Zoned Magma Chamber Research Articles

A history of violence: magma incubation, timing and tephra distribution of the Los Chocoyos supereruption (Atitlán Caldera, Guatemala)

ABSTRACTThe climactic Los Chocoyos (LCY) eruption from Atitlán caldera (Guatemala) is a key chronostratigraphic marker for the Quaternary period given the extensive distribution of its deposits that reached both the Pacific and Atlantic Oceans. Despite LCY tephra being an important marker horizon, a radioisotopic age for this eruption has remained elusive. Using zircon (U–Th)/He geochronology, we present the first radioisotopically determined eruption age for the LCY of 75 ± 2 ka. Additionally, the youngest zircon crystallization 238U–230Th rim ages in their respective samples constrain eruption age maxima for two other tephra units that erupted from Atitlán caldera, W‐Fall (130 +16/−14 ka) and I‐Fall eruptions (56 +8.2/−7.7 ka), which under‐ and overlie LCY tephra, respectively. Moreover, rim and interior zircon dating and glass chemistry suggest that before eruption silicic magma was stored for >80 kyr, with magma accumulation peaking within ca. 35 kyr before the LCY eruption during which the system may have developed into a vertically zoned magma chamber. Based on an updated distribution of LCY pyroclastic deposits, a new conservatively estimated volume of ~1220 ± 150 km3 is obtained (volcanic explosivity index VEI > 8), which confirms the LCY eruption as the first‐ever recognized supereruption in Central America.

Implications of Hf Isotopes for the Evolution of the Mantle Source of Magmas Associated with the Giant El Teniente Cu-Mo Megabreccia Deposit, Central Chile

We have determined the Hf isotopic compositions of 12 samples associated with the giant El Teniente Cu-Mo megabreccia deposit, central Chile. The samples range in age from ≥8.9 to 2.3 Ma and provide information about the temporal evolution of their magmatic sources from the Late Miocene to Pliocene. Together with previously published data, the new analysis indicates a temporal decrease of 10 εHf(t) units, from +11.6 down to +1.6, in the 12.7 m.y. from 15 to 2.3 Ma. These variations imply increasing incorporation of continental crust through time in the magmas that formed these rocks. The fact that the samples include mantle-derived olivine basalts and olivine lamprophyres suggests that these continental components were incorporated into their mantle source, and not by intra-crustal contamination (MASH). We attribute the increase, between the Middle Miocene and Pliocene, of crustal components in the subarc mantle source below El Teniente to be due to increased rates of subduction erosion and transport of crust into the mantle. The deposit formed above a large, long-lived, vertically zoned magma chamber that developed due to compressive deformation and persisted between ~7 to 4.6 Ma. Progressively more hydrous mantle-derived mafic magmas feed this chamber from below, providing heat, H2O, S and metals, but no unique “fertile” Cu-rich magma was involved in the formation of the deposit. As the volume of these mantle-derived magmas decreased from the Late Miocene into the Pliocene, the chamber crystallized and solidified, producing felsic plutons and large metal-rich magmatic-hydrothermal breccias that emplaced Cu and S into the older (≥8.9 Ma) mafic host rocks of this megabreccia deposit.

The caldera-forming eruption of the quaternary Payún Matrú volcano, Andean back-arc of the southern volcanic zone

The Payún Matrú caldera is a Quaternary Andean back-arc caldera (36° 25′ S, 69° 12′ 30” W), and represents the main volcanic edifice of the homonym Payún Matrú volcanic field, located within the Payenia Basaltic province. The Payún Matrú volcanic field is also characterized by the Payún Liso volcano, and 220 monogenetic basaltic cones and associated lava flows. The Pleistocene – Holocene Payún Matrú caldera presents a long lived pre-caldera stage, which built a multi-vent and shield-like volcanic edifice, mainly trachytic in composition. The syn-caldera stage is represented by the trachytic Portezuelo Ignimbrite (the caldera-forming eruption deposits) constrained between 148 and 82 ka, and the resulting caldera depression of 8.5 km in diameter. Afterwards the caldera event, volcanism continued mostly along the caldera rim.Stratigraphic sections of the Portezuelo Ignimbrite and its correlations are presented, in addition to mineral compositions and previous whole-rock chemical analyses. Only the extra-caldera deposits can be observed, while the intra-caldera ignimbrite is hidden by post-caldera volcanic products and unconsolidated modern sediments. Different facies within the ignimbrite were recognized, being most of them massive or eutaxitic and massive lapilli tuff facies, deposited by dense pyroclastic density currents. Fall deposits are found only proximally and restrained to the caldera margin. The absence of a widespread fall deposit suggests the lack of a sustained eruptive column at the onset of eruption. Massive tuff facies at the topmost deposits indicates the development of a co-ignimbritic plume as the pyroclastic density currents wanned.On the basis of color, size and shape, vesicularity and crystallinity, four distinct juvenile clasts types were defined: gray fiamme, black juvenile, dark gray and light gray pumice. Available chemical analysis of gray fiamme and black juvenile clasts indicate a similar trachytic composition, although their mineralogical composition presents some differences.Several eruptive units for the northern and southern outer slopes of the caldera were defined, on the basis of the juvenile types present and discontinuity surfaces. The onset of caldera collapse is not marked by the presence of a lithic breccia, but it is suggested by the appearance of dark gray and light gray pumice along with black juveniles in the topmost eruptive units, together with an increase in lithic content relative to previous units. The sequential appearance of different juvenile clasts types suggests a zoned magma chamber, in terms of crystal content and composition.

