Magma Chemistry Research Articles

Chemical feedbacks during magma degassing control chlorine partitioning and metal extraction in volcanic arcs

Hydrous fluids released from subducting oceanic lithosphere fuel arc magmatism and associated hydrothermal mineralization, including formation of porphyry copper deposits. Critical magma degassing parameters are the depth, chemistry and style of fluid release during magma ascent, notably the behaviour of chlorine, a key metal-transporting ligand. Currently, understanding is limited by restricted data on fluid-melt partitioning of chlorine as a function of pressure and magma chemistry, and the complex interplay between the two that occurs in polybaric magmatic systems. Here we present experimental determinations of chlorine partitioning as a function of fluid and melt composition at pressures from 50 to 800 MPa. We provide, for the first time, a quantitative understanding of chlorine and copper evolution that is valid for shallow, deep or transcrustal differentiation and degassing. Monte Carlo simulations using our new data reproduce the chemical evolution of melt inclusions from arc volcanoes and fluid inclusions from upper crustal intrusions and porphyry copper deposits. Our results not only provide a novel chemical framework for understanding magma degassing, but quantify the primacy of magmatic chlorine concentration at the point of fluid saturation in promoting efficient copper extraction from magmas.

Apatite Texture, Composition, and O-Sr-Nd Isotope Signatures Record Magmatic and Hydrothermal Fluid Characteristics at the Black Mountain Porphyry Deposit, Philippines

AbstractThe trace elemental and isotopic signatures in apatite can be modified during hydrothermal alteration. This study investigates the suitability of apatite as an indicator of the source, chemistry, and evolution of magma and hydrothermal fluids. In situ textural, elemental, and O-Sr-Nd isotope analyses were performed on apatite in thin sections, from fresh and propylitically altered pre- and synmineralized dioritic porphyries from the Black Mountain porphyry Cu deposit in the Philippines.All studied apatite crystals have similar subhedral to euhedral shapes and are homogeneous in the grayscale in backscattered electron images. In cathodoluminescence images, the apatite in fresh and altered rocks displays yellow to yellow-green and green to brown luminescence, respectively. Apatite in fresh rocks has a higher Cl and Mn content, and lower Fe, Mg, Sr, Pb, and calculated XOH-apatite, compared to apatite in altered rocks. The content of F, rare earth elements (REEs), Y, U, Th, and Zr, and the Sr-Nd isotope signatures of apatite from fresh and altered rocks are similar in all apatite grains (87Sr/86Sr = 0.7034–0.7042 vs. 0.7032–0.7043, εNd(t) = 5.3–8.0 vs. 5.1–8.4). The X-ray maps and elemental and oxygen isotope signatures across individual apatite crystals are typically homogeneous in apatite from both fresh and altered rocks. The distinct luminescence colors, coupled with distinct mobile element compositions (Cl, OH, Mn, Mg, Fe, Sr, Pb), indicate modification of primary magmatic apatite during interaction with hydrothermal fluids. The similarities in Sr isotope ratios (87Sr/86Sr = 0.7032–0.7043) but slight differences in O isotope signatures (δ18O = 6.0 ± 0.3‰ vs. 6.6 ± 0.3‰) in apatite from fresh and altered rocks are consistent with the magma and hydrothermal fluids having the same source and suggest significant phase separation in the hydrothermal fluids given that 18O preferentially fractionates into the residual liquid relative to 16O during phase separation. The similarity of immobile element (REE, Y, U, Th, and Zr) contents in both populations of apatite, consistency of textures and Nd isotope compositions, and absence of obvious dissolution-reprecipitation features all suggest that altered apatite retains some magmatic characteristics. The apatite in fresh rocks has oxygen isotope compositions similar to that of zircons from the same sample (δ18O = 5.9 ± 0.3‰), indicating little to no oxygen isotope fractionation between zircon and apatite and that apatite can be a good proxy for the oxygen isotope composition of the magma. Based on the Cl contents of the magmatic and replacement apatite, and assuming their equilibrium with high-temperature magma fluid and replacement hydrothermal fluid, respectively, the calculated Cl content of the early magmatic fluid and the later replacement fluid can be estimated to be 6.4 to 15.1 wt % and ~0.25 ± 0.03 wt %, respectively. This indicates a depletion of Cl from the early high-temperature fluid to the replacement fluid, consistent with phase separation.This study demonstrates that cathodoluminescence, elemental compositions (such as Cl, Mn, Mg, Fe, Sr, Pb) and Sr-O isotope signatures in apatite can be modified during hydrothermal alteration, whereas other components (REE, Y, U, Th, and Zr) and the Nd isotope composition are preserved. These features can be used to constrain the origin, chemistry, and evolution of the primary magma and ore-forming hydrothermal fluids.

