Mafic xenoliths in volcanic rocks can provide valuable information about the petrology and chemistry of unexposed subvolcanic basement, the composition of parental mafic magmas and differentiation processes during the early stage of magmatic evolution [e.g. 1–5]. Most of the mafic xenoliths in volcanic rocks from island arc systems fall into one of three categories: cognate xenoliths that are indicative of magma mixing [e.g. 6, 7], plutonic blocks derived from middle to lower crustal sources [e.g. 8, 9], or cumulate rocks formed by crystallization and precipitation in magma reservoirs [e.g. 10–12]. Among these, rare gabbroic xenoliths are found in rocks extruded from dominantly felsic (dacite to rhyolite) volcanoes (e.g. Mount St. Helens, Washington [13, 14]; Little Glass Mountain, California ).
Niijima volcano in the northern Izu-Bonin volcanic arc is a rare example of a volcano that has produced predominantly rhyolitic lava domes and pyroclastic rocks accompanied by only minor andesitic to basaltic flow. Its volcanic products contain various types of xenolith [e.g. 16], but include both volcanic xenoliths (andesitic to basaltic composition) in rhyolitic pyroclastic flow deposits and lavas [e.g. 16], and plutonic xenoliths (e.g. diorite, tonalite, and gabbro) in basaltic basal surge deposits and rhyolitic units [e.g. 16, 17]. However, there are few published discussions of the origin and process of formation of these types of xenolith and their relationships with erupting host magmas. In this paper, we present the results of new petrographical, petrological, and chemical investigations of the two types of gabbroic xenoliths found at Niijima, discuss the processes and temperature-pressure conditions of their formation, and consider possible parental magmas with the aim of understanding magmatic processes beneath Niijima volcano.
2 Geological background
Niijima volcano lies in the northern part of the Izu-Bonin arc about 20 km west of the volcanic front (Fig. 1a). Most of the frontal arc volcanoes in the arc produce basaltic magmas, but the Niijima volcano is composed mainly of monogenetic rhyolitic lava domes. Volcanism at Niijima probably started during the late Quaternary , and has produced at least 15 monogenetic, mostly rhyolitic eruptions [e.g. 16] (Fig. 1b). The rhyolitic volcanic episodes included three characteristic stages of lava production: a first stage of hypersthene-cummingtonite-hornblende rhyolite, a second stage of cummingtonite rhyolite, and a third stage of biotite rhyolite. Exceptions to this sequence were two eruptions of olivine-bearing basaltic magmas in the northern part of the island during the third stage. The first of these basaltic eruptions was in the Wakago area 2000 to 3000 years ago and the second was northwest of Atchiyama before the eruption that formed the Atchiyama lava dome [16, 18] (Fig. 1b). Andesite lavas are also in known at Niijima, but they are minor extent.
For this study, we collected samples of gabbroic xenoliths and their host rocks from the Atchiyama rhyolitic lava dome and from the Wakago basaltic pyroclastic deposit (Fig. 1b).
3 Sample description and petrography
Gabbroic xenoliths are rare in the volcanic units at both Wakago and Atchiyama (Fig. 1c). Large angular xenoliths (up to 30 cm diameter) are found in the volcanic units, whereas the small xenoliths (1 to 15 cm in diameter) are rounded to sub-rounded (Figs. 2a and 2b). On the basis of rock textures and mineral associations, we identified two groups of xenoliths: amphibole-free gabbroic xenoliths (A-type gabbro) and amphibole-bearing gabbroic xenoliths (B-type gabbro). The A-type xenoliths correspond to the xenoliths of Azuma and Yashima  and are found in both the Wakago basaltic pyroclastic deposits and the Atchiyama rhyolitic lava dome (Figs. 1 and 2). The B-type xenoliths are found only in the Atchiyama rhyolitic lava dome. We collected non-altered xenolith samples in the outcrops. In small size of samples (5~7 cm), core parts of the xenoliths were selectively used for observations and chemical analyses, because the rim parts of the xenoliths sometimes were altered, or interacted in parts with surrounding volcanic rocks. Very small sizes of gabbro (< 5 cm) (mostly B-type) were excluded for observation and chemical analyses.
