A model of reverse differentiation at Dikii Greben' Volcano,Kamchatka: progressive basic magma vesiculation in a silicic magma chamber

Dikii Greben' Volcano is the largest modern volcano with silicic rocks in the Kurile-Kamchatka island arc. It consists of many domes and lava flows of rhyodacite, dacite and andesite which were erupted in a reverse differentiation sequence. Non-equilibrium phenocryst assemblages (quartz + Mg-rich olivine, An-rich + An-poor plagioclase etc.), abundance of chilled mafic pillows in the dacites and andesites, and linear variations of rock compositions in binary plots are considered as mineralogical, textural and geochemical evidence for mixing. Mafic pillows in volcanics have a lower density (because of high porosity) and contain the same non-equilibrium phenocryst assemblages as the host rocks. Their groundmass contains skeletal microlites of plagioclase and amphibole proving that the groundmass as well as the pillows themselves formed from a water-rich basaltic magma at depth. They are considered as supercooled, vesiculated floating drops of a hot hybrid layer in the magma chamber which formed after refilling. The lower density of the inclusions allows them to float in the host magma and to concentrate at the top of the chamber prior to eruption. Magma mingling was effected by mechanical disintegration of the inclusions in the host magma during eruption. The rhyodacitic and basic end-members of the mixing series cannot be linked by low-P fractionation though high-P, amphibole-rich fractionation is not excluded.     

Origin of Phenocrysts and Compositional Diversity in Pre-Mazama Rhyodacite Lavas, Crater Lake, Oregon

Phenocrysts in porphyritic volcanic rocks may originate in avariety of ways in addition to nucleation and growth in thematrix in which they are found. Porphyritic rhyodacite lavasthat underlie the eastern half of Mount Mazama, the High Cascadeandesite/dacite volcano that contains Crater Lake caldera, containevidence that bears on the general problem of phenocryst origin.Phenocrysts in these lavas apparently formed by crystallizationnear the margins of a magma chamber and were admixed into convectingmagma before eruption. About 20 km³ of pre-Mazama rhyodacite magma erupted during arelatively short period between400 and 500 ka; exposed pre-Mazamadacites are older and less voluminous. The rhyodacites formedas many as 40 lava domes and flows that can be assigned to threeeruptive groups on the basis of composition and phenocryst content.Phenocryst abundance decreases (from 32 to 8 vol.%) and SiO₂content increases (from 68 to 73 wt.%) in the apparent orderof eruption. Phenocrysts (plagioclase, orthopyroxene, augite,and Fe-Ti oxides) are commonly fragmental or form polycrystallineaggregates with interstitial glass. Discrete phenocrysts withcomplete euhedral outlines are rare except for small elongatedcrystals. The abundance of discrete phenocrysts increases withthat of aggregates. The grain-size of minerals in the aggregatescovers the range of discrete phenocrysts (0.2–4.2 mm).Rim compositions of phenocrysts and the range of chemical zoningare almost uniform among the three rhyodacite groups, regardlessof whether crystals are discrete or in aggregates. However,a small fraction of phenocrysts, especially small elongatedcrystals, have different compositions: plagioclase with Fe-richcores and augite with Wo-poor cores, both of which are characteristicof crystals in undercooled andesite enclaves in the rhyodacites.The majority of phenocrysts were derived by disintegration ofpolycrystalline aggregates; rare, small phenocrysts crystallizedin andesitic magma similar to that represented by the andesiteenclaves. The modal and chemical compositions of the rhyodacites can beexplained by different degrees of admixing of crystals, representedby the aggregates, into magma having 4 vol.% ‘true’phenocrysts, mainly plagioclase. The aggregates may be partsof the rind formed by in situ crystallization near the walland roof of the magma chamber. The rind was disrupted duringor just before eruption, and pieces were variably disaggregatedand incorporated into erupting magma. The amount of rind incorporateddeclined during the sequence of eruptions. Owing to vesiculationof interstitial liquid and shearing during flow, crystals inthe aggregates were separated and became phenocrysts. Pre-Mazamarhyodacite was erupted dominantly as lava, as opposed to thecompositionally similar rhyodacite pumice of the Holocene caldera-formingeruption of Mount Mazama, apparently because its source chamberwas crystallizing inward rather than actively growing.     

