Shallow Crustal Magma Chambers Research Articles

Platinum-Group Elements and Nb-Ta geochemistry of the Deccan basalts from Kumbharli and other Ghat sections and the Koyna Seismic Zone, Maharashtra (India): Crustal contamination and implication for sulfide saturation of the magma

The Deccan basalts from Kumbharli, Nadlapur, Bijaynagar, Hazarmacchi, and Surli Ghat sections (surface samples) and the Koyna borehole-7 (KBH-7, sub-surface samples) from Maharashtra (western India) demonstrate a two-stage crystallization process. Olivine and chromite were crystallized earlier in a shallow crustal magma chamber while clinopyroxene and plagioclase phenocrysts were crystallized in the sub-volcanic conduits forming the crystal mush, and carried to the surface by the evolved Fe-Ti-rich ascending basaltic magma. The high modal abundance of plagioclase at the top and clinopyroxene at the bottom part of the Kumbharli Ghat section is responsible for an increase in Sr and a decrease in Sc content towards the upper flows. The surface and sub-surface basalt samples contain Pd = 5–31 ppb, Pt = 4–15 ppb, Ir = 1–4 ppb, Ru = 1–3 ppb, and Rh = 3–5 ppb. Negative relations of Pd/Ir, Pd/Ru, and Pd/Rh with MgO from both sets of samples indicate partitioning of Ir, Ru, and Rh to early fractionated cumulus chromite (or olivine). Magnetite is a late crystallizing phase from the evolved Fe-Ti-rich basaltic magma. With an increasing modal abundance of magnetite, there is an increase of FeO total , TiO 2 , and V with Pd indicating fractionation of Pd by magnetite. Geochemical characters show that the surface basalt samples are less contaminated compared to the sub-surface KBH-7 basalts. High Nb (6.8–13.6 ppm) and low Ta (0.2–0.37 ppm) content and elevated Nb/Ta ratios of the surface samples from the ghat sections indicate the possible interaction of the basaltic magma with the carbonate metasomatized subcontinental lithospheric mantle (SCLM). On the other hand, a high degree of partial melting with limited crustal contamination can also elevate the Nb/Ta ratios of the basalts from the ghat sections. Deccan basalts from the study areas do not bear the signature of sulfide supersaturation and subsequent immiscible sulfide segregation. This is evidenced by their high chalcophile element concentrations (Ni = 79–103 ppm, Cu = 121–259 ppm) and ratios like Cu/Zr (≥ 1), Ni/MgO (13.7–21.8) with low Cu/Pd (2140–29938) compared to other Ni-Cu (PGE) sulfide mineralized continental flood basalts. The lack of crustal sulfur in the various contaminants is perhaps the main reason for this sulfide-undersaturated character of the Deccan basalts of the current study.

Understanding the links between volcanic systems and epithermal ore formation: A case study from Conical Seamount, Papua New Guinea

The Tabar-Lihir-Tanga-Feni (TLTF) island chain in northeastern Papua New Guinea formed by tectonic and alkaline to shoshonitic magmatic activity since the Pliocene. Several volcanic centers are CuAu mineralized including the world-class Ladolam Au deposit and Conical Seamount south of Lihir. The latter has been recognized as a juvenile analogue to the Ladolam deposit located on-shore. Whereas the mineralization at Conical Seamount is reasonably well studied, the specific magmatic processes that promote epithermal mineralization at this seamount but not at others are poorly understood. Here, we present new petrological and geochemical data from Conical Seamount, and compare them with those from the barren (unmineralized) Edison, Tubaf and New World seamounts nearby. We focus on whole rock compositions and major and trace element analysis of melt inclusions and minerals including clinopyroxene, sulfide and magnetite. We combine our observations with modelled constraints on mantle source composition and partial melting as well as magma evolution. A first-stage melting leaves a residual mantle source enriched in Au. Second-stage melting of a previously subduction-metasomatized mantle generally promotes the transfer and concentration of metals and volatiles in the ascending melts. These magmas are unlikely to control ore formation as all seamounts show evidence for similar mantle sources and parental melt composition. However, the presence of a shallow crustal magma chamber is unique to Conical Seamount. It is characterized by frequent melt replenishments and extensive magma fractionation leading to sulfide and magmatic volatile saturation. These specific magma chamber processes lead to the pre-enrichment of the magma in chalcophile elements including Au, while sulfide saturation coeval with magmatic volatile exsolution provide the way for an effective Au transfer from the magmatic to the epithermal system.

