, based mainly on a radial exponential decrease in average pumice size, and is consistent with a general radial decrease in welding and degree of post-emplacement crystallization. Rheomorphic tuff is found to a radius of 40–60 km from the inferred source.Four facies of welding and four of post-emplacement crystallization are distinguishable. They are: non-welded, incipiently welded, partially welded and densely welded zones; and vapor phase, pervasively devitrified, spherulite and lithophysae zones. The vapor phase, pervasively devitrified and lithophysae zones are divided into macroscopically distinguishable subzones. At constant thickness (20±3 m), and over a distance of 1–3 km, nonrheomorphic sections can cary between two extremes: (a) entirely vitric sections grading from nonwelded to incipiently welded; and (b) highly zoned sections. Highly zoned sections have a basal non- to densely welded vitric tuff overlain by a spherulite zone that grades upward through a lithophysae-dominated zone to a zone of pervasive devitrification, which, in turn, is overlain by a zone of vapor-phase crystallization and is capped by partially welded vitric tuff. A three-dimensional welding and crystallization model has been developed based on integrating local and regional variations of 85 measured sections.Strong local variations are interpreted to be the result of threshold-governed welding and crystallization controlled by residence time above a critical temperature, which is achieved through differences in thickness and accumulation rate.     

The 322 ka Tiribí Tuff: stratigraphy, geochronology and mechanisms of deposition of the largest and most recent ignimbrite in the Valle Central, Costa Rica

The Tiribí Tuff covered much of the Valle Central of Costa Rica, currently the most densely populated area in the country (∼2.4 million inhabitants). Underlying the tuff, there is a related well-sorted pumice deposit, the Tibás Pumice Layer. Based on macroscopic characteristics of the rocks, we distinguish two main facies in the Tiribí Tuff in correlation to the differences in welding, devitrification, grain size, and abundance of pumice and lithic fragments. The Valle Central facies consists of an ignimbritic plateau of non-welded to welded deposits within the Valle Central basin and the Orotina facies is a gray to light-bluish gray, densely to partially welded rock, with yellowish and black pumice fragments cropping out mainly at the Grande de Tárcoles River Gorge and Orotina plain. This high-aspect ratio ignimbrite (1:920 or 1.1×10⁻³) covered an area of at least 820 km² with a long runout of 80 km and a minimum volume outflow of 25 km³ (15 km³ DRE). Geochemically, the tuff shows a wide range of compositions from basaltic-andesites to rhyolites, but trachyandesites are predominant. Replicate new ⁴⁰Ar/³⁹Ar age determinations indicate that widespread exposures of this tuff represent a single ignimbrite that was erupted 322±2 ka. The inferred source is the Barva Caldera, as interpreted from isopach and isopleth maps, contours of the ignimbrite top and geochemical correlation (∼10 km in diameter). The Tiribí Tuff caldera-forming eruption is interpreted as having evolved from a plinian eruption, during which the widespread basal pumice fall was deposited, followed by fountaining pyroclastic flows. In the SW part of the Valle Central, the ignimbrite flowed into a narrow canyon, which might have acted as a pseudo-barrier, reflecting the flow back towards the source and thus thickening the deposits that were filling the Valle Central depression. The variable welding patterns are interpreted to be a result of the lithostatic load and the influence of the content and size of lithic fragments.     

Flow directions in ash-flow tuffs: a comparison of geological and magnetic susceptibility measurements, Tshirege member (upper Bandelier Tuff), Valles caldera, New Mexico, USA

