Abstract
Felsic magmas produced at subduction zones have played an important role in the generation and evolution of the continental crust. For the origin of felsic magmas, processes such as fractional crystallization of mafic magmas, partial melting of crustal materials, partial melting of subducting slabs, and partial melting of pyroxenitic mantle wedge components have been proposed. Recent experimental studies have predicted that felsic melt can also be produced in the mantle wedge by the separation of slab-derived supercritical liquid beyond depths corresponding to the critical point. To date, however, the presence of felsic magma of this origin has not yet been reported. In this study, we investigated dacitic lavas and preceding calc-alkaline andesite lavas from the Rishiri Volcano, located at the rear of the Kuril arc. We show that hydrous felsic melt and aqueous fluid were separated from slab-derived supercritical liquid in the mantle wedge. The former erupted as dacitic magma whilst the aqueous fluid induced the generation of primary basaltic magma involved in creating calc-alkaline andesite magma. We infer that slab-derived supercritical liquid is an efficient transport medium for moving silicate-rich components from subducting slabs to the Earth’s surface, and that this process may have contributed to the growth of the continental crust.
Similar content being viewed by others
Introduction
Subduction-zone magmatism is considered to have played an important role in the generation and evolution of the continental crust, because the chemical composition of calc-alkaline andesites that are widespread in island arcs and active continental margins is similar to the average composition of the continental crust1,2. Primary mafic magmas are thought to be generated by melting of the mantle wedge with an influx of slab-derived water-rich materials3,4. Intermediate to felsic magmas (andesite to rhyolite) are also commonly produced in many subduction zones. Some felsic magmas have been emplaced as large granitic plutons, which contribute to the establishment of a buoyant continental crust5,6, and others have mixed with mafic magmas to produce calc-alkaline andesitic magmas and other intermediate magmas in crustal magma chambers7, producing geochemical variability in the crust2.
It has been suggested that felsic magmas found at subduction-zone volcanoes are produced by mechanisms including fractional crystallization with or without crustal assimilation8, partial melting of the crust9,10, and partial melting of metasomatised silica-excess pyroxenite mantle11,12. In addition, adakitic magmas with high-Sr/Y ratios13 and high [La/Yb]N (primitive mantle-normalised La/Yb) ratios14,15 can also be produced by partial melting of hot subducting slabs at young subduction zones13,14,15 and by partial melting of garnet-bearing lower part of the thick crust16,17.
Recently, experimental studies have suggested that intermediate to felsic magma may also be produced as hydrous melt through the separation of slab-derived supercritical liquids in the sub-arc mantle. At relatively shallow depths, the slab-derived materials are released as aqueous fluid or hydrous melt. However, with increasing depths of fluid release, the solubility of silicate materials in aqueous fluid tends to become close to those of water in silicate melt, and the slab-derived water-rich materials would eventually become supercritical liquids at the critical point. On the other hand, ascending slab-derived supercritical liquid would separate into aqueous fluid and hydrous melt at the critical point18,19. Because the pressure corresponding to the critical end-point for silicic supercritical liquid can be as low as <3 GPa, andesitic to dacitic magmas would be generated by the separation of slab-derived supercritical liquid in sub-arc mantle20. To date, however, the presence of magmas originating from this process has not yet been reported in nature.
Rishiri Volcano is a Quaternary stratovolcano at the rear of the southern Kuril arc, located 300 km above the Wadati-Benioff Zone (Fig. 1a). Previous studies have suggested that slab-derived supercritical liquid has been released from the subducting slab21,22, and that some dacitic lavas have adakitic geochemical signatures23. In this study, we conducted a detailed petrological, geochemical, and chronological study on the dacites. We suggest that the dacitic magmas represent hydrous melts generated through separation of slab-derived supercritical liquid.
Results
The Rishiri Volcano is characterised by the coexistence of alkali basalt, tholeiitic andesite to dacite, and calc-alkaline andesite to dacite24 (Fig. 2a and Supplementary Fig. S1a). Its volcanic activity can be divided into five stages: the Early-1, Early-2, Middle, Late-1, and Late-2 Stages25. The eruption rates of the individual stages are estimated to be >0.1 km3/ky, >0.4 km3/ky, >0.4 km3/ky, >0.35 km3/ky, and >0.09 km3/ky, respectively25. The main target of this study are the calc-alkaline dacitic lavas, which constitute the main edifice of the volcano together with the calc-alkaline andesite lavas and pyroclastics25 (Fig. 1b,c). These dacites and andesites belong to the Middle Stage of volcanic activity, characterised by the highest eruption rate over the lifetime of the volcano25. The petrogenesis of the calc-alkaline andesites was investigated in detail by ref. 23. The magmas were suggested to have been produced via mixing between crust-derived felsic magmas and primitive basaltic magmas generated through fluid-fluxed melting of the mantle. The dacitic lavas always overly the andesitic lavas in the area (Supplementary Fig. S2), indicating that the eruption of the andesitic magma predated that of the dacitic magma. New 40Ar/39Ar dating shows that activity leading to production of the dacite and andesite took place at 35.5 ± 1.4 and 34.6 ± 3.0 ka, respectively (Table S1 and Supplementary Fig. S3).
