Journal of Petrology Advance Access originally published online on November 7, 2006
Journal of Petrology 2007 48(1):79-111; doi:10.1093/petrology/egl055
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Hybridization of a Shallow ‘I-type’ Granitoid Pluton and its Host Migmatite by Magma-Chamber Wall Collapse: the Tokuwa Pluton, Central Japan
1Geological Institute, Graduate School of Environment and Information Sciences, Yokohama National University, Tokiwadai, Hodogaya-Ku, Yokohama 240-8501, Japan
2Geological Survey of Japan, 1-1-1 Higashi, Tsukuba 305-8567, Japan
RECEIVED NOVEMBER 20, 2003; ACCEPTED SEPTEMBER 1, 2006
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONThe Miocene Tokuwa pluton of ‘I-type’ granitoid affinity was emplaced discordantly into a Cretaceous to Paleogene accretionary complex and induced a contact aureole in which various thermally metamorphosed rocks were developed, including hornfels, metatexite, diatexite and cordierite-bearing tonalite (Crd-tonalite) of ‘S-type’ granite affinity. The thermally metamorphosed rocks show low-pressure reaction textures culminating in partial melting. Peak P–T conditions of
3 kbar at 780°C are estimated on the basis of the TWQ thermobarometer for the garnet-bearing rocks. The rocks in the contact aureole exhibit a gradual transition from hornfels, through metatexite and diatexite to Crd-tonalite. The Sr-isotopic composition at the time of Tokuwa pluton emplacement at 12 Ma decreases systematically from metatexite (0·7100–0·7112) through diatexite (0·7078–0·7094) to Crd-tonalite (0·7067–0·7068); this trend is interpreted in terms of mixing between the Tokuwa magma and the aureole migmatites. The field relationships, geochemical data, and isotopic data collectively suggest that the emplacement of the Tokuwa pluton triggered partial melting of the surrounding metasedimentary rocks. Subsequent hybridization of the Tokuwa magma with the metatexite in variable proportions produced the Crd-tonalite and diatexite. The hybridization was caused by invasion of the Tokuwa magma into the migmatite zone, accompanied by gravitational collapse of the previously crystallized wall of the magma chamber. The data presented demonstrate that even a relatively low-temperature, shallow, ‘I-type’ granitoid pluton can induce contact anatexis and hybrid ‘S-type’ granitoid formation at the intrusive contact. KEY WORDS: contact metamorphism; hybridization; magma–host-rock interaction; migmatite; ‘S-type’ granitoid
|
INTRODUCTION |
|---|
To understand the petrogenetic history of a granitic pluton it is essential to decipher not only the processes in the source region, but also the extent of the interaction of magma with its country rocks at the level of emplacement. It is well documented that granitoid magmas intruding partially melted middle crust interact with their anatectic host-rocks, forming hybrid facies along the contact (e.g. Ugidos & Recio, 1993
; Greenfield et al., 1996). However, in the case of shallow-level granitoid plutons evidence for contact anatexis induced by the plutons is scarce and occurs only where H2O is available to facilitate melting of the host-rocks. This is mainly a consequence of the relatively low temperature of the initial granitoid magmas, which are unlikely to cause extensive melting of country rocks that were initially at low temperature. Although contact anatexis around shallow-level granitoid plutons has been documented in some contact aureoles (e.g. Ballachulish aureole: Pattison & Harte, 1988; Holness & Clemens, 1999; Onawa aureole: Symmes & Ferry, 1995; Marchildon & Brown, 2001; Goat Ranch migmatite complex: Zeng et al., 2005a, 2005b), the mechanism of interaction between granitoid magmas and their anatectic host-rocks has received little attention in the literature.
The Miocene Tokuwa pluton is of ‘I-type’ granitoid affinity (Sato & Ishihara, 1983; Shimizu, 1986; Takahashi, 1989; Sato, 1991) and is exposed in the Izu arc collision zone, central Japan. The pluton exhibits a thermal aureole in which the rocks range from low-grade hornfels to migmatite; the migmatites locally exhibit a gradual transition from metatexite through diatexite to cordierite-bearing tonalite. The objectives of this study are (1) to describe the field relationships between the Tokuwa pluton and its host-rocks through the contact aureole, (2) to estimate the P–T conditions of emplacement of the Tokuwa pluton, and (3) to discuss the petrogenetic processes involved in magma–host-rock interaction within the contact aureole, based on isotopic and geochemical constraints.
