In this study, we investigate the effect of the availability of sites for heterogeneous nucleation on the microstructure of moderately to highly undercooled synthetic trachybasalts under anhydrous and hydrous (2 wt
Volcanism in the Izu rear-arc is bimodal in composition and occurs in an extensional zone associated with thickened arc crust. IODP Expedition 350 sampled a 4.0 to 4.4 Ma pumiceous felsic succession at 700 m below seafloor from Site U1437 in the Izu rear-arc. Named Unit II, this 45 m thick succession exhibits multiple homogeneous pumice-rich beds that are complexly graded and interbedded. Minor tuffaceous mudstone represents hiatus in eruptions when background sedimentation resumed. Using plagioclase crystal chemistry, we investigate the generation and evolution of the magmas and reconstruct the architecture of their corresponding magma plumbing system. Plagioclase is compositionally bimodal with peak frequencies at An 30–60 and An 75–95. Plagioclase crystals have 87Sr/86Sr of 0.703096 to 0.703380, and ẟ18O values of 5.2 to 6.4‰. We interpret the chemical zonation within individual crystals to represent fractional crystallisation, recharge, magma mixing and partial melting of altered crust. From the plagioclase-liquid crystallisation pressures, we identified deep mafic magma pockets in the oceanic lithosphere (30 to 38 km) and below the mantle-crust transition (20 to 24 km). During ascent, the magmas differentiated and stalled in the lower (10 to 12 km) and upper crust (4 to 6 km). We infer the source of the Unit II magmas to be depleted mantle wedge, associated with rifting of the Philippine Sea Plate, enriched by a uniform contribution of mélange and variable fluid flux from the subducted slab.
We present new experimental constraints on the pre-eruptive magma storage conditions at Nabro Volcano, Eritrea, prior to its 2011 eruption, and discuss the implications for sulphur emissions. We find that the magma was likely stored in the shallow crust prior to eruption, following an intrusion in March 2011. This reservoir was located at 7 km depth, and was recharged immediately prior to the eruption, from a deeper reservoir located at 20 km. The temperature of the magma was 1030 °C with 3 wt
The Kulanaokuaiki Tephra Unit 3 (900 C.E.) was a subplinian basaltic scoria eruption, the largest known to have occurred at Kīlauea’s summit. The initiation mechanism and cause(s) of this highly hazardous eruption remain poorly understood, particularly as this style of activity has not been observed in the modern period. We use olivine diffusion timescales from Fe-Mg and lithium zoning to assess magma storage timescales and evaluate evidence for mixing prior to eruption. Of the 200 crystals assessed, 55
Experimental determinations of trace element distribution between rocks/minerals and fluids at mantle pressures and temperatures have concentrated on eclogite and on fluids containing H2O with or without the addition of NaCl. Here we present data for websterite (pyroxenite) coexisting with five different fluids: H2O, H2O+CO2, H2O+NaCl, H2O+CO2+NaCl and H2O+NH3. All experiments were conducted at 1.5 GPa and 800˚C using glassy carbon fluid traps that were kept in a frozen state following experiments until trace elements in the fluid traps were analysed by laser cryoablation-ICP-MS. Results show that Rb, Ba, Sr, Pb and Li are fluid-mobile, with lower rock/fluid partition coefficients (DRock/Fluid) for all mixed fluids relative to pure H2O. Lowest values (0.0004–0.123) are found for H2O+NH3 fluids. Almost all for the REE are < 1, contrasting with results at higher pressures for eclogite, and show very little fractionation of LREE from HREE. The REE are least mobile in H2O+CO2+NaCl and H2O+NH3 fluids, as are the HFSE, which also show lower DRock/Fluid than for eclogites, with or without NaCl. Re and Pt are fluid-mobile, with DRe < DPt and lowest values for both in H2O+CO2 fluids. The first row transition elements show D-values ranging from 0.0004 to 2.88 with only 4 of 35 data points > 1. DRock/Fluid increase in the order Cu < Ni ≈ Zn < Co < Cr, V, Sc. Rock/fluid partitioning of trace elements for pyroxenite resembles peridotite more than eclogite due to the abundance of minerals with low DRock/Fluid,particularly low-Al clinpyroxenes.