Petrology of the prehistoric lavas and dyke of the Barren Island,Andaman Sea,Indian Ocean

Although Barren Island (Andaman Sea, Indian Ocean) witnessed several volcanic eruptions during historic times, the eruptions that led to the formation of this volcanic island occurred mainly during prehistoric times. It is still active and currently in the fumarolic stage. Its volcanic evolution appears to be characterized by a constructive phase with the piling up of lava flows and scoria deposits and Strombolian activities, followed by a sudden collapse of the main cone. Deposits of a possible caldera-forming eruption were not recognized earlier. After a period of peri-calderic hydromagmatic activity, whose deposits presently mantle inner and outer caldera walls, a new phase of intracalderic Vulcanian activities took place. A prominent dyke in the SE inner side of the caldera wall was recognized. Petrographically the lava flows and dyke are similar but they differ in their chemical composition (viz., SiO₂, MgO, Ni, Cr) significantly. Similarity in major, minor and trace element composition (viz., K/La, K/Nb, K/Rb, K/Ti ratios) of these rocks together with Chondrite normalized trace element (Rb, Ba, Sr, P, Zr, Ti and Nb) and REE (La, Ce, Nd and Y) patterns of the Barren Island prehistoric lava flows and dyke and low-K lavas of Sunda Arc indicates that Barren Island must have evolved from a source similar to that of Sunda Arc lavas during the Quaternary Period.     

Geochemical constraints on magma processes in a peralkaline system: The Paisano volcano,west Texas

Major and trace element data for a sequence of peralkaline silicic lavas and pyroclastic flows, exposed in the caldera wall of the Paisano volcano, west Texas, document systematic fractional crystallization during magmatic evolution and an open system, magma mixing event in the upper parts of the sequence. Stratigraphically lowest flows are comendite and comenditic quartz trachyte lavas and ash flow tufts. Overlying these units is a trachyte with compositional, textural and mineralogical features indicating that it is the product of magma-mixing; similar flows occur in other parts of the volcano at the same stratigraphic level. This composite trachyte is considered to be a mixture of mugearitic or mafic trachytic magma, derived from a similar source region which yielded the earlier caldera wall flows. Trace element concentrations of the post-trachyte comenditic quartz trachyte lavas suggest they were erupted from a chamber whose magma was diluted by an influx of mugearitic or mafic trachytic magma during a magma mixing event.Rayleigh fractionation calculations show that the comendites and comenditic quartz trachytes can be derived from a parental mugearite magma by 88% to 93% fractionation of dominantly plagioclase and alkali feldspar, with lesser amounts of clinopyroxene, magnetite and apatite. Zircon was not a significant fractionating phase. The composition, mineralogy and depth of the source region(s) which generated these magmas cannot be constrained from the present data set.     

Potential Hazards of Eruptions around the Tianchi Caldera Lake, China

Since the eruption of the Tianchi volcano about 1000 years ago, there have been at least 3 to 5 eruptions of small to moderate size. In addition, hazardous avalanches, rock falls and debris flows have occurred during periods between eruptions. A future eruption of the Tianchi volcano is likely to involve explosive interaction between magma and the caldera lake. The volume of erupted magma is almost in a range of 0.1-0.5 km3. Tephra fallout may damage agriculture in a large area near the volcano. If only 1% of the lake water were ejected during an eruption and then precipitated over an area of 200 km2, the average rainfall would be 100 mm. Moreover, lahars are likely to occur as both tephra and water ejected from the caldera lake fall onto flanks of the volcano. Rocks avalanching into the caldera lake also would bring about grave hazards because seiches would be triggered and lake water with the volume equal to that of the landslide would spill out of the existing breach in the caldera and cause flooding     