Linking magmatic processes and magma chemistry during the post-glacial to recent explosive eruptions of Ubinas volcano (southern Peru)

Understanding the links between the magma differentiation processes, the magma plumbing system and the magma composition at arc volcanoes is of paramount importance for volcanic hazard assessment. In this work we focus on the post-glacial, Holocene, historical, and recent eruption products of Ubinas volcano (Peru), which display an overall decrease in silica content from the older, plinian (VEI 3–5), rhyolitic eruptions (69–71 wt% SiO2) to the historical and recent (2006–2009, 2013–2017), vulcanian (VEI 1–2) basaltic andesitic eruptions (55–57 wt% SiO2). Based on a comprehensive study of the major and trace elements and the Sr-Nd-Pb isotopes, we conclude that this temporal pattern reflects the evolution of the Ubinas magmas in the middle-to-upper crust by a coupled Assimilation-Fractional Crystallization (AFC) process involving a cumulate composed of plagioclase, amphibole, clinopyroxene, orthopyroxene and Fe–Ti oxides, with minor amounts of olivine and biotite at the mafic and felsic end-members, respectively. Upper crustal assimilation is limited to 5–8 vol%, but the overall radiogenic Sr-Nd-Pb signature of the Ubinas magmas requires a larger crustal component, which must therefore occur at middle to lower crustal depths. The petrology of the Ubinas magmas also points to an overall increase in P-T conditions: the large Holocene dacitic and rhyolitic eruptions record temperatures ranging from 800 to 850 °C and pressures in the range of 200–400 MPa, whereas the historical and recent (2006–2009, 2013–2017) basaltic andesitic eruptions provide higher temperatures and pressures (1000 °C, >300–400 MPa). Overall, the thermo-barometry, phase equilibrium and geochemical constraints allow us to propose the existence of a middle-to-upper crust magma column composed of a highly crystalline magma mush containing batches of liquid magma, which seems to be continually recharged from deeper levels. On the basis of the petrological nature of the historical basaltic andesitic eruptions (1667 CE, 2006–2009, 2013–2017), we postulate that during the last centuries, Ubinas experienced a recharge-dominated process, with no evidence for a rejuvenation of the silica-rich reservoir that fed the large Holocene dacitic to rhyolitic eruptions. This study highlights the importance of detailed petrological studies of Holocene sequences at explosive arc volcanoes in order to constrain the magmatic processes and conditions that control large explosive eruptions.

The Long-Term Life-Cycle of Nevado de Toluca Volcano (Mexico): Insights Into the Origin of Petrologic Modes

The petrologic diversity of volcanic rocks reflects the dynamics of magma reservoirs and the temporal evolution of magma chemistry can provide valuable information for hazard assessment. While some stratovolcanoes monotonously produce intermediate magmas (55–68 wt% SiO2), dominantly erupted magma types (e.g., basaltic andesite, andesite or dacite) frequently differ even between neighboring volcanoes. If such differences arise due to thermal maturation processes over time or are predetermined by other properties of magmatic systems remains poorly understood. This study helps to elucidate the underlying factors modulating the chemistry of the magma preferentially erupted by Nevado de Toluca volcano in Central Mexico. We present a new dataset of bulk-rock and mineral chemistry spanning the entire 1.5 Million years of the volcanos’ eruptive history. The results reveal that Nevado de Toluca dacites and minor andesite originate in a stable configuration of pre-eruptive processes and plumbing system architecture by hybridization between an upper crustal silicic mush and deeper sourced basaltic andesite magmas. Yet, a subtle trend toward increasing silica content with time (2 wt% in 1.5 Ma) and episodicity in magma hybridization conditions are observed. We use thermal simulations of pulsed magma injection to probe the controlling variables on the temporal variation and compositional mode of magma geochemistry. The results show that the subtle temporal trend toward increasing bulk-rock SiO2 content is plausibly explained by slightly dropping recharge rates and continued upper crustal reservoir growth. Our modeling also shows that the dominant composition of eruptible magmas (“petrologic mode”) can shift as a function of magma flux, extrusive:intrusive ratio and temperature of the recharge magma. A comparison of SiO2 whole rock distributions for monotonous Mexican stratovolcanoes and their peripheral cones shows that their petrologic modes vary in concert, indicating that the recharge magma chemistry or temperature is a major control on the preferentially erupted magma composition for these volcanoes.

Volcanic ash ice-nucleating activity can be enhanced or depressed by ash-gas interaction in the eruption plume

Volcanic ash can trigger ice nucleation when immersed in supercooled water. This will impact several processes (e.g., electrification, aggregation, precipitation) in the eruption plume and cloud and in the wider atmosphere upon ash dispersal. Previous studies show that ash bulk properties, reflecting the chemistry and phase state of the source magma, likely contribute to the ice-nucleating activity (INA) of ash. However, it remains unexplored how interaction with magmatic gases in the hot eruption plume, which inevitably leads to altered ash surface properties, affects the ash INA. Here we demonstrate that the INA of tephra is raised by exposure to H2O(g) mixed with SO2(g) at both 800 and 400°C, but is substantially reduced by exposure to H2O(g) alone or mixed with HCl(g) at the same temperatures. In contrast, the INA of K-feldspar and quartz is reduced by all three eruption plume processing treatments. The decrease in INA of all silicates after heating with H2O(g) might relate to a loss of ice-active sites by surface dehydroxylation and/or oxidation. In the presence of HCl(g) or SO2(g), respectively, metal chloride or sulphate salts form on the tephra surfaces only. While NaCl and CaCl2 seem to have no effect on the tephra INA, CaSO4 is inferred to create ice-active sites, potentially through a particular combination of surface chemistry and topography. Overall, our findings suggest a complex interplay of bulk mineralogy and surface alteration in influencing ice nucleation by volcanic ash, and highlight the general sensitivity (enhancement or depression) of ash INA to interaction with magmatic gases in the eruption plume.