Crystallizing spodumene pegmatites exsolve hydrothermal fluids, which generate exomorphic metasomatic halos in their host rocks. Composition, timing and frequency of fluid release relate to the evolution of the pegmatite. At the Moylisha prospect of the Leinster pegmatite belt in SE Ireland, spodumene pegmatites intrude the granite of the Leinster Batholith. Narrow ( 10 cm) visible halos developed at the contact, but halo mineralization may occur meters away, isolated from pegmatite contacts. Two halo assemblages were identified using petrography, in situ chemistry and imaging of metasomatic minerals and textures. A rutile-present assemblage (ferroan muscovite-siderophyllite-feldspar-quartz-rutile) has rare alkali-enriched mica and contains accessory rutile. In the rutile-absent assemblage (ferroan muscovite–quartz–apatite ± siderophyllite ± feldspar) micas are Nb–Ta-enriched and have lower Nb/Ta ratios; apatite is abundant. The metasomatic evolution of both assemblages is modelled with a progressive reaction model as the interaction of the host rock with two distinct fluids. The rutile-present assemblage fluid is mainly enriched in rare alkalis and likely expelled after the onset of orthoclase and spodumene crystallization in the pegmatite but before resorption of primary minerals. The rutile-absent assemblage fluid additionally carries Nb, Ta, K and P and likely corresponds to the late hydrothermal stage of pegmatite evolution after partial resorption of primary minerals. Isolated halo mineralization has potential implications for mineral exploration as it can be detected distal to the pegmatite. The (mineral) chemistry of the halo has potential for distinguishing pegmatites where spodumene is preserved, from pegmatites that have undergone widespread albitization and spodumene resorption.
Titanium isotope systematics have been increasingly used as a tracer for magmatic processes. A knowledge gap is how Ti isotopes fractionate during chemical diffusion. The magnitude of diffusive fractionation is described by β, which correlates the diffusivities of isotopes to their inverse mass ratio. To determine βTi, we have conducted diffusion-couple experiments on Fe-bearing basaltic melts at 1380 °C and 1 GPa using a piston-cylinder press. The measured Ti isotopic variability arising from diffusion is 0.4‰ (± 0.05‰, 2 SD), and the best-fit βTi is 0.022 ± 0.003 (2 SD). This value agrees with published values determined using Fe-free basalt analogues. A low βTi is consistent with Ti being strongly bound to the silicate network in basaltic melts, even if present in coordination numbers greater than four. Using this experimentally constrained βTi, we modeled isotopic fractionation during rapid olivine growth and boundary layer development. Under realistic magmatic conditions, the maximum δ49Ti predicted is 0.13‰. Titanium isotopes therefore have the potential to detect kinetic effects occurring in natural basaltic melts.
The West Eifel Volcanic Field (Germany) is the type locality of maars, a common volcano type in continental intraplate volcanic fields. Timescales of eruptive recurrence and spatiotemporal clustering of maars and scoria cones are critical to assess controls of melting, magma ascent, and venting in such fields. Here, we overcome limitations of conventional geochronometers to constrain these timescales through zircon (U-Th)/He dating of crustal xenoliths accidentally scavenged during magma ascent. Zircon U-Th-Pb crystallization ages distinguish metaigneous and metasedimentary xenolith provenance from Palaeozoic basement. The absence of pre-eruptive ⁴He in zircon was verified by dating petrologically diverse xenoliths from individual vents, yielding consistent results. Zircon (U-Th)/He eruption ages are: Meerfelder Maar (75 ± 4 ka, all 2σ), Mosenberg (78 ± 9 ka), Emmelberg (48 ± 3 ka), Daun maar group (23 ± 1 ka), Pulvermaar (25 ± 2 ka), Wartgesberg (23 ± 3 ka), Oberwinkler Maar (195 ± 7 ka), and Facher Höhe (25 ± 2 ka). Oberwinkler Maar falls into a previously postulated volcanic hiatus, whereas two volumetrically dominant clusters at c. 25 and 75 ka coincide with deposition of the Central European Eltville and Rocourt tephra markers. The onset of volcanic activity and its peak coinciding with the last glacial maximum correlate with progressive marine regression in the North Sea basin. This contradicts previous models linking volcanic upsurges to interglacial warming and sea level rise. Zircon (U–Th)/He dating of pyrometamorphic crustal xenoliths thus provides a robust chronological framework to improve knowledge about eruptive pulsing in intraplate volcanic fields.