This study concludes that the elongation axis (K ₁) of the ellipsoid of anisotropic magnetic susceptibility (AMS) is a suitable proxy for flow axis in ashflow tuffs. 153 oriented samples (176 specimens) were studied from 18 sites in the 1.1 Ma Tshirege member of the Bandelier Tuff. These sites are distributed around the Valles caldera at distances of 5–25 km outside of the rim.K ₁ axes correlate well with postulated radial flow axes at 13 sites.K ₁ also agrees with measured geological flow indicators, mainly imbricated larger clasts, at 7 sites. At 2 of the 5 sites where significant disagreement is seen between theoretical radial flow directions and measuredK ₁ axes, theK ₁ axes correspond well with geological flow indicators, indicating that the divergence of flow from the predicted radial flow pattern is real. Two major topographic buttresses are suggested as the cause of flow divergence for the Tshirege ash flows: the San Pedro buttress northwest of the caldera, and the San Miguel buttress in the southeast. In situK ₁ axes plunge about 7° toward the source at two-thirds of the sites; therefore the plunge ofK ₁ is a plausible in situ indicator for thedirection of flow. Multiple flow zones in sections of several meters thickness indicate changes of flow direction that are both rapid and large during ash-flow emplacement. These observations raisre the question of how best to represent mean flow directions in ash-flow sheets: by eigenvector methods, by vector-sum methods, or by modes. A method for measuring imbrication of larger clasts using apparent dips in vertical joints is outlined. Imbrication, determined in this way at one-third of the sites, dips toward the source, i.e., up-flow. The minimum (K ₃) axis of the AMS ellipsoid correlates with the flow foliation rather than with the larger clast imbrication. The flow axes of ash flows correspond with theK ₁ axes, not with the declination ofK ₃ axes as suggested by some authors. Initial dip of the sampled ash flows is not large and does not affect the paleomagnetic remanence direction, which is reversed with a mean ofD=173.5°,I=-38.4°, ₉₅=3.4°N=18. This mean is not different at the 95% confidence level from that of earlier workers. The mean pole, at 098.0°E, 74.8°N,A ₉₅=3.3°,N=18, is about 15° far-sided relative to the expected time-averaged geomagnetic pole, suggesting a history of emplacement too short to adequately average secular variation.     

Flow directions in ash-flow tuffs: a comparison of geological and magnetic susceptibility measurements,Tshirege member (upper Bandelier Tuff), Valles caldera,New Mexico,USA

This study concludes that the elongation axis (K ₁) of the ellipsoid of anisotropic magnetic susceptibility (AMS) is a suitable proxy for flow axis in ashflow tuffs. 153 oriented samples (176 specimens) were studied from 18 sites in the 1.1 Ma Tshirege member of the Bandelier Tuff. These sites are distributed around the Valles caldera at distances of 5–25 km outside of the rim.K ₁ axes correlate well with postulated radial flow axes at 13 sites.K ₁ also agrees with measured geological flow indicators, mainly imbricated larger clasts, at 7 sites. At 2 of the 5 sites where significant disagreement is seen between theoretical radial flow directions and measuredK ₁ axes, theK ₁ axes correspond well with geological flow indicators, indicating that the divergence of flow from the predicted radial flow pattern is real. Two major topographic buttresses are suggested as the cause of flow divergence for the Tshirege ash flows: the San Pedro buttress northwest of the caldera, and the San Miguel buttress in the southeast. In situK ₁ axes plunge about 7° toward the source at two-thirds of the sites; therefore the plunge ofK ₁ is a plausible in situ indicator for thedirection of flow. Multiple flow zones in sections of several meters thickness indicate changes of flow direction that are both rapid and large during ash-flow emplacement. These observations raisre the question of how best to represent ‘mean’ flow directions in ash-flow sheets: by eigenvector methods, by vector-sum methods, or by modes. A method for measuring imbrication of larger clasts using apparent dips in vertical joints is outlined. Imbrication, determined in this way at one-third of the sites, dips toward the source, i.e., up-flow. The minimum (K ₃) axis of the AMS ellipsoid correlates with the flow foliation rather than with the larger clast imbrication. The flow axes of ash flows correspond with theK ₁ axes, not with the declination ofK ₃ axes as suggested by some authors. Initial dip of the sampled ash flows is not large and does not affect the paleomagnetic remanence direction, which is reversed with a mean ofD=173.5°,I=-38.4°, α₉₅=3.4°N=18. This mean is not different at the 95% confidence level from that of earlier workers. The mean pole, at 098.0°E, 74.8°N,A ₉₅=3.3°,N=18, is about 15° far-sided relative to the expected time-averaged geomagnetic pole, suggesting a history of emplacement too short to adequately average secular variation.     

