Pre-caldera Lavas Research Articles

Remobilization and eruption of an upper crustal cumulate mush at the Singkut caldera (North Sumatra, Indonesia)

Understanding the conditions and timescales of storage and remobilization of magma bodies in the upper crust is key to interpreting the signals of potential reawakening at active volcanoes. In this paper, we provide the first volcanological and petrochronological characterization of the Singkut caldera, a young volcanic system located in northern Sumatra (Indonesia), in close proximity to Medan, one of the country's most popoluous cities. Singkut formed at ∼44 ka during a VEI 6 explosive eruption that deposited at least ∼26 km3 of tephra (dense rock equivalent, DRE). The cataclysmic eruption was preceded by >200 ky of mostly effusive pre-caldera activity and followed by effusive to mildly explosive post-caldera activity. The lavas and pumices have high crystallinity (up to 62% crystals) with andesitic to dacitic bulk-rock composition and rhyolitic glass. Mineral textures and matrix glass compositions indicate resorption of quartz, plagioclase and zircon. Zircon crystallization ages show a complete overlap with the eruption ages in pre-caldera lavas, while a time gap in zircon crystallization (>50 ky) is identified in the caldera-forming tuff and post-caldera lavas. Ti-in-zircon thermometry shows that the Singkut magma body experienced a temperature increase starting approximately upon eruption of the pre-caldera lavas (∼254 ka). Such thermal perturbation determined progressive melting of mineral phases in the cumulate crystal mush, caused the resorption of the youngest zircon domains before the caldera-forming eruption, and hampered zircon crystallization between the caldera-forming eruption and the effusion of the post-caldera lavas (∼16 ka). Our data demonstrate how cumulate melting processes played a key role in leading the volcanic system towards a caldera-forming eruption and controlled the transitions in eruptive style between the effusive phases and the explosive climactic eruption.

Basaltic volcanism of Medvezhia caldera on the Iturup Island of Kurile Isles: impact of regional tectonics on subduction magmatismMartynov Yu.А., Rybin А.V., Chibisova М.V., Ostapenko D.S.,Davydova M.Yu

ABSTRACT We present newly acquired data on the trace element composition and Sr–Nd–Pb-O isotopes of Pliocene – Holocene basalts of the Medvezhiy caldera located on the Iturup Island, Kuril arc. With its unique rhenium mineralization and high-Mg basalt volcanism, it is the largest caldera on the volcanic front of the Kurile Island arc. The calc-alkaline suite of rocks spanning the range from high-Mg basalts to high-Al basalts, has typical arc rock features, such as Nb-Ta depletion and LILE enrichment. The whole set of the caldera geochemical data is divided into two age units, pre- and postcaldera basalts. The precaldera basalts typify low-K, high-Al, basaltic suite of the frontal zones of «cold» subduction. PRIMACALC2 code and Arc Basalt Simulator version 3.1 forward model points to ~21.7% of partial melting of moderately depleted mantle peridotite with relatively high water content (~0.34 wt%) at the pressure of ~1.37 GPa. With similar isotope and trace element signatures, Late Pleistocene–Holocene basalts are marked by the high abundances of MgO (up to 11 wt %), but low Al2O3. The primary magmas were generated from depleted mantle with lower H2O content (~0.13%) and degree of partial melting (~9.5%), but at higher pressure (~1.77 GPa). The decompression scheme reproduces the fractionation trends of postcaldera magnesian basalts for most major elements but fails to explain the composition variations of the precaldera lavas. Their origin has been the matter of some complex processes in the shallow magma chamber. The abrupt change in the type of eruptions and magma composition of the caldera in Late Pleistocene–Holocene were related to a transformation of both the magma-feeding and melt generation systems as the result of geodynamic reorganization of the vast territory adjacent to the Sea of Japan and Kamchatka.

Magmato-tectonic links: Ignimbrite calderas, regional dike swarms, and the transition from arc to rift in the Southern Rocky Mountains