Compositional gradients in large reservoirs of silicic magma as evidenced by ignimbrites versus Taylor Creek Rhyolite lava domes

The Taylor Creek Rhyolite of southwest New Mexico consists of 20 lava domes and flows that were emplaced during a period of a few thousand years or less in late Oligocene time. Including genetically associated pyroclastic deposits, which are about as voluminous as the lava domes and flows, the Taylor Creek Rhyolite represents roughly 100 km³ of magma erupted from vents distributed throughout an area of several hundred square kilometers. Major-element composition is metaluminous to weakly peraluminous high-silica rhyolite and is nearly constant throughout the lava field. The magma reservoir for the Taylor Creek Rhyolite was vertically zoned in trace elements, ⁸⁷Sr/⁸⁶Sr, and phenocryst abundance and size. Mean trace-element concentrations, ranges in concentrations, and element-pair correlations are similar to many subalkaline silicic ignimbrites. However, the polarity of the zonation was opposite that in reservoirs for ignimbrites, for most constituents. For example, compared to the Bishop Tuff, only ⁸⁷Sr/⁸⁶Sr and Sc increased upward in both reservoirs. Quite likely, a dominant but nonerupted volume of the magma reservoir for the Taylor Creek Rhyolite was zoned like that for the Bishop Tuff, whereas an erupted, few-hundred-meter-thick cap on the magma body was variably contaminated by roof rocks whose contribution to this part of the magma system moderated relatively extreme trace-element concentrations of uncontaminated Taylor Creek Rhyolite but did not change the sense of correlation for most element pairs. The contaminant probably was a Precambrian rock of broadly granitic composition and with very high ⁸⁷Sr/⁸⁶Sr. Although examples apparently are not yet reported in the literature, evidence for a similar thin contaminated cap on reservoirs for large-volume silicic ignimbrites may exist in the bottom few meters of ignimbrites or perhaps only in the pumice fallout that normally immediately precedes ignimbrite emplacement. ⁸⁷Sr/⁸⁶Sr in sanidine phenocrysts of the Taylor Creek Rhyolite is higher than that of their host whole rocks. Covariation of this isotope ratio with sanidine abundance and size indicates positive correlations for all three features with decreasing distance to the roof of the magma reservoir. The sanidine probably is more radiogenic than host whole rock because growing phenocrysts partly incorporated Sr from the first partial melt of roof rocks, which contained the highly radiogenic Sr of Precambrian biotite ± hornblende, whereas diffusion was too slow for sanidine to incorporate much of the Sr from subsequently produced less radiogenic partial melt of roof rocks, before eruption quenched the magma system. Disequilibrium between feldspar phenocrysts and host groundmass is fairly common for ignimbrites, and a process of contamination similar to that for the Taylor Creek Rhyolite may help explain some of these situations.     

Physical and Chemical Models of Zoned Silicic Magmas: The Loma Seca Tuff and Calabozos Caldera, Southern Andes

