Journal of Petrology Advance Access published online on December 4, 2008
Journal of Petrology, doi:10.1093/petrology/egn055
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
New Insights into Andesite Genesis: the Role of Mantle-derived Calc-alkalic and Crust-derived Tholeiitic Melts in Magma Differentiation beneath Zao Volcano, NE Japan
1Institute For Research On Earth Evolution (Ifree), Japan Agency For Marine–Earth Science And Technology (Jamstec), Yokosuka 237-0061, Japan
2Department Of Earth And Environmental Sciences, Yamagata University, Yamagata 990-8560, Japan
3Department Of Earth Sciences, Kanazawa University, Kanazawa 920-1192, Japan
Received April 5, 2008; Revised typescript accepted October 10, 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONOVERVIEW OF THOLEIITIC AND...GEOLOGICAL BACKGROUND OF ZAO...ANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Two distinctive differentiation trends, tholeiitic and calc-alkalic, are recognized in Zao volcano, which is located immediately behind the volcanic front of the NE Japan arc. The genetic relationship between these two magma series is critical for a better understanding of andesite genesis, because they often coexist in close spatial and temporal proximity in arc volcanoes. Petrographic features indicative of ‘disequilibrium’, such as reversely zoned pyroxene phenocrysts, the wide and bimodal compositional distribution in Ca/(Ca + Na) of plagioclase phenocrysts, honeycomb textures and dusty zones that these plagioclase phenocrysts often exhibit, and the presence of olivine–pyroxene pairs with different Mg/Fe, are observed exclusively in calc-alkalic rocks. In tholeiitic rocks the Sr isotopic ratios of plagioclase phenocrysts, determined by both micromilling combined with thermal ionization mass spectrometry, and laser-ablation inductively coupled plasma mass spectrometry techniques, are constant at 0·7042–0·7044. On the other hand, those in calc-alkalic rocks (0·7033–0·7042) show more complex characteristics, which can be best understood if at least three end-member components, a calc-alkalic basaltic melt, a tholeiitic basaltic melt and a tholeiitic felsic melt, contribute to the production of mixed calc-alkalic magmas. The 87Sr/86Sr and trace element compositions of the least-differentiated basalt magmas, which are inferred from the composition of the calcic plagioclase [Ca/(Ca + Na) >0·9], suggest that two types of basaltic magma, calc-alkalic and tholeiitic, exist beneath the volcano. The tholeiitic basalt magma has a higher 87Sr/86Sr than the calc-alkalic magma (0·7042 vs 0·7038) and a characteristic trace element signature consistent with the presence of plagioclase and amphibole as melting residues. This suggests that the tholeiitic magmas are produced via anatexis of amphibolitic crust caused by underplating and/or intrusion of mantle-derived calc-alkalic basalt magmas into the sub-Zao crust. The mantle-derived calc-alkalic basalt magma mixes with crust-derived tholeiitic melts to form calc-alkalic andesite magmas. The hypothesis proposed here requires revision (or even abandonment) of the general consensus that calc-alkalic magmas have greater contributions of a crustal component than tholeiitic magmas.
KEY WORDS: andesite; calc-alkalic: crust; mantle; tholeiitic
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONOVERVIEW OF THOLEIITIC AND...GEOLOGICAL BACKGROUND OF ZAO...ANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
How andesite is generated has long been a central question of igneous petrology. The reason for this is two-fold. First, andesite erupts in more than 80% of arc volcanoes, typifies subduction zone magmatism that creates over 20% of current terrestrial magmatic products, and is the dominant volcanic rock in mature continental arcs. Second, the continental crust, the most differentiated end-member among components within the solid Earth, is overall andesitic or intermediate in composition (e.g. Rudnick, 1995
; Taylor, 1995; Rudnick & Gao, 2003). Knowledge of andesite genesis should therefore provide key constraints on the origin of continental crust and differentiation processes during the evolution of the solid Earth. The following five models are currently favored for the production of andesitic magmas (sensu lato).
(1) Crystallization differentiation of mantle-derived basaltic magmas either in shallow crustal reservoirs or in the deep crust close to the Moho (Sisson & Grove, 1993; Müntener et al., 2001; Annen & Sparks, 2002; Pichavant et al., 2002; Prouteau & Scaillet, 2003).
(2) Anatexis of mafic lower crust by intrusion or underplating of mantle-derived basaltic magma (Takahashi, 1986; Smith & Leeman, 1987; Petford & Atherton, 1996; Kimura et al., 2002; Annen et al., 2006; Tatsumi et al., 2008).
(3) Open-system differentiation, such as mixing between felsic and mafic magmas, and crustal assimilation (Eichelberger, 1975; DePaolo, 1981; Sakuyama, 1981; Hildreth & Moorbath, 1988; Clynne, 1999; Tatsumi & Kogiso, 2003; Dungan & Davidson, 2004).
(4) Generation of melts in equilibrium with mantle peridotite under hydrous conditions as a result of either direct fluxing of slab-derived fluids or slab-melting and subsequent melt–mantle interaction (Kay, 1978; Tatsumi, 1981; Crawford et al., 1989; Pearce et al., 1992; Yogodzinski et al., 1994; Kelemen, 1995; Blatter & Carmichael, 2001; Tatsumi & Hanyu, 2003; Parman & Grove, 2004).
(5) Production of andesitic to more felsic melts by dehydration melting of the subducted oceanic crust (Kay, 1978; Martin, 1986; Stern & Kilian, 1986; Defant & Drummond, 1990).
Two distinctive differentiation trends, tholeiitic and calc-alkalic, are recognized in the sub-alkalic volcanic rocks, denoting the presence or absence of relative iron enrichment during magmatic differentiation (Wager & Deer, 1939; Nockolds & Allen, 1953; Kuno, 1959; Irvine & Baragar, 1971; Miyashiro, 1974). It has been well documented that tholeiitic rocks are dominant in juvenile oceanic arcs, whereas calc-alkalic rocks are the major magmatic products in mature continental arcs with thicker crust (e.g. Miyashiro, 1974; Ewart, 1982). In several arc–trench systems, however, tholeiitic and calc-alkalic andesites to dacites have been observed to coexist in close temporal and spatial proximity (Kuno, 1959; Sakuyama, 1981; Grove & Baker, 1984; Fujinawa, 1988; Brophy, 1990; Hunter & Blake, 1995; Hunter, 1998). Resolution of the genetic relationship between these two types of andesitic magmas should, therefore, provide a better understanding of andesite genesis and arc crust evolution.
The primary aim of this paper is to investigate the mechanism that produces tholeiitic and calc-alkalic magmas at a single volcano, by combining petrographical and geochemical data, including micro-analyses of the isotopic and trace element compositions of plagioclase phenocrysts.
|
OVERVIEW OF THOLEIITIC AND CALC-ALKALIC ANDESITE |
|---|
|
TOPABSTRACTINTRODUCTIONOVERVIEW OF THOLEIITIC AND...GEOLOGICAL BACKGROUND OF ZAO...ANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Definition and chemical characteristics
Bowen (1928
) proposed that silica content increases and iron content decreases with differentiation of sub-alkalic magmas. On the other hand, Fenner (1929) emphasized that some suites show iron enrichment during magmatic differentiation. These two differentiation trends have been referred to as calc-alkalic and tholeiitic, respectively (Wager & Deer, 1939). The identification of these two trends is commonly based on a ternary plot of Na2O + K2O, FeO* (total iron as FeO), and MgO (Irvine & Baragar, 1971). For a more quantitative distinction of the two magma series, FeO*/MgO vs SiO2 variation plots (Miyashiro, 1974) are often used; calc-alkalic and tholeiitic rock series show steeper and gentler slopes, respectively, than the straight line: SiO2 (wt %) = 6·4 x FeO*/MgO + 42·8. Unfortunately, however, Miyashiro's discriminant line is frequently misused and applied as a simple compositional discriminant outside this range rather than its intended use as a ‘trend slope’ comparison.
The terms medium-K and calc-alkalic series are often used interchangeably (Fig. 1). This misusage may be due to poor understanding of the following two observations. First, concentrations of incompatible elements such as K, Rb, and Nb in lavas at a constant SiO2 content increase with distance from the volcanic front or the height of a volcano above the slab surface (Fig. 1), which is generally known as the K–h relationship (Dickinson, 1975). Second, calc-alkalic series rocks are generally more enriched in incompatible elements than tholeiitic series rocks (e.g. Masuda & Aoki, 1979; Kimura & Yoshida, 2006). A schematic illustration showing the across-arc variation in magma series from tholeiitic, via calc-alkalic, to high-K or alkalic with distance from the volcanic front (Fig. 1a) is then often cited (e.g. Hess, 1989; Wilson, 1989), although calc-alkalic rocks do exist along the volcanic front and tholeiitic rocks belonging to the medium-K series often erupt at volcanoes behind the volcanic front as shown in Fig. 1b (Kuno, 1960; Kawano et al., 1961; Gill, 1981; Yoshida & Aoki, 1984; Tatsumi & Eggins, 1995; Tatsumi & Kogiso, 2003). The tholeiitic and calc-alkalic series should be defined by the presence and absence of iron enrichment, respectively, whereas low-, medium-, and high-K series should be defined on the basis of K2O concentrations.
|
Yoder & Tilley (1962) emphasized the presence of at least two types of basalt magmas, one with normative hypersthene differentiating to silica-saturated and the other with normative nepheline differentiating to silica-undersaturated liquids, and coined the terms olivine tholeiite and alkali basalt for these magmas, respectively. Tholeiites can be clearly defined based on normative compositions; however, it should be noted that Yoder & Tilley's tholeiites are identical to sub-alkalic magmas and thus include both tholeiitic and calc-alkalic rock series defined based on the presence or absence of iron enrichment during differentiation of sub-alkalic magmas.
Calc-alkalic and tholeiitic series can be well identified for rocks with intermediate compositions; in other words, it may be difficult to classify mafic (basaltic) or felsic (rhyolitic) rocks into these two rock series. We tentatively use the term calc-alkalic or tholeiitic for basalts and rhyolites, if they, together with andesites, form calc-alkalic or tholeiitic trends, respectively.
Petrographical characteristics
The calc-alkalic vs tholeiitic series should be defined exclusively on the basis of differentiation trends. However, the identification of such chemical trends is, in some cases, difficult because of a lack of sufficient data for defining the chemical trend. As a result, this can be supplemented with petrographical characteristics to identify these two magma series. On the basis of groundmass mineralogy Kuno (1950, 1959, 1968) divided sub-alkalic volcanic rocks into two series: hypersthenic and pigeonitic. These are distinguished by the presence or absence of orthopyroxene in the groundmass, and are synonymous with the calc-alkalic and tholeiitic series, respectively. Although it may not be easy to identify fine-grained groundmass pyroxenes under the microscope, orthopyroxene phenocrysts with a reaction rim of clinopyroxene, which occur solely in the pigeonitic rock series, can be recognized (e.g. Kawano et al., 1961). Furthermore, phenocrysts of hornblende and biotite are limited to the hypersthenic rock series (Kuno, 1950). It has been well established that Kuno's scheme is valid for Quaternary arc volcanoes along the trench-side volcanic chain in the NE Japan and Izu–Bonin arcs (Kuno, 1950; Kawano et al., 1961; Wada, 1981, 1985; Fujinawa, 1988, 1990), where tholeiitic rocks are broadly equivalent to the low-K series of Gill (1981).