Petrological insights into connections between the S- and I-type magmatic associations in metamorphic core complexes: a case study of the Çataldağ metamorphic core complex (NW Turkey)

As one of the Cordilleran-type metamorphic core complexes in the western Anatolian extensional province and the broader Aegean region, the Çataldağ metamorphic core complex is composed of a granite-gneiss-migmatite complex and a synkinematic pluton, bounded by a detachment fault zone. The plutonic rocks within the footwall of the Çataldağ metamorphic core complex consist of Eo-Oligocene anatectic S-type leucogranites associated with gneiss-migmatite complex forming the core rocks and an I-type synkinematic pluton that emplaced into the core rocks in the Early Miocene. This paper presents the petrography, mineral chemistry, major-trace element geochemistry, and Sr-Nd-Pb isotope compositions of the Çataldağ I-type synkinematic pluton. It discusses the magma source evolution from the Eo-Oligocene to the Early Miocene and their relations to the Çataldağ metamorphic core complex formation by comparing two contrasting granitic associations from a petrological perspective. The Çataldağ I-type synkinematic pluton primarily comprises porphyritic granodiorite, granite, and peripheral rocks including evolved granite and pegmatite dikes. Thermobarometry calculations estimate crystallisation conditions of approximately 795 °C at 1–2 kbar (averaging 1.5 kbar), corresponding to a shallow crustal magma chamber (⁓5 km deep). The pluton is metaluminous to slightly peraluminous and of a high-K calc-alkaline character. The 87Sr/86Sr(i) and εNd values of the studied samples range from 0.70684 to 0.70772 and from −6.1 to −2.3, respectively. Pb isotopic compositions are 18.74–19.29 for 206Pb/204Pb, 15.69–15.78 for 207Pb/204Pb, and 38.89–39.39 for 208Pb/204Pb. Geochemical modelling shows that a large amount of metasomatised lithospheric mantle-derived melt component (up to 70%) and a minor crustal component (<50%) contributed to the source region of the magma. The combination of isotopic data and thermodynamic modelling suggests that the compositional variations within the pluton were primarily the result of open system processes, predominantly the assimilation of crustal rocks with a dominance of fractional crystallisation. The Sr-Nd-Pb isotopic signatures and trace element characteristics of I-type pluton, as well as their comparisons with those of S-type leucogranites within the Çataldağ core complex, indicate heterogeneous magma sources, evolving from crust-dominated to mantle-dominated magma from Eo-Oligocene to the early Miocene, reflecting mantle upwelling beneath NW Anatolia. We infer that the long-lived, mantle-driven Eocene plutonic activity that preceded the development of core complexes may have led to thermal weakening, melting, and crustal flow in the lower-middle crust beneath NW Anatolia, facilitating the formation of the granite-gneiss-migmatite complex during the incipient phase of extension in the Eo-Oligocene. Coinciding with the rapid exhumation of the S-type granite and migmatite complex, the formation and emplacement of I-type synkinematic pluton along the detachment fault zone of the Çataldağ core complex were developed as a result of asthenospheric upwelling driven by lithospheric removal processes. This inferred lithospheric removal was triggered by Hellenic slab rollback affecting the entire western Anatolia since the Oligocene. The transition from S- to I-type granites observed in the Çataldağ core complex and other Aegean massifs may indeed be a manifestation of the intensity or accumulation of this crustal extension.

Seismic Imaging of a Magma Chamber and Melt Recharge of the Dormant Datong Volcanoes

AbstractThe Datong volcanoes erupted episodically from ca. 0.74 to ca 0.10 Ma with the last eruption at ca. 100 ka, and ever since has remained dormant. It remains elusive whether the Datong volcanoes are being recharged and would become active again in the near future. Here, we image the crustal structures beneath the Datong volcanoes using local earthquake data recorded at a newly deployed dense Datong Nodal Array. The high‐resolution 3‐D model reveals, for the first time, the geometry and melt fraction of a shallow crustal magma chamber at the depths of 10–20 km beneath the Datong volcanoes. Here, we suggest a magmatic system beneath the Datong volcanoes that could transport basaltic melt from the uppermost mantle to the shallow crustal magma chamber beneath the Datong volcanoes. Further, the melt‐poor crustal magma chamber (melt fraction of 5%–10%) can explain why the Datong volcanoes are inactive at present, and future eruptions can be expected if melt recharge from a mantle source is sustained.

T-P-fO2 conditions of sulfide saturation in magmatic enclaves and their host lavas

Physicochemical conditions under which early sulfide saturation occurs and the role of this process in the mineralising potential of magmas in different geodynamic settings have been the subject of interest in many recent studies. Here we present new temperature-pressure-fO2 data on sulfide-saturated magmatic enclaves and host lavas from volcanic systems in subduction, post-subduction and intraplate geodynamic settings, some of which are associated with porphyry and epithermal deposits. Petrographic investigations coupled with mineral chemistry of sulfide inclusions and their host minerals and with bulk chemistry of amphibole, plagioclase and pyroxene-rich enclaves and their host lavas indicate that sulfide occurrence, abundance, and composition depend on magma evolution. Sulfides are more abundant and have higher Ni/Cu bulk ratios when found in mafic enclaves compared to sulfide inclusions occurring in more evolved lavas. Thermo-oxibarometry estimates indicate that mafic hornblende-rich cumulates (SiO2 < 45 wt%) occurring at the studied intraplate setting evolve and reach sulfide saturation already in the mantle (61 ± 7 km) at high temperature and pressure conditions (17 kbar and 1135 ± 28 °C) and low ΔNNO (< −1). These cumulates are associated with a greater sulfide abundance (up to 0.23 area %) compared to other types of enclaves and have Cu-poor sulfides (mostly pyrrhotite, with Cu median = 0.16 wt%) with high Ni/Cu values (up to 65). In contrast, more evolved (SiO2 > 50 wt%) gabbroic plutonic enclaves found in arc settings saturate sulfides at shallower crustal levels (~26 ± 10 km, i.e. ≤ 7 kbar) and at lower temperature (1014 ± 28 °C) and higher ΔNNO (> +1), and are characterised by a lower sulfide abundance (down to 0.01 area %), dominated by Cu-rich sulfides (mostly chalcopyrite and bornite with Cu median = 50 wt%). By accounting for the sulfide volume, our results suggest that sulfides can retain significantly higher amounts of metals in the enclaves (Cu = 32–113 ppm) compared to sulfides in the host lavas (Cu < 8 ppm). Considered altogether, our results indicate that alkaline mafic, less oxidised, and barren magmas corresponding to intraplate and back-arc settings saturated in sulfides in the upper mantle. In contrast, calc-alkaline-felsic, oxidised and ore-related magmas corresponding to convergent margins started to be saturated in sulfides at the expected depth of the MASH zone and at shallower crustal levels. This seems to be a consequence of the investigated alkaline systems not forming shallow crustal magma chambers allowing magma differentiation, rather than the fact that late and shallow sulfide saturation is necessary for porphyry formation.