Oldest high-Ti basalt and magnesian crustal materials in feldspathic lunar meteorite Dhofar 1428

We conducted a thorough analysis of the feldspathic breccia meteorite Dhofar 1428 with the aim of better understanding the composition and evolution of lunar crust. This sample comprises a heterogeneous array of lithic fragments including magnesian and ferroan anorthositic granulites, mafic granulites/granulitic breccia, basalts, and different kinds of impact melt rocks. In which, a high-Ti basalt clast comprising large zoned pyroxene was observed. Based on equilibrium melt calculations of mineral zonations from this basalt, Mg-pyroxene cores were interpreted to be formed from a light rare earth element (LREE) enriched liquid, whereas the Fe-pyroxene rims grew from an LREE-depleted magma. We propose that LREE-depleted signature of Fe-pyroxene results from co-crystallization with apatite. The Mg-pyroxenes suggest that enriched liquids with higher REE contents and different REE patterns relative to KREEP existed within lunar interior. Oscillating Ti/Al ratios across pyroxene in this basalt may indicate several magma recharge events or crystal movement within a zoned magma chamber. This feature illustrates that magmas were derived from a variety of sources around the time of formation of this basalt. In situ U-Pb dating was conducted on apatite grains within this basalt, the excellent consistence between the U-Pb Concordia age (3941 ± 24 Ma, 2σ) and 207Pb/206Pb isochron age (3934 ± 24 Ma, 2σ) indicates the most likely crystallization age of this high-Ti basalt at ∼3940 Myr, making it one of the oldest high-Ti basalts formed on the Moon.Magnesian anorthositic granulites are mineralogically and geochemically similar to those trace element-poor magnesian anorthositic granulites in many lunar meteorites. These magnesian granulites cannot form from simple mixing of pristine Ferroan Anorthosite and lithologies from the Mg-Suite, and do not have any affinities with KREEP or the Procellarum KREEP Terrane, and they could be important components of farside highlands.

Field evidence, mineral chemical and geochemical constraints on mafic-felsic magma interactions in a vertically zoned magma chamber from the Chotanagpur Granite Gneiss Complex of Eastern India

The Nimchak granite pluton (NGP) of Chotanagpur Granite Gneiss Complex (CGGC), Eastern India, provides ample evidence of magma interaction in a plutonic regime for the first time in this part of the Indian shield. A number of outcrop level magmatic structures reported from many mafic-felsic mixing and mingling zones worldwide, such as synplutonic dykes, mafic magmatic enclaves and hybrid rocks extensively occur in our study domain. From field observations it appears that the Nimchak pluton was a vertically zoned magma chamber that was intruded by a number of mafic dykes during the whole crystallization history of the magma chamber leading to magma mixing and mingling scenario. The lower part of the pluton is occupied by coarse-grained granodiorite (64.84–66.61 wt.% SiO2), while the upper part is occupied by fine-grained granite (69.80–70.57 wt.% SiO2). Field relationships along with textural and geochemical signatures of the pluton suggest that it is a well-exposed felsic magma chamber that was zoned due to fractional crystallization. The intruding mafic magma interacted differently with the upper and lower granitoids. The lower granodiorite is characterized by mafic feeder dykes and larger mafic magmatic enclaves, whereas the enclaves occurring in the upper granite are comparatively smaller and the feeder dykes could not be traced here, except two late-stage mafic dykes. The mafic enclaves occurring in the upper granite show higher degrees of hybridization with respect to those occurring in the lower granite. Furthermore, enclaves are widely distributed in the upper granite, whereas enclaves in the lower granite occur adjacent to the main feeder dykes.Geochemical signatures confirm that the intermediate rocks occurring in the Nimchak pluton are mixing products formed due to the mixing of mafic and felsic magmas. A number of important physical properties of magmas like temperature, viscosity, glass transition temperature and fragility have been used in magma mixing models to evaluate the process of magma mixing. A geodynamic model of pluton construction and evolution is presented that shows episodic replenishments of mafic magma into the crystallizing felsic magma chamber from below. Data are consistent with a model whereby mafic magma ponded at the crust-mantle boundary and melted the overlying crust to form felsic (granitic) magma. The mafic magma episodically rose, injected and interacted with an overlying felsic magma chamber that was undergoing fractional crystallization forming hybrid intermediate rocks. The intrusion of mafic magma continued after complete solidification of the magma chamber as indicated by the presence of two late-stage mafic dykes.