|
GEOLOGICAL BACKGROUND |
|---|
The Izu Collision Zone (Taira et al., 1989
; Soh et al., 1991) is located near the Boso triple junction (Ogawa et al., 1989), where the northern Izu–Bonin–Mariana (IBM) arc system has been colliding with the Honshu arc since the middle Miocene (c. 15 Ma) (Niituma, 1989; Soh et al., 1991; Takahashi & Saito, 1997; Aoike, 1999) (Fig. 1a). The IBM arc is an immature intra-oceanic island arc in contrast to the mature Honshu arc. Neogene granitoid plutons are widely exposed in the Izu Collision Zone (Fig. 1a). The Kofu granitic complex (KGC) is the largest pluton among the Neogene granitoids in Japan and consists of several intermediate to felsic plutons including the Kinpusan, Tokuwa, Tonogi and Ashigawa plutons (Sato & Ishihara, 1983; Shimizu, 1986; Takahashi, 1989; Sato, 1991; Saito et al., 2004) (Fig. 1b). The Kinpusan pluton (northwestern part of KGC) consists of weakly peraluminous biotite-bearing granite, whereas the central part (Tokuwa pluton) predominantly consists of metaluminous hornblende–biotite-bearing granodiorite. The southern part of the KGC (Ashigawa and Tonogi plutons) is composed mainly of metaluminous hornblende-bearing tonalite and trondhjemite.
|
The Tokuwa pluton is the largest intrusive unit of the KGC. It intrudes discordantly into a Cretaceous to Paleogene accretionary complex that is composed mainly of shale and sandstone (Shimanto Belt, Fig. 2a). The internal part of the Tokuwa pluton corresponds to the magnetite-series of Ishihara (1977
), but the thick external part is composed mainly of ilmenite-series rocks (Fig. 2a). This external part is characterized by an abundance of sedimentary xenoliths and has been interpreted as a marginal facies generated by magma–country rock interaction (Shimizu, 1986; Takahashi, 1989; Sato, 1991). The main body of the Tokuwa pluton is medium-grained (2 mm grain size) and consists principally of plagioclase, quartz, K-feldspar, amphibole, biotite, and opaque minerals. It includes rocks ranging in composition from quartz-diorite, tonalite and granodiorites to granite (Fig. 3). The relatively mafic facies (< 19 modal % Qtz, < 1 modal % Kfs) are restricted to the external part; these have a large proportion of poikilitic hornblende (20–30 modal %) containing euhedral to subhedral plagioclase grains.
|
|
K–Ar dating for the KGC yields ages ranging from 15·7 to 7· 4 Ma (Kawano & Ueda, 1966
; Shibata et al., 1984; Saito et al., 1997), linking their genesis to the IBM arc–Honshu arc collision event. Although a number of petrological studies have been performed on the KGC, the petrogenesis of these plutons is as yet not fully understood. Sato (1991) suggested, based on Sr, O and S isotope data, that the range of petrographic and geochemical variations observed in the KGC was caused by variable degrees of assimilation of sedimentary materials into mantle-derived mafic magmas. In contrast, Saito et al. (2004) suggested that the southern plutons of the KGC (Ashigawa and Tonogi plutons) were formed by crystal fractionation processes from an intermediate parental magma that originated from the anatexis of basaltic crustal source materials. According to these workers, the collision process between the IBM and Honshu arcs caused the crustal anatexis that led to the formation of the Ashigawa and Tonogi plutons. |
ANALYTICAL METHODS |
|---|
Mineral analyses were carried out using an automated energy-dispersive electron microprobe (LINK QX2000J system) at the Geological Institute of Yokohama National University. The operating conditions were 15 kV and 0·155 nA on a cobalt standard. Data were processed using a LINK ZAF-4/FLS correction program. Representative mineral compositions are given in Table 1. Mineral abbreviations follow Kretz (1983
).
|
Major and trace element compositions were determined for thermally metamorphosed rocks in the contact aureole and granitoids from the Tokuwa pluton. The powder samples for whole-rock analysis were carefully prepared from 0·1–0·5 kg rock samples of the hornfels. The metatexite, diatexite, Crd-tonalite and granitoids are fine- to medium-grained rocks (< 3 mm) and the sample powders were prepared from 0·2–1· 0 kg rock samples. The analyzed samples of leucosome and schollen (10–50 g) were extracted from the diatexite. Crushed samples were pulverized using an agate ball mill. Major and trace element analyses were determined by X-ray fluorescence (XRF) (RIGAKU RIX-3000) at the National Institute of Polar Research, Japan. The analytical procedure followed the methods of Motoyoshi & Shiraishi (1995
), Motoyoshi et al. (1996) and Yamada et al. (1997).