The 2008–2010 Subsidence of Dallol Volcano on the Spreading Erta Ale Ridge: InSAR Observations and Source Models

In this work, we study the subsidence of Dallol, an explosive crater and hydrothermal area along the spreading Erta Ale ridge of Afar (Ethiopia). No volcanic products exist at the surface. However, a diking episode in 2004, accompanied by dike-induced faulting, indicates that Dallol is an active volcanic area. The 2004 diking episode was followed by quiescence until subsidence started in 2008. We use InSAR to measure the deformation, and inverse, thermoelastic and poroelastic modelling to understand the possible causes of the subsidence. Analysis of InSAR data from 2004–2010 shows that subsidence, centered at Dallol, initiated in October 2008, and continued at least until February 2010 at an approximately regular rate of up to 10 cm/year. The inversion of InSAR average velocities finds that the source causing the subsidence is shallow (depth between 0.5 and 1.5 km), located under Dallol and with a volume decrease between −0.63 and −0.26 × 106 km3/year. The most likely explanation for the subsidence of Dallol volcano is a combination of outgassing (depressurization), cooling and contraction of the roof of a shallow crustal magma chamber or of the hydrothermal system.

Central volcanoes and caldera collapses in the late Miocene – Late Pleistocene Tibesti Volcanic Province, northwest Chad

The Tibesti Volcanic Province (TVP) in northwest Chad represents the second largest of the five Gharyan–Tibesti volcanic provinces and covers an area around 29,000 km2. The other four provinces are in Libya, but all five provinces are from late Miocene to Quaternary and may have a common mantle source. The TVP, however, differs from the other four as regards volcano-tectonic processes, eruption style, and production of volcanic materials. The volcanic products of the TVP were erupted from the end of Miocene to late Pleistocene, range from basaltic to acidic, and suggest a double magma source – a shallow chamber fed by a deeper and larger reservoir. More specifically, field observations and numerical modelling results suggest that the basaltic magmas forming scoria cones, primarily at the periphery of the TVP, came from a deeper magma reservoir in the lower crust while the rhyolite and ignimbrites were fed by a shallow crustal magma chamber. By contrast, the volcanic products of the four volcanic provinces in Libya are primarily basalts and fed directly from single deep reservoirs. In the period from 8 Ma to 7−5 Ma, the evolution of the TVP was characterised by the formation of central volcanoes. Subsequently, in the period from 7−5 Ma to 0.43 Ma, the TVP was subject to caldera collapses that produced large-volume ignimbrites (>100 km3). Here we present numerical models with plausible loading conditions to understand better the volcano-tectonic evolution of the TVP and the spatial and temporal distribution of its major volcanic units. Our results suggest that the normal local stress field encouraged the formation of a central-volcano edifice (a volcanic cone) fed by radial dykes and inclined sheets and mostly erupting small volumes. By contrast, stress fields generated as a result of small crustal uplift or doming, due to the accumulation of magma in a deep-seated reservoir, triggered the formation of a ring-fault and the injection of a ring-dyke above the lateral margins of a shallow crustal magma chamber. Subsequently, the piston-like caldera subsidence helped to squeeze magma out of the shallow chamber resulting in large eruptions.

Structure of Lō‘ihi Seamount, Hawai‘i and Lava Flow Morphology From High-Resolution Mapping

High-resolution bathymetric data were collected using AUV Sentry at the summit and at two hydrothermal vent fields on the deep south rift of L¯o‘ihi Seamount. The summit map records a nested series of caldera and pit crater collapse events, uplift of one resurgent block, and eruptions that formed at least five low lava shields that shaped the summit. The earliest and largest caldera, formed ~5900 years ago, bounds almost the entire summit plateau. The resurgent block was uplifted slightly more than 100 m and has a tilted surface with a dip of about 6.5° towards the SE. The resurgent block was then modified by collapse of a pit crater centered in the block that formed West Pit. The shallowest point on L¯o‘ihi’s summit is 986 m deep and is located on the northwest edge of the resurgent block. Several collapse events culminated in formation of East Pit, and the final collapse formed Pele’s Pit in 1996. The nine mapped collapse and resurgent structures indicate the presence of a shallow crustal magma chamber, ranging from depths of ~1 km to perhaps 2.5 km below the summit, and demonstrate that shallow sub-caldera magma reservoirs exist during the late pre-shield stage. On the deep south rift zone are young medium- to high-flux lava flows that likely erupted in 1996 and drained the shallow crustal magma chamber to trigger the collapse that formed Pele’s Pit. These low hummocky and channelized flows had molten cores and now host the FeMO hydrothermal field. The Shinkai Deep hydrothermal site is located among steep-sided hummocky flows that formed during low-flux eruptions. The Shinkai Ridge is most likely a coherent landslide block that originated on the east flank of L¯o‘ihi.