The Oligocene Lund Tuff, Great Basin, USA: a very large volume monotonous intermediate

Unusual monotonous intermediate ignimbrites consist of phenocryst-rich dacite that occurs as very large volume (>1000 km³) deposits that lack systematic compositional zonation, comagmatic rhyolite precursors, and underlying plinian beds. They are distinct from countless, usually smaller volume, zoned rhyolite–dacite–andesite deposits that are conventionally believed to have erupted from magma chambers in which thermal and compositional gradients were established because of sidewall crystallization and associated convective fractionation. Despite their great volume, or because of it, monotonous intermediates have received little attention. Documentation of the stratigraphy, composition, and geologic setting of the Lund Tuff – one of four monotonous intermediate tuffs in the middle-Tertiary Great Basin ignimbrite province – provides insight into its unusual origin and, by implication, the origin of other similar monotonous intermediates.The Lund Tuff is a single cooling unit with normal magnetic polarity whose volume likely exceeded 3000 km³. It was emplaced 29.02±0.04 Ma in and around the coeval White Rock caldera which has an unextended north–south diameter of about 50 km. The tuff is monotonous in that its phenocryst assemblage is virtually uniform throughout the deposit: plagioclase>quartz≈hornblende>biotite>Fe–Ti oxides≈sanidine>titanite, zircon, and apatite. However, ratios of phenocrysts vary by as much as an order of magnitude in a manner consistent with progressive crystallization in the pre-eruption chamber. A significant range in whole-rock chemical composition (e.g., 63–71 wt% SiO₂) is poorly correlated with phenocryst abundance.These compositional attributes cannot have been caused wholly by winnowing of glass from phenocrysts during eruption, as has been suggested for the monotonous intermediate Fish Canyon Tuff. Pumice fragments are also crystal-rich, and chemically and mineralogically indistinguishable from bulk tuff. We postulate that convective mixing in a sill-like magma chamber precluded development of a zoned chamber with a rhyolitic top or of a zoned pyroclastic deposit. Chemical variations in the Lund Tuff are consistent with equilibrium crystallization of a parental dacitic magma followed by eruptive mixing of compositionally diverse crystals and high-silica rhyolite vitroclasts during evacuation and emplacement. This model contrasts with the more systematic withdrawal from a bottle-shaped chamber in which sidewall crystallization creates a marked vertical compositional gradient and a substantial volume of capping-evolved rhyolite magma. Eruption at exceptionally high discharge rates precluded development of an underlying plinian deposit.The generation of the monotonous intermediate Lund magma and others like it in the middle Tertiary of the western USA reflects an unusually high flux of mantle-derived mafic magma into unusually thick and warm crust above a subducting slab of oceanic lithosphere.     

Flooding and geomorphic impacts in a mountain torrent: Raise Beck,central Lake District,England

Raise Beck is a mountain torrent located in the central Lake District fells, northern England (drainage area of 1·27 km²). The torrent shows evidence of several major flood events, the most recent of which was in January 1995. This event caused a major channel avulsion at the fan apex diverting the main flood flow to the south, blocking the A591 trunk road and causing local flooding. The meteorological conditions associated with this event are described using local rainfall records and climatic data. Records show 164 mm of rainfall in the 24 hours preceding the flood. The peak flood discharge is reconstructed using palaeohydrological and rainfall–runoff methods, which provide discharge values of 27–74 m³ s^?1, and 4–6 m³ s^?1, respectively. The flood transported boulders with b‐axes up to 1400 mm. These results raise some important general questions about flood estimation in steep mountain catchments. The geomorphological impact of the event is evaluated by comparing aerial photographs from before and after the flood, along with direct field observations. Over the historical timescale the impact and occurrence of flooding is investigated using lichenometry, long‐term rainfall data, and documentary records. Two major historical floods events are identified in the middle of the nineteenth century. The deposits of the recent and historical flood events dominate the sedimentological evidence of flooding at Raise Beck, therefore the catchment is sensitive to high magnitude, low frequency events. Following the 1995 flood much of the lower catchment was channelized using rip‐rap bank protection, re‐establishing flow north towards Thirlmere. The likely success of this management strategy in containing future floods is considered, based on an analysis of channel capacities. It is concluded that the channelization scheme is only a short‐term solution, which would fail to contain the discharge of an event equivalent to the January 1995 flood. Copyright © 2002 John Wiley & Sons, Ltd.     