Sr-isotopic compositions for several representative samples of various rock types were determined using a VG Sector mass spectrometer at the Geological Survey of Japan, following the procedure described by Nakajima et al. (1990). 87Sr/86Sr ratios were normalized to 86Sr/88Sr = 0·1194. Repeated analyses of the NBS987 standard during this study gave a 87Sr/86Sr of 0·71025 ± 0·00002 (1
). As the K–Ar biotite age of the Tokuwa pluton in the studied area is about 12 Ma (Saito et al., 1997), ‘initial’ Sr-isotopic ratios (SrI) at a time of the Tokuwa pluton emplacement were recalculated back to 12 Ma. Representative whole-rock and Sr-isotope compositions are given in Table 2. Modal proportions (%) of minerals determined by point counting of thin sections are also listed in Table 2 for granitoids from the Tokuwa pluton, the diatexite and the Crd-tonalite.
|
|
THE TOKUWA CONTACT AUREOLE |
|---|
A metamorphic thermal aureole (1–3 km wide) defined by the biotite-in isograd (Shimizu, 1986
) is developed around the Tokuwa pluton (Fig. 2). Hornfels is the dominant rock type in the aureole. The hornfels consists of pelitic and quartzofeldspathic layers, which are inferred to represent original sedimentary bedding. Towards the intrusive contact the aureole rocks exhibit progressive changes in their mineral assemblage summarized as: Bt + Ms + Chl + Qtz + Pl, Crd + Bt + Ms + Qtz + Pl ± Chl, And + Crd + Bt + Ms + Qtz + Pl ± Chl, Grt + Crd + Bt + Ms + Qtz + Pl, Kfs + Crd + Bt + Qtz + Pl ± Ms, And + Kfs + Crd + Bt + Ms + Qtz + Pl, Sil + And + Kfs + Crd + Bt + Ms + Pl ± Qtz ± Crn, and Grt + Kfs + Crd + Bt + Qtz + Pl (Table 3). Representative microstructures in the hornfels are shown in Fig. 4.
|
|
Metatexites are exposed in the aureole within
15 m of the intrusive contact (Loc. 1, 2, 3, 4 and 5; Fig. 2; Table 3). These have a layered appearance consisting of mica-rich pelitic and quartz-rich psammitic layers, similar to the hornfels. Although the mineral assemblage of the metatexites is comparable with that in the hornfels, the metatexite has distinct petrographic characteristics indicative of partial melting, which were not identified in the hornfels. These include: (1) the presence of chemically zoned plagioclase; (2) the development of thin interstitial films composed of quartz and feldspars along grain boundaries (see Sawyer, 1999); (3) development of euhedral crystals of quartz, cordierite, plagioclase and/or K-feldspar (Fig. 5a) (see Vernon & Collins, 1988).
|
Contact aureole and Tokuwa pluton contact in the Magi River section
Along the Magi River, in the southeastern part of the contact aureole around the Tokuwa pluton (Loc. 1, Fig. 2b), there are extensive exposures of hornfels and metatexite. The intrusive contact of the Tokuwa pluton with the metatexites is sharp and well exposed (Fig. 6a). Dykes (50 cm to 2 m in width) derived from the Tokuwa pluton are locally present, cutting across the host-rock structures (Fig. 6b). Aplitic veins (
50 cm width) cut both the host-rock structures and the dykes (Fig. 6b). Thin quartz veins (< 1 cm width) occur in the hornfels and are concordant with the structures in the host-rocks, although in some places they cut the layering.
|
The Tokuwa pluton exposed in the Magi River section shows a wide range of petrographic variation that can be grouped into three main facies: (1) medium- to fine-grained, hornblende–biotite-bearing tonalite (Hb-tonalite); (2) fine-grained, hornblende-free, biotite-bearing porphyritic tonalite (Bt-tonalite); (3) medium- to fine-grained, hornblende–biotite-bearing quartz-diorite, locally developing a comb-layering structure (see Moore & Lockwood, 1973
). The Hb-tonalite consists of plagioclase (54–65 modal %; An27–79), quartz (17–29 modal %), hornblende (5–21 modal %), biotite (6–10 modal %), and opaque minerals (< 1 modal %) and overlaps the field of tonalite (Figs 3 and 6c). Aligned plagioclase laths are a characteristic feature of the fine-grained Hb-tonalite (Fig. 6d), suggesting a cumulate origin. This fine-grained type of Hb-tonalite is restricted to the marginal part of the pluton (up to 100 m from the intrusive contact). In the vicinity of the contact ( 1 m from the country rocks), the fine-grained Hb-tonalite contains orthopyroxene xenocrysts, which in places are surrounded by cummingtonite. The Bt-tonalite consists of plagioclase (46–54 modal %; An23–80), quartz (34–40 modal %), biotite (10–13 modal %) and opaque minerals (< 1 modal %). In the Qtz–Kfs–Pl diagram, this rock overlaps with the field for tonalite (Fig. 3). The Bt-tonalite is characterized by a porphyritic texture consisting of oscillatory-zoned (An23–80) phenocrystic plagioclase (0·5–1 mm in diameter) in a fine-grained groundmass (Fig. 6e). The plagioclase in the groundmass is normally zoned (An20–67). The rock contains angular xenoliths (1–30 cm in diameter) of fine-grained metapelite (< 0·1 mm grain size) (Fig. 6f), which contain the mineral assemblage biotite + plagioclase ± muscovite ± quartz ± andalusite ± corundum. The petrographic characteristics of the pelitic xenoliths are similar to those of the hornfels occurring within the contact aureole.