The A-type gabbro is a granular texture and its mineral assemblage is plagioclase, clinopyroxene, orthopyroxene, and Fe-Ti oxides with scarce olivine (Fig. 2c). Grain size ranges from 0.2 to 3 mm (average 1.0 mm). All of the samples contain various types and amounts of glasses. Most of the A-type gabbro has an adcumulate, even-grained mosaic texture [e.g. 19, 20]. Plagioclase, clinopyroxene, orthopyroxene, and magnetite are recognized as cumulus phases with unzoned post-cumulus overgrowth (Fig. 2c). Postcumulus interstitial magnetite is also present. However, narrowly zoned rims surrounding some crystals near their contact with patches of interstitial glass indicate mesocumulus growth. The B-type gabbros are clearly different from the A-type gabbros in both texture and modal mineralogy (Fig. 2d). Average grain size is less than 1 mm and the mineral assemblage is plagioclase, amphibole, quartz, Fe-Ti oxides, and rare minute clinopyroxene and orthopyroxene (Fig. 2d). The minerals are occasionally surrounded by unvesiculated interstitial glass. A peculiar feature is wide peripheral zonation in some plagioclases. This zoned rim is interpreted as the result of postcumulus crystallization in a closed system. The gabbroic cumulates of the B-type xenoliths are thus typical orthocumulates [e.g. 19, 21].
4 Analytical procedures
The chemical compositions of minerals from the gabbroic xenoliths and their host rocks were determined with an electron-microprobe analyzer (EPMA). Analyses were performed with an accelerating voltage of 15 kV and a probe current of 10 nA, and a beam diameter of 10 μm using JEOLJXA 8530F (wavelength dispersive) electron probe microanalyzer at the Chemical Analysis Division the Research Facility Center for Science and Technology of the University of Tsukuba. The data were regressed using an oxide-ZAF correction program supplied by JEOL, using natural minerals and synthetic materials of known compositions as standards. All Fe reported as FeO.
Whole rock major and trace element compositions were determined with an inductively-coupled plasma mass spectrometer (ICP-MS) at Activation Laboratories Ltd. in Canada, and with an X-ray fluorescence (XRF) spectrometer (PANalytical MagiX PRO) at the Kitakyushu Museum of Natural History and Human History, Japan. For the XRF, the analytical methods presented in Mori and Mashima  were followed.
5 Results and discussion
The representative chemical compositions of minerals from the gabbroic xenoliths and their host rocks were shown in Table 1. The anorthite content (An = 100 × Ca/(Ca+Na)) of plagioclases in the A-type gabbro ranged from An56 to An90; the most common composition was An80–90 (Fig. 3a). The cores and rims of most plagioclases in the A-type gabbro were similar compositions, consistent with adcumulus growth of the crystals. Some more sodic rims may represent postcumulus overgrowths in the mesocumulates. The core anorthite contents of plagioclases in the A-type gabbro are similar to those of the B-type gabbros, but were slightly lower than those in the host basalt (Wakago basalt) and mafic (basaltic) inclusions in the Atchiyama rhyolitic unit (Fig. 3a). The results suggest that plagioclase in the A-type gabbro might have precipitated from a slightly differentiated magma of the host basalt. In contrast, the anorthite content of plagioclases in the B-type gabbro showed a wider and bimodal compositional variation (An30–88) with peaks at An40–55 and An70–88 (Fig. 3b). Although the highest anorthite content of plagioclase in B-type gabbro was similar to that of the A-type gabbro, the lower peak of anorthite content in the B-type gabbro was clearly lower than that of the A-type gabbro. The bimodal anorthite distribution of the B-type gabbro is very similar to that of andesite (of minor exposure in Niijima) representing mixing of basaltic and rhyolitic magmas . The anorthite (An) contents of plagioclase from most of the B-type gabbros are higher than those in the host rhyolites (Fig. 3b). This result and the mineral assemblage of the B-type gabbros suggest that plagioclases in them were crystallized from intermediate (andesitic) magma, rather than rhyolitic magma.