Identification of a Kulshan caldera correlative tephra in the Palouse loess of Washington State, northwest USA

The Kulshan caldera formed at ∼1.15 Ma on the present-day site of Mt. Baker, Washington State, northwest USA and erupted a compositionally zoned (dacite-rhyolite) magma and a correlative eruptive, the Lake Tapps tephra. This tephra has previously been described, but only from the Puget Lowland of NW Washington. Here an occurrence of a Kulshan caldera correlative tephra is described from the Quaternary Palouse loess at the Washtucna site (WA-3). Site WA-3 is located in east-central Washington, ∼340 km southeast of the Kulshan caldera and ∼300 km east-southeast of the Lake Tapps occurrence in the Puget Lowland. Major- and trace element chemistry and location of the deposit at Washtucna within reversed polarity sediments indicates that it is not correlative with the Mesa Falls, Rockland, Bishop Ash, Lava Creek B or Huckleberry Ridge tephras. Instead the Washtucna deposit is related to the Lake Tapps tephra by fractional crystallisation, but is chemically distinct, a consequence of its eruption from a compositionally zoned magma chamber. The correlation of the Washtucna occurrence to the Kulshan caldera-forming eruption indicates that it had an eruptive volume exceeding 100 km3, and that its tephra could provide a valuable early-Pleistocene chronostratigraphic marker in the Pacific Northwest.

The origin of a zoned ignimbrite: Insights into the Campanian Ignimbrite magma chamber (Campi Flegrei, Italy)

Caldera-forming eruptions, during which large volumes of magma are explosively evacuated into the atmosphere from shallow crustal reservoirs, are one of the most hazardous natural events on Earth. The Campanian Ignimbrite (CI; Campi Flegrei, Italy) represents a classical example of such events, producing a voluminous pyroclastic sequence of trachytic to phonolitic magma that covered several thousands of squared kilometers in the south-central Italy around 39 ka ago. The CI deposits are known for their remarkable geochemical gradients, attributed to eruption from a vertically zoned magma chamber. We investigate the relationships between such chemical zoning and the crystallinity variations observed within the CI pyroclastic sequence by combining bulk-rock data with detailed analyses of crystals and matrix glass from well-characterized stratigraphic units. Using geothermometers and hygrometers specifically calibrated for alkaline magmas, we reconstruct the reservoir storage conditions, revealing the presence of gradients in temperature and magma water content. In particular, we observe a decrease in crystallinity and temperature and an increase in magma evolution and water content from the bottom to the top of the magma chamber. We interpret these features as the result of protracted fractional crystallization leading to the formation of a cumulate crystal mush at the base of the eruptible reservoir, from which highly evolved, crystal-poor, water-rich and relatively cold melts were separated. The extracted melts, forming a buoyant, easily eruptible cap at the top of the magma chamber, fed the initial phases of the eruption, until caldera collapse and eruption of the deeper more crystalline part of the system. This late-erupted, crystal-rich material represents remobilized portions of the cumulate crystal mush, partly melted following hotter recharge. Our interpretation is supported by: 1) the positive bulk-rock Eu anomalies and the high Ba and Sr contents observed in the crystal-rich units, implying feldspar accumulation; 2) the positive Eu anomalies in the matrix glass of the crystal-rich units, testifying to the presence of liquid derived from partial melting of low temperature mineral phases within the crystal mush (mostly feldspars); 3) the Ba and Sr-rich rims in the feldspars and positive Eu anomalies in clinopyroxene rims, suggesting late rim growth from a locally enriched melt following cumulate mush remelting and 4) the occurrence of An-rich plagioclase, relict from a more mafic recharge, which acted as a heat source. Our model reconciles many observations made over the years on zoned deposits of such high-magnitude explosive eruptions, and provides a framework to understand magma chamber processes leading up to cataclysmic events.