AbstractRadial and linear dike swarms in the eroded roots of volcanoes and along rift zones are sensitive structural indicators of conduit and eruption geometry that can record regional paleostress orientations. Compositionally diverse dikes and larger intrusions that radiate westward from the polycyclic Platoro caldera complex in the Southern Rocky Mountain volcanic field (southwestern United States) merge in structural trend, composition, and age with the enormous but little-studied Dulce swarm of trachybasaltic dikes that continue southwest and south for ∼125 km along the eastern margin of the Colorado Plateau from southern Colorado into northern New Mexico. Some Dulce dikes, though only 1–2 m thick, are traceable for 20 km. More than 200 dikes of the Platoro-Dulce swarm are depicted on regional maps, but only a few compositions and ages have been published previously, and relations to Platoro caldera have not been evaluated. Despite complications from deuteric alteration, bulk compositions of Platoro-Dulce dikes (105 new X-ray fluorescence and inductively coupled plasma mass spectrometry analyses) become more mafic and alkalic with distance from the caldera. Fifty-eight (58) new 40Ar/39Ar ages provide insight into the timing of dike emplacement in relation to evolution of Platoro caldera (source of six regional ignimbrites between 30.3 and 28.8 Ma). The majority of Dulce dikes were emplaced during a brief period (26.5–25.0 Ma) of postcaldera magmatism. Some northeast-trending dikes yield ages as old as 27.5 Ma, and the northernmost north-trending dikes have younger ages (20.1–18.6 Ma). In contrast to high-K lamprophyres farther west on the Colorado Plateau, the Dulce dikes are trachybasalts that contain only anhydrous phenocrysts (clinopyroxene, olivine). Dikes radial to Platoro caldera range from pyroxene- and hornblende-bearing andesite to sanidine dacite, mostly more silicic than trachybasalts of the Dulce swarm. Some distal andesite dikes have ages (31.2–30.4 Ma) similar to those of late precaldera lavas; ages of other proximal dikes (29.2–27.5 Ma) are akin to those of caldera-filling lavas and the oldest Dulce dikes. The largest radial dikes are dacites that have yet younger sanidine 40Ar/39Ar ages (26.5–26.4 Ma), similar to those of the main Dulce swarm.The older andesitic dikes and precaldera lavas record the inception of a long-lived upper-crustal magmatic locus at Platoro. This system peaked in magmatic output during ignimbrite eruptions but remained intermittently active for at least an additional 9 m.y. Platoro magmatism began to decline at ca. 26 Ma, concurrent with initial basaltic volcanism and regional extension along the Rio Grande rift, but no basalt is known to have erupted proximal to Platoro caldera prior to ca. 20 Ma, just as silicic activity terminated at this magmatic locus. The large numbers and lengths of the radial andesitic-dacitic dikes, in comparison to the absence of similar features at other calderas of the San Juan volcanic locus, may reflect location of the Platoro system peripheral to the main upper-crustal San Juan batholith recorded by gravity data, as well as its proximity to the axis of early rifting. Spatial, temporal, and genetic links between Platoro radial dikes and the linear Dulce swarm suggest that they represent an interconnected regional-scale magmatic suite related to prolonged assembly and solidification of an arc-related subcaldera batholith concurrently with a transition to regional extension. Emplacement of such widespread dikes during the late evolution of a subcaldera batholith could generate earthquakes and trigger dispersed small eruptions. Such events would constitute little-appreciated magmato-tectonic hazards near dormant calderas such as Valles, Long Valley, or Yellowstone (western USA).

A long term hydraulic stimulation study conducted at the Salak geothermal field

Awi 18-1 is an injection well drilled in the Cianten Caldera, near the western margin of the Salak (Awibengkok) reservoir in west Java, Indonesia. As the initial well injectivity was low, a long-term hydraulic stimulation was conducted to improve the permeability and establish a better connection to existing natural fractures. A geologic model of the area was built by integration of surface mapping, log and core data, and well performance information. The well penetrated pre- and post-caldera volcanics, the caldera ring fault intrusion and contact metamorphic zone, and marine sedimentary rocks. Permeability was found primarily near the caldera margin in pre-caldera lavas and ring fault intrusion along steeply dipping N to NW and NE trending fractures that were partially sealed by mineral precipitation. A geomechanical simulation model was developed from the geologic model to understand the injectivity evolution mechanisms and behaviours under different injection conditions, and also predict long term future injectivity performance. The model domain contains the completed interval of a deviated injection well (Awi 18-1) and covers the majority of microseismic events observed during the course of the stimulation. It simulates injectivity evolution using fully coupled processes of heat and mass transfers in poro-elastic media. The model was history-matched against the injection data (i.e. pressure) by calibrating rock mechanical and fracture properties. A few what-if scenarios under different operation conditions were simulated to evaluate the effects of injection pressure and temperature on the injectivity. These what-if scenario simulations indicate that both colder injection temperature and higher wellhead pressure lead to better injection efficiency. Also, the higher pressure had prolonged impacts while the colder temperature impacts are limited at the early time.