Sakuyama (1981) examined the petrographical characteristics of volcanic rocks from Myoko–Kurohime volcanoes, Central Japan, where both calc-alkalic and tholeiitic magmas have erupted from single vents, and divided these rocks into two types, N-type and R-type, on the basis of the absence and presence of reversely zoned mafic phenocrysts, respectively. The following ‘disequilibrium’ petrographical features characterize the R-type volcanic rocks:
(1) the presence of reversely zoned pyroxene phenocrysts with a lower Mg-number [= 100 x Mg/(Mg + Fe)] core surrounded by a higher Mg-number rim;
(2) the presence of groundmass pyroxenes with higher Mg-number than the phenocryst cores;
(3) bimodal distribution in the core compositions of plagioclase phenocrysts;
(4) disequilibrium phenocryst assemblages such as Mg-rich olivine and quartz;
(5) patchy groundmass with different colors and/or amount of mafic minerals.
These disequilibrium features are not observed in N-type rocks. It was emphasized by Sakutyama (1981) that the N- and R-type rocks are broadly equivalent to Kuno's pigeonitic and hypersthenic rock series, and hence more generally to tholeiitic and calc-alkalic series, respectively.
The consistency between Sakuyama's petrographical classification and bulk-rock chemical characteristics (i.e. N- vs R-type and tholeiitic vs calc-alkalic, respectively) has been well established at least for Quaternary NE Japan arc volcanoes (Wada, 1981; Sakuyama, 1983; Fujinawa, 1988, 1990). However, applied elsewhere there are exceptions. Particular calc-alkalic andesites (high-Mg andesites) from the Setouchi volcanic belt, SW Japan, characterized by unusually high MgO contents or high Mg-number and hence believed to represent least-differentiated mantle-derived magmas, are petrographically classified as N-type rocks (Tatsumi, 2006). On the other hand, high-Mg andesite from Mt. Shasta, USA, which is considered as representative of primitive andesite (Baker et al., 1994; Grove et al., 2002), has now been identified as R-type andesite, and hence is likely to be the product of magma mixing (Streck et al., 2007).
Occurrence
It has long been known that some volcanic arcs are characterized by either tholeiitic or calc-alkalic magmatism (Jakes & Gill, 1970; Plank & Langmuir, 1988). Calc-alkalic rocks are clearly dominant in continental arcs rather than oceanic arcs (Miyashiro, 1974; Ewart, 1982), and calc-alkalic/tholeiitic volume ratios tend to increase with increasing age or arc maturity (Baker, 1973) and crustal thickness (Gill, 1981). It should be stressed, however, that tholeiitic and calc-alkalic magmas do coexist in some single volcanic systems; e.g. Mt. Shasta, USA (Baker et al., 1994), Chichontepec, El Salvador (Bau & Knittel, 1993), Aso in SW Japan (Hunter, 1998), and Myoko–Kurohime in Central Japan (Sakuyama, 1981). Furthermore, along the volcanic front of the NE Japan arc about one-third of Quaternary volcanoes erupt both tholeiitic and calc-alkalic rocks (Kawano et al., 1961). Examining the geochemistry of the two coexisting magma series in the above volcanoes reveals that the calc-alkalic rocks are generally more enriched in both compatible elements such as Mg, Ni and Cr and incompatible elements such as Rb, K, Th and U (e.g. Masuda & Aoki, 1979; Kimura & Yoshida, 2006).
The misunderstanding caused by the interchangeable usage of the terms medium-K and calc-alkalic series (Fig. 1a) may further mislead us to a ‘likely’ conclusion that calc-alkalic magmas are more hydrous than tholeiitic magmas, because the H2O content in mantle-derived primary magmas may increase together with incompatible elements towards the back-arc side within a volcanic arc (Sakuyama, 1979; Tatsumi et al., 1983). The experimental constraints that calc-alkalic trends can be reproduced under hydrous conditions but tholeiitic trends under H2O-poor conditions (e.g. Grove & Baker, 1984; Sisson & Grove, 1993; Hamada & Fujii, 2008) may further reinforce this idea. However, it should be stressed here that there are no data available that demonstrate the difference in H2O contents in these magma series occurring in a single volcano.
|
GEOLOGICAL BACKGROUND OF ZAO VOLCANO |
|---|
|
TOPABSTRACTINTRODUCTIONOVERVIEW OF THOLEIITIC AND...GEOLOGICAL BACKGROUND OF ZAO...ANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
NE Japan arc
The NE Japan arc is formed by the subduction of the Pacific plate beneath the North American and Eurasian plates at the Japan Trench (Fig. 2) at a rate of
10 cm/year. The volcanic front of this arc, which is the trenchward boundary of the volcanic arc, runs parallel to the trench axis. The NE Japan arc exhibits the following tectonic and magmatic characteristics, which are common to most arc–trench system (Tatsumi & Eggins, 1995):
|
(1) dual volcanic chains are present, one trench-side and one on the back-arc side, known as the Nasu and Chokai chains, respectively;
(2) these chains are 110 km and 150–170 km above the top of the subducted oceanic lithosphere, respectively, at least in the central portion of the arc (Fig. 2);
(3) the number of volcanoes and the eruptive volume are greater in the trench-side volcanic chain;
(4) volcanic rocks in the back-arc side chain tend to be more enriched in incompatible elements than those in the trench-side chain.
These characteristics are well documented in the central to northern NE Japan arc, but not in the southern arc where the Philippine Sea plate is being subducted into the mantle wedge above the descending Pacific plate and the Izu–Bonin–Mariana arc is colliding with the Japanese islands (Fig. 2). This complicated tectonic setting may cause the unusual characteristics of the magmatism at the southern end of the NE Japan arc.
From north to south the pre-Tertiary basement rocks of the NE Japan arc consist of Cretaceous to Jurassic metamorphic rocks, Cretaceous sedimentary rocks and granitoids, and subduction zone and metamorphic complexes of Ordovician to Cretaceous age. Granitoids in these basements show spatial variations in isotopic composition, with more enriched isotopic signatures southwards (Kagami, 2005). Kersting et al. (1996) and Kimura & Yoshida (2006) further demonstrated the correlation between the Sr–Nd isotopic compositions of volcanic front lavas and their underlying basement rocks.
Zao volcano
A total of 55 volcanoes are distributed in the NE Japan arc. Zao volcano is one such Quaternary volcano, and is situated immediately behind the volcanic front in the Nasu trench-side chain (Fig. 2). The following summary of the geological history of Zao volcano is based on a synthesis of previously published studies (Oba & Konda, 1989; Takaoka et al., 1989; Sakayori, 1992).
The basement around Zao consists of Cretaceous granitoids of the Abukuma Terrane and Tertiary volcanic rocks. Magmatic activity, which commenced at 1· 0 Ma and has formed a volcanic edifice with a current volume of 25 km3, can be separated into four major stages (Fig. 3): Stage 1 (1· 0–0·6 Ma): formation of a small volcano consisting of basaltic to basaltic andesite pyroclastic rocks and dykes; Stage 2 (0·3 Ma): formation of a stratovolcano composed of andesitic to dacitic lavas and pyroclastic deposits; Stage 3 (0·3–0·1 Ma): eruption of basaltic andesite to basalt lava flows and pyroclastic deposits from two vents near the summit; Stage 4 (< 0·1 Ma): caldera-forming eruption (< 2·2 km in diameter), after which a pyroclastic cone composed of basaltic andesites built up within the caldera. The rocks of Stage 1 are classified as low-K tholeiitic series, whereas those of Stages 2–4 belong to the medium-K, calc-alkalic series. This study is based on 39 volcanic samples collected at 14 sites from Stages 1 and 3 of the evolution of Zao volcano (Fig. 3).
|
|
ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONOVERVIEW OF THOLEIITIC AND...GEOLOGICAL BACKGROUND OF ZAO...ANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Bulk-rock analyses
Rock samples for trace element and Sr–Nd–Pb isotope analyses were crushed to coarse chips (< 0·5 mm3) and fresh pieces were hand picked. To avoid surface contamination, the rock chips were washed with ethanol and then leached with 1· 0M HCl at room temperature for 1 h. These chips were rinsed three times with Milli-Q water an then dried in an oven. The chips were pre-pulverized using a tungsten carbide mortar. At this step, any remaining altered portions were removed. Finally, the coarse grain samples were pulverized in a tungsten carbide vibration mill.
Major and trace element (Ni, Cu, Zn, Rb, Sr, Y, Zr, Nb, Ba, Pb, and Th) compositions were measured on fused glass beads and pressed powder pellets, respectively, using RIGAKU® Simultix 12 and RIX3000 X-ray fluorescence (XRF) spectrometers. Analytical procedures, precision (mostly <1%), and accuracy (mostly <1%) have been fully described by Tani et al. (2006).
Concentrations of rare earth and 17 other trace elements (Sc, Co, Ni, Cu, Rb, Sr, Y, Zr, Nb, Cs, Ba, Hf, Ta, Tl, Pb, Th, and U) were analyzed using by inductively coupled plasma mass spectrometry (ICP-MS) using an Agilent 7500ce system fitted with PFA sample introducing and a Pt-inject torch system. The ICP-MS system was operated in no collision gas and multi-tune acquisition mode. This combination allowed a wide range of elements to be precisely determined using pulse counting detection with the hydrofluoric acid containing sample solution being delivered directly into the plasma. Sample dissolution, preparation and measurement were described by Chang et al. (2003) and Nakamura & Chang (2007). Analytical accuracy and precision for ICP-MS analyses, estimated from repeated measurements of international reference rocks (JB-1a of GSJ, BCR-2 and BIR-1 of USGS) were mostly better than 5% and 3%, respectively.
The analytical procedure used for chemical separation and mass spectrometry for Sr, Nd and Pb isotope determinations was outlined by Miyazaki et al. (2007). Total procedural blanks for Sr, Nd and Pb were less than 15 pg, 3 pg and 6 pg, respectively. Mass spectrometry was performed on a Thermo-Finnigan® Triton TI equipped with nine Faraday cups, using a static multi-collection mode. Normalizing factors used to correct for isotopic fractionation in Sr, Nd and Pb isotope analyses were 86Sr/88Sr = 0·1194, 146Nd/144Nd = 0·7219, and 0·148% per atomic mass unit, respectively. Measured isotopic ratios for standard materials were 87Sr/86Sr = 0·710262 ± 14 (2
) for NIST 987 (n = 40), 143Nd/144Nd = 0·512098 ± 13 (2), for JNdi-1 (n = 31), and 208Pb/204Pb = 36·717 ± 7 (2), 207Pb/204Pb = 15·497 ± 2 (2) and 206Pb/204Pb = 16·940 ± 2 (2) for NIST 981 (n = 23).