Generation of leucogranites via fractional crystallization: A case from the Late Triassic Luoza batholith in the Lhasa Terrane, southern Tibet

Leucogranites are important in understanding intracrustal differentiation and regional tectonic evolution, but how these rocks form remains a matter of much debate. The Luoza batholith in the central Lhasa subterrane provides an important opportunity to address these issues as this batholith consists of normal calc-alkaline S-type granodiorites (Group 1) and highly fractionated S-type leucogranites (Group 2). Laser Ablation Inductively Coupled Plasma Mass Spectrometry zircon U-Pb dating indicates that these rocks were emplaced at ca. 221–205 Ma. The Group 1 samples (ca. 205 Ma) have low SiO2 content (64–67 wt%) and low differentiation index values (DI = 75–76), and are slightly depleted in Ba, P, Sr, and Ti compared to the Group 2 samples (ca. 221 Ma and 213 Ma), which have high SiO2 content (74–75 wt%), high differentiation index values (91–94), and are significantly depleted in Ba, P, Sr, and Ti. Samples from Group 1 and Group 2 exhibit similar zircon εHf(t) values (−15.6 to −1.4), both of which are significantly higher than melts from the pure ancient basement in the central Lhasa subterrane (−20.5 to −10.6). The Group 1 samples can be interpreted as having been derived from the partial melting of metagreywacke within the ancient basement in the central Lhasa subterrane with contributions from mantle- or juvenile- crust-derived components, whereas the Group 2 samples may have formed via varying degrees of fractional crystallization of different minerals (e.g., plagioclase ± K-feldspar, biotite, zircon, allanite/monazite, titanite, and apatite) from the parental magmas, represented by the granodioritic samples in Group 1, in shallow crustal magma chambers. In combination with sedimentary records, the high-temperature and low-pressure conditions indicated by the Group 1 samples can be geodynamically linked to a back-arc extensional setting in response to the southward subduction of the Bangong-Nujiang Tethys Ocean seafloor. Our work implies that fractional crystallization is likely a feasible mechanism for the development of leucogranites and for explaining the compositional diversity of the granite during intracrustal differentiation.

Geochemical and zircon isotopic evidence for extensive high level crustal contamination in Miocene to mid-Pleistocene intra-plate volcanic rocks from the Tengchong field, western Yunnan, China

SHRIMP zircon Pb/U dating of Cenozoic volcanic rocks in the Tengchong area, western Yunnan Province, China, shows that the dacite and andesitic breccia lavas from Qushi village were intruded at 480±10ka and 800±40ka, respectively. Moreover, Pb/U dating of trachyandesite from Tuantian village and olivine basalt from Wuhe village give weighted mean 206Pb/238U ages of 2.82±0.08Ma and 12.28±0.30Ma. Corrections for initial 230Th disequilibrium of zircon were used for the former two younger ages.The Tengchong volcanic rocks have a large range of SiO2 (48.6–66.9wt.%) and mostly belong to a high-K calc-alkaline series. The lavas originated from heterogeneous sources and were modified by subsequent fractional crystallization. The REE and other trace element patterns of the Tengchong volcanic rocks resemble magmas having a large component of continental crust. All have similar degrees of LREE and HREE fractionation and are enriched in LILE, La, Ce and Pb, with depletions in Nb, Ta, Ti, Sr and P relative to primitive mantle. Zircon δ18O values of 6.96±0.17 and 7.01±0.24‰ and highly varied negative εHf(t) values of −1.5 to −11.0 and −10.3 to −13.7, as well as the presence of inherited zircon grains in the studied samples, indicate that the magmas contain crustal material on a large scale.The Tengchong volcanic rocks have HFSE ratios (e.g., Nb/Ta, La/Nb, Zr/Y) similar to continental flood basalts, indicative of an intra-plate extensional tectonic setting. Widespread distributed faults might have facilitated upwelling of mantle-derived melts and eruptions from shallow crustal magma chambers to form the large volcanic field.

Collapse-driven large eruptions

For a typical poroelastic shallow crustal magma chamber, about 0.1% of the mafic magma and about 4% of the felsic magma are erupted and/or injected during magma-chamber rupture. Magma chambers with volumes of the order of several tens to several hundred cubic kilometres, as are common, are thus sufficiently large to supply magma to small or moderate eruptions. For large eruptions, however, a much higher percentage of the magma volume must be squeezed out of the chamber. For an ordinary poroelastic chamber, the excess pressure in the chamber falls exponentially during the eruption. For a large eruption to be possible, however, the excess pressure must be essentially maintained until the very end of the eruption. Here I show that caldera collapse can maintain the excess pressure through forced magma-chamber volume reduction, in which case a resulting large eruption would be the consequence (not the cause) of the collapse. I also show that ring-fault dip partly controls the size and intensity (volumetric flow or effusion rate) of the eruption. If the ring-fault dips inward (a normal fault), the displacement is ‘stable’, the volumetric flow rate (intensity) remains essentially constant during the collapse, and the magma chamber remains active following the collapse. By contrast, if the ring-fault dips outward (a reverse fault), the displacement is ‘unstable’, the volumetric flow rate normally increases dramatically during the collapse, and the magma chamber may be entirely destroyed during the collapse.