The role of regional-scale faults in controlling a trapdoor caldera,Coromandel Peninsula,New Zealand

Regional gravity data from an eroded Miocene to Pliocene volcanic arc exposed in the Coromandel Peninsula, New Zealand, delineate a circular − 26-mGal, 15-km-diameter gravity anomaly. This anomaly, which has steep gradients on its northern and western margins but shallow gradients elsewhere, correlates with relatively young volcanic and volcaniclastic rocks within a broad topographic depression. Gravity modelling, using an exponentially decreasing density contrast with depth profile, requires very low-density rocks (ca. 2280 kg m^− 3) in the near-surface to account for the observed anomaly, giving a total depth of ca. 2.8 km for these rocks. The northern and western margins of this body dip steeply inward at 70°, whereas the southern and eastern margins have shallow inward dips (20–30°). The western margin coincides with the regional-scale Mangakino Fault, but the northern margin, recognizable only in the geophysical data (and named here the Ohinemuri Fault), is partially buried under younger volcanic rocks. We interpret these deep and steeply bounded, low-density volcanics in terms of a trapdoor caldera, faulted on its northern and western margins, with its hinge on the southern and eastern margins. Epithermal deposits are spatially associated with the Mangakino and Ohinemuri Faults, suggesting that both structures may have influenced hydrothermal fluid flow. These deposits pre-date caldera fill, indicating that caldera development followed pre-existing regional faults. These results delineate the subsurface geometry of a trapdoor caldera and highlight the role of pre-existing, regional-scale faults in controlling such caldera location and collapse.     

The campanian ignimbrite: a major prehistoric eruption in the Neapolitan area (Italy)

A geological, chemical and petrographical study of the Campanian ignimbrite, a pyroclastic flow deposit erupted about 30,000 years ago on the Neapolitan area (Italy), is reported. The ignimbrite covered an area of at least 7,000 km²; it consists of a single flow unit, and the lateral variations in both pumice and lithic fragments indicate that the source was located in the Phlegraean Fields area. Textural features, areal distribution and its morphological constraints suggests that the eruption was of the type of highly expanded low-temperature pyroclastic cloud. The original composition was strongly modified by post-depositional chemical changes involving most of the major and trace elements. No primary differences in the composition of the magma have been recognized. The Campanian ignimbrite is a nearly saturated potassic trachyte, similar to many other trachytes of the Quaternary volcanic province of Campania. Its chemistry indicates an affinity with the so-called «low-K association» of the Roman volcanic province.     

Karymshina, a giant supervolcano caldera in Kamchatka: Boundaries, structure, volume of pyroclastics

Data on a caldera discovered in 2006 are reported. The caldera formed in southern Kamchatka during Eopleistocene time (1.2 to 1.5 Ma). The caldera boundaries have been reconstructed and its dimensions determined (approximately 15 × 25 km). An uplifted block has been identified in the northwestern part of the caldera, the block is considered to be the result of emplacement of viscous rhyolite magmas at a later time (approximately 0.5–0.8 Ma), that is, as a resurgent uplift. We have reconstructed the boundaries of a large lacustrine basin that formed in the caldera after the appearance of the resurgent uplift. Calculations are provided yielding the volume of the pyroclastics ejected during caldera generation. It is shown that the caldera-forming eruption was a major one in Kamchatka in regard to its volume of ejected material, and ranks as a major eruption worldwide. We examined the structural controls of present-day hydrothermal systems and mineral occurrences situated in the area of study to demonstrate their relations to the caldera and the resurgent uplift.     

Propagation, linkage, and interaction of caldera ring-faults: comparison between analogue experiments and caldera collapse at Miyakejima, Japan, in 2000

The formation of ring faults yields important implications for understanding the structural and dynamic evolution of collapse calderas and potentially associated ash-flow eruptions. Caldera collapse occurred in 2000 at Miyakejima Island (Japan) in response to a lateral intrusion. Based on geophysical data it is inferred that a set of caldera ring faults was propagating upward. To understand the kinematics of ring-fault propagation, linkage, and interaction, we describe new laboratory sand-box experiments that were analyzed through Digital Image Correlation (DIC) and post-processed using 2D strain analysis. The results help us gain a better understanding of the processes occurring during caldera subsidence at Miyakejima. We show that magma chamber evacuation induces strain localization at the lateral chamber margin in the form of a set of reverse faults that sequentially develops and propagates upwards. Then a set of normal faults initiates from tension fractures at the surface, propagating downwards to link with the reverse faults at depth. With increasing amounts of subsidence, interaction between the reverse- and normal-fault segments results in a deactivation of the reverse faults, while displacement becomes focused on the outer normal faults. Modeling results show that the area of faulting and collapse migrates successively outward, as peak displacement transfers from the inner ring faults to later developed outer ring faults. The final structural architecture of the faults bounding the subsiding piston-like block is hence a consequence of the amount of subsidence, in agreement with other caldera structures observed in nature. The experimental simulations provide an analogy to the observations and seismic records of caldera collapse at Miyakejima volcano, but are also applicable to caldera collapse in general.     