Oxygen isotope ratios in zircon and garnet: A record of assimilation and fractional crystallization in the Dinkey Dome peraluminous granite, Sierra Nevada, California

Abstract The 119 Ma Dinkey Dome pluton in the central Sierra Nevada Batholith is a peraluminous granite and contains magmatic garnet and zircon that are complexly zoned with respect to oxygen isotope ratios. Intracrystalline SIMS analysis tests the relative importance of magmatic differentiation processes vs. partial melting of metasedimentary rocks. Whereas δ18O values of bulk zircon concentrates are uniform across the entire pluton (7.7‰ VSMOW), zircon crystals are zoned in δ18O by up to 1.8‰, and when compared to late garnet, show evidence of changing magma chemistry during multiple interactions of the magma with wall rock during crustal transit. The evolution from an early high-δ18O magma [δ18O(WR) = 9.8‰] toward lower values is shown by high-δ18O zircon cores (7.8‰) and lower δ 18O rims (6.8‰). Garnets from the northwest side of the pluton show a final increase in δ18O with rims reaching 8.1‰. In situ REE measurements show zircon is magmatic and grew before garnets. Additionally, δ18O in garnets from the western side of the pluton are consistently higher (avg = 7.3‰) relative to the west (avg = 5.9‰). These δ18O variations in zircon and garnet record different stages of assimilation and fractional crystallization whereby an initially high-δ18O magma partially melted low-δ18O wallrock and was subsequently contaminated near the current level of emplacement by higher δ18O melts. Collectively, the comparison of δ18O zoning in garnet and zircon shows how a peraluminous pluton can be constructed from multiple batches of variably contaminated melts, especially in early stages of arc magmatism where magmas encounter significant heterogeneity of wall-rock assemblages. Collectively, peraluminous magmas in the Sierran arc are limited to small <100 km2 plutons that are intimately associated with metasedimentary wall rocks and often surrounded by later and larger metaluminous tonalite and granodiorite plutons. The general associations suggest that early-stage arc magmas sample crustal heterogeneities in small melt batches, but that with progressive invigoration of the arc, such compositions are more effectively blended with mantle melts in source regions. Thus, peraluminous magmas provide important details of the nascent Sierran arc and pre-batholithic crustal structure.

Magma diversity reflects recharge regime and thermal structure of the crust

The chemistry of magmas erupted by volcanoes is a message from deep within the Earth’s crust, which if decrypted, can provide essential information on magmatic processes occurring at inaccessible depths. While some volcanoes are prone to erupt magmas of a wide compositional variety, others sample rather monotonous chemistries through time. Whether such differences are a consequence of physical filtering or reflect intrinsic properties of different magmatic systems remains unclear. Here we show, using thermal and petrological modelling, that magma flux and the thermal structure of the crust modulate diversity and temporal evolution of magma chemistry in mid to deep crustal reservoirs. Our analysis shows that constant rates of magma input leads to extractable magma compositions that tend to evolve from felsic to more mafic in time. Low magma injection rates into hot or deep crust produces less chemical variability of extractable magma compared to the injection of large batches in colder or shallower crust. Our calculations predict a correlation between magma fluxes and compositional diversity that resembles trends observed in volcanic deposits. Our approach allows retrieval of quantitative information about magma input and the thermal architecture of magmatic systems from the chemical diversity and temporal evolution of volcanic products.

Mineral equilibrium constraints on the feasibility of diffusive H2O‐fluxed melting in the continental crust

AbstractGeneration of granitic melt is believed to occur predominantly by melting through the breakdown of hydrous minerals. However, melting due to the influx of H2O has been recognized in anatectic amphibolite facies tonalitic grey gneisses, metagreywackes and low‐P metapelites, and has consequently been proposed as an alternative mechanism for the generation of granitic melt. Melting induced by H2O addition is recognized from voluminous melt production at relatively low temperature, where hydrous minerals are stable and anhydrous minerals are preferentially consumed during melting. Mineral equilibrium modelling to determine the P–T conditions, melt volumes, melting reactions and viable H2O sources reveals that the process is not restricted to specific compositions or P–T conditions, although lower pressure and lithologies with a low hydrous mineral content are more favourable. Melting reactions in all lithologies primarily consume quartz and feldspars to yield 5–6 mol.% melt for each mol.% of H2O added. remains constant at ~0.70 to 0.77 during progressive melting as long as alkali feldspar is present. Once alkali feldspar is exhausted, plagioclase becomes the main reactant, producing more tonalitic melt compositions with gradually higher . Our results demonstrate that, at the site of melting, melting is driven by diffusion of H2O into the target rock along chemical potential gradients, rather than the advective flow of a mechanically distinct water‐rich fluid phase. Melting will initiate and proceed as long as a gradient exists between the H2O source and target lithology. Our calculations show that an ordinary magma, such as an I‐type magma with typical H2O content, has a high enough to be a viable H2O source, allowing diffusive H2O‐fluxed melting to produce melt proportions and fertility comparable to that of dehydration melting. However, high degrees of partial melting require a considerable amount of H2O, which necessitates a continuously advecting H2O source such as a magma conduit or melt‐bearing shear zone. A magmatic H2O source at emplacement level will undergo a similar amount of crystallization as the melt fraction produced in the target rock such that there will be no net melt production. Considering that shear‐zone hosted magma conduits are localized features, diffusive H2O‐fluxed melting is likely to only be viable in a small fraction of the anatectic orogenic crust. Although it may play an important role in locally raising melt volumes and modifying magma chemistry through mingling and hybridization, it does not appear to, of itself, be able to generate significant volumes of granitic melt.