The phenocryst assemblage of the dacite lavas is composed of clinopyroxene, orthopyroxene, and plagioclase, some of which form crystal aggregates (Supplementary Fig. S4). All phenocrysts are clear under a polarised light microscope, and sieve-textured plagioclase is not found. The phenocrysts are homogeneous or normally zoned, with limited variation in their core composition (Fig. 3). The whole-rock SiO2 content of the dacites ranges from 64.1 to 65.3 wt.% (Table S2, Fig. 2a and Supplementary Fig. S1). The dacites and the preceding andesites belong to the calc-alkaline series in the Miyashiro diagram26 (Fig. 2a). The dacitic lavas have a relatively high MgO (2.3–2.7 wt.%; Supplementary Fig. S1e), Cr (25–42 ppm; Supplementary Fig. S1f), and Ni (26–39 ppm) content compared with other felsic lavas from the Rishiri Volcano. The Mg#’s [= Mg/(Mg + Fetotal) × 100] of the dacites of 52–54 are similar to the average continental crust (~55; ref. 2), and are lower than those of mantle-derived primary melt (>60). The dacitic lavas are also characterised by high Sr/Y (48–55) and [La/Yb]N (9–11) ratios, and low YbN (2.5–3.0) (i.e., adakitic signature13,14) compared with the andesitic lavas (Fig. 2b–d), and have been classified as high-SiO2 adakite15 (Fig. 2b and Supplementary Fig. S1e). No significant Eu anomaly is observed in the dacite lavas (Fig. 2d), with the exception of one outlier sample with extremely low P2O5, Y, and Yb content (Supplementary Fig. S5). The dacitic lavas have higher 87Sr/86Sr, 208Pb/204Pb and 206Pb/204Pb ratios and lower 143Nd/144Nd ratios than both the calc-alkaline andesitic and basaltic lavas from the volcano (Fig. 4a,b). The Pb isotopic compositions of the dacitic lavas are mostly on the binary mixing line between sediments (SED) and altered oceanic crust (AOC) (Fig. 4b). In a U–Th equiline diagram, age corrected compositions of the dacitic and andesitic lavas (Table S3) plot on the left-hand side of the equiline (Fig. 4c).
Discussion
All phenocrysts in the dacite lavas are clear under a polarised light microscope, and essentially homogeneous in composition or slightly normally zoned with unimodal distributions in their core composition (Fig. 3). In addition, sieve-textured plagioclase is not found. These observations suggest that the dacitic magma did not undergo mixing after phenocryst crystallisation. The phenocryst phases occasionally form crystal aggregates, showing evidence of simultaneous growth, while no Eu anomaly is evident in whole-rock samples (Fig. 2d). In addition, the 87Sr/86Sr ratios of the plagioclase phenocrysts (Table S4) are identical to those of the host lavas (Fig. 4a). This indicates that all the phenocrysts were grown in situ in the dacitic magmas with no Eu anomaly.
The dacitic lavas are characterised by high Sr/Y and [La/Yb]N ratios (Fig. 2b–d). One possible origin of dacitic melt with an adakitic signature is crystal fractionation, specifically of hornblende and garnet, from hydrous basaltic magmas27 or from calc-alkaline andesite magmas (with relatively lower [La/Yb]N ratios; Fig. 2d) that erupted just before the dacite magmas. If this was the case, the dacitic melt would have been in equilibrium with hornblende or garnet at depth. The water content and temperature of the dacitic melt are estimated to be 5.3 ± 1.1 wt.% and 980 °C, respectively, at 0.46 GPa, using plagioclase-melt equilibrium (see Methods). Experimental studies27 showed that, for a dacitic melt with major element composition similar to that of the Rishiri dacite, the melt with ~5.3 wt.% H2O cannot be in equilibrium with hornblende unless the temperatures are lower than 930 °C at 0.4 GPa and 900 °C at 0.96 GPa. The melt cannot crystallize garnet at any temperatures at ≤0.96 GPa. Therefore, it is not likely that hornblende and/or garnet crystallised as a primary liquidus phase from the dacitic melt with ~980 °C in the crust (<~0.8 GPa; ref. 28) beneath Rishiri. This is supported by the absence of hornblende and garnet phenocrysts in the dacitic lavas, as well as in the calc-alkaline andesite lavas. Therefore, we conclude that the dacitic magmas were not derived from hydrous basaltic magmas or calc-alkaline andesite magmas through crystal fractionation.