Micro-analyses
Mineral compositions of major and minor elements were analyzed by electron-probe micro-analysis (EPMA) using a JEOL JXA-8800 instrument, following the method described by Shukuno (2003). For all elements the excitation potential, specimen current, and analytical time of peak and background were: on olivine 20 kV, 25 nA, 20 s, and 10 s (except for Mn, Ca, and Ni, for which 25 kV, 20 nA, 100 s and 50 s were used); on spinel 15 kV, 12 nA, and 20 s; on pyroxene and plagioclase 15 kV, 15 nA, and 20 s. ZAF correction procedures were employed.
Trace element [Rb, Sr, Y, Zr, Nb, Ba, rare earth elements (REE), Hf, Ta, Pb, Th, and U] micro-analyses of plagioclase were performed by laser-ablation (LA)-ICP-MS on the same thick sections that were used for the Sr isotope analyses (see below). The 193 nm excimer laser system aerosol line was connected to a VG Elemental® PQ3 quadrupole-type ICP-MS system (Kimura et al., 2000). We used the 43Ca peak for internal standardization, which corrected any variation in the ablated sample volume. CaO contents were determined by EPMA prior to the LA-ICP-MS analysis. NIST 612 synthetic glass was used as a standard, adopting the reference values of Pearce et al. (1996). The element concentrations measured in NIST 612 were within 8·5% of the reference values. The overall homogeneity of the glass for most elements was better than 10% (2 SD), when 20 µm laser spots were used (Kimura et al., 2000). The LA conditions were set at 200 mJ laser source energy, 50 µm crater size, and repetition rate at 5 Hz, yielding signals of 2000 c.p.s./ppm at 115In. Ar gas blank was measured for 60 s for blank subtraction prior to sample analysis. Three spots on the NIST 612 standard were analyzed and averaged to minimize errors caused by heterogeneity. Typical analytical time for a single crater was 60 s, generating craters about 30 µm deep. Accuracy of the results can be affected by the matrix effect between the synthetic glass standard and the silicate minerals. However, accuracy is generally better than 15% when the aerosol loading to the plasma is controlled by ±50% (Kimura et al., 2000). Such accuracy is sufficient for most geochemical and petrological purposes.
Sr isotopes were analyzed in the plagioclases using two techniques; a combined micromilling followed by chemical separation and thermal ionization mass spectrometry (MM-TIMS), and in situ laser ablation multi-collector inductively coupled plasma source mass spectrometry (LA-MC-ICP-MS).
The MM-TIMS techniques we used were similar to those of Davidson & Tepley (1997), Tepley et al. (1999) and Charlier et al. (2006), and were fully described by Takahashi et al. (2005). The rock sample was cut into a wafer with a thickness of 1· 0 mm, which was then bonded onto a glass slide and polished. Micromilling of the sample was carried out using a New WaveTM MicroMillTM. The diameter at the tip of the drill used for sampling is 0·27 mm. After micromilling, the collected sample powder was decomposed with HF, HCl and HNO3. Sr selective extraction resin (Sr Resin from EICHROM Technologies Inc.) was used for chemical separation of Sr (Horwitz et al., 1992). Resin was charged into a modified pipette tip column with quartz wool filter. Bedded resin volume was 0·05 ml. Total procedural blanks for Sr in the MM-TIMS procedure were less than 10 pg. Methods used for Sr isotope MM-TIMS analyses are identical to those used for the bulk-rock analyses. Repeated analyses of NIST 987 (10 ng Sr) gave 87Sr/86Sr = 0·710260 ± 22 (2, n = 8). Additionally, NIST 610 glass was used for estimating the analytical precision. Analysis of NIST 610 by the bulk-rock analysis procedure and the MM-TIMS techniques gave 87Sr/86Sr = 0·709681 ± 07 (2, n = 4) and 87Sr/86Sr = 0·709696 ± 29 (2, n = 5), respectively. These values are almost identical to the value (87Sr/86Sr = 0·709699 ± 18) reported for NIST 610 (Woodhead & Hergt, 2001).
In situ Sr isotope micro-analyses were also performed by LA-MC-ICP-MS using a VG Elemental® Plasma 54 MC-ICP-MS system equipped with a dry and solution aerosol dual sample intake system. The dual intake system consists of an in-house 193 nm ArF excimer laser ablation (Kimura et al., 2000) with He carrier gas and an Aridus® desolvating nebulizer using Ar (+ trace N2) carrier gas. The two carrier gas lines were mixed in a 50 cm3 volume Teflon® mixing chamber prior to the ICP torch. While the laser aerosol alone was analyzed, the Milli-Q® deionized water was taken up by the Aridus® solution line, for which blanks were negligible. Standard solutions were introduced into the ICP torch while LA was unfired with the LA carrier gas on, for determination of instrumental mass bias and isobaric overlap correction factors (see below). We hereafter call this the ‘dual intake system’.
An on-peak background method was applied to subtract the blank from the Kr impurity (82Kr, 83Kr, 84Kr, and 86Kr) in the Ar plasma gas (Woodhead et al., 2005). This is necessary rather than peak stripping using mass bias factor by monitoring one of the Kr peaks, because mass bias for Kr cannot be determined because of complex interferences during the sample analyses (Vroon et al. 2008). Background signals were acquired for the first 15 s on the gas blank and the signal from the sample or standard was then collected, typically for 5 min, depending on the sample.
The 87Rb interference on 87Sr was corrected for by monitoring 85Rb using an empirical overlap correction factor, determined from a Rb-doped NIST 987 Sr isotope standard solution (20 ppb Rb in 100 ppb Sr, Rb/Sr = 0·2) from the Aridus® aerosol line. The mass bias factor of Rb is not identical to that of Sr (Hirata, 1996) and Rb has only two stable isotopes, which prevents internal mass bias correction. The mass bias factor for Rb may be different between solution and LA modes (Vroon et al., 2008). However, our simultaneous dual intake system cancels out the mass bias difference caused by the different ICP operating parameters in discrete LA aerosol or solution aerosol introduction. Therefore, our empirical Rb overlap correction method is an alternative to correction factor determination using a synthetic glass standard (Davidson et al., 2001). The difference between LA and Aridus® solution analyses with our dual intake system is in the presence or absence of matrix elements for the LA and the Aridus® aerosols, respectively. The non-spectral matrix effect (Barling & Weis, 2008) should be present and is noted below. The rate of 87Rb overlap on 87M during plagioclase analyses was typically less than 1% (Rb/Sr < 0·00025) and occasionally exceeded 10% (Rb/Sr > 0·025) with plagioclase containing glass inclusions. However, we did not see any problems caused by the change in the Rb overlap correction factor on the 87Sr/86Sr ratios of plagioclase within analytical precision (87Sr/86Sr = ±0·00005).
The CaAr and Ca dimmer molecular ion interferences on masses 82M, 84M, 86M, and 88M were subtracted by monitoring 82M using the correction method proposed by Woodhead et al. (2005). The effect of the correction was monitored by the most interference-sensitive isotope ratio, 84Sr/86Sr, the value of which was typically 0·0565 (Woodhead et al., 2005). We have confirmed the validity of the correction method by analyzing a Ca-doped NIST 987 standard solution. The reference isotope ratio of NIST 987 was reproducible within the analytical precision with a standard solution containing 50 ppm Ca and 100 ppb Sr (Ca/Sr = 500), in which Ca/Sr was more than that in the natural plagioclase (50–200) and comparable with natural carbonates (500).
Other interferences from doubly charged ions such as Yb2+ are not considered. Plagioclase crystals have low element abundances of REE (see trace element compositions of the plagioclase crystals below). The effect of the doubly charged heavy ions is negligible (Vroon et al., 2008).
Mass bias, which was not considered by internal correction using 86Sr/88Sr = 0·1194, caused by day-to-day basis changes in interface cone or operating parameters (gas flows, sampling depths, etc., Woodhead et al., 2001), was further corrected for using the beta factor (Pachett et al., 1981; Iizuka & Hirata, 2005) determined by analyses of a solution of NIST 987 with 87Sr/86Sr = 0·70125 from the Aridus® line. The dual sample intake system was again advantageous for performing this complex correction procedure, as it allows immediate switching between the two introduction lines without changing the condition of the plasma. Absence and presence of matrix elements may cause change in Sr mass bias between the LA and the standard solution aerosols. However, increase of the matrix element by the addition of 50 ppm Ca to the NIST 987 solution did not make any detectable change in 87Sr/86Sr, confirming previous reports (Woodhead et al., 2005; Vroon et al., 2008). The NIST 987 standard bracketing method was used to perform the beta correction after a series of 10 analyses on the plagioclase.
A source energy of 200 mJ was used for the laser, with a repetition rate of 20 Hz and a diameter of 200 µm. This yielded a stable signal of 2–3 V on 88Sr for plagioclase with 300 ppm Sr over 5 min by our LA system, with crater penetration to about 400–500 µm in the thick section. All the plagioclase analyses were performed in single spot mode rather than raster mode. The 87Sr/86Sr ratios of individual plagioclase spots were measured for 5 s each. The downhole Sr isotopic zoning was carefully rejected by observing the time resolving profiles. Integrated data from the homogeneous section of the profile were used for further statistical treatment, including average and two standard error calculations. Downhole element fractionation has been reported between U and Pb in a zircon crystal (Horn et al., 2000). On the other hand, such fractionation was not detectable during Hf isotope analysis of zircon crystal including the overlap correction factors of Lu and Yb on 176Hf (Woodhead et al., 2004). The downhole elemental fractionation may occur between Rb and Sr, likewise between Pb and U. However, we did not see any detectable change in the Sr isotopic ratio in homogeneous plagioclase crystals with sporadic melt inclusions at different depth levels. In fact, overall precision in the single spot analyses of the homogeneous crystals was better than ±0·00005 (2 SE), irrespective of the presence or absence of melt inclusions that contain Rb. This indicates that isotopic fractionation during downhole laser ablation is unlikely or at least not detectable with our analytical precision, for both Sr and Rb, which is similar to the case of Pb, Hf, Yb, and Lu isotopes in zircon (Horn et al., 2000; Woodhead et al., 2004).