Melting relations of anhydrous olivine-free pyroxenite Px1 at 2 GPa

Abstract. The reaction between melt derived by mafic heterogeneities and peridotites in an upwelling mantle may form hybrid olivine-free pyroxenites. In order to evaluate the impact of these lithologies on the chemistry of primitive magmas and their ability to give rise to new mantle heterogeneities, we experimentally investigate the melting relations at 2 GPa of the model olivine-free pyroxenite Px1 (XMg=0.81, SiO2=52.9 wt %, Al2O3 = 11.3 wt %, CaO = 7.6 wt %). The subsolidus assemblage consists of clinopyroxene, orthopyroxene, and garnet. At 2 GPa, the solidus of Px1 is located between 1250 and 1280 ∘C, at a temperature about 70 ∘C lower than the solidus of fertile lherzolite. At increasing melt fraction, the sequence of mineral phase disappearance is garnet–clinopyroxene–orthopyroxene. Across the solidus, partial melting of Px1 is controlled by reaction garnet + clinopyroxene = liquid + orthopyroxene, and above 1300 ∘C, once garnet is completely consumed, by reaction clinopyroxene + orthopyroxene = liquid. Orthopyroxene is the liquidus phase, and at 1480 ∘C olivine-free pyroxenite Px1 is completely molten indicating a melting interval of about 200 ∘C. Isobaric melt productivity is similar to garnet clinopyroxenites, and it is more than 3 times that of a fertile lherzolite at 1400 ∘C. Px1 partial melts cover a wide range of XMg (0.57–0.84), with SiO2, Al2O3 and Na2O decreasing and Cr2O3 increasing with the degree of melting. CaO content in partial melts increases as long as clinopyroxene is involved in melting reactions and decreases after its exhaustion. At 2 GPa and for melting degrees higher than 10 %, Px1 produces MgO-rich basaltic andesites matching the composition of eclogitic melts in terms of silica and alkali contents but with significantly higher XMg values. These melts differ from those derived from lherzolites at 2 GPa by higher SiO2 and lower CaO contents. Their high silica activity makes them very reactive with mantle peridotite producing hybrid orthopyroxene-rich lithologies and residual websterites. Melt–rock reactions likely prevent direct extraction of melts produced by olivine-free pyroxenites.

Zircon xenocryst evidence for crustal recycling at the Mid-Atlantic Ridge

The presence of zircon xenocrysts in Mid Atlantic Ridge (MAR) rocks poses essential questions concerning the path they have followed from their sources, how they have survived transport through the mantle, and why they are much more abundant than in any other ocean. Sixteen out of twenty-five seafloor samples collected between S 21° to N 42° contained xenocrysts, here defined as those grains that coexist with a significantly younger population or are significantly older than the band of the Atlantic seafloor age from which they were collected. The xenocrysts are of two types, young and old. Hf and O isotopes and trace elements indicate that the younger crystals (3.1 Ma to 25.6 Ma) are oceanic, derived from previous MAR rocks, whereas the older crystals (190 Ma to 3.2 Ga) are continental, derived from crust slivers located at the MAR or immediate vicinity. The high ∂18O (~7‰ to 11‰) of many older xenocrysts indicates ultrafast recycling through the mantle; otherwise, the fast diffusivity of oxygen would have driven the xenocrysts ∂18O to that of their mantle hosts, ∂18O ≈ 5.3‰. The incorporation of xenocrysts into MAR magmas mainly occurs after ridge jumps, when hot mafic magma pervades a new crustal section scavenging xenoliths, or locally melts felsic wall-rocks and mixes with the resulting zircon-laden low-T magmas. In either case, the fusible component of the xenocrysts source contaminates the upwelling mafic magma. When the source was oceanic, the resulting chemical and isotopic signatures of the contaminated magma would remain MORB-like, albeit enriched in incompatible elements. When the source was continental, the resulting magma would reflect a perceptible continental influence. Crustal contamination at the ridge explains the abundance of zircon xenocrysts and the distinctive chemistry of the MAR magmas compared with other MORBs. These features are ultimately related to the Atlantic slow-spreading because it promotes ridge jumping and, hence, active crustal recycling.