Hillslope–channel sediment transfer in a slope failure event: Wet Swine Gill,Lake District,northern England

This paper describes and analyses a hillslope–channel slope failure event that occurred at Wet Swine Gill, Lake District, northern England. This comprised a hillslope slide (180 m³, c. 203 ± 36 t), which coupled with the adjacent stream, resulting in a channelized debris flow and fluvial flood. The timing of the event is constrained between January and March 2002. The hillslope failure occurred in response to a rainfall/snowmelt trigger, on ground recently disturbed by a heather moorland fire and modified by artificial drainage. Slide and flow dynamics are estimated using reconstructed velocity and discharge values along the sediment transfer path. There is a rapid downstream reduction in both maximum velocity, from 9·8 to 1·3 m s^?1; and maximum discharge, ranging from 33·5 to 2·4 m³ s^?1. A volumetric sediment budget quantified a high degree of coupling between the hillslope and immediate channel (～92%: 167 m³), but virtually all of the sediment was retained in the first‐order tributary channel. Approximately 44% (81 m³) of the slide volume was retained in the run‐up deposit, and termination of the debris flow prior to the main river meant that the remainder did not discharge into the fluvial system downstream. These results suggest poor transmission of sediment to the main river at the time of the event, but importantly an increase in available material for post‐event sediment transfer processes within the small upland tributary. Copyright © 2007 John Wiley & Sons, Ltd.     

The Hasan Dagi stratovolcano (Central Anatolia, Turkey): evolution from calc-alkaline to alkaline magmatism in a collision zone

The Hasan Dagi volcano is one of the two large Plio-Quaternary volcanoes in Cappadocia (Central Anatolia, Turkey). Three stages of edifice construction have been identified for this volcano: Paleovolcano, Mesovolcano and Neovolcano. Most samples from Hasan Dagi volcano are calc-alkaline and define an almost complete trend from basaltic andesite to rhyolite. However, the more recent (Neovolcano) mafic samples are alkaline basalts. The mineralogical and geochemical characteristics of the oldest lavas (Keçikalesi (13 Ma) and Paleo-Hasan Dagi (7 Ma)) are significantly different from those of the younger lavas (Meso- and Neo-Hasan Dagi (<1 Ma)). Calcic plagioclase and pigeonite are typically observed in these older lavas. The Paleovolcano basalts are depleted in alkalis and display a tholeiitic tendency whereas the differentiated lavas are depleted in Na₂O but enriched in K₂O compared to younger lavas. There is an evolution through time towards higher TiO₂, Fe₂O₃^*, MgO, Na₂O and K₂O and lower Al₂O₃ and SiO₂ which is reflected in the basalt compositions. All the basalts display multi-element patterns typical of continental margin magmas with a significant enrichment in LILE (K, Rb, Ba and Th) and LREE and strong (Paleovolcano) to moderate (Meso- and Neovolcano) negative Nb, Zr and Ti anomalies. However, the younger basalts are the most enriched in incompatible elements, in agreement with their alkaline affinities and do not systematically display negative HFSE anomalies. REE data suggest an hydrous amphibole-bearing crystallization history for both Meso- and Neovolcano lavas. The distinction between the older and younger lavas is also apparent in trace element ratios such as Nb/Y, Ti/Y and Th/Y. These ratios indicate the role of a subducted component±crustal contamination in the genesis of the Hasan Dagi lavas, particularly for the oldest lavas (Keçikalesi and Paleo-Hasan Dagi). The decreasing influence of this component through time, over the last 6–7 m.y., has been accompanied by an increasing contribution of melt-enriched lithosphere. Although the range of variation of Sr, Nd and Pb isotopic ratios is small (0.70457–0.70515; 0.51262–0.51273; 18.80–18.94; 15.64–15.69; 38.87–39.10), it also reflects the evolution of the magma sources through time. Indeed, the youngest (Neovolcano) and most primitive basalts display significantly lower ⁸⁷Sr/⁸⁶Sr than the Paleo- and Mesovolcano basalts, whereas the Mesovolcano basalts display more radiogenic Pb than Paleovolcano samples. Magma mixing processes between initially heterogeneous and/or variably contaminated magmas may account for the genesis of the less differentiated and intermediate lavas (48–57% SiO₂). Meso- and Neovolcano differentiated lavas (60–68% SiO₂) are either derived from the analyzed basalts or from more primitive and more depleted magmas by fractional crystallization±some crustal contamination (AFC). Furthermore, the highly differentiated samples (72–75% SiO₂) are not strongly contaminated. The strong calc-alkaline character of Hasan Dagi lavas, in the absence of contemporaneous subduction, must reflect the heritage of the early subduction of the Afro–Arabian plate under the Eurasian plate. The evolution towards alkaline compositions through time is clearly related to the development of extensional tectonics in Central Anatolia in the Late Miocene.     