The accuracy of the LA-ICP-MS plagioclase method was tested by comparing MM-TIMS data from the same homogeneous plagioclase (J.-I. Kimura, unpublished data, and the present study); the results agreed within ±0·0001, which is almost the same level as those reported elsewhere for LA-MC-ICP-MS Sr isotope analyses (Vroon et al., 2008). This accuracy is acceptable for the purpose of this study.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONOVERVIEW OF THOLEIITIC AND...GEOLOGICAL BACKGROUND OF ZAO...ANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Bulk-rock chemistry
Major and trace element concentrations and Sr–Nd–Pb isotopic compositions of Zao volcanic rocks are reported in Table 1.
|
The volcanic rocks define two distinct chemical trends; tholeiitic and calc-alkalic (Fig. 4). These are generally regarded as equivalent to Kuno's pigeonitic and hypersthenic rock series, and Sakuyama's N-type and R-type rocks, respectively. Tholeiitic magmatism occurred solely during Stage 1, whereas the rocks of Stages 2–4 are classified as calc-alkalic. It should be stressed that the Zao volcanic rocks form chemical trends similar to those that typify the tholeiitic and calc-alkalic rocks of the trench-side Nasu volcanic chain (Fig. 4), suggesting that the genetic relation between the two magma series at Zao may be applied to the NE Japan arc in general. Among the major elements, the calc-alkalic rocks tend to be more depleted in Fe and more enriched in Mg and K than the tholeiitic rocks. As a result they belong to the medium-K series, whereas the tholeiitic rocks belong to the low-K series (Fig. 4).
|
The normal mid-ocean ridge basalt (N-MORB) normalized incompatible trace element patterns of the Zao volcanic rocks are shown in Fig. 5. Although elements with higher incompatibility during mantle melting tend to be more enriched, the high field strength elements such as Nb, Ta and Zr do not show such enrichment. This selective enrichment of particular elements results in strongly spiked patterns (Fig. 5), which have also been observed in other arc lavas (e.g. Tatsumi & Eggins, 1995
). Calc-alkalic rocks are characterized by higher concentrations of incompatible trace elements than tholeiitic rocks.
|
The Sr–Nd–Pb isotope compositions of the Zao volcanic rocks are also well within the range of the trench-side Nasu lavas (Fig. 6). It has been well established that volcanic rocks along the NE Japan volcanic front exhibit more enriched Sr–Nd isotopic signatures towards the south (Notsu, 1983
; Kimura & Yoshida, 2006). Zao volcano is situated in the southernmost part of Zone I of Kimura & Yoshida (Fig. 2) and has Sr–Nd isotopic characteristics typical of this zone (Fig. 6). The Pb isotopic ratios of Zao, and more generally of Quaternary volcanic rocks from the NE Japan arc, form a broad trend towards the compositions of Pacific sediments from the MORB field (Fig. 6).
|
Calc-alkalic and tholeiitic rocks from Zao volcano exhibit systematic differences in both major/trace element and isotopic characteristics. To emphasize the difference in element concentrations, element abundances are compared in Fig. 7 by normalizing the calc-alkalic compositions to the tholeiitic composition at 55 wt % SiO2. It should be stressed that the calc-alkalic rocks are more enriched in both highly incompatible elements (U, Rb, Th, and K) and compatible elements (Ni and Mg). Isotopically, the calc-alkalic rocks exhibit more depleted characteristics in terms of Sr, Nd and Pb isotopes (Fig. 6).
|
Petrography
Modal proportions of phenocrysts and representative mineral compositions are given in Tables 1–7
. Subsequent petrographic description of the Zao volcanic rocks (below) highlights the differences between the calc-alkalic and tholeiitic rocks.
|
|
|
|
|
|
Olivine phenocrysts (usually <1 vol. %) occur as euhedral to subhedral crystals unrimmed or rarely mantled by orthopyroxene in basaltic andesites to andesites from both the calc-alkalic and tholeiitic series. Although there is little difference in the Mg-number of the core compositions between the two magma series (65–85), NiO contents are higher in the calc-alkalic rocks at a constant Mg-number (Fig. 8).
|
Orthopyroxene phenocrysts are ubiquitous in the Zao volcanic rocks. Those in rocks from the tholeiitic series exhibit little or normal zoning with a higher Mg-number core mantled by a lower Mg-number rim (Fig. 9). In contrast, calc-alkalic rocks contain both normally and reversely zoned orthopyroxene phenocrysts (Fig. 9), with the cores of normally zoned orthopyroxene phenocrysts having higher Mg-number (70–80) than those of the reversely zoned phenocrysts (
60). Orthopyroxene phenocrysts with reactions rims of clinopyroxene (pigeonite) occur only in the tholeiitic rocks. Only in the calc-alkalic series is orthopyroxene present as a groundmass phase.
|
Clinopyroxene (augite) phenocrysts occur in all Zao volcanic rocks, although they are less abundant than orthopyroxene. Augite phenocrysts have a rather limited compositional range, both in tholeiitic and calc-alkalic rocks, (Mg-number
70; Fig. 9). Reversely zoned clinopyroxene phenocrysts are found only in calc-alkalic rocks (Fig. 9), whereas pigeonite exists only in the groundmass of tholeiitic rocks. Plagioclase is the most abundant phenocryst phase and generally makes up 10–30 vol. % of the analyzed samples. The plagioclase phenocrysts in the tholeiitic rocks have a limited compositional range, whereas in the calc-alkalic rocks they exhibit a broader range of compositions with a bimodal distribution (Fig. 10). Notably, the plagioclase phenocrysts in the calc-alkalic rocks commonly have honeycomb textures and dusty zones. Representative textures of plagioclase phenocrysts are shown in Fig. 11, together with compositional profiles across the crystal for Ca/(Ca + Na) ratio and Sr concentration.
|
|
Plagioclase microanalyses
Representative trace element concentrations and Sr isotopic compositions of plagioclase phenocrysts are listed in Table 5. The correlations of selected trace element abundances and 87Sr/86Sr with Ca/(Ca + Na) are shown in Figs 12 and 13.
|
Plagioclase phenocrysts in calc-alkalic rocks tend to be more enriched in Ba and Sr than those in tholeiitic rocks (Fig. 12), consistent with the chemical characteristics of the bulk-rocks (Fig. 7). Plagioclases from the two magma series show little difference in REE concentrations, whereas Y concentrations in Ca-rich plagioclase from calc-alkalic rocks are lower than those from tholeiitic rocks (Fig. 12).
Sr isotopic ratios in plagioclases from tholeiitic rocks are constant at 0·7042–0·7044 and show little correlation with the anorthite content (Fig. 13), whereas those in the calc-alkalic rocks show more complex characteristics. The core compositions of plagioclase phenocrysts in calc-alkalic rocks show a broad bimodal distribution in terms of anorthite content (Fig. 10). Importantly, the high-An plagioclase from the most mafic calc-alkalic basaltic andesites exhibit the lowest 87Sr/86Sr of 0·7034 (Fig. 13).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONOVERVIEW OF THOLEIITIC AND...GEOLOGICAL BACKGROUND OF ZAO...ANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Closed vs open processes in the tholeiitic series
Tholeiitic rocks from Zao volcano belong to the N-type rocks of Sakuyama (1981
) and Kuno's pigeonitic rock series, and exhibit little evidence for disequilibrium: (1) phenocryst phases such as pyroxenes and plagioclase are normally zoned (Figs 9 and 10); (2) olivine is in equilibrium with pyroxenes in terms of Fe–Mg partitioning (Fig. 14) at temperatures of 1050°C (Fig. 15). These observations, together with the relatively constant 87Sr/86Sr for plagioclase phenocryst cores (Fig. 13), suggest that the Zao tholeiitic magmas differentiate mainly in a closed system, via fractional crystallization; this is a mechanism that has been accepted as a general process of differentiation in arc tholeiites (e.g. Sakuyama, 1981; Fujinawa, 1988, 1990; Tatsumi & Kogiso, 2003; George et al., 2004; Villiger et al., 2007). It should be stressed here that differentiation by crystallization is not the only process that can occur during closed-system differentiation. Partial melting, or anatexis, for example, of basaltic lower crust could cause differentiation from mafic to felsic compositions, which will be discussed further below.
|
|
However, closed-system differentiation cannot solely explain the geochemical characteristics of the Zao tholeiitic rocks. For example, plagioclase rims tend to show higher 87Sr/86Sr (up to 0·7046) than plagioclase cores (Fig. 13). Furthermore, a positive correlation between SiO2 content and 87Sr/86Sr is observed for bulk-rock compositions (Fig. 16b). These observations suggest that an open-system process, such as shallow-level crustal contamination, plays a role in controlling the final compositions of the tholeiitic magmas. Assuming Abukuma granitic rocks (Gr in Table 8, data from Kamei et al., 2003
), which form the basement of the Zao volcano, to be a possible upper crustal contaminant, the observed variation in 87Sr/86Sr for the tholeiitic rocks can be explained by simple bulk contamination by upper crust (Fig. 16b).
|
|
Magma mixing in the calc-alkalic series
‘Disequilibrium’ petrographic features in the calc-alkalic rocks of Zao volcano indicate that they belong to the R-type rocks of Sakuyama (1981
). Such features include: the occurrence of reversely zoned pyroxene phenocrysts (Fig. 9); a wide compositional range in Ca/(Ca + Na) of plagioclase phenocrysts, which also show a bimodal compositional distribution (Fig. 10); the occurrence of plagioclase phenocrysts with honeycomb textures and dusty zones; and the presence of disequilibrium olivine–pyroxene pairs within a single specimen (Fig. 14).
Disequilibrium in the calc-alkalic rocks is further indicated by temperature estimates for pyroxene crystallization. Figure 15 summarizes the temperature estimates from contiguous crystals of clinopyroxene and orthopyroxene using a two-pyroxene geothermometer (Wells, 1977). This thermometry cannot be applied to the pyroxenes in the groundmass of the tholeiitic rocks in Zao volcano, because they do not contain orthopyroxene; they belong to Kuno's pigeonitic rock series. Figure 15 indicates that in calc-alkalic rocks the rim–rim pair and the groundmass pair of pyroxenes exhibit crystallization temperatures higher than those obtained for the core–core pair.
One possible process that can provide a comprehensive explanation of these disequilibrium textures would be mixing of magmas having different compositions and temperatures (e.g. Eichelberger, 1975; Sakuyama, 1981, 1983; Wada, 1985; Fujinawa, 1988, 1990; Hunter & Blake, 1995; Clynne, 1999; Streck et al., 2007). To better constrain the magma mixing process, especially to decode the chemical characteristics of the end-member components for the mixed calc-alkalic magmas, the Sr isotopic compositions recorded in plagioclase phenocrysts (Fig. 13) are considered in combination with the chemical composition of the magma. Element concentrations for magmas that crystallize plagioclase in the Zao volcanic rocks are obtained using element partitioning data (Bindeman et al., 1998; Bindeman & Davis, 2000). The panels down the right-hand side of Fig. 13 show the relationship between 87Sr/86Sr and Sr concentration in the melt in equilibrium with the plagioclase and allow the identification of four melt components that mix to form the calc-alkalic rocks. These are: a low 87Sr/86Sr (0·7034), Sr-rich (650–700 ppm) melt (I in Fig. 13), probably having a mafic/basaltic composition; and three components (II, III, and IV in Fig. 13) with similar rather high 87Sr/86Sr (0·7040), but with different Sr concentrations (500–400, 300, 200–150 ppm, respectively), that are probably intermediate in composition. The above four melt components may crystallize plagioclase with An>90, An80–90, An70, and An50–60, respectively (Figs 13 and 16a).