It has also been recognised that felsic magmas with an adakitic signature can be produced by the partial melting of the lower part of the thick (>30 km) crust, consisting of garnet-bearing mafic rocks16,17. However, the adakitic signature could not be produced in the crust beneath the Rishiri Volcano, because the crust is too thin (~25 km; ref. 28) for the garnet to be stable29,30. The isotopic composition of dacitic lavas are found to be significantly different from the crustal xenoliths at the volcano31 (Fig. 4a,b), also negating the potential for the dacite magma to have originated from partial melting of the crust.
Based on these considerations, we conclude that the dacitic melt with adakitic signature could not have been derived from crustal magmatic processes but instead originated in the upper mantle. As it is unlikely that primary melt with dacitic composition was produced by the partial melting of mantle peridotite, the possible origin of the dacitic melt would be (1) partial melting of the subducting slab13,14,15, (2) partial melting of a pyroxenite mantle source with an influx of slab-derived fluid11,12, or (3) separation of slab-derived supercritical liquid18,19,20. These hypotheses are evaluated below.
If scenario (1) was the case, extremely high temperature would have been required to melt the slab, because the depth of the slab is ~300 km beneath Rishiri. Experimental studies have suggested that extremely high temperatures of >1300 °C are required to induce hydrous partial melting of the slab at the depth of 300 km32,33. However, the slab surface temperature are estimated to be ~1000 °C at 300 km depth, based on the extrapolation of the subduction thermal model (D80)34. In addition, evidence for temperature elevation in the slab, such as a slab window or slab tear, has not been found beneath the Rishiri Volcano35,36, suggesting that this scenario (1) is unlikely.
In scenario (2), the generation of dacitic melt is explained by partial melting of silica-excess pyroxenites, formed by the reaction of a peridotite mantle with infiltrated silicic slab components, in the sub-arc mantle11,12. In this case, it is expected that the primary dacitic melt would have high Ni contents11,12. However, the Rishiri dacites have significantly lower Ni contents of 26–39 ppm than those of mantle-derived primary magma (>~200 ppm). High-Ni olivines, that characteristically occur in magmas generated by the melting of a pyroxenite mantle source11,12, are also not observed in the dacite lavas. The Pb isotopic ratios of the dacite lavas can be explained by binary mixing between AOC component and SED component with little contribution of the depleted-MORB mantle (DMM) component (Fig. 4b); this observation also suggests that the mantle component was not significantly involved in the genesis of the dacites. For these reasons, we conclude that scenario (2) cannot explain the genesis of the dacitic magmas at Rishiri.
At the depth of the subducting slab of ~300 km at the Rishiri Volcano, supercritical liquid rather than aqueous fluid is expected to be released from the slab20,32,33. In fact, some primitive alkali basalt magmas at Rishiri are considered to have been generated by the melting of the source mantle with an influx of slab-derived supercritical liquid21,22. These alkali basalt lavas show 230Th-excesses with respect to 238U21. This is contrary to the U-excess signatures of frontal-arc lavas, which reflect a much higher mobility of U in slab-derived aqueous fluids than Th. Slab-derived supercritical liquids produce a 230Th-excess21 because of a preferential partition of Th relative to U37,38 during the dehydration of the subducting slab. The dacite lavas, as well as the preceding calc-alkaline andesitic lavas, show Th-excesses (Fig. 4c), which is consistent with the inference that slab-derived supercritical liquid was involved in the generation of both the dacite and andesite magmas. The eruption ages of the andesitic and dacitic lavas of 34.6 ± 3.0 and 35.5 ± 1.4 ka, respectively, and the stratigraphic relationship between the two indicate that the andesitic magmas erupted just before the dacitic magmas. In addition, the primitive basaltic magmas involved in the generation of the calc-alkaline andesitic magmas are considered to have been generated by the influx of slab-derived fluids into the source mantle23. These observations suggest that, in the upper mantle directly under the volcano, slab-derived fluid was present alongside the dacitic melt. As aqueous fluid and hydrous silicate melt cannot be produced from a subducting slab at an identical temperature32,33, these two components were most plausibly generated by the separation of ascending slab-derived supercritical liquid in the subarc mantle (i.e., scenario 3). The similarity of 238U–230Th disequilibria in the andesite and dacite lavas is also consistent with the derivation of andesites and dacites by essentially the same slab-derived supercritical liquid. The adakitic signature of the dacites (i.e., the silicate melt component in the slab-derived supercritical liquid) is considered to reflect the presence of residual garnet in the slab, because the basaltic oceanic crust is in the eclogite-facies field at a depth of ~300 km39.