The quartz-diorite is a hornblende-rich, medium- to fine-grained rock (Fig. 3) consisting of plagioclase (52–62 modal %; An28–87), hornblende (18–33 modal %), biotite (2–12 modal %), quartz (5–13 modal %) and opaque minerals (< 3 modal %) (Fig. 6g). This rock exhibits cumulate textures defined by preferentially oriented plagioclase and large (up to 2 cm in diameter) poikiloblastic hornblende crystals. In places, a comb-layering structure is developed in the quartz-diorite (Fig. 6h). The comb-layers have a gabbroic composition (Fig. 3) and consist of long (up to 1 cm length) laths of plagioclase crystals ( 48 modal %; An74–87), coarse hornblende crystals ( 50 modal %), opaque minerals ( 2 modal %), and minor quartz and biotite grains (< 1 modal %).
|
MIGMATITES AND ASSOCIATED CORDIERITE-BEARING TONALITE AT LOC. 1-a |
|---|
Diatexite and cordierite-bearing tonalite occur together with metatexite in a narrow zone along a stream located in the northern part of the Magi River section (Loc. 1-a; Fig. 2b). Although thick soil cover and dense vegetation hamper the observation of a direct contact relationship between the migmatite and the Tokuwa pluton, the rocks at this locality provide an excellent opportunity to examine genetic relationships between the migmatites, cordierite-bearing tonalite, and Tokuwa pluton. The rocks at Loc. 1-a are grouped into three types based on their macroscopic textural characteristics: (1) metatexite containing leucosome; (2) diatexite containing discrete leucosomes and schollen; (3) cordierite-bearing tonalite (Crd-tonalite). There is gradational transition between these rock types (Fig. 7). The metatexite, diatexite, and Crd-tonalite have nearly identical mineral assemblages of quartz + plagioclase + biotite + cordierite ± K-feldspar ± muscovite. Zircon, apatite, monazite, pyrite, spinel, tourmaline, pyrrhotite and ilmenite are accessory minerals.
|
The migmatite terminology used throughout this study is defined as follows (see Mehnert, 1968
; Brown, 1973; Pattison & Harte, 1988; Sawyer, 1996): (1) metatexite: the migmatite that preserves pre-migmatization structures; (2) diatexite: the migmatite in which pre-migmatization structures are completely overprinted; (3) mesosome: the mesocratic part, with decussate and/or granoblastic textures, of the metatexite, which is finer-grained than the leucosome; (4) leucosome: the leucocratic plutonic part, poor in mafic minerals; (5) schollen: the melanocratic fragments within the diatexite.
Metatexite
The metatexite consists of mica-rich pelitic and quartz-rich psammitic layers with quartzofeldspathic leucosomes. The leucosome occurs within pelitic layers or at the interface between pelitic and psammitic layers. The relative proportion of leucosome in the metatexite increases towards the diatexite zone, and finally the rock develops an anastomosing network of leucosomes (Fig. 7b). Cordierite [0·53–0·60 XMg; XMg = Mg/(Mg + Fe)] in the leucosome occurs as inclusion-free euhedral to subhedral discrete grains (Fig. 5b). In the mesosome it has higher XMg (0·58–0·62) and occurs as poikiloblastic crystals containing inclusions of quartz, plagioclase, muscovite and biotite. Fine-grained (< 0·1 mm) hercynitic spinel inclusions occur locally in poikiloblastic cordierite. Plagioclase in both the leucosome and the mesosome exhibits normal chemical zoning. Biotite in the leucosome forms euhedral to subhedral grains and is relatively Fe-rich (0·38–0·41 XMg) compared with the subhedral to anhedral biotite grains in the mesosome (0·49–0·52 XMg). K-feldspar in the mesosome contains fine-grained quartz and biotite inclusions.