The next problem to address is how these four components are generated. Figure 16a summarizes the geochemical characteristics of the four calc-alkalic melt components (I–IV in Fig. 13) and the tholeiitic melts in Zao volcano inferred from the core compositions of plagioclase phenocrysts. To understand the characteristics of the Zao calc-alkalic melts, at least three end-member components may be required, L1, L2, and L3 in Fig. 16a and Table 8. L1 and L2 are a calc-alkalic and a tholeiitic basaltic melt, believed to be the host melts for Ca-rich (An>90) plagioclase found in the calc-alkalic and tholeiitic basaltic andesites at Zao volcano, respectively. The SiO2 contents of these basaltic melts are assumed to be 50 wt %; Sr concentrations and 87Sr/86Sr are deduced from plagioclase core compositions (Table 8). L3 is a tholeiitic felsic melt that is a liquid differentiated from L2 and is assumed to contain 70 wt % SiO2. The 87Sr/86Sr of L3 is simply assumed to be identical to that of L2; the Sr concentration in L2 is <200 ppm, and has been assumed to be 100 ppm. The reason for this is that the identifiable and the most differentiated tholeiitic melt in Zao volcano contains 200 ppm Sr and crystallizes An60 plagioclase (Fig. 13), which is more calcic than the plagioclase present in the calc-alkalic intermediate component (An50–60; Fig. 13). This suggests that a more differentiated tholeiitic felsic melt with a Sr concentration lower than 200 ppm is required as the felsic end-member for the calc-alkalic rocks.
Mixing between L1 and L2 (15% contribution of L1) produces a basaltic melt, L4, that has chemical characteristics consistent with a melt component inferred from An80–90 plagioclase in the calc-alkalic rocks (Fig. 16a and Table 8). A felsic melt component, L5, can be produced by 1:9 mixing of L1 and L3 (Fig. 16a and Table 8). Inferred intermediate calc-alkalic melt components can then be interpreted as mixing products between a felsic component (L5), and a mafic component (L4) as shown in Fig. 16a and Table 8.
Liquid mixing and cryptic mixing in calc-alkalic magmas
Isotopic and elemental compositions can be used to successfully identify the melt components that mixed to produce the variety of calc-alkalic rocks in Zao volcano. It should be stressed here that calc-alkalic andesites from Zao volcano, whose end-member components are produced by mixing between a calc-alkalic basaltic melt (L1 in Fig. 16a), and tholeiitic basaltic and felsic melts (L2 and L3, respectively), contain plagioclase crystallizing from mixed end-member components (L4 and L5), but not from L1, L2, nor L3 (see calc-alkalic andesite in Fig. 13). This observation implies that mixing of melts or liquids, not magmas containing plagioclase crystals, and subsequent crystallization of the mixed melts is the likely process that formed the calc-alkalic end-member components. On the other hand, the existing calc-alkalic andesites are produced by mixing of these end-member ‘magmas’, which contain plagioclase and other phenocryst phases.
The SiO2 contents and 87Sr/86Sr of the inferred mixed magmas formed by the above processes can be calculated using the compositions assumed for L1, L2, and L3 (Table 8 and Fig. 16b), and are plotted along the mixing curve between L4 and L5 in Fig. 16b. The existing calc-alkalic andesites have 87Sr/86Sr lower than the inferred mixed magma (Fig. 16b). One possible mechanism to explain this apparent dilemma would be that an above-liquidus liquid, not a sub-liquidus magma, with low 87Sr/86Sr, such as L1 (Fig. 16b), contributes to the existing calc-alkalic andesite formation. Zao calc-alkalic rocks, however, show no petrographic signs of such mixing. We thus propose to call this process ‘cryptic’ mixing of basaltic liquid or melt. A calc-alkalic magma that experiences cryptic mixing should not crystallize plagioclase phenocrysts. If this is the case, then such cryptic mixing must have taken place immediately before and could have caused the eruption.
One petrographical observation that characterizes calc-alkalic or Sakuyama's R-type andesites is the presence of pyroxenes in the groundmass that record higher temperatures than the phenocryst pyroxenes (e.g. Sakuyama, 1981, 1983; Wada, 1985; Fujinawa, 1988, 1990); this is the case for Zao calc-alkalic rocks (Fig. 15). This observation can be also explained by the cryptic mixing of a higher-T basaltic melt with a lower-T magma.
Genesis of tholeiitic vs calc-alkalic basaltic magmas: different fluid contributions
Petrological and geochemical observations suggest that at least two basaltic magmas are simultaneously present in the magma plumbing system of Zao volcano. One is a magma that differentiates to form the tholeiitic series, and the other a mafic end-member magma that mixes to form the calc-alkalic series rocks. Although basaltic andesites in Zao volcano are differentiated and do not represent primitive magma compositions, it is reasonable to assume that calcic plagioclase (An>90) in both tholeiitic and calc-alkalic mafic andesites crystallizes from different primitive magmas, as such plagioclase phenocrysts in the two rock series exhibit different concentrations of some trace elements, especially of Sr (Fig. 12), If so, then the geochemical characteristics of such primitive magmas, one tholeiitic and other calc-alkalic, can be inferred from the plagioclase compositions and element partitioning between plagioclase and silicate melt (Table 9 and Fig. 17).
|
|
Calc-alkalic rocks tend to be more enriched in incompatible trace elements than tholeiitic rocks, which is also the case for the inferred primitive magma compositions (Figs 7 and 17). This observation led Masuda & Aoki (1979
) to the conclusion that both calc-alkalic and tholeiitic primary magmas tap a common mantle source but the former is produced by lower degrees of melting. However, this study on volcanic rocks from Zao volcano indicates that the tholeiitic primitive magma has a higher 87Sr/86Sr than the calc-alkalic magma, suggesting a more radiogenic mantle source for the tholeiitic magmas. Such isotopic heterogeneity within the mantle wedge could be caused by larger contributions from isotopically enriched, slab-derived fluids to the tholeiitic magma source than to the calc-alkalic source.
To test this hypothesis quantitatively, geochemical modeling of element transport by slab-derived fluid addition to the mantle wedge and subsequent incremental fractional melting was conducted (Table 10 and Fig. 18). The model compositions, including H2O contents of subducted altered oceanic crust and terrigenous sediments, are from Tatsumi & Hanyu (2003). Experimental data for the mobility of Sr during dehydration of amphibolite (Kogiso et al., 1997) and sediment (Aizawa et al., 1999) were used to estimate the fluid compositions. Sr concentrations (357 and 664 ppm) and 87Sr/86Sr (0·70420 and 0·70343) in tholeiitic and calc-alkalic primary magmas, respectively, are based on the average compositions of melts in equilibrium with An>90 plagioclase phenocrysts (Fig. 13). The results (Table 10 and Fig. 18) indicate that the geochemical characteristics of the tholeiitic and calc-alkalic primary magmas can be explained by different contributions from slab-derived fluids (1% vs 0·2%) and different degrees of melting (40% vs 7%). It is suggested that these processes produce primary magmas containing 2·63–2·65 wt % H2O, which are acceptable values for arc magmas (e.g. Hauri et al., 2006).
|
|
Although the different contributions from enriched slab-derived fluids, together with different degrees of melting, could generate two types of primary magma in the mantle wedge, this may not be a plausible process to produce the tholeiitic and calc-alkalic magmas of Zao volcano. This process does not explain why magma mixing plays a major role in the differentiation of calc-alkalic magmas but not in the tholeiitic magmas; the two types of mantle-derived basalt magmas should have an equal chance of magma mixing.
Genesis of tholeiitic vs calc-alkalic basaltic magmas: crust vs mantle melts
When mantle-derived basaltic magmas are underplated and/or intruded into the arc crust they transfer heat into the overlying and surrounding crust, which can lead to partial melting of the wall-rocks (e.g. Hildreth, 1981; Raia & Spera, 1997; Annen & Sparks, 2002). However, herein lies a problem: whether or not basaltic magmas emplaced at the base of the lower crust could transfer enough heat to continue to cause crustal anatexis; heat transfer from the basaltic magma to the crust would cause a rapid temperature drop in the magma, which would lead to the magma being unable to further melt the crust (e.g. Marsh, 1989; Petford & Gallagher, 2001). This problem may be overcome if the temperature of the pre-existing crust is high enough; that is, the basaltic magma intrudes where crustal temperatures are near the basalt solidus or the crust is in a partially molten state (Couch et al., 2001; Tatsumi et al., 2006). Numerical simulations of heat transfer (Annen & Sparks, 2002; Annen et al., 2006) further suggest a model in which mantle-derived basalts emplaced as a succession of sills into the lower crust generate a deep crustal ‘hot zone’ where differentiated melts are produced from two distinct sources: crystallization of mantle-derived magma, and melting of crustal rocks.
A ‘hot zone’ is likely to occur beneath the NE Japan arc. This is believed to be the case for the following two reasons. First, the temperature of the NE Japanese mantle-derived magma is much higher than the solidus temperature of the lower crust. High-pressure melting experiments (Tatsumi et al., 1983) suggest that primary magmas beneath the volcanic front of this arc equilibrate with the mantle at 1400°C and 1· 0 GPa, a pressure equivalent to the depth immediately beneath the Moho underlying the NE Japan arc. Second, low-frequency tremors and micro-earthquakes, which may be caused by deformation associated with magma intrusion, are observed at depths of 30–50 km only beneath the Quaternary volcanoes of the volcanic front of the NE Japan arc (Obara, 2002; Katsumata & Kamaya, 2003).
The petrographic and geochemical data for the Zao volcanic rocks presented here suggest the presence of two distinct basaltic magmas beneath a single volcano; one a high 87Sr/86Sr tholeiitic magma and the other a low 87Sr/86Sr calc-alkalic magma. If the formation of a ‘hot zone’, resulting from basaltic underplating and subsequent generation of both crust-derived and mantle-derived magmas, is accepted, then it is reasonable to suggest that the isotopically enriched tholeiitic and depleted calc-alkalic magmas may be crust- and mantle-derived, respectively.
Contributions from melting of crustal and mantle materials to generate tholeiitic and calc-alkalic magmas, respectively, are now examined on the basis of their geochemical characteristics. Relative abundances of trace elements between tholeiitic and calc-alkalic primitive magmas, which are based on solid–melt partitioning between Ca-rich plagioclase and silicate melts (Table 9), are plotted as a function of the ionic radii of elements in Fig. 19. Systematic patterns of enrichment and depletion of certain elements can be observed for the two magma series. This pattern and a consideration of the crystal structure control on trace element partitioning between melts and solid phases (Matsui et al., 1977) may suggest that the tholeiitic basalt magma is more depleted in elements that are likely to be partitioned into plagioclase and possibly amphibole than the calc-alkalic basalt magma (Fig. 19). This can be understood as the result of buffering of particular elements by residual phases during partial melting; for example, Sr and Eu by plagioclase, and K and Ba by amphibole. If so, then the melting residue of the tholeiitic basalt magma is distinct from that of the calc-alkalic basalt magmas in the presence of plagioclase (and amphibole). These phases are commonly observed in the melting residues of hydrous mafic compositions such as amphibolite (e.g. Beard & Lofgren, 1991; Beard et al., 1993; Patiño Douce & Beard, 1995), but are unlikely to coexist with primary arc basalt magmas in a peridotite system (e.g. Tatsumi et al., 1983). Therefore, it is possible that the geochemical characteristics of the tholeiitic and calc-alkalic basaltic magma series in Zao volcano can be understood if they are produced by partial melting of amphibolitic lower crust and peridotitic upper mantle, respectively.
|
Subduction zone tholeiitic magmas have been considered to form from mantle-derived basaltic magmas via differential crystallization (e.g. Wada, 1981
; Sakuyama, 1983; Fujinawa, 1988, 1990; Tatsumi & Kogiso, 2003) for the following reasons: (1) tholeiitic rocks show little evidence of disequilibrium textures; (2) they exhibit systematic changes in both phenocryst compositions and assembl-ages; (3) the tholeiitic trend can be explained by the fractionation of phenocryst phases. On the other hand, these petrographic and compositional characteristics are also consistent with inverse differential crystallization of a parental basaltic magma (i.e. partial melting of a basaltic source) having affected the tholeiitic rocks. As a basaltic parental magma for the Zao tholeiitic rocks is generated by melting of mafic lower crust caused by heat transfer from an underplating mantle-derived, calc-alkalic basaltic magma, it is reasonable to suggest that differentiated tholeiitic melts are also created via crustal anatexis rather than crystallization of a mantle-derived basaltic magma.