Oscillatory zoning in blueschist and eclogitic garnets has been interpreted as indicative of fundamental processes that occur in subduction zones such as the infiltration of external fluids, fluctuations in the temperature and/or pressure, the accumulation and release of strain, or a change in the identities of reacting phases. Although these processes may and probably do occur, they are not necessarily required to produce oscillatory zoning in garnet from subduction zones. The present contribution describes a 2-D numerical model that generates oscillatory zoning of Mn in garnet (compensated by zoning in Fe and, to a lesser degree, Ca and Mg) based not on the invocation of external agents such as the influx of fluids but simply on the relative rates of garnet growth and elemental fluxes resulting from grain boundary diffusion. The number of oscillations is governed by the number of cycles in which the ratio of diffusion to reaction (D/R) changes. The magnitude of the compositional change is a function of the magnitude of the change in this ratio, and the width of the oscillations is a function of the duration of the change in the ratio. It is proposed that oscillatory zoning in blueschist and eclogitic garnet is common because the temperature of formation is appropriate for variations in the grain boundary diffusion of Mn to be manifested in the composition of garnet. At higher temperatures (e.g., Barrovian conditions), the grain boundary diffusivity of Mn and other major elements is sufficiently rapid to prevent the creation of major oscillations. However, several studies have shown that oscillations may still be preserved in more slowly diffusing trace elements even in the absence of major element oscillations. As a case study, we apply this model to evaluate the development of oscillatory zones in garnets from a retrogressed blueschist from the Tillotson Peak Complex, Vermont, USA. Results from flux calculations suggest that oscillatory zones can be formed on decadal timescales potentially consistent with paleo-seismic cycles.
The role of fractional crystallization in the formation of unzoned Li-mineralized pegmatites remains poorly understood. Apatite crystallizes throughout the cooling history of pegmatitic melts and can preserve critical information on the properties of the initial melt and its subsequent evolution. In this study, we report in situ element and Nd isotope compositions of apatite from unzoned Li-mineralized pegmatites in the Altyn Tagh Orogen, northern Tibetan Plateau, to investigate their formation and mineralization processes. These pegmatites have Nd isotope compositions similar to those of coeval S-type granites, suggesting that pegmatitic melts are likely fractionated products of granitic melts. Texturally early apatite (Ap1) has positive Eu anomalies (Eu/Eu* = 1.0–93), resulting from the depletion of Sm and Gd in the initial pegmatitic melts. This depletion may be due to fractionation of monazite and zircon within the nearby parent granitic melts prior to pegmatitic melt extraction, as supported by phase equilibrium modelling. Ap1 has elevated Li concentrations of > 6 μg/g, which can be taken to discriminate Li-mineralized from barren pegmatites, indicating that fractionation of granitic melts leads to initial Li enrichment. Compared to Ap1, which is closely associated with plagioclase and contains information about the properties of the initial pegmatitic melts, Ap2 formed in a late stage and coexists with spodumene, montebrasite, and muscovite. Texturally late apatite (Ap2) has negative Eu anomalies (Eu/Eu* = 0.001–0.9). The shift from positive to negative Eu anomalies in pegmatitic melts recorded in Ap1 and Ap2 reflects feldspar crystallization within the pegmatitic melts. Ap2 has higher Li concentrations than Ap1, indicating that in situ fractional crystallization further enriched Li, culminating in spodumene crystallization. Our study demonstrates that unzoned pegmatites experience significant in situ fractional crystallization after the emplacement of melts, which is crucial for their formation and Li-mineralization, akin to zoned pegmatites.
Apatite is a robust recorder of open-system magmatic evolution in granitoids, as it may preserve subtle geochemical signals of magma compositional changes. Here, we integrate textural, compositional, and O-isotopic analyses of magmatic apatite from Jurassic granodiorites and biotite granites in the Cathaysia Block, Southeast China, to better constrain the nature and evolution of open-system magmatic processes and refine the petrogenesis of these granitic rocks. In granodiorites, O-isotope disequilibrium between apatite and zircon documents open-system magmatism: low-δ18O apatite (Group 1; 5.7–7.1‰; Δδ18O_zircon–apatite = +1.31‰) indicates greater mantle-derived input, whereas high-δ18O apatite (Group 2; 8.0–9.0‰; Δδ18O_zircon–apatite = -0.84‰) reflects enhanced supracrustal assimilation. In biotite granites, apatite and zircon are in near isotopic equilibrium (Group 3; 6.3–9.0‰; Δδ18O_zircon–apatite = -0.08–0.09‰ ), and δ18O differences among samples likely reflect variable proportions of mantle- and crust-derived components. Apatite core–rim zoning provides additional evidence for complex magma interaction: apatite in granodiorites shows abrupt core-to-rim decreases in REE, Y, and SiO2, consistent with mafic recharge, while apatite in biotite granite displays opposite trends, reflecting hybridization with congenetic, more evolved felsic melts. Apatite trace-element systematics further track mineral–melt differentiation during magma evolution: decreasing Sr coupled with increasingly negative Eu anomalies reflects plagioclase fractional crystallization, whereas pronounced LREE depletion in apatite indicates early allanite crystallization. Together, these in situ textural, geochemical, and O-isotopic data highlight apatite as a powerful tracer of magma sources, open-system evolution, and differentiation processes in granitoids.