Mamaku Ignimbrite: a caldera-forming ignimbrite erupted from a compositionally zoned magma chamber in Taupo Volcanic Zone, New Zealand

Mamaku Ignimbrite was deposited during the formation of Rotorua Caldera, Taupo Volcanic Zone, New Zealand, 220–230 ka. Its outflow sheet forms a fan north, northwest and southwest of Rotorua, capping the Mamaku–Kaimai Plateau. Mamaku Ignimbrite can be divided into a partly phreatomagmatic basal sequence, and a main sequence which comprises lower, middle, and upper ignimbrite. The internal stratigraphy indicates that it was emplaced progressively from a pyroclastic density current of varying energy that became less particulate away from source. Gradational contacts between lower, middle, and upper ignimbrite are consistent with it being deposited during one eruptive event from the same source. Variations in lithic clast content and coexistence of different pumice types through the ignimbrite sequence indicate that caldera collapse occurred throughout the eruption, but particularly when middle Mamaku Ignimbrite was deposited and in the final stages of deposition of upper Mamaku Ignimbrite. Maximum lithic data and the location of lithic lag breccias in upper Mamaku Ignimbrite confirm Rotorua Caldera as the source. At least 120 m of geothermally altered intra-caldera Mamaku Ignimbrite occurs inside Rotorua Caldera. Pumice clasts in the Mamaku Ignimbrite are dacite to high-silica rhyolite and can be chemically divided into three types: high–silica rhyolite (type 1), rhyolite (type 2), and dacite (type 3). All are petrogenetically related and types 1 and 2 may be derived by up to 20% crystal fractionation from the type 3 dacite. All three types probably resided in a single, gradationally zoned magma chamber. Andesitic juvenile fragments are found only in upper Mamaku Ignimbrite and inferred to represent a discrete magma that was injected into the silicic chamber and is considered to have accumulated as a sill at the base of the magma chamber. The contrast in density between the andesitic and silicic magmas did not allow eruption of the andesitic fragments during the deposition of lower and middle Mamaku Ignimbrite. The advanced stage of caldera collapse, late in the main eruptive phase, created withdrawal dynamics that allowed andesitic magma to reach the surface as fragments within upper Mamaku Ignimbrite.     

Sources of chloride in hydrothermal fluids from the Valles caldera, New Mexico: a Cl study

The Valles caldera in New Mexico hosts a high-temperature geothermal system, which is manifested in a number of hot springs discharging in and around the caldera. In order to determine the fluid pathways and the origin of chloride in this system, we measured ³⁶Cl/Cl ratios in waters from high-temperature drill holes and from surface springs in this region. The waters fall into two general categories: recent meteoric water samples with low Cl⁻ concentrations (< 10 mg/L) and relatively high ³⁶Cl/Cl ratios [(300–1000) × 10⁻¹⁵]; and geothermal brines with high Cl⁻ concentrations (800–9400 mg/L) but low ³⁶Cl/Cl ratios [(11–26) × 10⁻¹⁵]. The ³⁶Cl/Cl ratios for meteoric waters are slightly higher than expected for this region, suggesting a small addition of anthropogenic ³⁶Cl. Because of low ³⁶Cl/Cl ratios and high Cl⁻ concentrations in the brines, chloride in these waters must be derived from subsurface sources. A comparison between the observed ³⁶Cl/Cl ratios in the brines and those calculated for potential source formations in this region indicates that the present host formations, mainly volcanic tuffs, cannot be major sources of chloride, and that formations at greater depth, such as the Paleozoic and Precambrian formations are more likely to be sources of chloride in the brines. The results suggest that brines are meteoric waters which penetrated into the basement where they derive chloride from leaching of basement rocks and/or from saline pore fluids trapped there, along with likely addition of chloride from Paleozoic strata. Although these fluids have since come to reside in the intracaldera volcanic sequence after convective upwelling, they do not derive much Cl⁻ from the volcanic strata; and residence times of fluids in the volcanics are < 100,000 years.     