Magma plumbing system beneath Zao volcano
The model for the generation of the two types of magmas, tholeiitic and calc-alkalic, beneath Zao volcano described below is shown schematically in Fig. 20.
|
A mantle-derived basaltic magma (L1), which finally equilibrates with the upper mantle immediately below the Moho (
1· 0 GPa; Tatsumi et al., 1983), underplates and transfers heat to the base of the crust, causing both its own crystallization and partial melting of the lower crust. Although the temperature and melt fraction gradually decrease upwards within the partially molten ‘hot zone’ formed at the base of the lower crust, the hot zone is simplified to consist of two sub-zones with higher and lower melt fractions that generate basaltic and felsic melts. These are melts L2 and L3, respectively, inferred from the isotopic compositions of plagioclase phenocrysts in the tholeiitic rocks (Figs 16 and 20). The boundary between the two sub-zones may be defined by the breakdown and melting of amphibole, which causes an abrupt increase in melt fraction (e.g. Foden & Green, 1992; Annen et al., 2006). If so, then the temperature at this boundary would be 1075°C (Müntener et al., 2001). This temperature estimate is consistent with the following petrographic and experimental constraints; first, the highest temperature estimate for tholeiitic magmas based on a two-pyroxene geothermometer is 1075°C (Fig. 15); second, experiments at 0·3 GPa on a basalt (SiO2 49 wt %; Al2O3 18 wt %; FeO* 11 wt %; MgO 7 wt %) in the presence of 0·5 wt % H2O yield a partial melt with a composition similar to that of the tholeiitic andesitic basalts (SiO2 52 wt %) at 1100°C by 40–50% of partial melting (Tatsumi & Suzuki, in preparation). A variable contribution of L2 vs L3 through liquid–liquid mixing yields melts with mafic to felsic compositions, which crystallize in shallow-level magma reservoirs to form the tholeiitic rocks (Fig. 20). Alternatively, differential crystallization of L2 may also contribute to the production of tholeiitic magmas with intermediate compositions. In addition to these processes, contamination of the magmas by upper crustal granitic rocks plays a role in controlling the Sr isotopic compositions of the tholeiitic magmas (Fig. 20), which is suggested by (1) the observation that plagioclase rims tend to have higher 87Sr/86Sr than cores in tholeiitic rocks (Fig. 13) and (2) bulk compositions exhibiting higher 87Sr/86Sr with increasing SiO2 (Fig. 16).
The generation of calc-alkalic magmas is distinct from tholeiitic magmas in that it involves a mantle-derived basaltic component, L1, either as a liquid or magma (Fig. 20). This contributes to the generation of mafic (L4) and felsic (L5) end-member magmas for andesites via mixing with crust-derived, tholeiitic basalt (L2) and felsic (L3) melts. Furthermore, the calc-alkalic primitive liquid L1 plays a role in the production of all calc-alkalic magmas via cryptic mixing (Fig. 16). On the other hand, a magma with the composition of L1, containing Ca-rich and low 87Sr/86Sr (0·7034) plagioclase phenocrysts, can be identified in the most mafic basaltic andesite in Zao volcano (Figs 16 and 20).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONOVERVIEW OF THOLEIITIC AND...GEOLOGICAL BACKGROUND OF ZAO...ANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
It is generally accepted that the differentiation of calc-alkalic magmas involves variable contributions from crustal components; for example, via wall-rock assimilation, or mixing with crust-derived felsic magma (Eichelberger, 1975
; DePaolo, 1981; Sakuyama, 1981; Hildreth & Moorbath, 1988; Clynne, 1999; Dungan & Davidson, 2004; Tatsumi & Kogiso, 2003), whereas tholeiitic magmas show more pristine mantle signatures. The hypothesis presented here, which proposes a crustal origin for the tholeiitic magmas and a mantle origin for the calc-alkalic basaltic magmas, requires that these models be revised and even in some cases discarded.
One aspect that we need to re-examine concerns the geochemical characteristics of the ‘mantle-derived’ basalt magmas that are used to understand the contribution of slab-derived components to arc magma generation. To minimize the effect of shallow-level crustal contamination and to assess the magma source characteristics, basaltic rocks and/or tholeiitic rocks tend to be examined (e.g. Notsu, 1983; Sakuyama & Nesbitt, 1986; Shibata & Nakamura, 1997; Kimura & Yoshida, 2006). We suggest that the tholeiitic basalt magmas in Zao volcano are derived from melting of mafic lower crust via underplating of calc-alkalic, mantle-derived basalt magmas and subsequent crustal anatexis. If so, then the tholeiitic basalt, although it is relatively primitive, does influence the geochemical signatures of Quaternary mantle-derived arc magmas, contributing to the production of calc-alkalic magmas. However, only by examining the phenocryst phases that crystallized from the least differentiated mantle-derived magmas can the effect of shallow-level magma mixing processes on the mantle signatures be hinted at. Therefore, analysis and examination of the compositions of minerals that crystallize from the primitive calc-alkalic basalt magma could provide the only chance to fully understand the geochemical characteristics of a mantle-derived magma, and hence the source mantle and slab-derived components.
|
ACKNOWLEDGEMENTS |
|---|
We thank Bogdan Vaglarov for analytical assistance, Miki Fukuda for preparing the manuscript and figures, and Richard Price, Bruce Charlier, Gene Yogodzinski, Alex Nichols, and the editor John Gamble for their critical and constructive comments on the manuscript. This work is partially supported by Grant-in-Aid for Creative Scientific Research (19GS0211).
*Corresponding author. E-mail: tatsumi{at}jamstec.go.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONOVERVIEW OF THOLEIITIC AND...GEOLOGICAL BACKGROUND OF ZAO...ANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Aizawa Y, Tatsumi Y, Yamada H. Element transport during dehydration of subducting sediments: implication for arc and ocean island magmatism. Island Arc (1999) 8:38–46.[CrossRef]
Annen C, Sparks R. SJ. Effects of repetitive emplacement of basaltic intrusions on thermal evolution and melt generation in the crust. Earth and Planetary Science Letters (2002) 203:937–955.[CrossRef][Web of Science]
Annen C, Blundy JD, Sparks R. SJ. The genesis of intermediate and silicic magmas in deep crustal hot zones. Journal of Petrology (2006) 47:505–539.
Baker MB, Grove TL, Price R. Primitive basalts and andesites from the Mt. Shasta region, N. California: products of varying melt fraction and water content. Contributions to Mineralogy and Petrology (1994) 118:111–129.[CrossRef][Web of Science]
Baker WE. The role of humic acids from Tasmanian podzolic soils in mineral degradation and metal mobilization. Geochimica et Cosmochimica Acta (1973) 37:269–281.[CrossRef][Web of Science]
Barling J, Weis D. Influence of non-spectral matrix effects on the accuracy of Pb isotope ratio measurement by MC-ICP-MS: implications for the external normalization method of instrumental mass bias correction. Journal of Analytical Atomic Spectrometry (2008) 23:1017–1025.[CrossRef][Web of Science]
Bau M, Knittel U. Significance of slab-derived partial melts and aqueous fluids for the genesis of tholeiitic and calc-alkaline island-arc basalts; evidence from Mt. Arayat, Philippines. Chemical Geology (1993) 105:233–251.[CrossRef][Web of Science]
Beard JS, Lofgren GE. Dehydration melting and water-saturated melting of basaltic and andesitic greenstones and amphibolites at 1, 3 and 6·9 kb. Journal of Petrology (1991) 32:365–401.
Beard JS, Abitz RJ, Lofgren GE. Experimental melting of crustal xenoliths from Kilbourne Hole, New Mexico and implications for the contamination and genesis of magmas. Contributions to Mineralogy and Petrology (1993) 115:88–102.[CrossRef][Web of Science]
Bindeman IN, Davis AM. Trace element partitioning between plagioclase and melt; investigation of dopant influence on partition behavior. Geochimica et Cosmochimica Acta (2000) 64:2863–2878.[CrossRef][Web of Science]
Bindeman IN, Davis AM, Drake MJ. Ion microprobe study of plagioclase–basalt partition experiments at natural concentration levels of trace elements. Geochimica et Cosmochimica Acta (1998) 62:1175–1193.[CrossRef][Web of Science]
Blatter DL, Carmichael I. SE. Hydrous phase equilibria of a Mexican high-silica andesite: A candidate for a mantle origin? Geochimica et Cosmochimica Acta (2001) 65:4043–4065.[CrossRef][Web of Science]
Bowen NL. The Evolution of the Igneous Rocks (1928) Princeton, NJ: Princeton University Press.
Brophy JG. Andesites from northeastern Kanga Island, Aleutians—implication for calc-alkaline fractionation mechanisms and magma chamber development. Contributions to Mineralogy and Petrology (1990) 104:568–581.[CrossRef][Web of Science]
Chang Q, Shibata T, Shinotsuka K, Yoshikawa M, Tatsumi Y. Precise determination of trace elements in geological standard rocks using inductively coupled plasma mass spectrometry (ICP-MS). Frontier Research on Earth Evolution (2003) 1:357–362.
Charlier BLA, Ginibre C, Morgan D, Nowell GM, Pearson DG, Davidson JP, Ottley CJ. Methods for the microsampling and high-precision analysis of strontium and rubidium isotope at single crystal scale for petrological and geochronological applications. Chemical Geology (2006) 232:114–133.[CrossRef][Web of Science]
Clynne MA. A complex magma mixing origin for rocks erupted in 1915, Lassen Peak, California. Journal of Petrology (1999) 40:105–132.[CrossRef][Web of Science]
Couch S, Sparks R. SJ, Carroll MR. Mineral disequilibrium in lavas explained by convective self-mixing in open magma chambers. Nature (2001) 411:1037–1039.[CrossRef][Web of Science][Medline]
Crawford AJ, Falloon TJ, Green DH. Classification, petrogenesis and tectonic setting of boninites. In: Boninites and Related Rocks—Crawford AJ, ed. (1989) London: Unwin Hyman. 1–49.
Davidson JP, Tepley FJ. Recharge in volcanic systems: evidence from isotope profiles of phenocrysts. Science (1997) 275:826–829.