Temporal magma source changes at Gaua volcano, Vanuatu island arc

Gaua Island (also called Santa Maria), from the central part of the Vanuatu arc, consists of a large volcano marked by a caldera that hosts the active Mount Garet summit cone. In this paper, a geochemical study including Sr, Nd, Pb and Hf isotopic compositions of 25 lavas emitted since 1.8Ma is presented, with a focus on the volcanic products that preceded (old volcanics, main cone and pyroclastic series) and followed (Mount Garet) the caldera forming event.All lavas show an island arc signature with enrichment in LILE and depletion in HFSE. Post-caldera lavas define a medium-K calc-alkaline trend, whereas lavas from the former main cone have high-K calc-alkaline compositions. Compared to the pre-caldera volcanic suite, the Mount Garet lavas have similar Th/Nb (~1.5), 143Nd/144Nd (~0.51295) and 176Hf/177Hf (~0.28316) ratios, but higher Ba/La (~42 vs. ~27) and 87Sr/86Sr (0.70417 vs. 0.70405) ratios and lower Ce/Pb (~2.7 vs. ~4.6), La/Sm (~2.5 vs. ~4.0) and 206Pb/204Pb (18.105 vs. 18.176) ratios. High Th/Nb and low Nd and Hf isotopic ratios compared to N-MORB suggest the contribution of ~2% of subducted sediment melt to the mantle source of Gaua magmas. Most of the observed differences between pre- and post-caldera lavas can be accounted for by the involvement of at least two portions of the mantle wedge, metasomatized by different slab-derived aqueous fluids. In addition, the lower La/Sm (at a given 143Nd/144Nd) ratios of Mount Garet lavas suggest a higher degree of partial melting (~10–15%) compared to the pre-caldera lavas (~5%). The Santa Maria Pyroclastic Series (SMPS) eruption probably triggered the caldera collapse, in response to emptying of the magmatic chamber. This event may have allowed new access to the surface for a geochemically distinct batch of magma issued from a separate magma chamber, resulting in the birth and construction of Mount Garet within the caldera. As both magmatic suites were emitted over a very short time, the storage of their parental magmas beneath the volcano is still possible.

Major ash eruptions of Barren Island volcano (Andaman Sea) during the past 72 kyr: clues from a sediment core record

Barren Island (Andaman Sea) is the northernmost active volcano of the Indonesian Arc. To construct the eruptive history of this little studied volcano, we measured 14C dates of inorganic carbon in sediment beds, and Sr and Nd isotopic ratios of seven discrete ash layers, in a marine sediment core collected from 32 km southeast of the volcano. The study reveals that the volcano had seven major ash eruptions at ~70, 69, 61, 24, 19, 15, and 10 ka. The ash layers erupted from 70 ka through 19 ka have highly uniform Nd isotopic composition, and since the ~15 ka eruption to the present the isotopic composition has been highly variable. Between ~24 ka and ~10 ka, the volcano had large ash eruptions spaced at 4,500 year intervals. Isotopically correlating the precaldera lavas and ash exposed on the volcano to the uppermost ash layer in the core, we infer that the caldera of Barren Island volcano is younger than 10 ka.

Isotope geochemistry (Sr–Nd–Pb) and petrogenesis of leucite-bearing volcanic rocks from “Colli Albani” volcano, Roman Magmatic Province, Central Italy: inferences on volcano evolution and magma genesis

The “Colli Albani” composite volcano is made up of strongly silica-undersaturated leucite-bearing rocks. Magmas were erupted during three main periods, but a complex plumbing system dominated by regional tectonics channelled magmas into different reservoirs. The most alkali-rich magmas, restricted to the caldera-forming period (pre-caldera), are extremely enriched in incompatible trace elements and display more radiogenic Sr (87Sr/86Sr = 0.71057–0.71067), with slightly less radiogenic Pb with respect to those of the post-caldera period. Post-caldera volcanic activity was concentrated in three different volcanic environments: external to the caldera, along the caldera edge and within the caldera. The post-caldera magmas produced melilite- to leucitite-bearing, plagioclase-free leucitites. In contrast to the pre-caldera lavas, they are characterised by lower incompatible trace element abundances and less radiogenic Sr (87Sr/86Sr = 0.71006–0.71039). Magmas evolved through crystal fractionation plus minor crustal assimilation in a large magma chamber during the pre-caldera period. The multiple caldera collapses dissected and partially obliterated the early magma chamber. During the post-caldera stage, magmas were channelled through several pathways and multiple shallow-level magma reservoirs were established. A lithospheric mantle wedge previously depleted in the basaltic component and subsequently enriched by metasomatic slab-derived component is suggested as the mantle source of Colli Albani parental magmas. Two different parental magmas are recognised for the pre- and post-caldera stages. The differences may be related to the interplay between smaller degrees of melting for the pre-caldera magmas and more carbonate-rich recycled subducted lithologies in the post-caldera magmas.