Effects of aseismic ridge subduction on the geochemistry of frontal arc magmas

Aseismic ridges are variably broad linear topographic features built by submarine volcanic activity that are thicker and more buoyant than the surrounding oceanic crust. Together with oceanic plateaus they constitute a non-trivial ∼30% of the oceanic floor. As such, aseismic ridge subduction significantly affects subduction dynamics as well as deformation of the overriding plate. Indeed, it is considered to exert major controls on volcanic activity, the generation of juvenile Earth's crust and the formation of porphyry Cu-Au deposits. Previous studies claimed that magma chemistry changes associated with aseismic ridge subduction are mainly due to source processes, such as melting of the subducting ridge, melting of the edges of the oceanic plate following slab tear, or subduction erosion of the continental margin with subsequent partial melting of the scraped and subducted continental material. Here we explore the role of the subducted aseismic Carnegie ridge on modulating frontal arc magma chemistry in the Ecuadorian-southernmost Colombian frontal arc. We show symmetric changes of several geochemical indices (Th, Nd, Sm, Nb, Th/La, Ba/Th) and evolutionary paths (Na2O-SiO2 correlations) in frontal volcanic arc rocks with respect to the equatorial latitude, which coincides with the projection of the crest of the subducting Carnegie ridge at the frontal arc. We explore the systematic along-arc changes of these geochemical indices through geochemical modelling using a Monte Carlo approach, and conclude that the trends are controlled to a significant extent by intracrustal processes. In particular we argue that magmatic systems associated with the central part of the Ecuadorian volcanic arc evolve through fractionation, partial melting of crustal rocks and mixing, at average greater depths than those in the northern and southern sides. We speculate about possible links between the systematic along-arc geochemical changes and the subduction of the Carnegie ridge. Our findings highlight a new facet of the controls that are exerted by the overriding plate on the geochemical variability of arc magmas.

Long-lived localized magmatism in central-eastern part of the Pernambuco-Alagoas Domain, Borborema Province (NE Brazil): Implications for tectonic setting, heat sources, and lithospheric reworking

Magmatic activity circumscribed to a restricted area lasting for tenths of millions of years allows to link changes in magma chemistry to deformation, and thus to determine the tectonic evolution of a region. Here we describe an area of about 180 × 40 km in eastern Borborema Province (northeastern Brazil) where several pulses of magmatism occurred in the time interval 656–562 Ma. The oldest pulses are represented by orthogneisses bearing a flat-lying foliation resulting from deformation and metamorphism of granitoids emplaced in three stages: 656–645 Ma, 632–618 Ma, and 606–597 Ma. The oldest and most important unit (656–652 Ma-old) has large length/width ratio (c. 10), alkali-calcic signature, lacks pronounced negative Ta anomalies whereas showing positive Hf and Zr anomalies in normalized trace element diagrams, and yields crystallization age similar to the deposition age of surrounding, dominantly clastic metasedimentary units. These characteristics point to emplacement in an extensional setting and most probably fit in a rift/sag basin model, with thermal thinning of the lithosphere generating regional subsidence and magmatism. Rocks related to the 632–618 Ma-old stage are represented by orthogneisses with petrographic and geochemical characteristics akin to I-type granites, indicating a shift to an orogenic setting at around 640–630 Ma. Increased radioactive heating during continued thickening promoted anatexis of metasedimentary rocks at mid-crustal levels and segregation of small volumes of peraluminous melts at 606–597 Ma. The subsequent period of magmatic activity at 587–581 Ma witnesses the transition to a transcurrent regime, with intrusion of large granitic and syenitic batholiths with high-K calc-alkalic to shoshonitic affinity. Intrusion of these plutons added to heat resulting from strain heating to produce peraluminous granites at 573–566 Ma through partial melting of metasedimentary sources. The final magmatic episode at c. 562 Ma suggests involvement of the lithospheric mantle and crustal sources, perhaps related to localized delamination. Generally, younger orthogneisses and plutons contain zircons inherited from previous magmatic events, pointing to important crustal reworking. Together with the geochemistry of the rocks, these results imply only modest addition of juvenile material and thus no net crustal growth in eastern Borborema Province during the 660–560 Ma time interval.

Ages and glass compositions for paired large-volume eruptions from the Acigöl volcanic complex, Cappadocia (Turkey)

Acigol volcanic complex is a bimodal volcanic field located at the western margin of the Neogene–Quaternary Cappadocian volcanic province in Central Anatolia (Turkey). Its most voluminous eruptions are preserved as a pair of widespread tuffs known as Lower Acigol Tuff (LAT) and Upper Acigol Tuff (UAT). Both are nearly aphyric high-silica rhyolites that are compositionally identical; they can be distinguished, however, by the presence of obsidian lithics in UAT and the absence thereof in LAT. Here, we report the discovery of medial deposits of these tuffs at ~ 60–65 km distance from their source, and re-estimate the volume for LAT in the range of 21.4–26.7 km3 (similar constraints are lacking for UAT, but a subequal volume is assumed). Development of paleosol between LAT and UAT provides evidence for a hiatus between both events which is further supported by radiometric dating using combined U–Th and (U–Th)/He zircon geochronology. The revised age for LAT based on sequence modeling of U–Th zircon crystallization and (U–Th)/He zircon eruption ages is 190 ± 11 ka, whereas new and published (U–Th)/He zircon ages indicate eruption of UAT at 164 ± 4 ka (uncertainties generally stated at 1σ). The compositional similarity between both eruptions previously known from whole-rock analyses also extends to glass major, minor, and trace element compositions. Moreover, U–Th zircon crystallization ages are also indistinguishable between LAT and UAT in proximal and medial locations. U–Pb zircon ages for crystals in secular equilibrium indicate a provenance from Cretaceous granitic basement and Miocene volcanic overburden typical for Cappadocia, and such xenocrysts appear to be slightly more common in LAT compared to UAT. The compositional similarity between LAT and UAT along with the apparent recycling of zircon from the non-erupted LAT magma during the UAT event indicates sequential tapping of rhyolite melt from a magma system that maintained thermal and compositional uniformity over a protracted time interval. This can be explained by rhyolite melt extraction from a thermally buffered crystal mush that remained viable to produce voluminous high-silica eruptions for few 10s of ka. The subsequent migration of eruptive activity in the Acigol volcanic complex from east to west also led to a change in magma chemistry and triggered a new pulse of zircon crystallization, indicating abandonment of the previous magmatic focus, and a shift of magma input to a new location. The identification of sizable medial deposits of LAT and UAT implies that both tephra units are potentially useful marker horizons for marine isotope stage (MIS) 6 sedimentary sequences in and around Anatolia. The recent discovery of UAT in a Black Sea sediment core supports this view.