Generation and Rejuvenation of a Supervolcanic Magmatic System: a Case Study from Mangakino Volcanic Centre, New Zealand

The Kidnappers [∼1200 km3 dense rock equivalent (DRE)] and Rocky Hill (∼200 km3 DRE) caldera-forming events in the Taupo Volcanic Zone, New Zealand, were erupted in close succession from the Mangakino volcanic centre. They have identical radiometric ages at ∼1 Ma, yet erosion along the contact between the two deposits suggests that some years to decades separated the two eruptions. Field constraints and the similarities of crystal textures and compositions and glass chemistries of both eruption deposits demonstrate that they came from one overall magmatic system with a common crystal mush source. However, second-order variations in these parameters confirm that the Kidnappers and Rocky Hill deposits represent distinct events and are not the products of a single zoned magma chamber. The systematically zoned Kidnappers fall deposits provide evidence for the tapping of three discrete magma bodies, whereas the succeeding Kidnappers ignimbrite is compositionally more diverse. The transition from fall to flow deposition marks a change in the style of caldera collapse and the simultaneous evacuation of discrete but compositionally diverse melts, each of which underwent a distinct evolution and was held at slightly different P–T conditions prior to eruption. Contrasting plagioclase and orthopyroxene zonation patterns are present in pumices originating from three discrete magma bodies. Less evolved mafic melts interacted with the system, which mobilized portions of the final erupted melt through heating and volatile or chemical exchange in the mush. The two largest Kidnappers melt-dominant bodies were re-tapped in modified form, or re-established from their common mush source, prior to the Rocky Hill event. Rocky Hill pumices contain common, fluid-affected antecrystic crystal clots derived from chamber wall material. Amphibole compositions from each eruption reflect melt evolution processes and, in particular, the contemporaneous crystallization of biotite and breakdown of orthopyroxene. Plagioclase and orthopyroxene from Rocky Hill pumices share common zonation patterns with those from the two largest magma bodies in the Kidnappers. The rapid production of new melt-dominant bodies and the triggering of the Rocky Hill eruption reflect the ability of the magmatic system to rejuvenate on a geologically short timescale. The Mangakino centre did not follow a typical cycle of decreased activity after the supervolcanic Kidnappers event, instead producing a second caldera-forming eruption, within years to decades from the same system.

Formation of the trachybasalt–trachyte bimodal series of the Malo-Khamardaban volcanotectonic complex, southwestern Transbaikalia: Role of fractional crystallization and magma mixing

The Malo-Khamardaban volcanogenic association is one of the largest and typomorphic structures for the Jurassic–Cretaceous rift magmatism of the Western Transbaikalia. It is mainly made up of bimodal volcanogenic trachybasalt–trachyte series with subordinate amount of trachyandesites, trachyrhyodacites, and trachyrhyolites. Geochemical modeling gives grounds to conclude that the formation of trachytic melts was controlled by crystallization fractionation of trachybasaltic magma. Mineralogical–petrographic, mineralogical–geochemical, and geochemical features of trachyandesites and some syenite porphyries (subvolcanic analogues of trachytes) indicate their hybrid nature. The trachyandesites were formed by mixing of trachybasaltic magma with trachytic magmas (endohybridisation) and trachybasaltic magma with contaminated trachytic magma. Hybrid syenite porphyries were derived by contamination of trachytic magma with host rocks (Late Paleozoic granitoid rocks of the Zaza complex). It is suggested that the trachybasaltic magma prior to generation of daughter trachytic melt and contamination of the latter by host granitoids evolved in a zoned magma chamber, where differentiated salic melt of the apical part interacted with the framework rocks. The hybrid trachyandesite magma was produced by mixing of basaltic and trachytic magmas during their simultaneous injection in one magma chamber.