Magma Evolution of the Sete Cidades Volcano, São Miguel, Azores

The Sete Cidades volcano (Sao Miguel, Azores) is situated at the eastern end of the ultraslow spreading Terceira rift axis. The volcano comprises several dominantly basaltic pre-caldera eruptions, a trachytic caldera-forming stage and a post-caldera stage consisting of alternating trachytic and basaltic eruptions. The post-caldera flank lavas are more primitive (> 5 wt % MgO) than the pre-caldera lavas, implying extended fractional crystallization and longer crustal residence times for the pre-caldera, shield-building lavas. Thermobarometric estimates show that the ascending alkali basaltic magmas stagnated and crystallized at the crust-mantle boundary (similar to 15 km depth), whereas the more evolved magmas mainly fractionated in the upper crust (similar to 3 km depth). The caldera-forming eruption was triggered by a basaltic injection into a shallow trachytic magma chamber. Lavas from all stages follow a single, continuous liquid line of descent from alkali basalt to trachyte, although slight differences in incompatible element (e.g. Ba/Nb, La/Nb) and Sr isotope ratios imply some heterogeneity of the mantle source. Major and trace element data suggest similar partial melting processes throughout the evolution of the volcano. Slight geochemical differences between post- and pre-caldera stage lavas from the Sete Cidades volcanic system indicate a variation in the mantle source composition with time. The oxygen fugacity increased from the pre-caldera to the post-caldera stage lavas, probably as a result of the assimilation of crustal rocks; this is supported by the presence of crustal xenoliths in the lavas of the flank vents. The lavas from the Sete Cidades volcano generally have low Sr isotope ratios; however, rocks from one post-caldera vent on the western flank indicate mixing with magmas resembling the lavas from the neighbouring Agua de Pau volcano, having higher Sr isotope ratios. The different magma sources at Sete Cidades and the adjacent Agua de Pau volcano imply that, despite their close proximity, there is only limited interaction between them.

Transition from mildly-tholeiitic to calc-alkaline suite: the case of Chichontepec volcanic centre, El Salvador, Central America

The Chichontepec volcano is a Plio-Pleistocene composite volcano that erupted lavas ranging from high-alumina basalts to dacites. It experienced a caldera-forming paroxysmal eruption during the early Pleistocene. Pre-caldera lavas are mildly tholeiitic and they evolved mainly by low pressure crystal fractionation, notwithstanding the fact that most mafic lavas (low-MgO high-alumina basalts) retain traces of polybaric evolution. Conversely, post-caldera lavas, which are mainly pyroxene andesites, are clearly calc-alkaline, having evolved by open-system crystal fractionation. Sr–Nd isotopic data and trace elements characteristics indicate that the same mantle source was involved in the petrogenesis of these series. Modelling the AFC process showed that it did not play any role in the petrogenesis of these rocks; a crystal fractionation model is considered to be more relevant. A slight variation in the fractionating assemblage could have caused the transition from an early mildly tholeiitic trend to a late calc-alkaline one. Mineralogical evidence, mass-balance calculations and elemental chemistry support this hypothesis, assuming that the greater amount of pyroxene on the liquidus is at the expense of plagioclase; this would have prevented the trend in iron enrichment.

Time scales of large volume silicic magma systems: Sr isotopic systematics of phenocrysts and glass from the Bishop Tuff, Long Valley, California

The initial Sr isotopic compositions of glass and mineral separates from the 0.74 Ma Bishop Tuff ashflow in eastern California were determined to investigate the time scales of magmatic processes in a large silicic system. It was found that there is substantial isotopic heterogeneity, both between eruptive units and between glass and phenocryst phases of individual units. The frist-erupted, lower temperature units generally have higher initial 87Sr/86Sr than later crupted, higher temperature units. Within each unit, feldspar phenocrysts have the lowest 87Sr/86Sr, associated glass has higher 87Sr/86Sr, and biotite phenocrysts have the highest 87Sr/86Sr. These isotopic differences were produced by processes in the magma chamber and not by post-eruptive alteration. Two samples were similar Fe−Ti oxide temperatures but from widely separated localities have nearly identifical Sr isotopic characteristics, indicating the existence of compositionally uniform layers of substantial volume within the chamber. Trace element data indicate that the feldspars crystallized from a liquid represented by the associated glass, and that the feldspar-glass pairs are not accidental. The rhyolitic liquids of the Bishoptuff magma chamber apparently experienced increasing 87Sr/86Sr at a rate too fast for feldspar phenocrysts to remain in isotopic equilibrium. The increasing 87Sr/86Sr is caused primarily by radioactive decay of 87Rb in the high-Rb/Sr liquids and not primarily by assimilation of radiogenic wall-rock material. A self-consistent model can be constructed to account for all of the isotopic data except for those on biotite phenocrysts. The time scale for evolution of the system is bounded on the high side at about 500 ky by observations made on precaldera lavas, and on the low side at approximately 300 ky by the time necessary to establish homogeneous layers in an actively differentiating chamber. The deduced time scale is consistent with model Rb−Sr ages, which date the differentiation of low temperature liquids from higher temperature liquids, and is compatible with the observed isotopic disequilibrium between feldspars and glass because of the low diffusivity of Sr in fieldspars ( 3x105 years. The silicic-magma production rate within the Bishop Tuff magma chamber is estimated to be 10-3km3/y. The growth rate of alkali feldspar is estimated to be about 10-14 cm/s based on the Sr isotopic difference between sanidine and glass of the lower Bishop Tuff. The biotite population is inferred to be partially (>50 ppm) xenocrystic, the xenocrysts being introduced to the chamber less than one year prior to eruption.