From these considerations, this study concludes that the dacitic melt was most plausibly produced by the separation of slab-derived supercritical liquid18,19,20, rather than by the partial melting of the subducting slab13,14,15 or the partial melting of the pyroxenite mantle11,12. Slab-derived supercritical liquid ascending in the mantle wedge separated to produce hydrous melt and aqueous fluid beyond depths corresponding to the critical point. The former is represented by the dacitic melt and the latter induced generation of the primary basaltic magma involved in the generation of the calc-alkaline andesite magma (Fig. 5). It is difficult to estimate the depth at which the superciritical liquid was separated into the aqueous fluid and hydrous felsic melt, because of the scarcity of experimental studies. However, the high-K/Na alkali basalt magmas (Nozuka lavas) at Rishiri are considered to have been generated at ~2.3 GPa through an influx of slab-derived supercritical liquid into the melting region22. This observation suggests that the depth of the critical point for the supercritical liquid would have been shallower than the depth corresponding to ~2.3 GPa. If so, the separation of the slab-derived supercritical liquid for the dacite magmas might have occurred at depths corresponding to <~2.3 GPa.
The concentrations of fluid-mobile elements, such as Rb, Sr, Ba, and Pb, are significantly depleted in the dacitic lavas compared with the estimated concentrations of possible slab-derived supercritical liquid (see Supplementary Methods and Fig. S6). For example, the Rb/Zr and Pb/Hf ratios of the dacitic lavas of ~0.2 and ~1.6, respectively, are much lower than those of the supercritical liquid at 1.5 and 5.3, respectively. These features are consistent with preferential partitioning of fluid-mobile elements into an aqueous fluid relative to the remaining hydrous silicate melt during the separation of supercritical liquid40, although these features may also be attributed to a low content of fluid-mobile elements in the slab-derived supercritical liquid, which would have resulted from depletion in the subducting slab by removal through dehydration processes before reaching ~300 km depth.
Based on experimental studies on the behavior of slab-derived supercritical liquid18,19, Kawamoto et al.20 suggested the generation of two different types of primary mafic magmas in the mantle wedge beyond depths corresponding to the critical points; mafic magma from melting of the mantle with an influx of separated aqueous fluid and mafic magma from a reaction of separated hydrous melt with mantle peridotite. The occurrence of mafic magmas of these origins was suggested at the Mariana arc41. As such, this study suggests for the first time that hydrous felsic melt generated through the separation of slab-derived supercritical liquid can erupt without significant interaction with the mantle wedge.
Felsic magmas with high Sr/Y and [La/Yb]N ratios have occurred as adakites in modern hot subduction zones13. It has been experimentally suggested that the minimum depth at which a supercritical liquid can occur in subducting slabs tends to become shallower with increasing proportions of sediment components in the slab20. The proportion of sediment components in adakites, such as those from the Southwestern Japan arc above the Philippine Sea slab, are ~40%42, significantly higher than studied dacites of ~5% (Supplementary Methods) above the Pacific slab. Moreover, the Philippine Sea slab is significantly hotter than the Pacific slab43. Thus, the supercritical liquid may have occurred at much shallower levels in the hot modern subducting slabs such as the Philippine Sea slab than the Pacific slab beneath the Rishiri Volcano. If so, some modern adakitic magmas may represent felsic magmas generated through the separation of slab-derived supercritical liquid.
Supercritical liquid may ascend in the mantle wedge more effectively than silicate melt, because of the reduced viscosity. The large 230Th-excess (Fig. 4c) of dacites from the Rishiri Volcano is consistent with the rapid transport of slab materials as supercritical liquid, although the timescale of transport cannot be estimated due to the uncertainty of the U/Th ratio of the slab source materials. We propose that slab-derived supercritical liquid is an efficient transport medium for silicate-rich components from the subducting slab through the mantle wedge to the surface, and it may have contributed to the growth of the continental crust.
Methods
40Ar/39Ar age analyses
For reliable 40Ar/39Ar dating, we collected samples with holocrystalline groundmass. The 40Ar/39Ar age for groundmass of dacite and preceding andesite are obtained at the Argon Geochronology Laboratory, Oregon State University via incremental heating of hand-picked holocrystalline groundmass samples with laser irradiation. The 40Ar/39Ar isotopes are analysed by a multi-collector mass spectrometer (Thermo Scientific Model ARGUS-VI) and decay constants in this study from ref. 44 with 2σ internal errors.