Temporal evolution of granitic magmas in the Luanchuan metallogenic belt, east Qinling Orogen, central China: Implications for Mo metallogenesis

The Luanchuan metallogenic belt, located within the eastern part of the Qinling Orogen, central China, hosts a number of world-class Mo deposits that are closely related to small late Mesozoic granitic plutons. Zircon U–Pb dating of distinct plutons in the Luanchuan metallogenic belt has yielded ages of 153±1, 154±2, 152±2, and 148±1Ma. Molybdenite Re–Os isotopic compositions of Yuku ore district in the southern part of Luanchuan metallogenic belt has yielded an isochron age of 146±1Ma, which is consistent with the large-scale mineralization ages in the northern part of the Luanchuan metallogenic belt. A combination of previous studies and new geochronological and isotopic data show a concordant temporal and genetic link between granitic magmatism and Mo mineralization in the Luanchuan metallogenic belt, suggesting that this mineralization episode formed the most extensive Mo mineralization belt in the east Qinling Orogen. Zircon grains from Mo-related granitic plutons show similar trace element distributions. High-precision Multi Collector–Inductively Coupled Plasma–Mass Spectrometry (MC–ICP–MS) Pb isotope analysis of K-feldspar megacrysts from mineralization-related granites suggest that they were derived from the lower crust. Similarly, the Pb isotopic compositions of pyrite coprecipitated with molybdenite also suggest that the metals were derived form the lower crust, with probably minor mantle contribution.A continuum mineralization model that describes the sourcing of Mo from an evolving granitic magma over successive differentiation events, possibly in separate but connected magma chambers, could explain the remarkable Mo enrichment in the Luanchuan metallogenic belt. The volatile- and Mo-bearing granitic magmas ascended as diapirs from the deep crust, and were emplaced as dikes in the upper crust. Lithological differences between these Mo-bearing granites may relate to different stages in the evolution of individual magmas. Finally, ore-forming fluids were exsolved from the granitic melts in shallow-crustal magma chambers, resulting in the formation of the Mo deposits of the Luanchuan metallogenic belt. This model may account for the widespread late Mesozoic Mo deposits of eastern China.

Petrogenesis and metallogenesis of the Xinjie layered mafic–ultramafic intrusion, China: Modeling of recharge, assimilation and fractional crystallization

The Xinjie layered mafic–ultramafic intrusion in the central Emeishan large igneous province (ELIP), SW China, hosts Fe–Ti–V oxide ore in the upper part and Ni–Cu–platinum-group element (PGE) sulfide deposits in the lower part. In this study, we use published Sr–Nd isotopic data to simulate and evaluate the energy-constrained recharge, assimilation, and fractional crystallization (EC-RAFC) model with a view to track the petrogenesis and mineralization. In contrast to the energy-constrained assimilation fractional crystallization (EC-AFC) model, the EC-RAFC modeling shows that the Xinjie intrusion may represent a shallow crustal magma chamber system, where it experienced the RAFC processes. In the early stage of the magmatic process (Tm=1460°C), a pulse of magma was injected into an actively evolving magma chamber. Minor melting (Ma∗=0.257) and assimilation of the wallrock (underlying Emeishan basalts) occurred when the temperature of magma (Tm=1245°C) decreased close to the equilibration temperature (Teq=1165°C). The mass–temperature plots indicate that fractional crystallization was significantly affected by the recharge and assimilation processes, whereas the contaminated magma recharge favored the early crystallization of Fe–Ti oxides as well as the formation of the PGE-bearing immiscible melt in the lower part of the intrusion. In contrast, the formation of the Fe–Ti oxides ores in the upper part occurred probably through fractional crystallization and/or immiscibility of the evolved Fe–Ti-rich magma, resulting in the paragenesis of Fe–Ti–V oxides and Ni–Cu–PGE sulfides in the Xinjie layered intrusion.