Comment on “Evidence for long residence times of rhyolitic magma in the Long Valley magmatic system: the isotopic record in precaldera lavas of Glass Mountain” by A.N. Halliday, G.A. Mahood, P. Holden, J.M. Metz, T.J. Dempster and J.P. Davidson

Comment and replies on the paper of A.N. Halliday, G.A. Mahood, P. Holden, J.M. Metz, T.J. Dempster and J.P. Davidson (1989) - Earth Planet. Sci. Lett. 94, 274-290

Reply to comment of R.S.J. Sparks, H.E. Huppert, and C.J.N. Wilson on “Evidence for long residence times of rhyolitic magma in the Long Valley magmatic system: the isotopic record in precaldera lavas of Glass Mountain”

Reply to comment of R.S.J. Sparks, H.E. Huppert, and C.J.N. Wilson on “Evidence for long residence times of rhyolitic magma in the Long Valley magmatic system: the isotopic record in precaldera lavas of Glass Mountain”

Second reply to comment of R.S.J. Sparks, H.E. Huppert and C.J.N. Wilson on “Evidence for long residence times of rhyolitic magma in the Long Valley magmatic system: the isotopic record in the precaldera lavas of Glass Mountain”

Second reply to comment of R.S.J. Sparks, H.E. Huppert and C.J.N. Wilson on “Evidence for long residence times of rhyolitic magma in the Long Valley magmatic system: the isotopic record in the precaldera lavas of Glass Mountain”

Evidence for long residence times of rhyolitic magma in the Long Valley magmatic system: the isotopic record in precaldera lavas of Glass Mountain

The high-silica rhyolite lavas of Glass Mountain, California, provide a detailed record of the evolution of the Long Valley magmatic system during the 1.4 m.y. prior to the catastrophic eruption of the voluminous zoned rhyolitic Bishop Tuff at 0.73 Ma. The older lavas of Glass Mountain are extremely evolved, and were erupted from 2.1 to 1.2 Ma with 87Sr/ 86Sr of 0.707–0.739 and ε Nd of −3 to −4, whereas the younger lavas are slightly less evolved and were erupted between 1.2 and 0.79 Ma with 87Sr/ 86Sr of 0.706–0.707 and ε Nd close to −1, essentially identical to the Sr and Nd isotopic compositions of the Bishop Tuff. Neither the older nor younger lavas display a relationship between the isotopic compositions of Sr and Nd. The Pb isotopic compositions are effectively uniform at 206Pb/ 204Pb= 19.13–19.17 . A group of older lavas that outcrop in a northwest-trending band near the topographic rim of the caldera collectively define an apparent Rb—Sr isochron age (t)of2.09 ± 0.06Ma with an intercept of 87Sr/ 86Sr= 0.7060 ± 3 , where analyzed older lavas outcropping outboard of this band define an isochron of 1.90 ± 0.02Ma( 87Sr/ 86Sr) t = 0.7063 ± 4) These isochron ages are identical to the K—Ar ages for oldest erupted rhyolites in the two regions, respectively. The younger lavas away from the caldera define an apparent Rb—Sr isochron age (t)of1.14 ± 0.08Ma with the same ( 87Sr/ 86Sr) t as the older lavas. The agreement between the 87Sr/ 86Sr ratios of the rhyolite magmas when emplaced in the Long Valley magmatic system (0.706) and that of basaltic lavas in the region (0.7059–0.7062) suggests that the Sr isotopic compositions of these Sr-poor rhyolites may be totally dominated by Sr from mantle-derived components. K—Ar ages and stratigraphic relations for both the older and younger lavas seem to rule out the possibility that the Rb—Sr isochron ages record the times of eruption. The very low Sr concentrations of all the lavas (down to 0.1 ppm) cannot be produced simply by crustal fusion; the extremely high and variable Rb/Sr ratios (up to 2000) and low Sr concentrations must be due to extensive fractional crystallization. We interpret the isochrons as dating this process. To produce the observed volume of highly evolved rhyolite requires differentiation of a sizeable body of magma; thus the older lavas may have been derived from an evolved roof zone of a differentiated magma chamber that was present in the Long Valley region as early as 2.1 Ma. The similarity of the isotopic compositions of the younger Glass Mountain lavas and the Bishop Tuff, as well as the isochronous relationship of most of the younger lavas, indicate that the chamber containing the magma later to erupt as the younger lavas of Glass Mountain and the Bishop Tuff was already formed, isotopically homogenized, and had developed an evolved roof zone by about 1.1 Ma. There is no evidence for significant assimilation of crustal rocks or input of mantle-derived magma into the upper reaches of the system after this time. The preservation of isochronous relationships between lavas (and their constituent glasses) implies that most of the crystallization of the magma (i.e., that which affected Rb/Sr ratios) took place in discrete dateable events. The fact that the three isochrons have the same ( 87Sr/ 86Sr) t requires that the magmas did not exist for long as highly evolved liquids prior to the differentiation events that established the range of Rb/Sr ratios and the isochrons, as such magmas would increase in 87Sr/ 86Sr by 0.0002–0.005 per 100,000 years (depending on Rb/Sr ratio) due to in-situ decay. Comparing the isochron ages of the Glass Mountain lavas with times of their eruption yields residence times of silicic magma as great as 0.7 m.y. To prevent the rhyolitic magma from cooling and crystallizing during these intervals, heat must have been supplied to the system, presumably in the form of new additions of magma. This magma must not have reached the upper regions of the magma reservoir; otherwise the isochrons would have been destroyed by mixing in these Sr-poor rocks. Our interpretation of the isotopic data for Glass Mountain is at variance with recent models that invoke repeated melting of the deep crust rather than episodic tapping of magma chambers as mechanisms for producing repeated rhyolitic eruptions spanning periods of 10 5–10 6 years. The evidence for Long Valley is rather that major differentiation accompanies fresh inputs of magma into the system and that magma is maintained in stable zones in the chamber for hundreds of thousands of years.