The chemical compositions of olivine in the A-type gabbros showed a narrow range of Mg# (Mg# = Mg/(Mg+Fe2+)) values (0.73~0.76) and overlapped the compositions (Mg# = 0.67~0.77) of olivines in the host basalts (Table 1). Clinopyroxene from the A-type gabbro was calcic augite (Wo42−45En42−44Fs11−13) (Wollastonite, Wo = 100 × Ca/(Ca+Mg+Fe2+); Enstatite, En = 100 × Mg/(Ca+Mg+Fe2+), Ferrosilite, Fs = 100 × Fe2+/(Ca+Mg+Fe2+)), whereas that from the B-type gabbro ranged from augite to sub-calcic augite (Wo37−44En42−47Fs13−16) (Table 1 and Fig. 4). Clinopyroxenes in the A-type gabbro had slightly higher Mg and Al contents than that in the B-type gabbro. The AlIVcontent of clinopyroxene (calculated on the basis of cations per formula unit) in the A-type gabbro was 0.04-0.11, which was higher than that (0.01-0.05) in the B-type gabbro. The different AlIV contents in clinopyroxene in both types of gabbro suggest a difference of crystallization pressures [e.g. 24, 25]. The compositional difference between cores and rims in orthopyroxenes was similar for both types of xenolith.
We calculated temperature and pressure conditions for the crystallization of the two types of gabbro under the assumption of crystallization at equilibrium within a magma chamber. Clinopyroxene-orthopyroxene pairs from the A-type gabbros yielded crystallization temperatures of 899 to 955°C, and pressures of 3.6 to 5.9 kbar using the method of geothermo- and geobarometry by Putirka . The best results of these calculations were for mafic systems where the Mg# for clinopyroxene was > 0.75 . For the B-type gabbro, the temperature was estimated using the amphibole-plagioclase thermometry by Holland and Blundy , and then based on the obtained temperature, pressure was calculated by the method of Anderson and Smith . As a result, the B-type gabbro yielded crystallization temperatures of 687 to 824° C and pressures of 0.8 to 3.6 kbar.
Under silica-saturated conditions, the estimated temperature and pressure of crystallization for the B-type gabbros was 722 to 824° C at 0.8 to 2.3 kbar, whereas for a system under a silica-undersaturated, the estimates were 687 to 817° C at 1.0 to 3.6 kbar. Although the B-type gabbros contain interstitial quartz, it might represent a later stage of crystallization. Nevertheless, our results indicate that the A-type gabbros were derived from higher-temperature magmas and solidified at deeper levels than the B-type gabbros. These results (low pressure for B-type gabbro) are similar to those of Matsui et al. , who estimated the crystallization pressure for the biotite rhyolite to be 0.5-1.5 kbar.
For analyses of whole rock chemistry, we newly selected seven A-type gabbro and four B-type gabbro for chemical analyses. Also, some of the host basalt and rhyolite, andesite and mafic basaltic inclusions were determined for comparison. Whole rock chemistry data (Table 2) show that the SiO2 contents (wt. %) of the A- and the B-type gabbros are 42 to 45 % and 51 to 57 %, respectively (Fig. 5). MgO contents were 5.5-10.3 % for the A-type and 3.8-5.1 % for the B-type gabbro. V and Zr contents also differ among the two types of gabbro, and their host rocks, and minor andesite units (Fig. 5), and each of these, with the exception of the rhyolite, shows a degree of linearity.
Chondrite-normalized rare earth element (REE) patterns of the A-type gabbros (Fig. 6) are characterized by depletion of light rare-earth elements (LREEs) pattern, and a minor positive Eu anomaly, which together suggest the accumulation of plagioclase. In contrast, the REE patterns of the B-type gabbros have a flatter pattern, with only slight depletion of LREEs. A weak positive Eu anomaly is also apparent for B-type gabbros. The REE patterns of the B-type gabbros clearly differ from those of the host rhyolites, which have much higher LREE contents (Fig. 6). It is possibly assumed that the B-type gabbro was formed from andesitic host magma, but not from basaltic and rhyolitic. Considering mineral assemblages and chemistries, the whole rock chemistry for the two types of gabbro and their host rocks supports crystallization of the A-type gabbro from a mafic (basaltic) magma and the B-type gabbro from a mafic to intermediate (andesitic composition) magma.