Diatexite
The diatexite is a medium- to fine-grained rock, containing leucosomes (up to 5 cm long) and schollen (up to 15 cm long). This rock lacks the layered appearance of the metatexite, implying that it has undergone extensive structural homogenization, which overprinted the original sedimentary structures (see Brown, 1973; Bea, 1991; Sawyer, 1996). However, it still has an heterogeneous appearance, comprising a fine-grained migmatitic part with a granoblastic texture and a coarser-grained part with an equigranular and/or poikilitic texture. The rock exhibits a gradational change in grain size and texture (Fig. 7c). Towards the Crd-tonalite zone, the diatexite changes from a fine-grained rock ( 0·5 mm grain size) with a granoblastic texture to a medium-grained rock (1 mm grain size) with an equigranular or poikilitic texture. In a Qtz–Kfs–Pl diagram (Fig. 3), the diatexites straddle the boundary between the granodiorite and granite fields. The mineral assemblage of the diatexite is quartz (31–49 modal %) + plagioclase (23–37 modal %) + K-feldspar (3–19 modal %) + biotite (7–13 modal %) + cordierite (1–10 modal %) ± muscovite (< 2 modal %). The diatexite contains chemically zoned plagioclase. The plagioclase has a rounded-shape or sieve-structured calcic core (An75–88) mantled by an overgrowth with oscillatory or normal zoning (An19–58) (Fig. 5d and f;). Fine-grained biotite inclusions occur in the plagioclase (Fig. 5e and f). These biotite inclusions have a wide range of compositions (0·24–0·55 XMg). Cordierite (0·51–0·63 XMg) occurs as discrete subhedral to euhedral crystals and has two forms; one is inclusion-free (Fig. 5g) and the other contains small amounts of quartz, biotite and/or muscovite inclusions. In rare cases, cordierite contains fine hercynitic spinel grains (< 0·05 mm). The diatexite lacks poikiloblastic cordierite, which is common in the hornfels and metatexite. Biotite in the diatexite matrix occurs predominantly as subhedral to anhedral crystals (0·38–0·45 XMg). Euhedral to subhedral biotite (0·41–0·45 XMg), round-shaped plagioclase (An21–28), round-shaped quartz, and inclusion-free cordierite (0·55–0·63 XMg) occur as inclusions in coarse (10 mm) poikilitic K-feldspar crystals (Fig. 5h).
The leucosomes in the diatexite are coarse-grained (1–2 mm grain size) consisting of K-feldspar (18–40 modal %), quartz (34–41 modal %), plagioclase (13–25 modal %), euhedral to anhedral biotite (7–10 modal %; 0·33–0·41 XMg) (Fig. 5i), inclusion-free cordierite (1–3 modal %; 0·51–0·56 XMg), and muscovite (2–4 modal %). The leucosome falls in the granite field in the modal Qtz–Kfs–Pl diagram (Fig. 3). Plagioclase in the leucosome has either a rounded-shape or a sieve-structured calcic core (An65–80) that is mantled by a chemically zoned overgrowth (An18–60). In most cases, the leucosomes are developed around schollen. The schollen exhibit a granoblastic texture (< 0·2 mm grain size) and consist mainly of plagioclase (60 modal %; An20–26), biotite (27 modal %; 0·35–0·40 XMg) and cordierite (9 modal %; 0·54–0·58 XMg) (Fig. 5j). Biotite in the schollen is subhedral to anhedral and shows a preferred orientation. Fine-grained inclusions of hercynitic spinel (< 0·05 mm) are locally observed in the cordierite grains.
Discrete clots (2 mm long) enriched in biotite occur in the diatexite (Figs 5k, l, m and 7c). These consist of high-Mg and low-Ti biotite (0·51–0·52 XMg; < 0·4 wt % TiO2), zoned plagioclase (An31–88), quartz, and opaque minerals (Fig. 5k and l). The high-Mg and low-Ti biotite has a sponge-like texture (Fig. 5k and l). The clot is surrounded by cordierite-rich regions consisting of inclusion-free cordierite (0·57–0·62 XMg), zoned plagioclase (An19–90), euhedral to anhedral biotite (0·41–0·46 XMg; 2·4–5·2 wt % TiO2), and quartz (Fig. 5m).
Cordierite-bearing tonalite
The Crd-tonalite is a homogeneous, medium- to coarse-grained rock (2 mm grain size) with an equigranular texture (Fig. 5n). It consists mainly of quartz (40–45 modal %), plagioclase (31–39 modal %), biotite (15–16 modal %), cordierite (< 6 modal %), muscovite (< 2 modal %), and K-feldspar (< 1 modal %). In the Qtz–Kfs–Pl diagram, it overlaps the field for tonalite (Fig. 3). As with the diatexite, there are two types of plagioclase in the Crd-tonalite. One has a calcic core (An80–95) mantled by an oscillatory or normally zoned (An20–80) plagioclase overgrowth and the other exhibits normal compositional zoning (An20–30) and lacks a calcic core. The plagioclase with calcic cores contains fine-grained biotite inclusions. Cordierite (0·56–0·60 XMg) occurs as subhedral grain. In places it contains biotite and/or muscovite inclusions. Biotite (0·39–0·44 XMg) occurs as subhedral to anhedral crystals.
|
P–T CONDITIONS AT THE EMPLACEMENT LEVEL OF THE TOKUWA PLUTON |
|---|
Metamorphic P–T conditions were calculated for the garnet-bearing rocks in the contact aureole (Loc. 1-b, Loc. 2 and Loc. 3, Fig. 2). The rock from Loc. 1-b is a hornfels consisting of Grt + Crd + Bt + Qtz + Pl + Ms (Fig. 4c). The rocks from Loc. 2 and Loc. 3 are metatexites with the assemblage Grt + Kfs + Crd + Bt + Qtz + Pl. The mesosome of the migmatites at Loc. 2 and Loc. 3 shows a decussate and granoblastic texture, respectively.