The late Pleistocene Calabozos ash-flow and caldera complexlies in central Chile in a section of the Andean cordillerathat is transitional between dominantly andesitic-to-rhyoliticvolcanism to the north and mafic andesitic and high-aluminabasaltic volcanism to the south. The Calabozos rocks range incomposition from basaltic andesite to rhyodacite and definea high-K calcalkalic suite. They contain 2–25% phenocrystsof plagioclase, clinopyroxene, orthopyroxene, Fe-Ti oxides,and apatite, ? minor biotite or amphibole. More than 1000km³ of rhyodacitic to dacitic magma erupted atthe Calabozos caldera complex as three major compositionallyzoned ash-flow sheets, Unit L (0?8 Ma), Unit V (0? 30Ma), andUnit S (0?15 Ma) of the Loma Seca Tuff. Phenocryst modes, trace-elementcontents, inferred magmatic volatile contents, and oxygen fugacitiesvary systematically with major-element composition in the tuffs.In the cases of Units V and S, it is possible to reconstructcompositional, thermal, and volatile gradients that existedin density-stratified magma chambers shortly prior to theireruption. The magma graded from crystal-poor, water-rich, andbiotite-bearing rhyodacite in the upper reaches of the chamberto more crystal-rich, water-poor, and amphibole-bearing daciteat deeper levels. Fe-Ti oxide equilibration temperatures are800 to 900?C for rhyodacite and 900 to 950?C for dacite. Magmathat erupted as Unit S was slightly hotter and more oxidizedthan magma that gave rise to Unit V. More mafic magmas wereassociated with the voluminous rhyodacitic to dacitic magmareservoir, as indicated by the presence of andesite and basalticandesite lava flows and by scoria inclusions in Unit V. The compositional suite from basaltic andesite to rhyodacitecan be simulated satisfactorily by crystal-fractionation calculationsbased on major-element phenocryst and rock compositions, andis consistent with modes of the Calabozos rocks. Rhyodacitesof Units V and S, however, are enriched in elements such asRb, Ba, and Zr relative to trace-element contents predictedby crystal-fractionation models. The enrichment can be achievedby assimilation of wall rock or a partial melt of the wall rock.The latter requires that the ratio of assimilation rates tocrystallization rates be between 0?1 and 0?3. Rates of assimilationversus crystallization were greater for Unit S than for UnitV, which is consistent with the lower Fe-Ti oxide temperaturesand less oxidized state of the latter. The Loma Seca Tuff is similar in bulk composition to sanidine-bearingash-flow sheets erupted on ‘mature’ continentalcrust, but it is mineralogically akin to ash-flow tuffs eruptedon ‘immature’ crust. The difference is attributed,in part, to the effect of the density of the crust on the rateof magma ascent at shallow levels. The ascent of large bodiesof silicic magma is slower in silicic (less dense) crust thanin mafic crust, causing the magmas to be erupted at a laterpoint in the crystallization history.     

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.     

Geological development of Heard Island,Central Kerguelen Plateau

We report ⁴⁰Ar–³⁹Ar laser step-heating age determinations on 15 stratigraphically controlled lava flows and intrusive rocks from Heard Island, Central Kerguelen Plateau (Indian Ocean). The island history began with uplift of pelagic limestone intruded by 22 Ma gabbro sills. Subaerial and wave erosion levelled the early island, producing an unconformity onto which pillow lavas, tuffaceous sediments and shallow-water, fossiliferous marine siltstone (Drygalski Formation) were deposited, beginning in late Miocene time. Two volcanic systems then formed in the late Quaternary. Big Ben dominates the larger southeast part of the island, while Mount Dixon occupies the northwest Laurens Peninsula. Feeder dykes and the early lava flows in both systems are 400–200 ka. Lava flows with evolved compositions (trachytes, trachyandesites) erupted 100–20 ka. Well-preserved parasitic cones exposed at low elevations are 15–10 ka and younger. Mawson Peak, near the summit of Big Ben, has erupted lava flows as recently as 2007. Heard Island, and nearby active McDonald Island, are subaerial features of a larger Neogene volcanic region of Central Kerguelen Plateau that includes several large sea knolls and recently identified submarine fields of small cones. This broadly distributed volcanic activity is linked to incubation of plume material at the base of the nearly stationary overlying Central Kerguelen Plateau.     