These results are concordant with our petrographic results and provide strong evidence that both types of gabbro represent cumulates removed from parental magmas during crystallization and fractionation. Whole rock chemistry suggests that the A-type gabbro was derived from a basaltic parental magma similar to that erupted at Wakago, and that the B-type gabbro was not derived directly from either the basaltic or rhyolitic magmas that host the xenoliths, although it is possible that the parental magma of the B-type gabbro was a fractionation product of the basaltic magma. Our results (especially mineral assemblages and mineral chemistries) suggest that an intermediate (andesitic) magma was the parental magma for the B-type gabbro.
Similar results have been reported for xenoliths of cumulate gabbro elsewhere, including dacite erupted at Mount St. Helens (Washington) from 1980 to 1983 [e.g. 14]. Several types of gabbro-norite cumulates have been interpreted to be derived from mafic plutons formed by multiple emplacements of basaltic magma at middle to upper crustal levels beneath a volcano. Gabbroic cumulates within rhyolites at Little Glass Mountain (California) have been interpreted to represent a two-stage fractionation model on the basis of compositional gaps in plagioclase zoning in the rhyolite . These studies of cumulate gabbro in felsic (dacite-rhyolite) volcanic products strongly suggest that complex processes of magma intrusion and crystal-fractionation involving parental magma with various compositions have occurred at relatively in shallow crustal levels beneath volcanoes.
The evolution of magmatic system in Niijima volcano inferred from the stratigraphic succession of lavas [16, 30] is consistent with the development of fractionated lavas by open-system magmatic processes. Identification of the cumulate from which mafic xenoliths originated may prove that crystal fractionation has played a significant role in the evolution of the magmatic system. It is likely that these mafic xenoliths are products of liquid that was cognate with the effusive basalt and andesite. It is therefore probable that an ultramafic cumulate residuum remains in the middle-lower crust beneath the volcanic edifice, its presence is consistent with crustal structure modeled on the basis of seismic profiles [e.g. 31]. Therefore, the petrographical and chemical characteristics, and their assumed crystallization processes of gabbroic xenoliths at the rhyolite dominated Niijima volcano suggest that the existence of various cumulate bodies at different levels beneath the volcano, and that the effect of mafic to intermediate magmas beneath rhyolitic volcanoes can have considerable influence on the erupted products of these volcanoes.
Two different types of gabbroic xenoliths (A- and B-type xenoliths) in volcanic products among the rhyolite dominated Niijima volcano, northern Izu-Bonin volcanic arc (central Japan) were investigated on the basis of petrography, petrology and geochemistry. A-type gabbroic xenoliths consisting of plagioclase, clinopyroxene, and orthopyroxene with an adcumulate texture were found in both basalt and rhyolite units, and B-type gabbroic xenoliths consisting of plagioclase and amphibole with an orthocumulate texture were found only in rhyolite unit. The A-type gabbro were estimated to have formed at higher temperatures (899-955° C) and pressures (3.6-5.9 kbar) than the B-type gabbro (687-824°C and 0.8-3.6 kbar). These calculation modelling, and mineral and whole-rock chemistries suggest different parental magmas for the two types of gabbro. The A-type gabbro was assumed to have been produced from basaltic magma, whereas the B-type gabbro was formed from an intermediate (andesitic) magma. The results for the gabbroic xenoliths in the Niijima volcano indicate that mafic to intermediate cumulate bodies of different origins and some different fractionation processes existed beneath the rhyolite dominated volcano.