New constraints from Central Chile on the origins of enriched continental compositions in thick-crusted arc magmas

Magmas from continental arcs built on thick crust have elevated incompatible element abundances and “enriched” radiogenic isotope ratios compared to magmas erupted in island and continental arcs overlying thinner crust. The relative influence of the slab, mantle, and upper plate on this variability is heavily debated. The Andean Southern Volcanic Zone (SVZ; 33–46°S) is an ideal setting to investigate the production of enriched continental arc compositions, because both crustal thickness and magma chemistry vary coherently along strike. However, the scarcity of primitive magmas in the thick-crusted northern SVZ has hindered previous regional studies. To better address the origin of enriched continental compositions, we investigate the geochemistry (major and trace element abundances, 87Sr/86Sr and 143Nd/144Nd ratios) of new mafic samples from Don Casimiro and Maipo volcanoes within the Diamante-Maipo Caldera Complex of the northern SVZ. While evolved Diamante-Maipo samples show evidence for crustal assimilation, the trace element and isotopic enrichment of the most mafic samples cannot result from crustal processing, as no known regional or global basement lithologies are enriched in all of the necessary incompatible trace elements. Subduction erosion models similarly fail to account for the enriched isotopic and trace element signature of these samples. Instead, we suggest that the enrichment of northern SVZ magmas is derived from an enriched ambient mantle component (similar to EM1-type ocean island basalts), superimposed on a northward decline in melt extent. A substantial, but nearly uniform contribution of melts from subducting sediment and altered oceanic crust are required at all latitudes. The EM1-like enrichment may arise from recycling of metasomatized subcontinental lithospheric mantle (M-SCLM), as the isotopic trajectory of primitive rear-arc monogenetic cones trend towards the compositions of SCLM melts sampled across South America. Isotopic data from spatially distributed rear-arc centres demonstrate that the arc-parallel variations in the degree of EM1-type enrichment observed in arc-front samples are also present up to 600 km behind the trench in the rear-arc. Rear-arc trace element systematics require significant but variable quantities of slab melts to be transported to the mantle wedge at these large trench distances. Overall, we show that a unified model incorporating variable mantle enrichment, slab additions, and melt extents can account for along and across-arc trends within the SVZ. The recognition that mantle enrichment plays a key role in the production of enriched continental compositions in the SVZ has important implications for our understanding of the chemical evolution of the Earth. If ambient mantle enrichment is not taken into account, petrogenetic models of evolved lavas may overestimate the role of crustal assimilation, which, in turn, may lead models of continental crust growth to overestimate the amount of continental material that has been recycled back into the mantle.

Generation of coeval metaluminous and muscovite-bearing peraluminous granitoids in the same composite pluton in West Qinling, NE Tibetan Plateau

There is considerable controversy over the nature of (strongly) peraluminous granitoids and their relationship with coeval metaluminous granitoids. The origin of peraluminous granitoids is of great importance for an understanding of granite petrogenesis, as well as the formation and evolution of the continental crust. This study reveals the presence of contemporaneous metaluminous granitoids and muscovite-bearing peraluminous granitoids in the Guomaying composite pluton (GCP) at approximately 243–242 Ma in West Qinling in the NE Tibetan Plateau. The Middle Triassic granitoids in the GCP can be petrographically and geochemically divided into two groups. The group I samples are characterized by slightly higher K2O/Na2O ratios and are metaluminous with lower Zr-saturation temperatures (740–756 °C). In contrast, the group II samples have lower K2O/Na2O ratios and mostly strongly peraluminous compositions, with marginally higher Zr-saturation temperatures (762–778 °C). Both groups share similar trace elements and Sr–Nd–Pb–Hf isotopic compositions, pointing to middle to lower crustal origins with an amphibolitic residue. The metaluminous granitoids are interpreted as partial melting of amphibolites in the middle to lower crust beneath West Qinling, and the muscovite-bearing peraluminous granodiorites were also generated by partial melting of an amphibolitic source, with addition of a feldspar-rich component in the source. The hydrous basaltic melts derived from partial melting of a metasomatized mantle wedge source were also incorporated into the magma mush. It is inferred that disparate magma sources, entrained peritectic assemblages, and various degrees of antecryst recycling collectively exerted influence on the granitic magma chemistry of the GCP. Consequently, our data suggest that muscovite-bearing strongly peraluminous granitoids, which are widely regarded as the products of recycled metasedimentary rocks, can also originate from an amphibolite-dominated source with involvement of a feldspar-rich component.