Davidson JP, Tepley FJ, Palacz Z, Meffan-Main S. Magma recharge, contamination and residence times revealed by in situ laser ablation isotopic analysis of feldspar in volcanic rocks. Earth and Planetary Science Letters (2001) 184:427–442.[CrossRef][Web of Science]
Defant MJ, Drummond MS. Derivation of some modern arc magmas by melting of young subducted lithosphere. Nature (1990) 347:662–665.[CrossRef][Web of Science]
DePaolo DJ. Trace element and isotopic effects of combined wallrock assimilation and fractional crystallization. Earth and Planetary Science Letters (1981) 53:189–202.[CrossRef][Web of Science]
Dickinson WR. Potash–depth (K–h) relations in continental margin and intra-oceanic magmatic arcs. Geology (1975) 3:53–56.
Dungan MA, Davidson J. Partial assimilative recycling of the mafic plutonic roots of arc volcanoes: An example from the Chilean Andes. Geology (2004) 32:773–776.
Eichelberger JC. Origin of andesite and dacite: evidence of mixing at Glass Mountain in California and at other Circum-Pacific volcanoes. Geological Society of America Bulletin (1975) 86:1381–1391.
Ewart A. The mineralogy and petrology of Tertiary–Recent orogenic volcanic rocks with special reference to the andesitic–basaltic compositional range. In: Andesites—Thorpe RS, ed. (1982) Chichester: John Wiley. 26–87.
Fenner CN. The crystallization of basalts. American Journal of Science (1929) 18:225–253.[Web of Science]
Foden JD, Green DH. Possible role of amphibole in the origin of andesite; some experimental and natural evidence. Contributions to Mineralogy and Petrology (1992) 109:479–493.[CrossRef][Web of Science]
Fujinawa A. Tholeiitic and calc-alkaline magma series at Adatara Volcano, Northeast Japan: 1. Geochemical constraints on their origin. Lithos (1988) 22:135–158.[CrossRef][Web of Science]
Fujinawa A. Tholeiitic and calc-alkaline magma series at Adatara Volcano, Northeast Japan: 2. Mineralogy and phase relations. Lithos (1990) 24:217–236.[CrossRef][Web of Science]
George R, Turner S, Hawkesworth C, Bacon CR, Nye C, Stelling P, Dreher S. Chemical versus temporal controls on the evolution of tholeiitic and calc-alkaline magmas at two volcanoes in the Alaska–Aleutian arc. Journal of Petrology (2004) 45:203–219.
Gill JB. Orogenic Andesites and Plate Tectonics (1981) Berlin: Springer-Verlag. 390.
Grove TL, Baker MB. Phase equilibrium controls on the tholeiitic versus calc-alkaline differentiation trends. Journal of Geophysical Research (1984) 89:3253–3274.
Grove TL, Parman SW, Bowring SA, Price RC, Baker MB. The role of an H2O-rich fluid component in the generation of primitive basaltic andesites and andesites from the Mt. Shasta region, N California. Contributions to Mineralogy and Petrology (2002) 142:375–396.[Web of Science]
Hamada M, Fujii T. Experimental constraints on the effects of pressure and H2O on the fractional crystallization of high-Mg island arc basalt. Contributions to Mineralogy and Petrology (2008) 155:767–790.[CrossRef][Web of Science]
Hauri EH, Shaw AM, Wang J, Dixon JE, King PL, Mandeville C. Matrix effects in hydrogen isotope analysis of silicate glasses by SIMS. Chemical Geology (2006) 235:352–365.[CrossRef][Web of Science]
Hess PC. Origin of Igneous Rocks (1989) Cambridge, MA: Harvard University Press.
Hildreth W. Gradients in silicic magma chambers: implications for lithospheric magmatism. Journal of Geophysical Research (1981) 86:10153–10192.
Hildreth W, Moorbath S. Crustal contributions to arc magmatism in the Andes of central Chile. Contributions to Mineralogy and Petrology (1988) 98:455–489.[CrossRef][Web of Science]
Hirata T. Lead isotopic analysis using multiple collector-inductively coupled plasma mass spectrometry coupled with modified external correction method for mass discrimination effect. Analyst (1996) 121:1407–1411.[CrossRef]
Horn I, Rudnick RL, McDonough WF. Precise elemental and isotopic ratio determination by simultaneous solution nebulization and laser ablation-ICP-MS: application to U–Pb geochronology. Chemical Geology (2000) 164:281–301.[CrossRef][Web of Science]
Horwitz EP, Chiarizia R, Deitz ML. A novel strontium-selective extraction chromatographic resin. Solvent Extraction and Ion Exchange (1992) 10:313–336.[CrossRef][Web of Science]
Hunter AG. Intracrustal controls on the coexistence of tholeiitic and calc-alkaline magma series at Aso Volcano, SW Japan. Journal of Petrology (1998) 39:1255–1284.[CrossRef][Web of Science]
Hunter AG, Blake S. Petrogenetic evolution of a transitional tholeiitic–calc-alkaline series: Towada volcano, Japan. Journal of Petrology (1995) 36:1579–1605.
Iizuka T, Hirata T. Improvements of precision and accuracy in in situ Hf isotope microanalysis of zircon using the laser ablation-MC-ICPMS technique. Chemical Geology (2005) 220:121–137.[CrossRef][Web of Science]
Irvine TN, Baragar W. RA. A guide to the chemical classification of the common volcanic rocks. Canadian Journal of Earth Sciences (1971) 8:523–548.
Jakes P, Gill J. Rare earth elements and the island arc tholeiitic series. Earth and Planetary Science Letters (1970) 9:17–28.[CrossRef][Web of Science]
Kagami H. Formative periods and source materials of Cretaceous–Paleogene granitoids from Honshu Arc. Journal of the Geological Society of Japan (2005) 111:441–457.
Kamei A, Takagi T, Kubo K. Geology and petrography of the Abukuma Granites in the Hiyama District, Fukushima Prefecture, NE Japan. Bulletin of the Geological Survey of Japan (2003) 54:395–409.
Katsumata A, Kamaya N. Low-frequency continuous tremor around the Moho discontinuity away from volcanoes in the southwest Japan. Journal of Geophysical Research (2003) 30(1):1020. doi: 10.1029/2002GL015981.
Kawano Y, Yagi K, Aoki K. Petrography and petrochemistry of the volcanic rocks of Quaternary volcanoes of northeastern Japan. Science Reports, Tohoku University, Series 3: Mineralogy, Petrology, and Economic Geology (1961) 7:1–46.
Kay RW. Aleutian magnesian andesites: melts from subducted Pacific Ocean crust. Journal of Volcanology and Geothermal Research (1978) 4:117–132.[CrossRef][Web of Science]
Kelemen PB. Genesis of high Mg-number andesites and the continental crust. Contributions to Mineralogy and Petrology (1995) 120:1–19.[Web of Science]
Kersting AB, Arculus RJ, Gust DA. Lithospheric contributions to arc magmatism: isotope variations along strike in volcanoes of Honshu, Japan. Science (1996) 272:1464–1468.[Abstract]
Kimura J-I, Yoshida T. Contributions of slab fluid, mantle wedge and crust to the origin of Quaternary lavas in the NE Japan arc. Journal of Petrology (2006) 47:2185–2232.
Kimura J-I, Danhara T, Iwano H. A preliminary report on trace element distributions in zircon and apatite crystals using excimer laser ablation inductively coupled plasma mass spectrometry (ExLA-ICP-MS). Fission Track Newsletter (2000) 13:11–20.
Kimura J-I, Yoshida T, Iizumi S. Origin of low-K andesite lavas at Nekoma volcano, NE Honshu arc, Japan: geochemical constraints for lower-crustal melts. Journal of Petrology (2002) 43:631–661.
Kogiso T, Tatsumi Y, Nakano S. Trace element transport during dehydration processes in the subducted oceanic crust: 1, Experiments and implications for the origin of ocean island basalts. Earth and Planetary Science Letters (1997) 148:193–205.[CrossRef][Web of Science]
Kuno H. Petrology of Hakone volcano and the adjacent areas, Japan. Geological Society of America Bulletin (1950) 61:957–1014.
Kuno H. Origin of Cenozoic petrographic provinces of Japan and surrounding areas. Bulletin of Volcanology (1959) 20:37–76.
Kuno H. High-alumina basalt. Journal of Petrology (1960) 1:121–145.
Kuno H. Differentiation of basalt magmas. In: Basalts—The Poldervaart Treatise on Rocks of Basaltic Composition, Volume 2—Hess HH, Poldervaart A, eds. (1968) New York: Interscience. 623–688.
Marsh BD. Magma chambers. Annual Review of Earth and Planetary Sciences (1989) 17:439–474.[CrossRef]
Martin E. Effect of steeper Archean geothermal gradient on geochemistry of subduction-zone magmas. Geology (1986) 14:753–756.
Masuda Y, Aoki K. Trace element variations in the volcanic rocks from the Nasu Zone, Northeast Japan. Earth and Planetary Science Letters (1979) 44:139–149.[CrossRef][Web of Science]
Matsui Y, Onuma N, Nagasawa H, Higuchi H, Banno S. Crystal structure control in trace element partition between crystal and magma. Bulletin de la Société Française de Minéralogie et de Cristallographie (1977) 100:315–324.
Miyashiro A. Volcanic rock series in island arcs and active continental margins. American Journal of Science (1974) 274:321–355.[Abstract]
Miyazaki T, Takahashi T, Hirahara Y, Chang Q, Vaglarov BS, Suzuki K, Tatsumi Y. Low blank and precise Sr–Nd–Pb analysis and in-situ Sr analysis of minerals in volcanic rocks. Frontier Research on Earth Evolution (2007) 3:72–76.
Müntener O, Kelemen PB, Grove TL. The role of H2O during crystallization of primitive arc magmas under uppermost mantle conditions and genesis of igneous pyroxenites: an experimental study. Contributions to Mineralogy and Petrology (2001) 141:643–658.[Web of Science]
Nakamura K, Chang Q. Precise determination of ultra-low (sub-ng g–1) level rare earth elements in ultramafic rocks by quadrupole ICP-MS. Geostandards and Geoanalytical Research (2007) 31:185–197.[CrossRef][Web of Science]
Nockolds SR, Allen R. The geochemistry of some igneous rock series [Part 1]. Geochimica et Cosmochimica Acta (1953) 4:105–142.[CrossRef][Web of Science]
Notsu K. Strontium isotope composition in volcanic rocks from the Northeast Japan arc. Journal of Volcanology and Geothemal Research (1983) 18:531–548.[CrossRef]
Oba Y, Konda T. Geology and petrology of central Zao Volcano, Yamagata Prefecture. Bulletin of the Yamagata University, Natural Science (1989) 12:199–210.
Obara K. Nonvolcanic deep tremor associated with subduction in Southwest Japan. Science (2002) 296:1679–1681.
Pachett PJ, Kuovo O, Hedge CE, Tatsumoto M. Evolution of continental crust and mantle heterogeneity; evidence from Hf isotopes. Contributions to Mineralogy and Petrology (1981) 78:279–297.[Web of Science]
Parman SW, Grove TL. Harzburgite melting with and without H2O: Experimental data and predictive modeling. Journal of Geophysical Research (2004) 109. paper number B02201.
Patiño Douce AE, Beard JS. Dehydration-melting of biotite gneiss and quartz amphibolite from 3 to 15 kbar. Journal of Petrology (1995) 36:707–738.