We are grateful to N. Nishida and N. Chino for their kind helps in microprobe analyses. We thank A. Takada at the Geological Survey of Japan and T. Tsunogae at the University of Tsukuba for their fruitful comments and discussions. We deeply acknowledge to two anonymous reviewers and Dr. J. Barabach, (Managing Editor) for their critical comments and advices that significantly improved our manuscript.
Bacon C.R., Metz J., Magmatic inclusions in rhyolites, contaminated basalts, and compositional zonation beneath the Coso volcanic field, California. Contrib. Mineral. Petrol., 1984, 85, 346-365. CrossrefGoogle Scholar
Beard J.S., Characteristic mineralogy of arc-related cumulate gabbros: implications for the tectonic setting of gabbroic plutons and for andesite genesis. Geology, 1986, 14, 848-851.CrossrefGoogle Scholar
Hickey-Vargas R., Abdollahi M.J., Parada M.A., López-Escobar L., Frey F.A., Crustal xenoliths from Calbuco volcano, Andean Southern Volcanic Zone: implications for crustal composition and magma-crust interaction. Contrib. Mineral. Petrol., 1995, 119, 331-344.CrossrefGoogle Scholar
Brown S.J.A., Burt R.M., Cole J.W., Krippner S.J.P., Price R.C., Cartwright I., Plutonic lithics in ignimbrites of Taupo Volcanic Zone, New Zealand; sources and conditions of crystallisation. Chem. Geol., 1998, 148, 21-41.CrossrefGoogle Scholar
Costa F., Dungan M.A., Singer B.S., Hornblende- and phlogopite-bearing gabbroic xenoliths from Volcán San Pedro (36° S), Chilean Andes: evidence for melt and fluid migration and reactions in subduction-related plutons. J. Petrol., 2002, 43(2), 219-241. CrossrefGoogle Scholar
Coulon C., Clocchiatti R., Maury R.C., Westercamp D., Petrology of basaltic xenoliths in andesitic to dacitic host lavas from Martinique (Lesser Antilles): evidence for magma mixing. Bull. Volcanol.,1984, 47(4), 705-734. CrossrefGoogle Scholar
Shimazu M., Kawano Y., Kaji K., Igarashi S., Chemical compositions and Sr, Nd isotope ratios of gabbroic xenoliths in calc-alkali andesites of Naeba and Torikabuto volcanoes, North Fossa Magna, central Japan. J. Mineral. Petrol. Econ. Geol., 1991, 86, 53-64. CrossrefGoogle Scholar
Yasui M., Togashi S., Shimomura Y., Sakamoto S., Miyaji N., Endo, K., Petrological features and origin of the gabbroic fragments contained in the 1707 pyroclastic fall deposits, Fuji Volcano. Bull. Volcanol. Soc. Japan, 1998, 43(2), 43-59 (in Japanese with English abstract). Google Scholar
Pallister J.S., Hoblitt R.P., Crandell D.R., Mullineaux D.R., Mount St Helens a decade after the 1980 eruptions: magmatic models, chemical cycles, and a revised hazards assessment. Bull. Volcanol., 1992, 54, 126-146. CrossrefGoogle Scholar
Brophy J.G., Dorais M.J., Donnelly-Nolan J., Singer B.S., Plagioclase zonation styles in hornblende gabbro inclusions from Little Glass Mountain, Medicine Lake volcano, California: implications for fractionation mechanisms and the formation of composition gaps. Contrib. Mineral. Petrol., 1996, 126, 121-136. CrossrefGoogle Scholar
Isshiki N., Geology of the Nii Jima district. With Geological Sheet Map at 1:50,000, Geological Survey of Japan, Tsukuba, Japan, 1987 (in Japanese with English abstract). Google Scholar
Azuma K., Yashima R., A leucocratic xenolith from Niizima, Izu. Science Reports of the Faculty of Edu. Fukushima Univ., 1984, 34, 25-27 (in Japanese with English abstract). Google Scholar