Whole-rock compositional analyses
Concentrations of whole-rock major elements and some trace elements (Sc, V, Cr, Co, and Ni) are determined via X-ray fluorescence (XRF) using a Spectoris MagiX PRO at the Graduate School of Science, Hokkaido University, and a Phillips PW2400 at the Pheasant Memorial Laboratory, Institute for Planetary Materials, Okayama University at Misasa45. Additional trace elements are analysed through inductively coupled plasma mass spectrometry (ICP-MS), using a Thermo Fisher Scientific X-series instrument at Hokkaido University and a quadrupole-type Yokogawa Agilent 7500cs system at Okayama University. The isotopic ratios for Sr, Nd, and Pb are determined using a multiple collector (MC)-ICP-MS (Neptune plus, Thermo Fisher Scientific) at Hokkaido University and a thermal ionisation mass spectrometer (TIMS) (Finnigan MAT 262) at Okayama University. Isotope analyses of U and Th are conducted by a TIMS (Finnigan MAT262 with RPQplus) at Okayama University. In this paper, isotopic ratios in parentheses represent activity ratios. Decay constants of the U and Th nuclides46,47,48 used for calculations in this study are λ238U = 1.55125 × 10−10, λ234U = 2.8263 × 10−6, λ232Th = 4.9475 × 10−11, and λ230Th = 9.158 × 10−6. The details of the analytical procedures at Okayama University, including those of the chemical separations and data correction, reflect those presented by ref. 21, and those at Hokkaido University are essentially similar to those described in ref. 49. The composition of the Geological Survey of Japan reference material JB-3 is measured during the course of this study. Measured and reference values are listed in Table A1 of ref. 49.
Electron microprobe analyses
Mineral compositions are determined using a JEOL JXA-8800 electron microprobe at Hokkaido University. For mafic minerals, an accelerating voltage of 15 kV, a beam current of 20 nA, and a counting time of 20 s are adopted. An accelerating voltage of 15 kV, a beam current of 10 nA, a counting time of 10 s, and a beam diameter of 10 μm are the operating conditions used for the plagioclase. Both oxide and natural mineral standards are used, and data are obtained using the ZAF correction method.
Water contents of the dacitic melt
The water content of the dacitic magma is estimated using the plagioclase-melt hygrometer given in ref. 50. Plagioclase phenocrysts in the dacitic lavas are considered to have crystallised in situ, as it is observed that the lavas have no Eu anomaly (Fig. 2d), and because the 87Sr/86Sr ratios of the plagioclase phenocrysts are identical to those of the lavas (Fig. 4a). Therefore, it is assumed that the plagioclase phenocrysts with the highest An contents of 63.8 are in equilibrium with the melt that had a whole-rock composition of OS-2. The pressure condition is estimated using the compositions of the clinopyroxene and orthopyroxene phenocrysts, suggested to have grown simultaneously. By applying the two-pyroxene geobarometer given in ref. 50, we obtained a crystallisation pressure of 0.46 ± 0.4 GPa. The water content of the melt and the crystallisation temperature of plagioclase phenocrysts are then estimated by simultaneously solving Equations (25b) and (26) given in ref. 50 at a pressure of 0.46 GPa, to obtain a water content and a temperature of 5.3 ± 1.1 wt.% and 980 °C, respectively. The water contents of the dacites are also estimated by the plagioclase-melt hygrometer given in ref. 51, yielding 4.3 wt.%. This result is within the ranges of the estimates of 5.3 ± 1.1 wt.%, obtained by the hygrometer given in ref. 50.
References
Taylor, S. R. & White, A. J. R. Geochemistry of andesites and the growth of continents. Nature 208, 271–273 (1965).
Rudnick, R. L. & Gao, S. Composition of the Continental Crust in The Crust (ed. Rudnick, R. L.) Treatise on Geochemistry 3, 1–64 (Oxford, 2003).
Gill, J. B. Orogenic Andesites and Plate Tectonics . (Springer, 1981).
Elliott, T. Tracers of the slab in Inside the Subduction Factory (ed. Eiler, J.) Geophys. Monogr. 138, 23–45 (AGU, 2003).
Arndt, N. T. Formation and evolution of the continental crust. Geochem. Perspect 2, 405–533 (2013).
Petford, N., Cruden, A. R., McCaffrey, K. J. W. & Vigneresse, J.-L. Granite magma formation, transport and emplacement in the Earth’s crust. Nature 408, 669–673 (2000).
Eichelberger, J. C. Origin of andesite and dacite: evidence of mixing at Glass Mountain in California and at other circum-Pacific volcanoes. Geol. Soc. Am. Bull 86, 1381–1391 (1975).