Mineral chemistry and thermobarometry of Eocene monzogabbroic stocks from the Bafra (Samsun) area in Turkey: implications for disequilibrium crystallization and emplacement conditions

Monzogabbro stocks including felsic enclaves (monzosyenite) around the Bafra (Samsun) area at the western edge of the Eastern Pontides cut Eocene-aged volcanic and sedimentary units. The monzogabbros contain plagioclase, alkali feldspar, clinopyroxene, olivine, hornblende, biotite, apatite, and iron-titanium oxides, whereas the felsic enclaves contain alkali feldspar, plagioclase, hornblende, biotite, clinopyroxene, and iron-titanium oxides. Mineral chemistry data suggest that magmas experienced hydrous and anhydrous crystallization in deep and shallow crustal magma chambers. Several thermobarometers were used to estimate temperatures of crystallization and emplacement for the mafic and felsic magmas. Clinopyroxene thermobarometry yielded 1100–1232 C and 5.9–8.1 kbar for monzogabbros, and 931–1109 C and 1.8–6.9 kbar for felsic enclaves. Hornblende thermobarometry and oxygen fugacity estimates reveal 739–971°C, 7.0–9.2 kbar and 10−9.71 for monzogabbros and 681–928°C, 3.0–6.1 kbar and 10−11.34 for felsic enclaves. Biotite thermobarometry shows elevated oxygen fugacity varying from 10−18.9–10−11.07 at 632–904°C and 1.29–1.89 kbar for monzogabbros, to 10−15.99 –10−11.82 at 719–873°C and 1.41–1.77 kbar for felsic enclaves. The estimated zircon and apatite saturation temperatures are 504–590°C and 693–730°C for monzogabbros and 765–775°C and 641–690°C for felsic enclaves, respectively. These data imply that several phases in the gabbroic and syenitic magmas did not necessarily crystallize simultaneously and further indicate that the mineral compositions may register intervals of disequilibrium crystallization. Besides, thermobarometry contrasts between monzogabbro and felsic enclave may be partly a consequence of extended interactions between the mafic and felsic magmas by mixing/mingling and diffusion. Additionally, the hot felsic magma was close to liquidus conditions (crystallinity < 30%) when injected into cooler mafic magma (crystallinity > 50%), and thus, the monzogabbro stocks reflect hybrid products from the mingling and incomplete mixing of these two magmas.

Origin of the ca. 90 Ma magnesia-rich volcanic rocks in SE Nyima, central Tibet: Products of lithospheric delamination beneath the Lhasa-Qiangtang collision zone

Bulk-rock major and trace element, Sr–Nd–Hf isotope, zircon U–Pb age, and zircon Hf isotopic data of the Late Cretaceous Zhuogapu volcanic rocks in the northern Lhasa subterrane provide a new insight into tectonic processes following the collision of the terrane with the Qiangtang zone. SHRIMP zircon U–Pb dating reveals that the Zhuogapu volcanic rocks crystallized at ca. 91Ma, postdating the development of a regional angular unconformity between the Upper Cretaceous and the underlying strata in the Lhasa–Qiangtang collision zone. Compared to the Andean arc-type andesites and dacites, the Zhuogapu volcanic rocks are characterized by higher MgO of 2.78–5.86wt.% and Mg# of 54–64 for andesites and MgO of 2.30–2.61wt.% and Mg# of 55–58 for dacites. Eight andesite samples have whole-rock (87Sr/86Sr)i of 0.7054–0.7065, εNd(t) of −3.2 to −1.7, and εHf(t) of +3.8–+6.4, similar to those of the three dacite samples with (87Sr/86Sr)i=0.7056–0.7060, εNd(t) of −2.7 to −2.2, and εHf(t) of +5.6–+7.0. Thirteen analyses from a dacite sample give positive zircon εHf(t) of +5.6 to +8.7. These signatures indicate that the Zhuogapu Mg-rich andesites were most likely derived from partial melting of a delaminated mafic lower crust (including the lowermost crust straddling the northern and central Lhasa subterranes) that led to the generation of the Zhuogapu primary melts with adakitic signatures and small negative εNd(t). Such melts subsequently experienced interaction of melt-asthenospheric mantle peridotite followed by the modification of highly fractionated magmas in shallow crustal magma chamber. Hornblende-controlled fractionation results in the change of geochemical composition from Mg-rich andesitic to Mg-rich dacitic magmas. Field observations, together with geochronological and geochemical data, indicate that the Zhuogapu Mg-rich volcanic rocks and coeval magmatism in the northern Lhasa subterrane may be the result of thickened lithospheric delamination following the final Lhasa–Qiangtang amalgamation.

Magma flow and palaeo-stress deduced from magnetic fabric analysis of the Álftafjörður dyke swarm: implications for shallow crustal magma transport in Icelandic volcanic systems

Abstract Neogene regional mafic dykes extending north of the Álftafjörður central volcano in east Iceland are studied to test models of dyke swarm emplacement at spreading ridges. This is accomplished by using anisotropy of magnetic susceptibility to define fossilized magma flow regimes. The imbrication of the foliation plane, defined by the minor susceptibility axis, is used as an indicator of the flow direction. Contemporaneous shear resolved on the dyke walls may modify a pure flow-induced fabric and such shear regimes are therefore retracted. The magma flow and palaeo-stress resolved on the dykes are determined in 13 of 24 dykes. The magma flow is interpreted as subhorizontal and northwards directed away from the central volcano for nine dykes, and found to be vertical in three cases. The preferentially subhorizontal magma flow in the Álftafjörður swarm suggests that dyke propagation in this type of Icelandic volcanic system originates in shallow crustal magma chambers. The regional tectonic palaeo-stress field is deduced to cause oblique spreading across the Álftafjörður dyke swarm and govern a subhorizontal dextral shear component on the dyke planes during propagation. This interpretation is not in conflict with the left-stepping en echelon trend distribution of individual dykes relative to the trend of the swarm.