Evolution and Eruption Processes of a Highly Porphyritic Silicic Magma System: Petrology of the Historical Eruptive Stage of Hokkaido-Komagatake Volcano, Japan

After about 5000 years of dormancy, the Hokkaido-Komagatake volcano started its historical eruptive stage with a plinian eruption in AD 1640. During this eruption, a zoned magma chamber was formed with upper mafic and lower silicic layers. This study focuses on the temporal evolution and eruption processes of the magma chamber based on petrological investigations of subsequent eruptions in AD 1694, 1856, 1929 and 1942. In all the eruptive products the majority of the clasts are strongly porphyritic white pumice (silicic andesite), which was derived from the silicic end-member magma. These are associated with small amounts of mafic ejecta (mafic to intermediate andesite). Based on density calculations for the magmas, the layered structure of the zoned magma chamber is likely to have been stable throughout the historical eruptive stage. In each eruption since AD 1694, relatively mafic magma erupted initially followed by more silicic magma. In addition, although there is petrological evidence for a mafic magma injection into a resident silicic magma chamber prior to the AD 1640 eruption, all eruptions from AD 1694 onwards appear to have taken place without new mafic injections. This suggests that since the AD 1694 eruption magmas have been erupted sequentially from the upper mafic part of the magma chamber. The composition of scoria, the most mafic ejecta, changed from relatively mafic magma in the AD 1694 eruption to gray pumice in later eruptions, which represent a hybrid magma. This temporal change is interpreted to indicate that mafic magma mixed gradually with more silicic magmas during repeated eruptions. In addition, the volume of the mafic and hybrid magmas decreased because of repeated eruptions without new mafic magma injections, so that the AD 1942 eruption was dominated by the eruption of silicic magma. We consider the present chamber to be filled with highly porphyritic, viscous silicic magma.

On the anatomy of magma chamber and caldera collapse: The example of trachy-phonolitic explosive eruptions of the Roman Province (central Italy)

Textural and compositional features of pyroclastic products erupted during caldera-forming events often reveal the tapping of different portions of variably zoned magma chambers due to changing geometries of the conduit/vent systems. Here we report on ultrapotassic trachytic–phonolitic explosive eruptions of the Roman Province (central Italy), which show remarkable changes of textural features and glass compositions in the juvenile material, even if the bulk chemical composition is essentially constant. In each example, the lower eruption sequence contains whitish, crystal-poor (leucite-free), highly vesicular pumice, emplaced by early Plinian fallout and/or pyroclastic currents; upsection, the eruption sequence contains black, low porphyritic (sanidine+leucite-bearing), moderately vesicular, scoria or spatter, emplaced by major pyroclastic flows (red tuff with black scoria) and associated co-ignimbrite, coarse lithic-rich breccias. This suggests a shift from a central feeder conduit, tapping the central part of the magma chamber, to a ring fracture vent system, tapping the peripheral portions of the magma chamber, during caldera collapse. Key features of these evacuating magma chambers are the thermal and volatile concentration (Xvol) gradients that produce the observed textural and compositional spectrum of trachy-phonolitic rock types. In particular, the degrees of freedom during the crystallization of these ultrapotassic magmas are increased by the variation of the leucite stability field at different PH2O conditions. Both leucite-free and leucite-bearing differentiated ultrapotassic rock types can be produced in the course of individual eruptions, as a result of pre-eruptive conditions in the feeder magma, with no need to invoke different differentiation suites related to mantle source heterogeneities of parental magmas.

U-series zircon age constraints on the plumbing system and magma residence times of the Changbai volcano, China/North Korea border

article i nfo Th disequilibrium ages of zircons from lavas and ignimbrites of the pre-caldera (~4 ka), syn-caldera (1 ka), and post-caldera (~0.3 ka) events. The zircon isochron ages are 12.2 ± 1.1 ka (2σ) for the 4-ka comendite lava and 12.2 ± 1.7 ka for the 1-ka comendite ignimbrite. Zircons from the 0.3-ka trachyte ignimbrite exhibit 3 respective peaks at 2.6 ± 1.8 ka, 130 ± 10 ka and N230 ka. The in- distinguishable zircon ages for the 4-ka pre-caldera eruption of comendite lava and the 1-ka caldera-forming eruption of comendite pumice and ignimbrite suggest that the 4-ka lava provides an early sampling of a much larger magma body at depth and thus serves as a kind of petrologic early-warning signal. In addition, the 4-ka lava may represent the lowest-temperature magma in the roof zone of a thermally zoned magma chamber that usually escapes first. The distinctive multi-modal zircon age distributions of the 0.3-ka trachytic eruption, however, reveal that this post-caldera eruption tapped a different magma body and indicate that Changbai's magmatic plumbing system had changed after the 1-ka caldera-forming climactic eruption. Our results suggest very short zircon and magma residence times for the Changbai volcano (8 kyr for the 4-ka eruption, 11-12 kyr for the 1-ka eruption and 2.3 kyr for the 0.3-ka eruption).