DePaolo, D. J. Trace element and isotopic effects of combined wallrock assimilation and fractional crystallization. Earth Planet. Sci. Lett. 53, 189–202 (1981).
Huppert, H. E. & Sparks, R. S. J. The generation of granitic magmas by intrusion of basalt into continental crust. J. Petrol. 29, 599–624 (1988).
Annen, C., Blundy, J. D. & Sparks, R. S. J. The genesis of intermediate and silicic magmas in deep crustal hot zones. J. Petrol 47, 505–539 (2006).
Straub, S. M. et al. Formation of hybrid arc andesites beneath thick continental crust. Earth Planet. Sci. Lett. 303, 337–347 (2011).
Straub, S. M., LaGatta, A. B., Martin‐Del Pozzo, A. L. & Langmuir, C. H. Evidence from high-Ni olivines for a hybridized peridotite/pyroxenite source for orogenic andesites from the central Mexican Volcanic Belt. Geochem. Geophys. Geosyst. 9, Q03007 (2008).
Defant, M. J. & Drummond, M. S. Derivation of some modern arc magmas by melting of young subducted lithosphere. Nature 347, 662–665 (1990).
Martin, H. Effect of steeper Archean geothermal gradient on geochemistry of subduction-zone magmas. Geology 14, 753–756 (1986).
Martin, H., Smithies, R. H., Rapp, R., Moyen, J.-F. & Champion, D. An overview of adakite, tonalite-trondhjemite-granodiorite (TTG), and sanukitoid: relationships and some implications for crustal evolution. Lithos 79, 1–24 (2005).
Atherton, M. P. & Petford, N. Generation of sodium-rich magmas from newly underplated basaltic crust. Nature 362, 144–146 (1993).
Ma, Q., Zheng, J.-P., Xu, Y.-G., Griffin, W. L. & Zhang, R.-S. Are continental “adakites” derived from thickened or foundered lower crust? Earth Planet. Sci. Lett. 419, 125–133 (2015).
Shen, A. H. & Keppler, H. Direct observation of complete miscibility in the albite-H O system. Nature 385, 710–712 (1997).
Bureau, H. & Keppler, H. Complete miscibility between silicate melts and hydrous fluids in the upper mantle: experimental evidence and geochemical implications. Earth Planet. Sci. Lett 165, 187–196 (1999).
Kawamoto, T., Kanzaki, M., Mibe, K., Matsukage, K. N. & Ono, S. Separation of supercritical slab-fluids to form aqueous fluid and melt components in subduction zone magmatism. Proc. Natl. Acad. Sci. USA 46, 18695–18700 (2012).
Kuritani, T., Yokoyama, T. & Nakamura, E. Generation of rear-arc magmas induced by influx of slab-derived supercritical liquids: Implications from alkali basalt lavas from Rishiri Volcano, Kurile arc. J. Petrol. 49, 1319–1342 (2008).
Kuritani, T. & Nakagawa, M. Origin of ultra rear-arc magmatism at Rishiri Volcano, Kuril Arc. Geochem. Geophys. Geosyst. 17, 4032–4050 (2016).
Taniuchi, H., Kuritani, T. & Nakagawa, M. Generation of calc-alkaline andesite magma through crustal melting induced by emplacement of mantle-derived water-rich primary magma: Evidence from Rishiri Volcano, southern Kuril Arc. Lithos 354–355, 105362 (2020).
Ishizuka, Y. & Nakagawa, M. Petrological evolution of Rishiri volcano, northern Hokkaido, Japan [in Japanese with English abstract]. Jour. Mineral. Petrol. Econ. Geol. 94, 279–294 (1999).
Ishizuka, Y. Eruptive history of Rishiri Volcano, northern Hokkaido, Japan [in Japanese with English abstract]. Bull. Volcanol. Soc. Japan 44, 23–40 (1999).
Miyashiro, A. Volcanic rock series in island arcs and active continental margins. Am. J. Sci 274, 321–355 (1974).
Prouteau, G. & Scaillet, B. Experimental constraints on the origin of the 1991 Pinatubo dacite. J. Petrol. 44, 2203–2241 (2003).
Niu, X., Zhao, D., Li, J. & Ruan, A. P wave azimuthal and radial anisotropy of the Hokkaido subduction zone. J. Geophys. Res. Solid Earth 121, 2636–2660 (2016).
Rapp, R. P. & Watson, E. B. Dehydration melting of metabasalt at 8-32 kbar: Implications for continental growth and crust-mantle recycling. J. Petrol. 36, 891–931 (1995).