We used the TWQ/202 variant of the TEEWQ thermobarometer of Berman (1991) for the P–T estimation. Solid solution models employed in the calculations are those given by Berman & Aranovich (1996) for garnet and cordierite, by Chatterjee & Froese (1975) for muscovite, and by R. D. Berman & L. Ya. Aranovich (unpublished data, 1997) for biotite. The observed mineral assemblages were used. The calculations were performed in the KFMASH system including almandine, pyrope, annite, phlogopite, Mg-cordierite, quartz, muscovite and H2O for a rock from Loc. 1-b, and almandine, pyrope, annite, phlogopite, Mg-cordierite, K-feldspar, quartz and H2O for rocks from Loc. 2 and Loc. 3. The compositions of garnet, biotite, cordierite and muscovite used in the calculations are given in Table 1. The compositions of the rims of the grains were used in the calculation.
The two independent sets of reactions intersect at 597°C and 3·0 kbar for Loc. 1-b, at 660°C and 3·1 kbar for Loc. 2, and at 705°C and 3·1 kbar for Loc. 3 (Fig. 8). The estimated pressure conditions are very close for all the three locations and can therefore be taken to represent the pressure at the emplacement level of the Tokuwa pluton (10 km depth).
|
The estimated P–T conditions are shown (Fig. 9) in the KFMASH petrogenetic grid of Pattison & Harte (1997
). Two isopleths of Mg/(Mg + Fe) in cordierite (0·55 and 0·65) are shown because the average XMg(Crd) at Loc. 1-a ranges from 0·55 to 0·65. High activity of water conditions might be caused by the migration of exsolved volatiles from the crystallizing Tokuwa magma into the thermal aureole, or of internal water provided by dehydration reactions of the sedimentary protolith in the aureole, as indicated by the abundance of quartz veins within the hornfels. The P–T conditions estimated by the TWQ calculation for the hornfels at Loc. 1-b are subsolidus, but those calculated for the metatexites at Loc. 2 and 3 occur in the supra-solidus field defined by the H2O-saturated solidus curve of quartzofeldspathic rocks (Pl + Kfs + Qtz + H2O = L) (Johannes & Holtz, 1996), consistent with the petrographic observations.
|
Temperature is estimated for the migmatite in Loc. 3 by employing the Grt–Bt thermometer (TWQ/202, Berman, 1991
), in which the compositions of the central part of the garnet and the biotite in the matrix are used (Table 1). The temperature estimated by using the core composition of garnet is 777°C at 3 kbar (Fig. 9), significantly higher than the temperature (705°C) estimated by using the rim compositions of garnet in the same rock sample (Fig. 8). The estimated P–T (780°C and 3 kbar) are interpreted as the peak metamorphic conditions, which are located in the supra-solidus field in the Qtz–Ab–Or–H2O system under relatively high water activity [a(H2O) > 0·5, Fig. 9]. Therefore, we consider that the temperatures estimated by the multi-equilibrium calculations using the garnet rim compositions (Fig. 8) represent minimum temperatures reflecting some degree of Fe–Mg exchange after the thermal peak. |
MICROSTRUCTURES AND INFERRED REACTIONS |
|---|
Microstructures in hornfels
The Crd-bearing hornfels consists of Crd + Bt + Ms + Qtz + Pl ± Chl. The cordierite occurs as inclusion-rich poikiloblastic crystals containing quartz, chlorite, muscovite and biotite (Fig. 4a), suggesting the cordierite-forming reaction
|
| (1) |
|
| (2) |
|
| (3) |
|
| (4) |
|
| (5) |
Microstructures in the metatexite
At Loc. 2, the leucosome in the metatexite consists of quartz and feldspar in which the quartz develops crystal faces against the K-feldspar (Fig. 5a). The absence of euhedral ferromagnesian minerals within the leucosome suggests the following congruent melting reaction:
|
| (6) |
An0) in Fig. 9. We suggest that reaction (6) represents the onset of incipient partial melting in the Tokuwa contact aureole.
At Loc. 1-a, the leucosomes in the metatexite consists predominantly of K-feldspar, quartz and inclusion-free cordierite ( 0·55 XMg) and subordinate subhedral to euhedral biotite, zoned plagioclase, and muscovite. The occurrence of inclusion-free cordierite and euhedral biotite in the leucosome suggests that these minerals are peritectic products of a mica-breakdown melting reaction. The biotite in the mesosome is rich in Mg ( 0·50 XMg) compared with that in the leucosome ( 0·40 XMg), suggesting the following melting reaction:
|
| (7) |
3 kbar) is lower than that for these reactions in the P–T field (Fig. 9).