Pearce JA, Van der Laan SR, Arculus RJ, Murton BJ, Ishii T, Peate DW, Parkinson IJ. Boninite and harzburgite from Leg 125. (Bonin–Mariana forearc): a case study of magma genesis during the initial stages of subduction. In: Proceedings of the Ocean Drilling Program, Scientific Results, 125—Fryer P, Pearce JA, Stokking LB, et al, eds. (1992) College Station, TX: Ocean Drilling Program. 623–659.
Pearce NJG, Perkins WT, Westgate JA, Gorton MG, Jackson SE, Near CR, Chenery SP. A compilation of new and published major and trace element data for NIST SRM610 and NIST SRM612 glass reference materials. Geostandard Newsletters (1996) 21:115–144.
Petford N, Atherton M. Na-rich partial melts from newly underplated basaltic crust: the Cordillera Blanca batholith, Peru. Journal of Petrology (1996) 37:1491–1521.
Petford N, Gallagher K. Partial melting of mafic (amphibolitic) lower crust by periodic influx of basaltic magma. Earth and Planetary Science Letters (2001) 193:483–499.[CrossRef][Web of Science]
Pichavant M, Martel C, Bourdier JL, Scaillet B. Physical conditions, structure, and dynamics of a zoned magma chamber: Mont Pelée (Martinique, Lesser Antilles Arc). Journal of Geophysical Research (2002) 107:2093.[CrossRef]
Plank T, Langmuir CH. An evaluation of the global variations in the major element chemistry of arc basalts. Earth and Planetary Science Letters (1988) 90:349–370.[CrossRef][Web of Science]
Prouteau G, Scaillet B. Experimental constraints on the origin of the 1991 Pinatubo dacite. Journal of Petrology (2003) 44:2203–2241.
Raia F, Spera FJ. Simulations of crustal anatexis; implications for the growth and differentiation of continental crust. Journal of Geophysical Research (1997) 102:22629–22648.[CrossRef]
Rudnick RL. Making continental crust. Nature (1995) 378:571–578.[CrossRef][Web of Science]
Rudnick RL, Gao S. The composition of the continental crust. In: The Crust—Rudnick RL, ed. (2003) Oxford: Pergamon. 1–64.
Sakayori A. Geology and petrology of Zao Volcano. Journal of Mineralogy, Petrology and Economic Geology (1992) 87:433–444.
Sakuyama M. Lateral variation of H2O contents in Quaternary magmas of north-eastern Japan. Earth and Planetary Science Letters (1979) 43:103–111.[CrossRef][Web of Science]
Sakuyama M. Petrological study of the Myoko and Kurohime volcanoes, Japan: crystallization sequence and evidence for magma mixing. Journal of Petrology (1981) 22:553–583.
Sakuyama M. Phenocryst assemblages and H2O content in Circum-Pacific arc magmas. In: Geodynamics of the Western Pacific–Indonesian Region. American Geophysical Union, Geodynamics Series—Hilde T. WC, Uyeda S, eds. (1983) 11:143–158.
Sakuyama M, Nesbitt RW. Geochemistry of the Quaternary volcanic rocks of the Northeast Japan Arc. Journal of Volcanology and Geothermal Research (1986) 29:413–450.[CrossRef][Web of Science]
Shibata T, Nakamura E. Across-arc variations of isotope and trace element compositions from Quaternary basaltic volcanic rocks in northeastern Japan: Implications for interaction between subducted oceanic slab and mantle wedge. Journal of Geophysical Research (1997) 102:8051–8064.[CrossRef]
Shukuno H. Quantitative analysis of rock-forming minerals and volcanic glasses by electron probe microanalyzer. Frontier Research on Earth Evolution (2003) 1:129–136.
Sisson TW, Grove TL. Experimental investigations of the role of H2O in calc-alkaline differentiation and subduction zone magmatism. Contributions to Mineralogy and Petrology (1993) 113:143–166.[CrossRef][Web of Science]
Smith DR, Leeman WP. Petrogenesis of Mount St. Helens dacitic magmas. Journal of Geophysical Research (1987) 92:10313–10334.[CrossRef]
Stern CR, Kikian R. Role of the subducted slab, mantle wedge and continental crust in the generation of adakites from the Andean Austral volcanic zone. Contributions to Mineralogy and Petrology (1986) 123:263–281.
Streck MJ, Leeman WP, Chesley J. High-magnesian andesite from Mount Shasta; a product of magma mixing and contamination, not a primitive mantle melt. Geology (2007) 35:351–354.
Sun SS, McDonough WF. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Magmatism in the Ocean Basins. Geological Society, London, Special Publications—Saunders AD, Norry MJ, eds. (1989) 42:313–345.
Takahashi E. Genesis of calc-alkali andesite magma in a hydrous mantle–crust boundary: petrology of lherzolite xenoliths from the Ichinomegata Crater, Oga Peninsula, Northeast Japan; Part II. Journal of Volcanology and Geothermal Research (1986) 29:355–395.[CrossRef][Web of Science]
Takahashi T, Yoshikawa M, Shibata T, Tatsumi Y, Shimizu N. Sr isotopic micro analyses of plagioclase in andesites from Zao Volcano, NE Japan. Frontier Research on Earth Evolution (2005) 2. Sam-18.
Takaoka N, Konno K, Oba Y, Konda T. K–Ar dating of lavas from Zao Volcano, north-eastern Japan. Journal of the Geological Society of Japan (1989) 95:157–170.
Tani K, Kawabata H, Chang Q, Sato K, Tatsumi Y. Quantitative analyses of silicate rock major and trace elements by X-ray fluorescence spectrometer: Evaluation of analytical precision and sample preparation. Frontier Research on Earth Evolution (2006) 2:8.
Tatsumi Y. Melting experiments on a high-magnesian andesite. Earth and Planetary Science Letters (1981) 54:357–365.[CrossRef][Web of Science]
Tatsumi Y. High-Mg andesites in the Setouchi volcanic belt, southwestern Japan: analogy to Archean magmatism and continental crust formation? Annual Review of Earth and Planetary Sciences (2006) 34:467–499.[CrossRef][Web of Science]
Tatsumi Y, Eggins S. Subduction Zone Magmatism (1995) Boston, MA: Blackwell Science.
Tatsumi Y, Hanyu T. Geochemical modeling of dehydration and partial melting of subducting lithosphere: Towards a comprehensive understanding of high-Mg andesite formation in the Setouchi volcanic belt, SW Japan. Geochemistry, Geophysics, Geosystems (2003) 4:1081. doi:10.1029/2003GC000530.[CrossRef]
Tatsumi Y, Kogiso T. The subduction factory: its role in the evolution of the Earth's crust and mantle. Intra-Oceanic Subduction Systems: Tectonic and Magmetic Processes. Geological Society, London, Special Publications—Larter RD, Leat PT, eds. (2003) 219:55–80.
Tatsumi Y, Sakuyama M, Fukuyama H, Kushiro I. Generation of arc basalt magmas and thermal structure of the mantle wedge in subduction zones. Journal of Geophysical Research B (1983) 88:5815–5825.[CrossRef]
Tatsumi Y, Kawabata H, Sato K, Miyazaki T, Chang Q, Takahashi T, Tani K, Shibata T, Yoshikawa M. The petrology and geochemistry of Oto-Zan composite lava flow on Shodo-Shima Island, SW Japan: remelting of a solidified high-Mg andesite magma. Journal of Petrology (2006) 47:595–629.
Tatsumi Y, Shukuno H, Tani K, Takahashi N, Kodaira S, Kogiso T. Structure and growth of the Izu–Bonin–Mariana arc crust: II. The role of crust–mantle transformation and the transparent Moho in arc crust evolution. Journal of Geophysical Research (2008) 113:B02203. doi:10.1029/2007JB005121.[CrossRef]
Taylor SR. Cosmochemistry; potassium tells a tale. Nature (1995) 376:20–21.
Tepley FJ, Davidson JP, Clynne MA. Magmatic interactions as recorded in plagioclase phenocrysts of Chaos Crags, Lassen Volcanic Center, California. Journal of Petrology (1999) 40:787–806.[CrossRef][Web of Science]
Villiger S, Ulmer P, Müntener O. Equilibrium and fractional crystallization experiments at 0·7 GPa; the effect of pressure on phase relations and liquid compositions of tholeiitic magmas. Journal of Petrology (2007) 48:159–184.
Vroon PZ, van der Wagt B, Koornneef JM, Davies GR. Problems in obtaining precise and accurate Sr isotope analysis from geological materials using laser ablation MC-ICPMS. Analytical and Bioanalytical Chemistry (2008) 390:465–476. doi:10.1007/s00216-007-1742-9.[CrossRef][Web of Science][Medline]
Wada K. Contrasted petrological relations between tholeiitic and calc-alkalic series from Funagata Volcano, Northeast Japan. Journal of the Japanese Association of Mineralogists, Petrologists and Economic Geologists (1981) 76:215–231.
Wada K. Magma mixing process of calc-alkalic andesites from Funagata Volcano. Journal of the Japanese Association of Mineralogists, Petrologists and Economic Geologists (1985) 80:467–483.
Wager LR, Deer WA. Geological investigations in East Greenland. Part III. The petrology of the Skaergaard intrusion, Kangerdlugssuaq, East Greenland. Meddelelser om Gronland (1939) 105:352.
Wells PRA. Pyroxene thermometry in simple and complex systems. Contributions to Mineralogy and Petrology (1977) 62:129–139.[CrossRef][Web of Science]
Wilson M. Igneous Petrogenesis (1989) London: Unwin Hyman.
Woodhead J, Hergt J. Strontium, neodymium and lead isotope analyses of NIST glass certified reference materials: SRM 610, 612, 614. Geostandards Newsletter (2001) 25:261–266.[CrossRef][Web of Science]
Woodhead J, Hergt J, Shelley M, Eggins S, Kemp R. Zircon Hf-isotope analysis with an excimer laser, depth profiling, ablation of complex geometries, and concomitant age estimation. Chemical Geology (2004) 209:121–135.[CrossRef][Web of Science]
Woodhead J, Swearer S, Hergt J, Maas R. In situ Sr-isotope analysis of carbonates by LA-MC-ICP-MS: interference corrections, high spatial resolution and an example from otolith studies. Journal of Analytical Atomic Spectrometry (2005) 20:22–27.[CrossRef][Web of Science]
Woodhead JD, Hergt JM, Davidson JP, Eggins SM. Hafnium isotope evidence for ‘conservative’ element mobility during subduction zone processes. Earth and Planetary Science Letters (2001) 192:331–346.[CrossRef][Web of Science]
Yoder HS, Tilley CE. Origin of basalt magmas; an experimental study of natural and synthetic rock systems. Journal of Petrology (2003) 3:342–529.
Yogodzinski GM, Volynets ON, Koloskov AV, Seliverstov NI, Matvenkov VV. Magnesian andesites and the subduction component in a strongly calc-alkaline series at Piip Volcano, far western Aleutians. Journal of Petrology (1994) 35:163–204.
Yoshida T, Aoki K.-i. Geochemistry of major and trace elements in the Quaternary volcanic rocks from northeast Honshu, Japan. Science Reports, Tohoku University, Series 3: Mineralogy, Petrology, and Economic Geology (1984) 16:1–34.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||