Petrologic evolution of divergent peralkaline magmas from the Silent Canyon Caldera Complex, Southwestern Nevada Volcanic Field

The Silent Canyon volcanic center consists of a buried Miocene peralkaline caldera complex and outlying peralkaline lava domes. Its location has been corroborated by geophysical data and more than 50 drill holes. Two widespread ash flow sheets, the Tub Spring and overlying Grouse Canyon members of the Miocene Belted Range Tuff, were erupted from the caldera complex and have volumes of 60–100 km3 and 200 km3, respectively. Eruption of the ash flows was preceded by widespread extrusion of precaldera comendite domes and was followed by extrusion of postcollapse peralkaline lavas and tuffs within and outside the caldera complex. Lava flows and tuffs were also deposited between the two major ash flow sheets. Rocks of the Silent Canyon center vary significantly in silica content and peralkalinity. The most mafic rocks are precollapse and postcollapse trachytes (65–69% SiO2). Low‐silica comendites (69–73% SiO2) were erupted as the mafic upper part of the chemically zoned Grouse Canyon Member and as postcollapse lavas. The lower part of the Grouse Canyon Member and the underlying rhyolite of Split Ridge are moderately peralkaline comendite (PI is molar ratio Na + K/Al is 1.17–1.26). These comendites have major element characteristics and trace element enrichments approaching those of pantellerites. The Tub Spring Member, by contrast, is a weakly peralkaline chemically unzoned silicic comendite (75–76% SiO2) ash flow tuff. Weakly peralkaline silicic comendites (PI 1.0–1.1) are the most abundant precaldera lavas. Postcollapse lavas range from trachyte to silicic comendite; some have anomalous light rare earth element (LREE) enrichments. Silent Canyon rocks follow a common petrologic evolution from trachyte to low‐silica comendite; above 73% SiO2, compositions of the moderately peralkaline comendites diverge from those of the weakly peralkaline silicic comendites. These contrasting differentiation paths are shown in the behavior of Fe and other transition metals, Al, Na, K; the trace elements Ba, Zr, Nb; and probably F and Cl. Weakly peralkaline silicic comendites show a LREE/heavy REE crossover in early erupted/late erupted rocks; moderately peralkaline comendites are enriched in all REE. The development of divergent peralkaline magmas, toward both pantelleritic and weakly peralkaline compositions, is unusual in a single volcanic center.

Major, trace element and Sr isotopic composition of lavas from Vico volcano (Central Italy) and their evolution in an open system

Major, trace element and Sr isotopic compositions have been determined on 21 lava samples from Vico volcano, Roman Province, Central Italy. The rocks investigated range from leucite tephritic phonolites to leucite phonolites and trachytes. Trace element compositions are characterized by high enrichments of incompatible elements which display strong variations in rocks with a similar degree of evolution. Well-defined linear trends are observed between pairs of incompatible trace elements such as Th-Ta, Th-La, Th-Hf. A decrease of Large Ion Lithophile (LIL) elements abundance contemporaneously with the formation of a large central caldera is one of the most prominent characteristics of trace element distribution. Sr isotope ratios range from 0.71147 to 0.71037 in the pre-caldera lavas and decreases to values of 0.70974–0.70910 in the lavas erupted after the caldera collapse. Theoretical modelling of geochemical and Sr isotopic variations indicates that, while fractional crystallization was an important evolutionary process, AFC and mixing also played key roles during the evolution of Vico volcano. AFC appears to have dominated during the early stages of the volcanic history when evolved trachytes with the highest Sr isotope ratios were erupted. Mixing processes are particularly evident in volcanites emplaced during the late stages of Vico evolution. According to the model proposed, the evolution of potassic magmas emplaced in a shallow-level reservoir was dominated by crystal fractionation plus wall rock assimilation and mixing with ascending fresh mafic magma. This process generated a range of geochemical and isotopic compositions in the mafic magmas which evolved by both AFC and simple crystal liquid fractionation, producing evolved trachytes and phonolites with variable trace element and Sr isotopic compositions.