Magmatic evolution biases basaltic records of mantle chemistry towards melts from recycled sources

The chemistry of erupted magmas provides a crucial window into the composition and structure of Earth's convecting mantle. However, magmatic evolution in the crust makes it challenging to reconstruct mantle properties from volcanic rocks in important but incompletely understood ways. Here we investigate how mantle-derived compositional variability in primary oceanic basalts determines their phase equilibria relations and the nature of the geochemical signals they record. By performing experiments on synthetic analogues of compositionally extreme primitive lavas from the Reykjanes Peninsula of Iceland at realistic magma storage conditions (300 MPa, 1140–1260°C), we show that melts from enriched mantle domains retain higher melt fractions as they cool than those generated by melting of typical fertile lherzolite (i.e. they crystallise less mass over any interval of decreasing temperature). These melt fraction differences arise because plagioclase crystallisation is suppressed in Na- and H2O-rich but Ca- and Al-poor liquids derived from enriched source lithologies. Thus, compositional characteristics inherited from the mantle have a first-order control on the efficiency with which cooling basalts crystallise. This means that enriched melts will be more likely to survive crustal processing than depleted melts. Basalt chemistry will therefore be disproportionately influenced by melts from volumetrically minor enriched lithologies compared with melts from the upper mantle's most common lithology, lherzolite, systematically biasing basaltic records towards melts from recycled mantle sources.We combine our experimental observations from Iceland with thermodynamic simulations on mid-ocean ridge basalt compositions and show that mantle-derived variability in crystallisation efficiency can explain two enigmatic features of the global oceanic basalt record: firstly, the anomalous over-enrichment of incompatible elements during the differentiation of mid-ocean ridge basalts, which may reflect a progressive bias towards enriched compositions as differentiation proceeds; and secondly, the frequently documented cargoes of highly anorthitic plagioclase crystals carried by evolved and enriched liquids from which they cannot have crystallised. These crystals can now be understood as the solidified remnants of depleted, lherzolite-derived melts that have been entrained into melt mixtures from more enriched sources. Increases in the degree of enrichment of cumulate rocks sampled from progressively shallower horizons of the oceanic crust can also be interpreted in terms of enriched melts surviving crustal processing in preference to depleted melts.

Evolution in magma storage conditions beneath Kick-'em-Jenny and Kick-'em-Jack submarine volcanoes, Lesser Antilles arc

Kick-'em-Jenny and Kick-'em-Jack are two submarine volcanoes located in the southern Lesser Antilles arc. We present data on the temporal and spatial compositional variations in lavas and melt inclusions from Kick-'em-Jenny and four lavas from Kick-'em-Jack. Analyses of crater samples from Kick-'em-Jenny show an evolution in magma compositions between 1972 and 2013/14. Rock types have progressed from predominantly basalts in 1972 to basaltic andesites in 2013/14, while melt inclusions have also become more silicic over time, ranging basaltic to andesitic in 1972 and basaltic andesite to dacitic in 2013/14. Hornblende-phyric basalts are unusually rich in MgO (≤13 wt%) and Ni (≤230 ppm) and contain rare olivine phenocryst cores of Fo90–92, suggestive of a primitive character. Other phenocryst phases include spinel, clinopyroxene and plagioclase. Conversely, 2013/14 landslide deposit samples contain orthopyroxene as an additional phenocryst phase and are distinct from crater deposits in terms of phenocryst and bulk rock chemistry. We propose that 1972 and 2013/14 crater deposits sample Kick-'em-Jenny's current edifice, whereas 2013/14 landslide deposits likely represent “proto Kick-'em-Jenny” (an edifice existing pre 43.5 kyr BP). Geochemical data for Kick-'em-Jack support the notion that it is part of the same volcanic complex as Kick-'em-Jenny and probably represents part of the now collapsed proto Kick-'em-Jenny.Low eruption temperatures (~1000 °C), estimated from geothermometry, coupled with whole-rock magma chemistry suggest an origin of the most mafic lavas of Kick-'em-Jenny by dehydration melting of hornblende-peridotite just below the Moho. Subsequent injection of these cool, oxidised, wet primitive basalts into the deep crust was accompanied early fractionation of hornblende-dominated assemblages. Large, zoned hornblende phenocrysts track rapid ascent of magmas through the crust along a trajectory sub-parallel to the hornblende liquidus curve. Melt inclusion data provide clear evidence of volatile-rich (≤5 wt% H2O and <3200 ppm CO2) pre-eruptive magma storage 7–12 km below Kick-'em-Jenny. This new understanding of the architecture of Kick-'em-Jenny's sub-volcanic system will be useful for interpreting future seismic signals and mitigating volcanic hazards.