大别山金刚台国家地质公园的火山岩相构造

通过对大别山金刚台国家地质公园火山岩的岩性特征、火山岩相、相模式和火山机构的研究,认为火成岩斑状结构的含碎屑—碎屑状岩石可以有深成、浅成、侵出、喷溢或爆发等不同产状。以大斑、多斑、富碎斑和熔蚀斑为特征,发育涌动流纹、塑性集块和假堆积等构造,并与时差很小的大斑花岗岩基、大斑花岗斑岩脉呈侵入接触关系的非层状火山岩,预示其成因为深成端元属性。各种碎屑粒度越小和含量愈高,则指示更强的隐爆强度。提出景区内出露的火山岩均为潜火山岩,早期火山机构为偏心式脉动膨胀—隐爆潜火山岩体,晚期为爆破角砾岩筒。代表爆发相的火山碎屑岩埋藏在北部中—新生界盆地之中,在深剥蚀的大别山腹地不可能保存。     

Evolution of the Miocene Tejeda magmatic system,Gran Canaria,Canary Islands

This paper is a companion to Clark (1988; hereafter Part I) which described the evolution of the Tejeda Magmatic System (TMS), a Miocene caldera complex, Gran Canaria, Spain, based on geochronologic, paleomagnetic and field data. In this study, petrochemical data are used to corroborate the history out-lined in Part I. Geochemical discriminant analysis shows that whereas the Extra-Caldera (EC) Mogan/Fataga volcanics are separated by a composition gap, no composition gap exists within the Intra-Caldera (IC) sequence. IC ignimbrites change rapidly but progressively from pantellerites and comendites to comenditic trachytes and finally to trachytes in a 0.47 Ma time interval. Significantly, the lower pantelleritic part of the IC series is similar to the EC pantellerites (units B, C and D) as expected based on results from Part I. The appearance of a compositional gap in the EC sequence is the result of flows having been trapped within the caldera during the 0.47 Ma Mogan-Fataga transition interval. The transitional IC sequence may be geochemically modelled by mixing of Mogan comendites and Fataga trachytes. The mixing was most probably induced by the high discharge of magma from the compositionally-zoned Tejeda magma body. The rate of change in erupted composition is best explained by imagining a continuous influx of Fataga or parental Fataga magma into a chamber whose previous silicic component (Mogan composition) was no longer being replenished and that the two magmas did not convectively mix prior to eruption. Repose times between successive eruptions in the lower to middle Mogan (from P1/T1 to A) were of order 30 000 a; the upper Mogan pantellerites and comendites/comenditic trachytes (B to F?) erupted once every 125 000 years or so. The longer repose time for the upper units is consistent with their more differentiated character.     

A volcano under China's Great Wall

Cone sheets are rare in China but are found in the Houshihushan sub‐volcanic Ring Complex (HRC) near the town of Shanhaiguan, Hebei Province. The HRC rises above a narrow coastal plain as the Houshihushan mountain massif (Fig. 1 ) in a scenic region about 400 km ENE of Beijing by the Bohai Gulf, where the Great Wall reaches the sea. It has a discontinuous outer ring‐dyke of porphyritic quartz syenite, separated from an inner stock of alkaline granite by a discontinuous screen of pyroclastic rocks down‐faulted to the present erosion level by caldera subsidence. The pyroclastic rocks, which have similar petrographic and geochemical characteristics to intrusive rocks of the ring‐dyke, cone‐sheets and central stock, were erupted from the same volcano and subsequently foundered during caldera subsidence.     

Volcanoes and rocks of St Andrew Strait,Papua New Guinea

When Tuluman volcano, in St Andrew Strait, northern Bismarck Sea, erupted between 1953 and 1957, it produced acid rocks similar in major element chemistry to those of three other islands in the Strait — Lou, Pam Lin, and Pam Mandian. These acid rocks — termed the TLP’ series — are thought to represent magmas, or to be derivatives of a parental magma, produced by melting of crust (about 25 km thick beneath St Andrew Strait). TLP rocks have agpaitic indices ranging between 0.86 and 0.96. Acid lava also makes up 3 of the 4 Fedarb Islands at the northern end of St Andrew Strait, but its composition appears to be unrelated to that of the TLP series, and its origin is uncertain.