Garnet is a key mineral of several metamorphic and igneous rock types in the lithosphere. Despite being traditionally considered the archetypal cubic mineral, crystallizing in Ia 3 d space group, there is increasing evidence that almandine and spessartine garnets from low-temperature metapelites and metabasites are optically anisotropic, and therefore possibly not cubic. Here, we study the best-known occurrence of optically anisotropic garnet - the metabasites of the Franciscan Complex (Cazadero, USA) - integrating polychromatic polarization microscopy (PPM), electron microprobe analysis (EMPA) with and without flank method for Fe3+/ΣFe estimation, field emission scanning electron microscopy coupled with focus ion beam (FIB−FESEM), electron energy-loss spectroscopy (EELS), single-crystal X-ray diffraction (SCXRD) and electron backscatter diffraction (EBSD). The result of this multi-analytical approach is a comprehensive optical, chemical and crystallographic characterization of these birefringent garnets. The Cazadero garnet has optical sector zoning according to twelve pyramidal sectors forming a rhombic dodecahedron. Within sectors, a concentric oscillatory zoning is also observed. From a chemical point of view, no sector zoning is detected, whereas the concentric oscillatory zoning is maintained. Chemical maps and profiles show a typical growth distribution characterized by a bell-shaped distribution of Mn. Considering the Fe3+ contents measured by both the EMPA flank method and EELS, a distinctive chemical zoning ranges from Alm53Sps20Grs16Pyr01Adr10 in the core to Alm66Sps04Grs20Pyr03Adr07 in the rim. Therefore, the analyzed garnets are almandine with a non-negligible andradite component, that reaches up to 10 mol
Magma reservoirs in middle to upper crust play a crucial role in the evolution of continents and volcanic activity, yet their architecture, compositional evolution, and melt extraction mechanisms remain incompletely understood. The Valle Fértil batholith (VF), an exhumed section of the Ordovician Famatinian arc, provides an exceptional opportunity to investigate these processes in situ. This study integrates bulk-rock and mineral geochemistry (plagioclase, amphibole, quartz and apatite) with textural observations to reconstruct the magmatic evolution of the middle to upper part of the Famatinian arc section. Amphibole compositions reveal a systematic compositional evolution from Intermediate to Silicic lithological units, reflecting progressive differentiation. Trace element variations in amphibole indicate fractional crystallization as the dominant control on magma evolution. Apatite compositions record relative changes in oxygen fugacity of the system, and they capture the chemical evolution of the magmas as well as the amphibole. Chemometric reconstructions of amphibole equilibrium melts indicate derivation from dacitic to high-silica rhyolitic compositions, closely resembling silicic volcanic rocks from the Famatinian arc. These findings suggest that the middle crust of the Famatinian arc functioned as a vertically stratified magmatic distillation column, linking more mafic lower crustal units with high-silica magmas found in the upper crust, confirming the dominant role of trans-lithospheric differentiation columns in the evolution of arc crusts.
Moissanite (SiC) and other super-reducing minerals have been reported from supra-subduction zone (SSZ) ophiolites, yet their origin remains controversial given the oxidized conditions expected in forearc mantle. In this study, hundreds of moissanite grains have been recovered from mantle rocks of the Early Permian Dun Mountain Ophiolite, New Zealand. We report the first occurrence of moissanite found in direct association with disordered carbonaceous material (CM), which records low peak formation temperatures of < 334 °C. Carbon isotope analyses reveal characteristic light compositions for both moissanite (δ¹³C = − 33.7 to − 19.2‰; mean − 25.0‰) and associated CM (δ¹³C = -39.2 to -29.0‰; mean − 35.3‰). The light C isotope signatures of both phases are consistent with moissanite formation via fractionation of ultra-reduced, carbon-rich fluids derived from organic-rich sediments carried by a subducting lithospheric slab. Preservation of moissanite in an oxidized forearc mantle could be aided by low temperatures and relatively short mantle residence times. Whilst the total mass of moissanite formed under these settings is low, these findings highlight the occurrence of localised ultra-reducing conditions at shallow depths in SSZ environments and identify an under-recognised pathway for carbon cycling in subduction zones.