Unravelling the complexity of magma plumbing at Mount St. Helens: a new trace element partitioning scheme for amphibole

Volcanoes at subduction zones reside above complex magma plumbing systems, where individual magmatic components may originate and interact at a range of pressures. Because whole-rock compositions of subduction zone magmas are the integrated result of processes operating throughout the entire plumbing system, processes such as mixing, homogenisation and magma assembly during shallow storage can overprint the chemical signatures of deeper crustal processes. Whereas melt inclusions provide an effective way to study the uppermost 10–15 km of the plumbing system, challenges remain in understanding magma intrusion, fractionation and hybridisation processes in the middle to lower crust (15–30 km depth), which commonly involves amphibole crystallisation. Here, we present new insights into the mid-crustal plumbing system at Mount St. Helens, USA, using multiple regression methods to calculate trace element partition coefficients for amphibole phenocrysts, and thus infer the trace element compositions of their equilibrium melts. The results indicate vertically distributed crystal fractionation, dominated by amphibole at higher pressures and in intermediate melts, and by plagioclase at lower pressures. Variations in Nb, Zr and REE concentrations at intermediate SiO2 contents suggest repeated scavenging of partially remelted intrusive material in the mid-crust, and mixing with material from geochemically diverse sources. Amphibole is an effective probe for deep crustal magmatism worldwide, and this approach offers a new tool to explore the structure and chemistry of arc magmas, including those forming plutonic or cumulate materials that offer no other constraints on melt composition.

Incremental Growth of Mid- to Upper-Crustal Magma Bodies During Arabia–Eurasia Convergence and Collision: A Petrological Study of the Calc-Alkaline to Shoshonitic Meghri–Ordubad Pluton (Southern Armenia and Nakhitchevan, Lesser Caucasus)

The composite Meghri-Ordubad pluton (MOP) is located in the southernmost Lesser Caucasus and outcrops over ~1000 km2 in southern Armenia and Nakhitchevan. It is characterized by nested intrusions incrementally emplaced over ~30 Myr from the Middle Eocene to Early Miocene, coeval with the closure of the Neotethyan Ocean and the regional Arabia-Eurasia continental collision. We recognize three compositionally distinct intrusive series in the MOP: a medium- to high-K calc-alkaline series 1 dominated by gabbro, quartz diorite and tonalite, a shoshonitic to high-K calc-alkaline series 2 including gabbro, monzogabbro, monzodiorite, monzonite and syenite, and a high-K calc-alkaline adakitic series 3 characterized by the emplacement of lamprophyre dikes followed by porphyritic granodioritic intrusions and dikes. Thermobarometry calculations together with detailed petrography, mineral chemistry and experimental mineral stability P-T diagrams indicate a polybaric crystallization history and magma emplacement at mid- to upper crustal levels (0.1-0.3 GPa). Mineral textures and compositions reveal open-system magmatic processes such as mafic magma replenishment and reactivation of crystal mushes prior to transport towards shallower crustal levels in the MOP. The evolution of magma chemistry (calc-alkaline vs shoshonitic) and mineral assemblages (amphibole-bearing vs amphibole-free) indicate fluctuation in primary melt alkali and water contents over time. Such chemical evolution is ascribed to a decrease in the degree of partial melting of a chemically heterogeneous, but isotopically homogeneous, mantle source. Overall, the MOP is characterized by the successive emplacement of hydrous magma batches in the mid- to upper crust and petrological processes that were dominated by fractional crystallization (± crustal assimilation) upon cooling and, or, decompression (± degassing). We propose a new temporal and spatial petrogenetic model for the MOP, encompassing three long-lived, chemically distinct magmatic series incrementally assembled over 30 Myr within a pre-, syn- and post-collisional geodynamic setting.

Did the circum-Rodinia subduction trigger the Neoproterozoic rifting along the Congo–Kalahari Craton margin?

Early Neoproterozoic metaigneous rocks occur in the central part of the Kaoko–Dom Feliciano–Gariep orogenic system along the coasts of the southern Atlantic Ocean. In the Coastal Terrane (Kaoko Belt, Namibia), the bimodal character of the ca. 820–785 Ma magmatic suite and associated sedimentation sourced in the neighbouring pre-Neoproterozoic crust are taken as evidence that the Coastal Terrane formed as the shallow part of a developing back arc/rift. The arc-like chemistry of the bimodal magmas is interpreted as inherited from crustal and/or lithospheric mantle sources that have retained geochemical signature acquired during an older (Mesoproterozoic) subduction-related episode. In contrast, the mantle contribution was small in ca. 800–770 Ma plutonic suites in the Punta del Este Terrane (Dom Feliciano Belt, Uruguay) and in southern Brazil; still, the arc-like geochemistry of the prevalent felsic rocks seems inherited from their crustal sources. The within-plate geochemistry of a subsequent, ca. 740–710 Ma syn-sedimentary volcanism reflects the ongoing crustal stretching and sedimentation on top of the Congo and Kalahari cratons. The Punta del Este–Coastal Terrane is interpreted as an axial part of a Neoproterozoic “Adamastor Rift”. Its opening started in a back-arc position of a long-lasting subduction system at the edge of a continent that fragmented into the Nico Perez–Luis Alves Terrane and the Congo and Kalahari cratons. The continent had to be facing an open ocean and consequently could not be located in the interior of the Rodinia. Nevertheless, the early opening of the Adamastor Rift coincided with the lifetime of the circum-Rodinia subduction system.