Microstructures in the diatexite
The Mg-rich biotite (0·47–0·50 XMg) in the biotite-enriched clots shows a sponge-like texture that is considered to be a breakdown reaction product of Mg-rich biotite involving melt (Fig. 5k and l; see Holtz & Johannes, 1991). The clot is surrounded by cordierite-rich regions consisting of inclusion-free cordierite ( 0·60 XMg), plagioclase (An19–90), quartz, and Fe-rich biotite (0·39–0·41 XMg) (Fig. 5m). These petrographic features suggest the following melting reaction:
|
| (8) |
) (Fig. 9). The plagioclase in the diatexite has irregular-shaped cores with extremely high anorthite contents (An80–95). The P–T conditions (780°C and 3 kbar) estimated using TWQ/202 (Berman, 1991) for the aureole plot in the supra-solidus field above the solidus reaction Qtz–Ab–Or–An100–H2O (Johannes & Holtz, 1996) (Fig. 9). This supports the conclusion that the peak metamorphic conditions in the aureole reached supra-solidus conditions and led to anatexis (Fig. 9) |
GEOCHEMISTRY |
|---|
Major and trace element data and Sr-isotope compositions
Selected major and trace element data and alumina saturation index (A.S.I.) values are plotted against SiO2 for the Tokuwa pluton samples (Fig. 10a) and for the hornfels, migmatites and Crd-tonalite (Fig. 10b). The initial 87Sr/86Sr (SrI) of these rocks is indicated in Fig. 10c.
|
Tokuwa main pluton
Samples from the Tokuwa main pluton (Fig. 2a) contain 55–73 wt % SiO2. The Al2O3, FeO(total) + MgO and CaO contents are negatively correlated with SiO2, whereas the Na2O, K2O, and Ba contents exhibit a positive correlation. The rocks in the Tokuwa main pluton have metaluminous to slightly peraluminous compositions (A.S.I. = 0·77–1· 08) and belong to the ‘I-type’ granitoid affinity defined by Chappell & White (1974
). The A.S.I. values show a positive correlation with SiO2 (Fig. 10a). In the Ab–An–Or diagram (Fig. 11), most of the samples of the Tokuwa main pluton fall in the granodiorite field, with a few samples (mafic samples, 55–57 wt % SiO2) plotting in the tonalite field. Most of the mafic samples are rich in poikilitic hornblende, suggestive of a cumulate origin. The SrI of the Tokuwa main pluton ranges from 0·7040 to 0·7051. The lowest SrI (0·7040) sample of the Tokuwa main pluton (MKG-05b, 55 wt % SiO2) was collected 1 km NNE of the Loc. 1-a.
|
Marginal rocks of the Tokuwa pluton in the Magi River section
The SiO2 contents of rocks in the marginal part of Tokuwa pluton in the Magi River section (Fig. 2b) range from mafic (quartz-diorite, 49–56 wt %) through intermediate (Hb-tonalite, 61–65 wt %) to felsic (Bt-tonalite, 70–72 wt %). The samples plot in the tonalite field on the An–Ab–Or diagram and show chemical characteristics that are clearly distinct from those of the rocks of the Tokuwa main pluton (Fig. 11). The major and trace element contents of the quartz-diorite are characterized by negative trends vs SiO2 for Al2O3, FeO(total) + MgO and CaO, and positive trends for Na2O, K2O, Ba and Nb (Fig. 10a).
Compared with the granitoids in the Tokuwa main pluton, the Hb-tonalite is characterized by higher Na2O and lower K2O at a given SiO2 content. It shows a decrease of Al2O3, FeO(total) + MgO, CaO and Na2O, but increases in K2O, Ba and Nb with increasing SiO2. The Bt-tonalite is characterized by a narrow range of compositional variation in most elements. It has higher Na2O and lower K2O contents at a given SiO2 wt % than the Tokuwa main granite. The quartz-diorite and Hb-tonalite are metaluminous (A.S.I. 0·70–0·89 and 0·86–0·97, respectively) and their SrI is in the range 0·7043–0·7044 and 0·7040–0·7045, respectively. The Bt-tonalite, in contrast, is slightly peraluminous (1· 05) and characterized by higher SrI (0·7054) compared with the quartz-diorite and Hb-tonalite (Fig. 10c).