Q‐normative basalts (quartz tholeiites) make up the fourth island of the Fedarb group, and ol‐ and hy‐normative basalts crop out on Baluan Island at the southern end of the Strait. These basalt types do not appear to be directly related to one another, although both may have been derived from parents that originated in the upper mantle. Andesitic rocks have not been found on any of the islands in St Andrew Strait.

Tuluman and the volcanoes of Lou Island form an arc which may be part of a developing (or completed) ring fracture whose centre coincides with the line between Baluan Island, the Pam Islands, and the Fedarb Islands. The ring fracture may be the result of sagging of crust above a zone of crustal melting that produced the TLP magmas. It is possible that collapse could take place along this ring fracture, producing a caldera.     

南海新生代碱性玄武岩中斜长石矿物的化学成分及意义

南海新生代碱性玄武岩中存在两种不同粒径的斜长石矿物.其一为斜长石斑晶,常见熔蚀麻点,是岩浆上升、压力降低时发生熔蚀作用并在骤冷条件下形成的;其二斜长石微晶,半定向或杂乱分布于火山玻璃中,其中空骸晶结构表明斜长石微晶是在淬冷条件下迅速结晶形成的.斜长石斑晶具弱成分环带,斑晶边部的An值稍高或接近于斜长石微晶.微晶斜长石An值与岩浆喷出后的水深以及喷发位置距离岩浆主通道的远近存在一定联系.本区的斜长石斑晶形成温度明显低于冲绳海槽地区,而类似于东海陆架地区;斜长石微晶的结晶温度类似于冲绳海槽,表明两地区在岩浆喷出海底后淬火结晶的物理化学条件相似.结合同样品中橄榄石斑晶研究结果以及已有的地球物理学和岩石学方面的资料,可能反映了地幔柱快速上涌使早期部分熔融及结晶分异作用较弱,岩浆本身温度高提供了早期结晶形成的斑晶与寄主岩浆进一步充分反应的热量.计算的斜长石斑晶温度不能反映源区温度特征,后者应高于本文所计算的斜长石斑晶的结晶温度.     

藏东南碧土带瓦浦组火山岩形成的大地构造环境

首次对藏东南原称的瓦浦组进行系统的岩石化学研究 ,发现它包括了两套不同时代和大地构造环境下形成的火山岩。瓦浦组火山熔岩由下部的玄武岩夹玄武安山岩和上部的流纹岩组成 ,是古特提斯洋盆中的洋岛火山岩 ,其时代初定为早二叠世—晚二叠世早期。在觉马—巴格和扎西所见的岩层是以钙质浊积岩为主的火山 -沉积岩系 ,火山岩为岛弧拉斑玄武岩 ,属晚三叠世早期活动大陆边缘产物。上述发现为碧土带是复杂的造山带拼贴体、古特提斯主洋盆是开阔的多岛洋和晚三叠世活动大陆边缘可能属马里亚纳型提供了重要证据     

Caldera-related volcanic rocks in the Shammar Group, northern Arabian Shield

Volcanic formations of the ca 630-620 Ma old Shammar Group in the Tuluhah area in the northern Arabian Shield occupy an oval area some 8×12 km. They overlie sedimentary rift-fill of the Kuara Formation and are interpreted as related to the formation of a caldera, here named the Awad Caldera. The earliest of the volcanic formations, the Dabsah Tuff, is more than 450 m thick in the south and wedges out in the north. It is composed of silicic, medial to proximal pyroclastic flow rocks that record an eruption during which an initial caldera is interpreted to have formed by probably trapdoor-style collapse. The Nijab Basalt, more than 200 m thick and present as flows overlying the Kuara Formation to the north of the caldera, is presumed to have originated outside the study area during an interval between periods of silicic volcanic activity, and to have flowed onto the Dabsah Tuff in the first-stage caldera. The succeeding Mindassa Megabreccia contains large rafts of the older Shammar rocks, mainly Nijab Basalt, in a tuff matrix, and is regarded as probably a caldera collapse and fallback megabreccia formed during a silicic eruption that led to the second stage of caldera development. The megabreccia is overlain by the post-collapse Sutayih Tuff, more than 450 m thick, composed of proximal pyroclastic flow units.     