Alonso-Perez, R., Müntener, O. & Ulmer, P. Igneous garnet and amphibole fractionation in the roots of island arcs: experimental constraints on andesitic liquids. Contrib. Mineral. Petrol. 157, 541–558 (2009).
Kuritani, T., Kitagawa, H. & Nakamura, E. Assimilation and fractional crystallization controlled by transport process of crustal melt: Implications from an alkali basalt-dacite suite from Rishiri Volcano, Japan. J. Petrol 46, 1421–1442 (2005).
Mibe, K., Kawamoto, T., Matsukage, K. N., Fei, Y. & Ono, S. Slab melting versus slab dehydration in subduction-zone magmatism. Proc. Natl. Acad. Sci. USA 108, 8177–8182 (2011).
Kessel, R., Ulmer, P., Pettke, T., Schmidt, M. W. & Thompson, A. B. The water-basalt system at 4 to 6 GPa: Phase relations and second critical endpoint in a Kfree eclogite at 700 to 1400 °C. Earth Planet. Sci. Lett. 237, 873–892 (2005).
Syracuse, E. M., van Keken, P. E. & Abers, G. A. The global range of subduction zone thermal models. Phys. Earth Planet. Inter 183, 73–90 (2010).
Miller, M. S., Kennett, B. L. N. & Gorbatov, A. Morphology of the distorted subducted Pacific slab beneath the Hokkaido corner, Japan. Phys. Earth Planet. Inter 156, 1–11 (2006).
Kennett, B. L. N. & Furumura, T. Tears or thinning? Subduction structures in the Pacific plate beneath the Japanese Islands. Phys. Earth Planet. Inter 180, 52–58 (2010).
Kessel, R., Schmidt, M. W., Ulmer, P. & Pettke, T. Trace element signature of subduction-zone fluids, melts and supercritical liquids at 120–180 km depth. Nature 437, 724–727 (2005).
Portnyagin, M., Hoernle, K., Plechov, P., Mironov, N. & Khubunaya, S. Constraints on mantle melting and composition and nature of slab components in volcanic arcs from volatiles (H2O, S, Cl, F) and trace elements in melt inclusions from the Kamchatka Arc. Earth Planet. Sci. Lett. 255, 53–69 (2007).
Hacker, B. R., Peacock, S. M., Abers, G. A. & Holloway, S. D. Subduction factory 2. Are intermediate-depth earthquakes in subducting slabs linked to metamorphic dehydration reactions? J. Geophys. Res 108, 2030 (2003).
Kawamoto, T. et al. Large-ion lithophile elements delivered by saline fluids to the sub-arc mantle. Earth Planets Space 66, 61 (2014).
Tamura, Y. et al. Mission immiscible: Distinct subduction components generate two primary magmas at Pagan Volcano, Mariana arc. J. Petrol. 55, 63–101 (2014).
Pineda-Velasco, I., Kitagawa, H., Nguyen, T.-T., Kobayashi, K. & Nakamura, E. Production of High-Sr Andesite and Dacite Magmas by Melting of Subducting Oceanic Lithosphere at Propagating Slab Tears. J. Geophys. Res. Solid Earth 123, 3698–3728 (2018).
Peacock, S. M. & Wang, K. Seismic consequences of warm versus cool subduction metamorphism: Examples from southwest and northeast Japan. Science 286, 937–939 (1999).
Min, K., Mundil, R., Renne, P. R. & Ludwig, K. R. A test for systematic errors in 40Ar/39Ar geochronology through comparison with U/Pb analysis of a 1.1- Ga rhyolite. Geochim, Cosmochim. Acta 64, 73–98 (2000).
Nakamura, E. et al. Comprehensive geochemical analyses of small amounts (100 mg) of extraterrestrial samples for the analytical competition related to the sample return mission MUSES-C. Institute of Space Astronautical Science Report SP 16, 49–101 (2003).
Le Roux, L. J. & Glendenin, L. E. Half-life of 232Th in Proceedings of the National Meeting on Nuclear Energy, Pretoria, South Africa 83–94 (1963).
Jaffey, A. H., Flynn, K. F., Glendenin, L. E., Bentley, W. C. & Essling, A. M. Precision measurement of half-lives and specific activities of 235U and 238U. Physic. Rev C4, 1889–1906 (1971).
Cheng, H. et al. The half-lives of uranium-234 and thorium-230. Chem. Geol. 169, 17–33 (2000).
Kuritani, T., Sakuyama, T., Kamada, N., Yokoyama, T. & Nakagawa, M. Fluid-fluxed melting of mantle versus decompression melting of hydrous mantle plume as the cause of intraplate magmatism over a stagnant slab: implications from Fukue Volcano Group, SW Japan. Lithos 282–283, 98–110 (2017).