Volcanic facies and mineral chemistry of Tertiary volcanics in the northern part of the Eastern Pontides, northeast Turkey: implications for pre-eruptive crystallization conditions and magma chamber processes

Tertiary volcanics in the northern zone of the Eastern Pontides are characterized by subaerial and shallow-subaqueous facieses, and are divided into three volcanic suites: Eocene aged (1) basalt-trachybasalt-basaltic trachyandesite (BTB) and (2) trachyte-trachyandesite (TT), and Miocene aged (3) basanite-tephrite (BT) suites. Clinopyroxene is a common phase in all three volcanic suites, and has different compositions with Mg# varying from 0.57 to 0.91 in BTB suite and 0.57–0.84 in TT suite to 0.65–0.90 in BT suite. Feldspars in all suites generally exhibit wide range of compositions from sanidine to albite or anorthite and have weak normal and reverse compositional zoning. Olivines in BTB and BT suites have Fo60–92. Hornblendes in BTB, TT and BT suites are commonly magnesio-hastingsite and rare pargasite in composition (Mg#: 0.67–0.90). Brown mica is mainly phlogopite with Mg# ranging from 0.56 to 0.92 in the BTB suite, 0.59–0.84 in the TT suite, and 0.75–0.93 in the BT suite. Analcime is present only in the BT suite rocks. Fe–Ti oxides in all suites are mainly composed of magnetite and titanomagnetite. Textural petrographic and mineral chemical data suggest that magmas had undergone hydrous and anhydrous crystallizations in deep-, mid-, and shallow-crustal magma chambers. Clinopyroxene thermobarometric calculations show that Eocene magma chambers were characterized by temperature ranging from 1,100 to 1,244 °C and pressure ranging from 1.84 to 5.39 kbar. Similarly, the Miocene magma chambers were characterized by temperature ranging from 1,119 to 1,146 °C and pressure ranging from 4.23 to 4.93 kbar. Hornblende thermobarometry, oxygen fugacity, and hygrometer reveal that the crystallization temperature of Eocene volcanics range from 956 to 959 °C at pressure ranging from 6.49 to 6.52 kbar. Eocene volcanics were characterized by water content ranging from 7.83 to 8.57 wt.% and oxygen fugacity of 10−9.36 to 10−9.46 (ΔNNO+2). Miocene volcanics had crystallization temperature ranging from 970 to 978 °C at pressure ranging from 8.70 to 9.00 kbar with water content ranging from 8.04 to 8.64 wt.% and oxygen fugacity ranging from 10−8.75 to 10−8.87 (ΔNNO+2). Brown mica thermobarometric data show that Eocene volcanics were characterized by relatively high oxygen fugacity varying from 10-10.32 to 10-12.37 (HM) at temperature ranging from 858 to 953 °C and pressure ranging from 1.08 to 1.41 kbar. Miocene volcanics were crystallized at highly oxidized conditions, which are characterized by high oxygen fugacity of 10−12.0 (HM) at temperature of 875 °C and pressure of 2.09 kbar. The wide range of obtained temperatures for clinopyroxenes of the suites denotes that the equilibration of clinopyroxene crystals initiates from depth until close to the surface before magma eruption. The compositional variations, resorbed core and reverse zoning patterns in clinopyroxene phenocrysts, as well as variable pressures of crystallization, further indicate that the magmas that formed the suites were polybaric in origins and were composite products of more than one petrogenetic stage. The observed range of phenocryst assemblage and different compositional trends possibly originated from fractionation of magmas with different initial water contents under variable pressures of crystallization. The repeated occurrence of magmas from different suites during a single period of activity suggests that the magmatic system consists of several conduit systems and that magma reservoirs are dispersed at different levels of crustal magma chambers.

Modeling the compositional evolution of recharging, evacuating, and fractionating (REFC) magma chambers: Implications for differentiation of arc magmas