The quaternary caldera volcano emuruangogolak,kenya rift,and the petrology of a bimodal ferrobasalt-pantelleritic trachyte association

Emuruangogolak is a Quaternary basalitrachyte volcano situated in the Suguta graben of the northern Kenva rift, and probably erupted last early in this century. Following the construction of an early trachytic shield volcano, two episodes of caldera collapse occurred. each preceded by explosive pvroclastic activity. Post-calelera volcanism consisted of alternating phases of basalt and trachyte eruption. The basic lavas are high-Ti ferrobasalts of a mildly alkaline ‘transitional’ composition and the trachytes are peralkaline and oversaturated. A distinct compositional bimodality exists and no rocks in the range 49–59°. SiO, have been found. Major and trace element analyses suggest that the trachytes are genetically related to the basalts. Associations of almost identical lavas occur in Ethiopia. Pantelleria and the Azores but with the presence of intermediate terms Fractional crystallization is the mechanism currently preferred to account for the origin of the trachytes. The ‘Daly gap’ may be a consequence of a crystallization process which limits the volume of intermediate magma available at any time. In addition, the physical properties and spatial distribution of the different magmas probably discriminate against the cruption of lavas of intermediate composition.     

The transition from calc-alkaline to potassic orogenic magmatism in the Aeolian Islands,Southern Italy

The Aeolian volcanic arc displays a wide range of magmatic products. Mafic lavas range from hypersthene normative calc-alkaline basalts to silica-undersaturated potassic absarokites, although the former are spatially and temporally dominant, consistent with the subduction-zone tectonic setting. In addition, intermediate and acidic members of the various fractionation series may be recognised. Large variations in trace element and isotope ratios accompany the rapid calc-alkaline to potassic transition, and it is argued that these may be largely explained in terms of subduction-zone mantle enrichment involving components derived from both basaltic ocean crust and subducted sediments. In addition, it seems that the mantle wedge itself was substantially heterogeneous prior to the onset of subduction zone processes. Not only are these subduction components similar to those proposed in a number of island arcs, but they also resemble those recognised in the ultra-potassic lavas of the Roman province, supporting recent subduction-related petrogenetic models of the Roman magmatism. Although subducted sediment plays an important role in the generation of some potassic magmatism, it is not uniquely responsible for K₂O-rich lavas, which are also produced without a large sediment contribution.     

Geology of a submarine volcanic caldera in the Tonga Arc: Dive results

A submersible dive conducted on Volcano #1 located near 21° 09′S–175° 45′W on the Tonga Arc showed that the volcanic edifice with a caldera floor area of 30 km² located at and 450 m deep (b.s.l.=below sea level) was constructed recently during episodic volcanism. The sequential volcanic events are recorded along a faulted terrain formed in response to the collapse of the caldera wall. The post-caldera events are marked by occasional eruptions that have built scoriaceous cones associated with low-temperature hydrothermal venting and localized small-scale collapse features. The stratigraphy of the caldera wall indicates that the volcano was built by explosive volcanism alternating with quieter eruptive events. The repeated, violent explosive events formed ≤ 20 m thick sequences composed of alternating fine-grained ash beds and sand- to boulder-sized pyroclastic layers. During quieter volcanic events, dykes and massive flows intruded and/or accompanied the eruption of the volcaniclastic deposits throughout the sections of the wall explored. Massive columnar-jointed flows consist of viscous, silica-rich lavas forming tabular and giant radial-jointed (GRJ) flows formed in large (> 8 m in diameter) conduits and extruded onto the sea floor. In addition, massive lava flows forming sill-like complexes were observed underneath and near the giant radial-jointed columnar flows. Also, an intermittent quiet type of eruption produced vesicular lava flows, which are interbedded within the pyroclastic layered deposits. The massive and vesicular lavas consist of andesites and dacites with Ca-depleted (pigeonite) and Ca-enriched (salite) pyroxene, and intermediate (andesine-labradorite) to calcic (bytownite) plagioclase. They are depleted in total alkalis (Na₂O + K₂O < 3%), K₂O (< 1%), Zr/Y (< 1.8), Nb/Zr (< 0.01) and light Rare Earth Elements. We interpret that these andesite–dacite series were erupted after undergoing crystal-liquid fractionation in a magma chamber located underneath the caldera floor.