Structure and Dynamics of a Silicic Magmatic System Associated with Caldera-Forming Eruptions at Batur Volcanic Field, Bali, Indonesia

The Batur volcanic field (BVF), in Bali, Indonesia, underwenttwo successive caldera-forming eruptions that resulted in thedeposition of silicic ignimbrites. The magmas erupted duringand between these eruptions show a broad range of compositionsfrom low-SiO₂ andesite to high-SiO₂ dacite. On the basis oftheir geochemistry and mineralogy these magmas may be assignedto six groups: (1) homogeneous andesites with phenocryst compositionsessentially in equilibrium with the whole-rock composition;(2) remobilized crystal-rich low-SiO₂ andesites with resorbedphenocrysts in equilibrium with the whole-rock composition;(3) mixed low-SiO₂ dacite with a relatively large range of phenocrystcompositions, with most phenocrysts slightly too evolved tobe in equilibrium with the whole-rock; (4) extensively mixedlow-SiO₂ dacites with a very large and discontinuous range ofphenocryst compositions, with most phenocrysts either more Mg-richor more evolved than the equilibrium compositions; (5) remobilizedcrystal-rich low-SiO₂ dacites with resorbed and euhedral phenocrysts;(6) homogeneous high-SiO₂ dacites lacking evidence for magmamixing and showing narrow ranges of phenocryst compositionsin equilibrium with the whole-rock composition. This range ofsilicic magmas is interpreted to reflect a combination of closed-and open-system fractional crystallization, magma mixing andremobilization of cumulate piles by heating. The variety ofmagmas erupted simultaneously during the caldera-forming eruptionssuggests that the magmatic system consisted of several independentreservoirs of variable composition and degree of crystallization.The magmatic evolution of individual reservoirs varied fromclosed-system fractional crystallization to fully open-systemevolution, thereby resulting in simultaneous production of magmaswith contrasted compositions and mineralogy. Extensive emptyingof the magmatic system during the caldera-forming eruptionsled to successive or simultaneous eruption of several reservoirs. KEY WORDS: caldera; ignimbrite; magmatic chambers; magma mixing; petrology; Sunda Arc     

Kinematic link between episodic trapdoor collapse of the Negra Muerta Caldera and motion on the Olacapato-El Toro Fault Zone, southern central Andes

A combined geochronological and structural analysis of the Miocene Negra Muerta Caldera was designed to better understand caldera formation associated with prominent faults on the central Andean plateau. Rb–Sr ages of the caldera outflow facies indicate that caldera formation occurred in two volcano-tectonic episodes. The first episode commenced with explosive eruption of the 9.0±0.1 Ma andesitic Acay Ignimbrite followed by a period of volcanic quiescence and moderate tectonic activity. Dominant volcanic and tectonic activity occurred during the second episode, which is bracketed by eruption of the 7.6±0.1 Ma rhyolitic Toba 1 Ignimbrite and effusive discharge of the 7.3±0.1 Ma rhyodacitic to andesitic lava flows. Structural relationships between rocks of the Negra Muerta Volcanic Complex and collapse-induced normal faults, notably NE-striking normal faults, agree with simultaneous volcanic activity and floor subsidence of the caldera during the second episode. Floor subsidence was achieved by tilting on an outward dipping reverse fault to the northwest of the caldera floor around a hinge zone located south of the caldera floor. This induced horizontal extension of the caldera floor and was accomplished by fragmentation of, and intrusion of dikes into, the floor. Collapse-induced and post-collapse fault populations of the caldera do not differ significantly in the directions of their axes of maximum extension and are in this respect kinematically compatible with left-lateral slip on the nearby Olacapato-El Toro Fault Zone. This furnishes evidence for a kinematic control by prominent faults on the formation of collapse calderas in the central Andes. The structural analysis of the Negra Muerta Caldera shows that collapse calderas can serve as deformation markers that contribute in elucidating the regional kinematic regime and the time of activity of prominent dislocations genetically related to collapse calderas.     