Optimizing pre-eruptive temperature estimates in thermally and chemically zoned magma chambers

We present a method to minimize the error of temperature estimate when multiple discrete populations of glass and clinopyroxene occur in a single heterogeneous eruptive unit. As test data we have used ~1100 clinopyroxene–melt pairs from isothermal and thermal gradient experiments. These latter are characterized by the crystallization of multiple modes of clinopyroxene as frequently documented for chemically and thermally zoned magma chambers. Equilibrium clinopyroxene–melt pairs are identified through the difference between predicted and measured components in clinopyroxene. The use of these equilibrium compositions as input data for one of the most recent clinopyroxene-based thermometers demonstrates that the error of temperature estimate is minimized and approaches to the calibration error of the thermometric model. To emphasize the paramount importance of this method for predicting the crystallization temperature of heterogeneous magmas, we have tested for equilibrium ~480 and ~150 clinopyroxene–melt pairs from the explosive eruptions of the Sabatini Volcanic District (Latium Region, Central Italy) and the Campi Flegrei Volcanic Field (Campanian Region, Southern Italy), respectively. These eruptions were fed by zoned magma chambers, as indicated by the occurrence of multiple modes of clinopyroxene in the eruptive units. Results from calculations demonstrate that clinopyroxene–melt pairs in equilibrium at the time of eruption are effectively captured by our method and, consequently, the error of temperature estimate is significantly reduced.

Stratigraphy, geochemistry and geochronology of a Quaternary pyroclastic sequence of the Chichontepec volcano, El Salvador

In this paper we present sedimentological, chemical and geochronological data on a pyroclasticsequence deposited at Chichontepeque Volcano (El Salvador) during a caldera-forming Plinian phase.Granulometric analyses and width distribution patterns of airfall pyroclastics showed that these are the result ofat least two eruptions, the oldest being the more complete. The Basal Fall, which marks the beginning of thePlinian eruption, shows the typical features of Plinian air-fall deposits.Chemical trends mirror roughly a downward tapping from a zoned magma chamber; inside each fall unit crys-tal fractionation trends can be recognized. 39Ar/40Ar dating of lavas underlying and overlying the pyroclasticsequence show that the Plinian phase occurred between 0.59 ±0.09 and 0.55±0.04 Ma before present.

The pre-eruption conditions for explosive eruptions at Merapi volcano as revealed by crystal texture and mineralogy

We analyze the textures and mineralogy of Merapi tephra generated during explosive VEI 3–4 eruptions over the past 2000years, and compare these data with those observed for Merapi dome and flow lavas. We find that the Merapi pumiceous tephra and lava textures differ significantly with respect to small-size crystal populations, but that phenocryst textures are generally similar. A similar initial phase of crystallization is indicated for tephras and lavas in mid-crustal (>10km depth) reservoirs. Subsequent textural differences are mainly affected by ascent rate and degassing during ascent, and, for dome lavas, with temporary storage in shallower reservoirs. These differences also correspond to different eruptive styles. Our analyses include study of pumices, lava, and breadcrust-bomb samples, including samples from some of the most recent explosive episodes prior to the 2010 eruption. Textural analyses of youthful breadcrust bomb samples yield insights on one type of transition between effusive and explosive eruptive styles, involving pressure build-up under a degassed crystalline shallow-conduit plug. In general, comparison of the crystal size distributions and calculated residence times among the effusive and explosive eruptive styles suggests that the two main magma-product types resided for similar lengths of time in a mid-crustal reservoir, before ascending toward the surface and either erupting explosively (tephra), or stagnating in a shallow magma chamber prior to extrusion (lava). The interpretation is supported by the occurrences of amphibole in pristine condition in tephra and in altered state in lava. Finally the 1872 and 2010 explosive eruptions are examined and compared with others over the past three millennia. The 2010 bulk-rock compositions overlap with products of other major explosive Merapi eruptions, such as the Ngrangkah, Tegalsruni, Temusari, and Kepuharjo tephras. The 2010 products show a gradational late-stage mafic enrichment, dissimilar to 1872 which showed little (<1%) compositional variation, but similar to the Ngrangkah, Tegalsruni, and Temusari tephras. Such variation implies conventional withdrawal from a zoned magma chamber, with the more-evolved products being erupted first.

Formation of a Compositionally Reverse Zoned Magma Chamber: Petrology of the ad 1640 and 1694 Eruptions of Hokkaido-Komagatake Volcano, Japan

that a zoned magma body was formed during the AD 1640 eruption and that the two contrasting magmas mixed until the AD 1694 eruption. The chamber was compositionally reverse zoned and consisted of upper mafic and lower silicic zones with hybrid layers between them; the silicic magma was strongly porphyritic and thus denser than the aphyric mafic magma. The sequence of the AD 1694 eruption can be explained by sequential ejection from the upper part of the zoned chamber without a new injection of mafic magma. This is consistent with the view that the AD 1694 banded pumice was the product of mingling between the silicic and hybrid magmas and that mafic magma had not interacted with the silicic magma.