Origin of metaluminous and alkaline volcanic rocks of the Latir volcanic field, northern Rio Grande rift, New Mexico

Volcanic rocks of the Latir volcanic field evolved in an open system by crystal fractionation, magma mixing, and crustal assimilation. Early high-SiO2 rhyolites (28.5 Ma) fractionated from intermediate compositionmagmas that did not reach the surface. Most precaldera lavas have intermediate-compositions, from olivine basaltic-andesite (53% SiO2) to quartz latite (67% SiO2). The precaldera intermediate-composition lavas have anomalously high Ni and MgO contents and reversely zoned hornblende and augite phenocrysts, indicating mixing between primitive basalts and fractionated magmas. Isotopic data indicate that all of the intermediate-composition rocks studied contain large crustal components, although xenocrysts are found only in one unit. Inception of alkaline magmatism (alkalic dacite to high-SiO2 peralkaline rhyolite) correlates with, initiation of regional extension approximately 26 Ma ago. The Questa caldera formed 26.5 Ma ago upon eruption of the >500 km3 high-SiO2 peralkaline Amalia Tuff. Phenocryst compositions preserved in the cogenetic peralkaline granite suggest that the Amalia Tuff magma initially formed from a trace element-enriched, high-alkali metaluminous magma; isotopic data suggest that the parental magmas contain a large crustal component. Degassing of water- and halogen-rich alkali basalts may have provided sufficient volatile transport of alkalis and other elements into the overlying silicic magma chamber to drive the Amalia Tuff magma to peralkaline compositions. Trace element variations within the Amalia Tuff itself may be explained solely by 75% crystal fractionation of the observed phenocrysts. Crystal settling, however, is inconsistent with mineralogical variations in the tuff, and crystallization is thought to have occurred at a level below that tapped by the eruption. Spatially associated Miocene (15-11 Ma) lavas did not assimilate large amounts of crust or mix with primitive basaltic magmas. Both mixing and crustal assimilation processes appear to require development of relatively large magma chambers in the crust that are sustained by large basalt fluxes from the mantle. The lack of extensive crustal contamination and mixing in the Miocene lavas may be related to a decreased basalt flux or initiation of blockfaulting that prevented pooling of basaltic magma in the crust.

Post-caldera dacites from the Santorini volcanic complex, Aegean Sea, Greece: an example of the eruption of lavas of near-constant composition over a 2,200 year period

The post-caldera Kameni islands of the Santorini volcanic complex, Aegean Sea, Greece are entirely volcanic and were formed by eleven eruptions between 197 B.C. and 1950. Petrographic, mineral chemical and whole-rock major and trace element data are presented for samples of lava collected from the products of seven eruptive cycles which span the entire period of activity. The main phenocryst phases are plagioclase, clinopyroxene, orthopyroxene and titaniferous magnetite, which are weakly zoned (e.g. plagioclase — An55 to An42). The lavas are typical calc-alkaline dacites and show a restricted range of composition (from 64.1 to 68.4 wt. % SiO2). The phenocrysts were in equilibrium with the melts at temperatures of 960–1012 °C, pressures of 800–1500 bars and oxygen fugacities of 10−9.6-10−9.9 bars. The pre-eruptive water content of the magmas was 3–4 wt. % but since the lavas contain only 0.1–0.4 wt. % H2O, a considerable amount (about 0.01–0.015 km3) of water was lost prior to or during eruption. This indicates that the magmas rose to the surface gradually allowing the (largely) non-explosive loss of volatiles. The lavas were probably extruded initially from more or less cylindrical conduits which developed into fissures as the eruptions proceeded. The post-caldera lavas evolved from more mafic parental magmas (basalt-andesite) via fractional crystallization. The small range of compositional variation shown by these lavas can be explained in terms of near-equilibrium crystallization. Analyses of samples of lavas belonging to single eruption cycles and to individual flows indicate that the underlying magma chamber is compositionally zoned. The average composition of erupted magma has remained approximately constant since 1570 A.D. but that fact that the 197 B.C. magma was sligthly richer in SiO2 provides additional evidence that the magma chamber is compositionally zoned. Crystal settling has not affected the composition of the magma over a 2,200 year period of time which indicates that the melts do not behave as Newtonian fluids. Zonation was thus probably established prior to the 197 B.C. eruption though it is possible that it is developed and maintained by crystal-liquid differentiation processes other than crystal settling (e.g. boundary layer crystallization). The data indicate that there has been no significant cooling during 2,200 years; the maximum amount of cooling is <50 °C and is probably less than ∼30 °C. Two hypotheses are considered to explain the thermal and chemical buffering of the post-caldera magma chamber: (i) The magma chamber is large and heat losses due to conduction are largely compensated by latent heat supplied by thick, partially crystalline cumulate sequences. (ii) Periodic influx of hot mafic magma, which does not mix with the dacitic magma, inhibits cooling. The second alternative is favored because the post-caldera lavas differ geochemically from the pre-caldera lavas which signifies that a new batch of magma was formed and/or emplaced after the catastrophic eruption of 1390 B.C., and hence that mafic magmas may still be reaching upper crustal levels.