Putirka, K. D. Thermometers and barometers for volcanic systems. Rev. Mineral. Petrol 69, 61–120 (2008).
Waters, L. E. & Lange, R. A. An updated calibration of the plagioclase-liquid hygrometer-thermometer applicable to basalts through rhyolites. Am. Mineral. 100, 2172–2184 (2015).
Kita, S., Okada, T., Hasegawa, A., Nakajima, J. & Matsuzawa, T. Anomalous deepening of a seismic belt in the upper-plane of the double seismic zone in the Pacific slab beneath the Hokkaido corner: Possible evidence for thermal shielding caused by subducted forearc crust materials. Earth Planet. Sci. Lett. 290, 415–426 (2010).
Ishizuka, Y. Mantle diaper model for polygenetic volcanoes: Geological and petrological study of Rishiri Volcano, northern Hokkaido, Japan. PhD thesis, Grad. Sch. of Sci., Hokkaido Univ. Sapporo (2000).
Sun, S.-S. & McDonough, W. F. Chemical and isotopic systematics of oceanic basalts: Implications for mantle composition and processes in Magmatism in the Ocean Basins (ed. Saunders, A. D. & Norry, M. J.) Geol. Soc. London Spec. Publ. 42, 313–345 (1989).
Zindler, A. & Hart, S. Chemical geodynamics. Annu. Rev. Earth Planet. Sci. 14, 493–571 (1986).
Kimura, J.-I. & Yoshida, T. Contributions of slab fluid, mantle wedge and crust to the origin of Quaternary lavas in the NE Japan Arc. J. Petrol 47, 2185–2232 (2006).
Hauff, F., Hoernle, K. & Schmidt, A. Sr–Nd–Pb composition of Mesozoic Pacific oceanic crust (Site 1149 and 801, ODP Leg 185): Implications for alteration of ocean crust and the input into the Izu-Bonin-Mariana subduction system. Geochem. Geophys. Geosyst. 4, 8913 (2003).
Cousens, B. L. & Allan, J. F. A Pb, Sr, and Nd isotopic study of basaltic rocks from the Sea of Japan, LEGS 127/128 in Proceedings of Ocean Drilling Program Scientific Results (ed. Tamaki. K. et al.) 127–128, 805–817 (Ocean Drill. Program, 1992).
Kuritani, T., Yokoyama, T. & Nakamura, E. Rates of thermal and chemical evolution of magmas in a cooling magma chamber: A chronological and theoretical study on basaltic and andesitic lavas from Rishiri Volcano, Japan. J. Petrol. 48, 1295–1319 (2007).
Acknowledgements
We are grateful to A. Matsumoto, M. Miyasaka for supporting our XRF and EPMA analyses, and H. Nomura and K. Nakamura for preparing the thin sections. Editorial handling by Lucia Pappalardo and constructive review and fruitful comments by Georg Zellmer and two anonymous reviewers are greatly appreciated. H.T. is indebted to S. Yoshimura, Y. Ishizuka, K. Tani, and C. Conway for their important advice. H.T. also acknowledges I. Matsuoka, M. Sato, and staff of Rishiri-Umineko Guesthouse for helping with the fieldwork. This work was supported by the Rishiri Research Project and with funding from the Fukada Geological Institute to H.T., by a research grant from JSPS KAKENHI (Grant Nos. 19J11030 to HT, 16H04071 and 25120006 to TK), and by the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan, under its Earthquake and Volcano Hazards Observation and Research Program and the Integrated Program for Next Generation Volcano Research and Human Resource Development.
Author information
Authors and Affiliations
Contributions
H.T. and T.K. conceived this study, and performed the field work, sampling, petrographic analysis, and geochemical analyses. T.K., T.Y. and E.N. performed the U-Th isotopic analysis. H.T. took the lead in preparing the manuscript with inputs from T.K., T.Y., E.N. and M.N.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Taniuchi, H., Kuritani, T., Yokoyama, T. et al. A new concept for the genesis of felsic magma: the separation of slab-derived supercritical liquid. Sci Rep 10, 8698 (2020). https://doi.org/10.1038/s41598-020-65641-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-020-65641-6
- Springer Nature Limited
This article is cited by
-
Compositional evolution of slab-derived fluids during ascent: implications from trace-element partition between hydrous melts and Cl-free or Cl-rich aqueous fluids
Contributions to Mineralogy and Petrology (2024)
-
Introduction to advances in the study of supercritical geofluids
Science China Earth Sciences (2023)
-
Petrogenesis of isotopically enriched Quaternary magma with adakitic affinity associated with subduction of old lithosphere beneath central Myanmar
Scientific Reports (2022)