Rhyolites associated to Ethiopian CFB: Clues for initial rifting at the Afar plume axis

A comprehensive tectono-magmatic model based on new geochemical and field data is discussed in order to highlight the significance of the high-TiO 2 bimodal picrite basalt/rhyolite association in the north-eastern sector of the Ethiopian Plateau, which is considered to be the axial zone of the 30 Ma Continental Flood Basalt activity related to the Afar plume (Beccaluva et al., 2009). In this area the volcanic sequence consists of approximately 1700 m of high TiO 2 (4–6.5%) picrite basalts, covered by rhyolitic ignimbrites and lavas, with an average thickness of 300 m, which discontinuously extend over an area of nearly 13,500 km 2 (ca. 3600 km 3). Petrogenetic modelling, using rock and mineral chemical data and phase equilibria calculations by PELE and MELTS, indicates that: 1) picrite basalts could generate rhyolitic, sometimes peralkaline, residual melts with persistently high titanium contents (TiO 2 0.4–1.1%; Fluorine 0.2–0.3%; H 2O 2–3%; density ca. 2.4) corresponding to liquid fractions 9–16%; 2) closed system fractional crystallisation processes developed at 0.1–0.3 GPa pressure and 1390–750 °C temperature ranges, under QFM fO 2 conditions; 3) the highest crystallisation rate – involving 10–13% of Fe–Ti oxide removal – in the temperature range 1070–950 °C, represents a transitory (short-lived) fractionation stage, which results in the absence of erupted silica intermediate products (Daly gap). The eruption of low aspect ratio fluorine-rich rhyolitic ignimbrites and lavas capping the basic volcanics implies a rapid change from open- to closed-system tectono-magmatic conditions, which favoured the trapping of parental picrite basalts and their fractionation in upwardly zoned magma chambers. This evolution resulted from the onset of continental rifting, which was accompanied by normal faulting and block tilting, and the formation of shallow – N–S elongated – fissural chambers parallel to the future Afar Escarpment. The eruption of large volumes of rhyolitic magmas generated from underlying picrite basalts appears to be a common geological process acting in most plume-related LIP, which in turn is related to the inversion of the stress regime from regional extension lithosphere doming to localised continental rifting.

Origin of crystal-poor, differentiated magmas: insights from thermal gradient experiments

Crystal-poor, differentiated magmas are commonly erupted from shallow, thermally zoned magma chambers. In order to constrain the origin of these magmas, we have experimentally investigated crystallization, differentiation and crystal-melt separation in presence of a thermal gradient. Experiments have been designed taking advantage of the innate temperature gradient of the piston cylinder apparatus and carried out on a phonolitic system at 0.3 GPa and temperature ranging from 1,050 to 800°C. Crystallization degree and melt composition in experimental products vary as a function of the temperature gradient. In particular, melt composition differentiates from tephri-phonolite (starting material) to phonolite moving from the hotter, glassy zone (T ≤ 1,050°C) towards the cooler, heterogeneously crystallized zone (T ≤ 900°C) of the charge. The heterogeneously crystallized zone is made up of: (1) a crystal-rich, mushy region (crystallinity >30 vol%), (2) a rigid crystal framework (crystallinity ≤80 vol%) and (3) glassy belts of phonolitic glass at the top. Thermal gradient experiments picture crystallization, differentiation and crystal-melt separation processes occurring in a thermally zoned environment and reveal that relatively large volumes of crystal-poor melt (glassy belts) can originate as a consequence of the instability and collapse of the rigid crystal framework. Analogously, in thermally zoned magma chambers, the development and collapse of a solidification front may represent the controlling mechanism originating large volumes of crystal-poor, differentiated magmas.