Hornfels and metatexite
The hornfels (64–77 wt % SiO2) are characterized by the following elemental variations and trends (Fig. 10b): a negative trend of Al2O3, FeO(total) + MgO, K2O, Ba and Nb vs SiO2 and constant contents of CaO and Na2O for variable SiO2 ranges. They are strongly peraluminous (A.S.I. between 1·79 and 2·20) and their SrI ratio ranges from 0·7085 to 0·7104. The SiO2 contents of the metatexites range from 61 to 78 wt %. Their Al2O3, FeO(total) + MgO, K2O, Ba and Nb contents define negative correlations vs SiO2, whereas Na2O and CaO contents are almost constant across the range of SiO2 contents. The whole-rock compositions of the metatexite are broadly comparable with those of the hornfels except for differences in Na2O and A.S.I., suggesting that the metatexite preserves most of the geochemical signature inherited from the hornfels. The metatexites have A.S.I. ranging from 1·29 to 1·94, which show a good negative correlation with SiO2. The SrI of the metatexite ranges from 0·7100 to 0·7112.
Diatexite and Crd-tonalite
The SiO2 content of the diatexite ranges from 68 to 72 wt %. Compared with the metatexite and hornfels, the diatexites contain lower amounts of Al2O3 and K2O and higher CaO and Na2O. Their A.S.I. (1·16–1·34) shows a positive correlation with SiO2 but is lower than that of the metatexite and hornfelses. The SrI of the diatexite (0·7078–0·7094) is lower than that of the metatexite (0·7100 and 0·7112). The leucosomes (73–77 wt % SiO2) are strongly peraluminous (A.S.I. 1·21–1·23) and enriched in K2O and Ba, but depleted in Al2O3, FeO(total) + MgO, CaO, Na2O and Nb compared with the diatexite. The schollen, on other hand, have lower SiO2 (55–62 wt %), K2O and Rb, but higher Al2O3, FeO(total) + MgO, CaO and Na2O than the diatexite. Their A.S.I. ranges from 1·16 to 1·30. The SrI for the leucosome and the schollen is 0·7090 and 0·7091, respectively.
The Crd-tonalites (71 SiO2 wt %) are strongly peraluminous (A.S.I. 1·16–1·17) and belong to the ‘S-type’ granitoid affinity defined by Chappell & White (1974). They have intermediate compositions between the diatexite and the Bt-tonalite. Their Al2O3, FeO(total) + MgO, Na2O, and K2O contents are comparable with those of the diatexite, although their CaO, Ba and Nb contents are intermediate between the diatexite and the Bt-tonalite. The SrI ratio of the Crd-tonalite ranges from 0·7067 to 0·7068, significantly higher than the values for the Bt-tonalite, but lower than those of the metatexite and diatexite.
|
DISCUSSION |
|---|
Magmatic process in the Magi River section
The quartz-diorite, Hb-tonalite, and Bt-tonalite in the Magi River section (Fig. 2b) have comparable SrI ratios (0·7040–0·7054), suggesting that they are genetically related. The Hb-tonalite and quartz-diorite, most of which exhibit various types of cumulate texture, have nearly identical SrI ratios (0·7040–0·7045; Table 2). We attribute the lithological and chemical variations in these two rock types to crystal accumulation processes involving the mineral constituents in these rocks. The medium-grained homogeneous tonalite (MGZ-14c), which occurs 200 m from the intrusive contact (Loc. 1-c, Fig. 2b) lacks any cumulate texture (Fig. 6c). This rock is considered to represent the parental magma composition. A least-squares mass-balance calculation successfully explains the major-element compositional variation in the fine-grained Hb-tonalites and quartz-diorites by accumulation of Pl + Hb ± Bt from the assumed parental magma composition (MGZ-14c, Table 4; Fig. 12). The mineral compositions used for the least-squares mass-balance calculations are listed in Table 5.
|
|
|
Compositional variations in the Bt-tonalite can be explained by fractional crystallization processes (F = 0·69–0·77) involving removal of plagioclase, hornblende and biotite from the parental magma composition (MGZ-14c, Table 6; Fig. 12). However, the SrI of the Bt-tonalite (
0·7054) is slightly higher than that of the assumed parental magma composition (MGZ-14c, 0·7045), suggesting a significant role for country rock assimilation in its derivation. Assimilation and fractional crystallization (AFC) modeling of DePaolo (1981) was used to model the variations in whole-rock and SrI compositions in the Bt-tonalites. The hornfels (MKG-31a, Table 2) that occurs in the Magi River section (Loc. 1-d, Fig. 2b) was selected as a likely contaminant. The F value was estimated by a least-squares major-element mass-balance calculation (Table 5). The liquid–crystal partition coefficients used for this modeling are given in Table 7. The AFC modeling reproduces the whole-rock compositional variation in the Bt-tonalites well (Table 8). The r value (ratio of the mass of material assimilated to the mass of material crystallized) ranges from 0·37 to 0·44, corresponding to the assimilation of relatively small amounts of hornfels (10·3–11·3 wt %) into the fractionated melt.
|
|
|
The genetic lin