We investigate the origin and evolution of quartz populations in tourmaline‑bearing rocks of the Cornubian Batholith, SW England, to constrain fluid sources and Sn‑mineralization processes. Our study focuses on massive quartz–tourmaline rocks (MQT) because of their spatial and genetic association with Sn mineralization. MQT occur as small stock‑ to dyke‑like bodies (typically < 300 m) exemplified by Roche Rock and Porth Ledden. Fifteen quartz samples from MQT and comparative quartz from biotite granite, tourmaline granite, pegmatite, tourmaline breccia and veins (Porth Ledden, Porthmeor Cove, Roche Rock, Tresayes, Wheal Remfry) were subject to SEM‑cathodoluminescence and oxygen‑isotope (δ18O) analysis. SEM‑CL reveals multiple quartz generations: primary magmatic quartz (granites, aplites, MQT), pegmatitic quartz, secondary fracture‑fill quartz, oscillatory hydrothermal overgrowths on magmatic cores (typical in MQT), and complexly zoned vein quartz. All δ18O values are relatively high (+ 11.5 to + 27.7‰), mostly clustering between + 12 and + 15‰. Magmatic quartz (granites, aplites) range from + 11.5 to + 14.8‰, overlapping pegmatite quartz, implying incorporation into the granitic melts of high‑δ18O metasedimentary material, consistent with local Devonian metasediments. Hydrothermal quartz spans + 12.7 to + 27.7‰; two anomalously high values (+ 25.6, + 27.7‰) in crystal rims suggest late addition of formation waters. We infer that MQT at Porth Ledden and Roche Rock formed by partial metasomatic replacement of tourmaline granite due to infiltration and entrapment of pneumatolytic, B‑rich fluids in the roof zone. Metasomatism dissolved K‑feldspar, mobilizing K, Rb, Ba, Sr, Cs, Pb and notably Sn, producing cavities later infilled by hydrothermal quartz and tourmaline. A progressive increase in δ18O from magmatic to late hydrothermal quartz suggests fluid cooling and possible mixing with oxidizing formation waters, rather than a component of meteoric-derived waters that contributed to the precipitation of cassiterite.
As the slowest species, the diffusion of silicon controls the deformation of olivine and plays a crucial role in Earth dynamics. The diffusion of silicon in olivine remains controversial as literature diffusivities spread over several orders of magnitude. Furthermore, the extent of enhancement by hydrogen, observed in both olivine and forsterite (iron-free olivine) remains unclear. We performed experiments of volume diffusion of silicon in dry forsterite. We conclude to a significant effect of hydrogen as diffusivities from previous hydrous experiments with up to 1000 wt. ppm H2O are up to 3 log unit higher than our anhydrous experiments. We analyze our results together with literature data for both forsterite and olivine, and provide laws for silicon volume diffusion in olivine (Ol) D_Si^Ol = ( 10^ - 10.50 + 10^ - 8.78 C_H_2 O^0.42)e^ - 316000 + 1460 P/8.314 T , and in forsterite (Fo) D_Si^Fo = ( 10^ - 11.64 + 10^ - 9.38 C_H_2 O^0.42)e^ - 316000 + 1460 P/8.314 T , where the diffusivity DSi is in m2/s, temperature T in K, pressure P in GPa and hydrogen concentration C_H_2 O in wt. ppm H2O. Hydrogen starts enhancing the diffusion of silicon in olivine at very low concentrations. Diffusivity is already increased by 1.7 log unit at 1 wt. ppm H2O. Consequently, the diffusion of silicon virtually always occurs in the hydrous regime in mantle olivine. However, variable mantle hydration is not expected to result in drastic variation in diffusivity. A change in olivine hydrogen concentration by a factor 10 modifies diffusivity by a factor 2.6 only. As the deformation mechanism of olivine that should prevail in the asthenosphere, i.e. dislocation creep, depends linearly on silicon diffusivity, olivine viscosity will primarily relate to temperature rather than hydration.