Mineral and chemical variations within an ash-flow sheet from Aso caldera,Southwestern Japan

Although products of individual volcanic eruptions, especially voluminous ash-flow eruptions, have been considered among the best available samples of natural magmas, detailed petrographic and chemical study indicates that bulk compositions of unaltered Pleistocene ash-flow tuffs from Aso caldera, Japan, deviate significantly from original magmatic compositions.The last major ash-flow sheet from Aso caldera is as much as 150 meters thick and shows a general vertical compositional change from phenocryst-poor rhyodacite upward into phenocryst-rich trachyandesite; this change apparently reflects in inverse order a compositionally zoned magma chamber in which more silicic magma overlay more mafic magma. Details of these magmatic variations were obscured, however, by: (1) mixing of compositionally distinct batches of magma during upwelling in the vent, as indicated by layering and other heterogeneities within single pumice lumps; (2) mixing of particulate fragments—pumice lumps, ash, and phenocrysts—of varied compositions during emplacement, with the result that separate pumice lenses from a single small outcrop may have a compositional range nearly as great as the bulk-rook variation of the entire sheet; (3) density sorting of phenocrysts and ash during eruption and emplacement, resulting in systematic modal variations with distance from the caldera; (4) addition of xenocrysts, resulting in significant contamination and modification of proportions of crystals in the tuffs; and (5) ground-water leaching of glassy fractions during hydration after cooling.Similar complexities characterize ash-flow tuffs under study in southwestern Nevada and in the San Juan Mountains, Colorado, and probably are widespread in other ash-flow fields as well. Caution and careful planning are required in study of the magmatic chemistry and phenocryst mineralogy of these rocks.Publication authorized by the Director, U. S. Geological Survey.     

Origin of the Felsic and Basaltic Dikes and Flows in the Rajula-Palitana-Sihor Area of the Deccan Traps,Saurashtra, India: A Geochemical and Geochronological Study

Rhyolite, trachyte, pitchstone, and granophyre dikes are associated with mafic dolerite dikes and basaltic flows of the northwestern part of the Deccan flood basalt province in the Saurashtra Peninsula, India. Felsic dikes, exposed in the Rajula area of Saurashtra, are similar in age to the basaltic flows of neighboring Palitana. The ages of both the felsic and mafic rocks straddle the ~65 Ma Cretaceous-Tertiary boundary and correspond to the main Deccan flood basalt episode. Palitana is centered on an elongated gravity high whose major axis is NE-SW, and Rajula is located on its southwestern flank. Unlike the younger Bombay felsic rocks from the western coast of India, which have been explained as partial melts of gabbros in deep crustal sills or previously erupted basalts, the incompatible-element characteristics of the Rajula rocks indicate that the Rajula rhyolites, trachytes, and dacites may have been generated by an almost complete melting of upper crustal rocks at the southwestern flank of the Rajula-Palitana-Sihor magmatic body. High potential temperatures of the Deccan plume, quick migration of the hot basaltic parent magma through lithospheric weak trends, and collection and residence of magma in upper-crustal magma chambers before eruption may have produced the right conditions to melt the upper crust in the vicinity of the Rajula-Palitana-Sihor magma chamber. On the other hand, the andesite located northeast of the magmatic body possibly evolved by assimilation of upper-crustal wall rocks accompanied by 5-10% crystallization of a Rajula-type basalt near the wall of the magma chamber. The Sihor rhyolites may also have been derived from the Sihor basalts through fractional crystallization accompanied by crustal assimilation. The Rajula granophyres, however, do not show any involvement of the upper crust in their genesis. These may have a history similar to that of the Bombay rocks and may have erupted in response to rifting along the Cambay rift.