Equations are presented to describe the compositional evolution of magma chambers undergoing simultaneous recharge (R), evacuation (E), and fractional crystallization (FC). Constant mass magma chambers undergoing REFC will eventually approach a steady state composition due to the “buffering” effect of recharging magma. Steady state composition is attained after ∼3/(Dαx+αe) overturns of the magma chamber, where D is the bulk solid/melt partition coefficient for the element of interest and αx and αe are the proportions of crystallization and eruption/evacuation relative to the recharge rate. Steady state composition is given by Cre/(Dαx+αe). For low evacuation rates, steady state concentration and the time to reach steady state scale inversely with D. Compatible (D>1) elements reach steady state faster than incompatible (D<1) elements. Thus, magma chambers undergoing REFC will eventually evolve towards high incompatible element enrichments for a given depletion in a compatible element compared to magma chambers undergoing pure fractional crystallization. For example, REFC magma chambers will evolve to high incompatible element concentrations for a given MgO content compared to fractional crystallization. Not accounting for REFC will lead to over-estimation of the incompatible element content of primary magmas. Furthermore, unlike fractional crystallization alone, REFC can efficiently fractionate highly incompatible element ratios because the fractionation effect scales with the ratio of bulk D’s. By contrast, in pure fractional crystallization, ratios fractionate according to the arithmetic difference between the bulk D’s.The compositional impact of REFC should be most pronounced for magma chambers that are long-lived, have low rates of eruption/evacuation, and/or are characterized by high recharge rates relative to the mass of the magma chamber. The first two conditions are likely favored in deep crustal magma chambers where confining pressures are high and warm country rock decrease the cooling rates of magma chambers. By contrast, REFC should be less significant in shallow crustal magma chambers, which erupt and cool more efficiently due to lower confining pressures, colder country rock, and the cooling effects of hydrothermal systems. We thus speculate that the effects of REFC will be small in mid-ocean ridge settings and most pronounced in arc settings, particularly mature island arcs or continental arcs, where magma chambers >10km depth are possible. This begs the question of whether high Fe3+, H2O and CO2 (all of which can be treated as incompatible “elements”) in arc basalts could be enhanced by REFC processes and thus not just reflect inheritance from the mantle source. We show that REFC can plausibly explain observed enrichments in Fe3+ and H2O in arc melts without significant depletion in MgO. Because the difference between calc-alkaline and tholeiitic differentiation series is mostly likely due to higher water and oxygen fugacity in the former, it may be worth considering the effects of REFC. Thus, if REFC is more pronounced in deep crustal magma chambers, mature island arcs and continental arcs would tend towards calc-alkaline differentiation, whereas juvenile island arcs would be more tholeiitic.To fully test the significance of REFC will require detailed analysis of other highly incompatible elements, but presently the relative differences in bulk D of such elements may not be constrained well enough. The equations presented here provide a framework for evaluating whether REFC should be considered when interpreting geochemical data in differentiated magmas. For completeness, we have also provided the more general equations for a magma chamber undergoing recharge (R), evacuation (E), crustal assimilation (A) and fractional crystallization (FC), e.g., REAFC, for constant mass, growing and dying magma chambers.

Catastrophic caldera-forming eruptions: Thermomechanics and implications for eruption triggering and maximum caldera dimensions on Earth

Approximately every 100,000years the Earth experiences catastrophic caldera-forming “supereruptions” that are considered to be one of the most hazardous natural events on Earth. Utilizing new temperature-dependent, viscoelastic numerical models that incorporate a Mohr-Coulomb failure criterion, we find that eruptive failure of the largest magma chambers is a function of the geometry of the overlying roof and the location of the brittle-ductile transition. In particular, the ductile halo created around the hot magma chamber buffers increasing overpressures and prevents pressure relief via magmatic injection from the magma chamber. The numerical results indicate that as chamber volume increases, the higher temperatures in the host rock and the decrease in the roof aspect ratio cause a shift from reservoir-triggered eruption to an external roof-triggered mechanism. Specifically, as overpressure increases within the largest magma chambers, extensive uplift in the overlying roof promotes the development of through-going faults that may trigger eruption and caldera collapse from above. We find that for magma chamber volumes>103km3, and roof aspect ratios (depth/width)<0.3, moderate magma chamber overpressures (<30MPa) will cause extensive through-going fault development in the overlying roof. This result indicates an external mechanism, caused by fault propagation in the roof, is a likely trigger for the largest caldera forming eruptions. The thermomechanical models also provide an estimate of the maximum size of magma chamber growth in a pristine host material and, thus, an estimate of the maximum size of the resultant caldera. We find a maximum reservoir volume range of 104–105km3 for shallow crustal magma chambers emplaced at depths to the top of the magma chamber of 3–7km. These volumes produce maximum caldera areas of 103–104km2, comparable to the largest calderas observed on Earth (e.g., Toba). These thermomechanical models offer critical new insight into the mechanics of catastrophic caldera collapse and provide a numerical construct for predicting how eruption is triggered in the largest crustal magma chambers.

The invisible hand: Tectonic triggering and modulation of a rhyolitic supereruption

Identifying the processes that initiate and control large explosive eruptions on time scales of human interest is challenging. We present evidence for syneruptive, subsurface interactions between rhyolite magmas derived from distinctive sources that reflect tectonic processes starting, then temporarily stopping, the 27 ka Oruanui supereruption (530 km3 magma; Taupo, New Zealand). Subordinate (3%–16%) biotite-bearing pumice lapilli occur in fall deposits from phases 1 and 2 (of 10) among the biotite-free “normal” pumices in the Oruanui deposits. The biotite-bearing magma is sourced to an independent magmatic system 10–15 km northeast of Taupo. Lateral migration of the biotite-bearing magma was initiated during a rifting episode that also triggered the Oruanui eruption. Limited physical mingling indicates that the biotite-bearing and Oruanui biotite-free magmas interacted in the conduit simultaneously with eruption. After phase 1, a time break of months elapsed before phase 2 explosive activity resumed from the same vent, again with intermixed biotite-bearing magma. A rift-related tectonic event, with its accompanying stress change, is inferred to have initiated diking from northeast to southwest, transporting biotite-bearing magma and triggering the Oruanui eruption. Stress relaxation following the rifting event terminated phase 1 before a tectonic event caused renewed diking and activity in phase 2. Tectonics may trigger large silicic eruptions through rupturing of shallow crustal magma chambers and lateral magma transport, especially in extensional regimes, without necessarily leaving any discernible geological imprint. Only the presence of a “foreign” magma in the Oruanui deposits allows the tectonic control to be identified.