Formation of a zoned magma chamber and its temporal evolution during the historic eruptive activity of Tarumai Volcano, Japan: Petrological implications for a long-term forecast of eruptive activity of an active volcano

Tarumai Volcano started a series of historic eruptive activity in AD 1667 after a dormancy of approximately 2000 years. The historic juvenile ejecta are mainly silicic andesite pumice associated with scoria, banded pumice and dome lava (SiO 2 = 55–63%), and are mixing products of two or three end-member magmas. In the initial largest plinian eruptions (AD 1667 period), simple mixing between two end-member magmas, silicic andesite (SA) and basalt, occurred. Large plinian eruptions (AD 1739 period) and the latest intermittent eruptions (AD 1804–AD 1909: latest period) also produced mixed magmas including both the SA, intermediate-SiO 2 andesite (IA), and basalt. Magmatic temperatures of the SA and IA magmas are 900–950 °C and approximately 1000 °C, respectively. The rocks of each period form linear trends in oxide–oxide diagrams, suggesting that mixing of two end-member magmas occurred in each period. Thus, it can be estimated that the IA magma was formed by mixing between the basaltic and SA magmas. These relations suggest that the injection of the basaltic magma into the SA magma occurred before the AD 1667 period, resulting in the formation of a zoned magma chamber. These two magmas were then withdrawn to mingle, during the AD 1667 period. After the period, the zoned chamber was composed of an upper SA magma and a lower mixed IA magma. Chemical compositions of the basaltic magma have been slightly different in each period since AD 1667. In addition, the phenocrystic minerals of the IA magma also have changed as a consequence of re-equilibration with the more mafic IA bulk magma compositions present from AD 1739 to AD 1909. Thus, distinct basaltic magma has repeatedly injected into the zoned chamber before each eruption. Although the scale of eruptions became much smaller after the plinian eruptions of AD 1739, the ratio of IA magma in the latest eruptive materials is much larger than that in AD 1739, suggesting that a larger amount of the lower part (IA magma) of the zoned magma chamber was effectively withdrawn in the latest period. However, withdrawal depth should be much shallower in the latest activity compared with the activities in AD 1667 and AD 1739 because the tapping depth from a chamber strongly depends on eruption rate. Thus, it is hypothesized that most of the SA magma in the upper part of the zoned magma chamber was consumed in the AD 1667 and AD 1739 periods. The temporal change of the magma system suggests that a large plinian eruption similar to what occurred in AD 1667 or AD 1739 is unlikely in the near future because most of the major magma (SA magma) has already erupted. Because it took considerable time (ca. 2000 yr) to accumulate a large volume of SA magma previously, we forecast that the volcano will enter a prolonged dormant period to accumulate voluminous SA magma for a next eruptive stage, or it will erupt only IA magma similar to the small eruptions that occurred during AD 1804–1909.

The exhumed Eocene Sultepec-Goleta Volcanic Center of southern Mexico: record of partial collapse and ignimbritic volcanism fed by wide pyroclastic dike complexes

The Sultepec-Goleta area in southern Mexico hosts one of the major Eocene silicic volcanic centers that make up the Cenozoic volcanic record of the north-central part of the Sierra Madre del Sur. This center is represented by a partially exhumed NNE-SSW trending ignimbritic field that covers an area of ∼400 km2 with a preserved volume of ∼200 km3. A remarkable feature of volcanic center is the exposure of pyroclastic dike complexes up to 1 km wide that extend almost continuously along the western and southern flanks of the volcanic field. The ignimbritic record comprises four units with different textural features as well as variable proportions of the main components (i.e. phenocrysts, pumice and lithic clasts). The most extensive unit is the basal Goleta Ignimbrite, of which the middle and upper members are richer in phenocrysts than the lower member. It has a thickness that ranges from ∼200 m in the north to at least 600 m in the south of the volcanic center. The overlying ignimbrite units (Cienega, Lobera and Potrero ignimbrites) are phenocryst-poor and are distributed in the northern sector of the volcanic field. The semi-curvilinear trend of the pyroclastic dikes along the southern flank of the Goleta Range, coupled with the greater thickness of the Goleta Ignimbrite encircled by the dikes, is indicative of the development of a partial-collapse caldera. Based on spatial relationships as well as analogies in the nature and abundances of the components, it has been recognized that the central and southern dike complexes fed the relatively phenocryst-rich Goleta Ignimbrite, whereas the phenocryst-poor units of the northern sector were extruded through the pyroclastic conduits distributed around Sultepec. We suggest that the stirring produced by the partial collapse of the magma chamber roof beneath the southern sector of the study area triggered the tapping of phenocryst-rich portions of the zoned magma chamber, resulting in an increase of phenocryst content in the upper members of the Goleta Ignimbrite. The presence of breccias and lithic-rich facies dominated by wallrock fragments at the borders of the pyroclastic dikes, including those of the northern extra-caldera sector, indicate that erosion of the conduit walls above the fragmentation level contributed to widening of the conduits and the construction of wide dike complexes.