The mineralogy and petrology of compositionally zoned ash flow tuffs, and related silicic volcanic rocks, from the McDermitt Caldera Complex, Nevada‐Oregon

The McDermitt caldera complex, Nevada‐Oregon, is a composite collapse structure formed following eruption of three ash flow tuffs from a single compositionalIy zoned magma chamber. The major and early erupted portions of the tuffs are mildly peralkaline high‐silica comendite. Later erupted portions of two ash flow sheets are metaluminous low‐silica rhyolite. Associated with the caldera complex are precaldera dacite to rhyolite lava flows and postcaldera rhyolite domes and intrusives. Systematic variations in mineral compositions and whole‐rock chemistry throughout the entire silicic volcanic suite record the petrogenesis of highly silicic comendite through fractional crystallization of dacitic or rhyodacitic parental magma. Fractionation from daciterhyodacite to metaluminous rhyolite is characterized by initially decreasing light and middle rare earth elements; little change in Zr and Hf; and increasing U, Th, K, Rb, and Ba; controlled by removal of plagioclase, low and high‐Ca pyroxenes and significant (5%) amounts of apatite, magnetite, and ilmenite. Fractionation from rhyodacite‐low‐silica rhyolite to high‐silica comendite is characterized by increasing rare earth elements, U, Th, K, Rb, Zr, and Hf, with drastically decreasing Eu and Ba, controlled by removal of ternary‐alkali feldspar, quartz, ferroaugite‐ferrohedenbergite, fayalite and lesser ilmenite, magnetite, and apatite. Heavy rare earth element enrichment (HREE) from dacite to rhyolite is pronounced in ash flow tuffs and postcaldera volcanic rocks but not in precaldera lavas; this is possibly due to the variable presence of F as a HREE complexing agent during crystal fractionation in the caldera magma chamber. Fe‐Ti oxide compositions and phase equilibria indicate progressively decreasing magmatic temperature and oxygen fugacity: 865–940°C in dacite‐low‐silica rhyolite, with fo2 below the fayalite‐magnetite‐quartz buffer, and approximately 800°C in high‐silica comendite, with fo2 slightly above the wustite‐magnetite buffer. In the ash flow tuffs, observed patterns of vertical compositional zonation and phenocryst distributions reveal preeruptive processes important for the genesis of mildly peralkaline silicic magmas. Influx of heat is required to maintain crystallization of relatively mafic, high‐temperature mineral assemblages, notably with ternary feldspar, and prohibits further crystallization of derivative melts. Heat transfer occurred throughout a density‐stratified magma chamber via convection within compositionally discreet zones of low‐ and high‐silica rhyolite.

Structure of hakone caldera as revealed by drilling

Hakone caldera, now 8 by 12 km in diameter, was formed by collapse of a center of a volcano probably 2700 m high. The collapse took place at two separate periods each of which was followed by periods of deep denudation. The central part of the caldera has been covered by a thick pile of lavas of post-caldera cones and domes. For the purpose of finding thermal spring, drilling to depths of a few hundred to one thousand meters was carried out at various points within the caldera except for its central region. The study of the drill cores revealed that the average amount of subsidence at points 2 and 3 km away from the base of the present caldera wall is 600 m and 1200 m respectively, and probably more than 1800 m in the middle of the caldera. Within the caldera, the pre-caldera lavas and pyroclastic rocks are either lacking or much thinner than would be expected. It is concluded therefore that the present topographic depression of the caldera owed its origin to both subsidence and denudation. It is inferred that the subsidence took place along a complicated system of concentric faults combined with tilting of individual fault blocks toward the middle of the caldera. The magma reservoir into which the fault blocks sank probably had a shape of a cupola with a diameter comparable to or a little smaller than the diameter of the caldera.