Carbonatitic melt–rock interactions in the crust provide key insights into element mobility, metasomatic processes, and the formation of rare mineral assemblages in high-grade metamorphic terrains. Ekanite-containing calc-silicate dykes at Ampegama, southwestern Highland Complex, Sri Lanka, provide compelling evidence for the interaction of crustal-derived carbonatitic melt with charnockitic gneiss wall rocks, forming an “antiskarn”-type assemblage with ekanite. The dykes, reaching up to 2 m in width, contain variously oriented, irregularly shaped fragments of charnockitic gneiss and form gradational boundaries and reaction zones with wall rocks / enclosed fragments. Field occurrence, textures and variations in mineralogical and chemical composition from charnockitic gneiss over contact zone toward calc-silicate indicate formation of silicate minerals – such as wollastonite, scapolite, clinopyroxene, titanite and ekanite (Ca2Th0.9U0.1Si8O20) – via metasomatic reactions between carbonatitic melt and wall rocks. Other minerals (e.g. K-feldspar) reflect assimilation from wall rocks. In addition, there are reaction textures that indicate late-stage overprint by CO2 and F-rich fluids. Unaltered ekanite is bottle-green and transparent and may have gem quality. It is metamict (glassy) but still contains crystallograpically oriented inclusions. The Th-U-Pb age of ekanite, determined by electron probe micro-analyser (EPMA) chemical dating, is 524.4 ± 6.4 Ma (2σ), assigning primary ekanite growth to late-stage regional metamorphism in the Highland Complex. Phase equilibria modeling of the host charnockitic gneiss gives the peak pressure-temperature conditions of ca. 850 ± 50 °C and 6 ± 1 kbar and the carbonatitic melt infiltration has occurred at or close to the peak conditions. Thorium necessary for ekanite formation is proposed to have been taken up from monazite- and thorite-bearing wall rocks by the ascending carbonatitic melt. This melt was likely derived from the anatexis of crustal carbonate rocks during regional metamorphism.
Ancient evaporitic borate deposits are rare in the geological record, and the mechanisms governing their formation and long-term preservation remain poorly constrained. The Paleoproterozoic Liao–Ji orogenic belt of the North China Craton hosts several metamorphosed stratiform borate deposits that provide a unique opportunity to address these issues. Heavy δ11B (+ 7.5 to + 9.8‰) of borate minerals (suanite, ludwigite and szaibelyite) and δ34SV−CDT (+ 8.7 to + 16.1‰) of pyrrhotite demonstrate marine contributions, whereas whole-rock geochemistry, mineral chemistry, and negative εNd values (mostly between − 2 and − 7) reflect substantial crustal input. MORB-like iron isotopes (-0.049 to + 0.292‰) indicate a dominantly hydrothermal source of iron. These features point to boron concentration within an evaporitic basin at the continent margin that received mixed marine and terrigenous contributions. Phase modelling and Zr-in-titanite temperatures constrain peak P–T conditions of regional granulite-facies metamorphism of the Wengquangou deposit to 750–800 °C at 0.6–1.0 GPa. In situ LA–ICP–MS U–Pb dating of zircon defines two age populations at ca. 2.15 Ga and 1.87–1.84 Ga, corresponding to the ages of deposition and granulite-facies metamorphism, respectively. Apatite U–Pb ages of ca. 1.71–1.64 Ga record a later post-peak thermal or fluid-related overprint. Prograde dehydration reactions converted soluble hydrous borates into stable anhydrous assemblages, fundamentally modifying mineralogy while largely retaining primary isotopic characteristics. This high-grade metamorphic overprint played a key role in stabilizing the deposit against post-depositional dissolution. Our data show that ancient evaporitic borate deposits formed by similar processes as modern ones, involving the leaching and mobilization of B from volcanic and sedimentary rocks, transport of B into a continent-margin basin, and concentration of B by evaporation. The rare preservation of ancient borate deposits does not indicate that they rarely formed, but that they are generally not preserved as regions hosting major evaporitic borate deposits seldom undergo subsequent high-grade metamorphism.
The condition of a crystal-rich, mushy magma reservoir plays a crucial role in governing both the processes leading up to volcanic eruptions and the style of eruption that follows. However, constraining its pre-eruptive state remains a significant challenge. In this study, we use crystal clots as a petrological tool to investigate the thermal and compositional evolution of the mushy magma reservoir beneath Unzen volcano prior to the 1991–1995 eruption. We analyzed the major-element compositions of crystals and interstitial glass within the crystal clots, with particular focus on amphibole and plagioclase, to estimate the depth of formation, temperature, and chemical composition of the magma reservoir. Our results indicate that the physicochemical conditions of the mushy reservoir remained largely stable throughout the eruption sequence. Furthermore, textural evidence of amphibole breakdown suggests that heating of the reservoir prior to magma mixing occurred on a timescale of approximately between > 24 h and < 48 h.