Tsukui M., Saito K., Hayashi K., Frequent and intensive eruptions in the 9th century, Izu Islands, Japan: Revision of volcanostratigraphy based on tephras and historical document. Bull. Volcanol. Soc. Japan., 2006, 51(5), 327-338 (in Japanese with English abstract). Google Scholar
Tollan P.M.E., Bindeman I., Blundy J.D., Cumulate xenoliths from St. Vincent, Lesser Antilles Island Arc: a window into upper crustal differentiation of mantle-derived basalts. Contrib. Mineral. Petrol., 2012, 163, 189–208. Web of ScienceCrossrefGoogle Scholar
Jeffery A.J., Gertisser R., Troll V.R., Jolis E.M., Dahren B., Harris C., Tindle A.G., Preece K., O’Driscoll B., Humaida H., Chadwick J.P., The pre-eruptive magma plumbing system of the 2007–2008 dome-forming eruption of Kelut volcano, East Java, Indonesia. Contrib. Mineral. Petrol., 2013, 166, 275-308. Web of ScienceCrossrefGoogle Scholar
Mori Y., Mashima H., X-ray fluorescence analysis of major and trace elements in silicate rocks using 1:5 dilution glass beads. Bulletin of the Kitakyushu Museum of Natural History and Human History, 2005, Ser A (3), 1-12. Google Scholar
Endo D., Petrological and petrochemical characteristics of the volcanic products in the Niijima volcano, northern Izu-Ogasawara arc. Master thesis, University of Tsukuba, Japan, 2012. Google Scholar
Jankovics M.É., Dobosi G., Embey-Isztin A., Kiss B., Sági T., Harangi S., Ntaflos T., Origin and ascent history of unusually crystal-rich alkaline basaltic magmas from the western Pannonian Basin. Bull. Volcanol., 2013, 75, 749, CrossrefWeb of ScienceGoogle Scholar
Leake B.E., Woolley A.R., Arps C.E.S., Birch W.D., Gilbert M.C., Grice J.D., Hawthorne F.C., Kato A., Kisch H.J., Krivovichev V.G., Linthout K., Laird J., Mandarino J.A., Maresch W.V., Nickel E.H., Rock N.M.S., Schumacher J.C., Smith D.C., Stephenson N.C.N., Ungaretti L., Whittaker E.J.W., Youzhi G., Nomenclature of amphiboles: Report of the Subcommittee on Amphiboles of the International Mineralogical Association, Commission on New Minerals and Mineral Names. Can. Mineral., 1997, 35, 219-246. Google Scholar
Putirka K.D., Thermometers and barometers for volcanic systems. In: Putirka K.D., Tepley F.J. (Eds.), Minerals, Inclusions and Volcanic Processes, Reviews Mineral. Geochem 69. Mineralogical Society of America, Washington, D.C., 2008, 61-120. Google Scholar
Matsui R., Nakamura M., Yoshiki K., Kuritani T., Yoshida T., Suzuki Y., Nagahashi Y., A petrological study on the magma plumbing systems in the Niijima and Shikinejima volcanoes, Izu Islands. Abstracts of Japan Geoscience Union Meeting, Makuhari, Japan, 2009, V160-P005. Google Scholar
Kodaira S., Sato T., Takahashi N., Ito A., Tamura Y., Tatsumi Y., Kaneda Y., Seismological evidence for variable growth of crust along the Izu intraoceanic arc. J. Geophys. Res., 2007, 112, B05104, CrossrefWeb of ScienceGoogle Scholar
Endo D., Arakawa Y., Ikehata K., Oshika J., Shinmura T., Mori Y., Two types of gabbroic xenoliths from Niijima volcano, Izu-Bonin arc: petrological and geochemical constraints. Abstracts of Annual Meeting of Japan Association of Mineralogical Sciences, Kumamoto, Japan, 2014, R7-08. Google Scholar
Sun S.S., McDonough W.F., Chemical and isotopic systematics of oceanic basalts: Implications for mantle composition and processes. In: Saunders A.D., Norry M.J. (Eds.), Magmatism in the Ocean Basins, Geological Society, London, Special Publications, 1989, 42, 313–345. Google Scholar
About the article
Published Online: 2017-03-01