The Izu Peninsula, Japan: Zircon geochronology reveals a record of intra-oceanic rear-arc magmatism in an accreted block of Izu–Bonin upper crust

Page 1

Author's personal copy

The Izu Peninsula, Japan: Zircon geochronology reveals a record of intra-oceanicrear-arc magmatism in an accreted block of Izu–Bonin upper crust

Kenichiro Tani a,⁎, Richard S. Fiske b, Daniel J. Dunkley c, Osamu Ishizuka a,d, Teruki Oikawa d,Ichiyo Isobe e, Yoshiyuki Tatsumi a

a Institute for Research on Earth Evolution, Japan Agency for Marine-Earth Science and Technology, 2-15, Natsushimacho, Yokosuka, 237-0061, Japanb Smithsonian Institution MRC-119, Washington, DC 20013-7012, USAc National Institute of Polar Research, 10-3, Midoricho, Tachikawa, Tokyo, 190-8518, Japand Institute of Geology and Geoinformation, Geological Survey of Japan/AIST, Central 7, 1-1-1, Higashi, Tsukuba, Ibaraki, 305-8567, Japane Niijimamura Museum, 2-36-3, Honson, Niijima, Tokyo, 100-0402, Japan

a b s t r a c ta r t i c l e i n f o

Article history:Received 26 August 2010Received in revised form 24 December 2010Accepted 28 December 2010Available online 4 February 2011

Editor: T.M. Harrison

Keywords:zircon U–Pb agesIzu PeninsulaIzu–Bonin arcrear-arcupper crust development

The Izu Peninsula, central Japan, is situated in a zone where the active intra-oceanic Izu–Bonin arc has beencolliding end-on with the mainland Honshu arc for the past 15 million years. As a result of this arc–arccollision, parts of the submarine Izu–Bonin upper crustal sequences have been accreted and uplifted to formthe Izu Peninsula, exposing seafloor volcaniclastic deposits, associated lava flows, and coeval intrusive bodies.Parts of this sequence have been subjected to extensive hydrothermal alteration, and these altered rocks havepreviously been interpreted as representative of hypothetical widespread Middle Miocene basement thatpresumably underlay northern Izu–Bonin arc volcanoes. New zircon U–Pb ages presented here, however,show that both fresh and altered volcanic sequences exposed in Izu Peninsula are broadly contemporaneousand were products of the same Late Miocene to Pleistocene magmatism. Geochemical characteristics of thesesequences show them to have formed in the Izu–Bonin rear-arc environment, providing an unusualopportunity to investigate in detail the growth and architecture of a rear-arc region in an active intra-oceanicarc. Moreover, zircon ages from altered basal units of Kozushima and Niijima, Quaternary volcanic islands inthe northern Izu–Bonin rear-arc, show that these islands rest on units only slightly older (b1 Ma) than themain body of these subaerial edifices, not, as previously believed, part of a regional older Miocene basement.The near-continuum growth of these arc volcanoes and their underlying successions, plus the absence of adistinctly older basement underlying the Izu Peninsula and northern Izu–Bonin arc, provide new insight intoupper crust development in an intra-oceanic, convergent margin environment.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Intra-oceanic arc systems are believed to be sites where juvenilecrust of intermediate (i.e. andesitic)bulk composition is created throughsubduction zone magmatism in the absence of preexisting continentalbasement (e.g. Tatsumi et al., 2008). Because the average continentalcrust is of intermediate composition, it has also been proposed that thecrustal formation in the intra-oceanic arc environment is fundamentalto continental growth (e.g. McLennan and Taylor, 1982). This hasyielded the generally acceptedhypothesis that continents grow throughsuccessive accretion of intra-oceanic arcs. Rear-arc magmatism, inparticular, has recently been postulated to have stronger geochemicalaffinities with continental crust, such as enrichment of incompatibleelements, compared to the composition of the volcanic front igneous

rocks (Gill et al., 2009). The study ofmodern intra-oceanic arcs and theirsuccessive modification during arc–arc collisions may provide insightsto the genesis of continental crust.

The Izu–Bonin arc is one of the world's most intensively studiedintra-oceanic arc systems, where the recent geophysical surveysrevealed three seismically defined layers within the arc crust; a high-velocity lower crust lying above the Moho discontinuity, an interme-diate-velocity middle crust, and a low-velocity upper crust (e.g. Kodairaet al., 2007; Takahashi et al., 2009). In situ seismic velocity measure-ments of representative Izu–Bonin rocks suggest the Izu–Bonin lowercrust and middle crust are composed chiefly of gabbroic andintermediate tonalitic rocks, respectively (Kitamura et al., 2003). Theprocesses operating to produce the gabbroic lower crust and interme-diate middle crust through subduction zone magmatism have beeninvestigated experimentally (Nakajima and Arima, 1998) and bypetrological and geochemical modeling (Kawate and Arima, 1998;Tatsumi et al., 2008). Much less is known, however, about thedevelopment of upper crust sequences in intra-oceanic arcs, despite

Earth and Planetary Science Letters 303 (2011) 225–239

⁎ Corresponding author. Tel.: +81 46 867 9766; fax: +81 46 867 9625.E-mail address: [email protected] (K. Tani).

0012-821X/$ – see front matter © 2011 Elsevier B.V. All rights reserved.doi:10.1016/j.epsl.2010.12.052

Contents lists available at ScienceDirect

Earth and Planetary Science Letters

j ourna l homepage: www.e lsev ie r.com/ locate /eps l

Page 2

Author's personal copy

the fact that these are the shallowest and easiest to access.Manyof thesesequences have been weathered and diagenetically altered, makinggeochemical, geochronological, and tectonic features difficult todecipher. In the Izu–Bonin arc system, the upper crust forms thebasement for the Quaternary arc volcanoes, and volcaniclastic depositswithin this basement record a long-term volcanic history of the arc andits magmatic evolution.

The northern part of the Izu–Bonin arc is colliding with the Honshuarc, a tract of continental crust, in the Izu collision zone (Taira et al.,1989; Fig. 1). This is the only place in the world where the active intra-oceanic arc is currently colliding and accreting end-on with the maturearc, providing an unusual opportunity to study the fundamentalgeologic processes during the arc–arc collision. A recent thermogeo-chronological study of a young tonalite plutonic complex in the Izucollision zone revealed rapid syncollisional tonalitic magma formationand one of the fastest exhumation rates on Earth, indicating veryvigorous tectonomagamatic processes (Tani et al., 2010). The ongoingcollision is also a fundamental interest to the modern society as thiscollision is responsible for themagnitude 7.9, Great Kanto earthquake of1923, one of the most disastrous earthquakes in 20th century, killing~105,000 people.

The Izu Peninsula, the main focus of this study, is located at thenorthern end of the Izu–Bonin arc (Fig. 1), where juvenile arc crust iscurrently migrating into an accreted block within the Izu collision zone.As a result of uplift associated with this collision, the Izu–Bonin uppercrust sequence is subaerially exposed on the Izu Peninsula, providingaccess to sequences that, just a short distance to the south, lie below sealevel. We apply zircon U–Pb geochronology to upper crustal rocks fromthe Izu Peninsula and to the basal units of nearby Quaternary volcanoesin the northern Izu–Bonin arc. The newly obtained ages and reinter-pretationof thegeochemical characteristic of thevolcanic rocks revealedthe nature of the upper crustal sequences that underlie the Izu–Boninarc volcanoes, as well as providing key insights on the tectonicinterpretation of the Quaternary volcanism in the Izu collision zone.

2. Geology

2.1. Izu Peninsula

The Izu Peninsula, located ~130 km southwest of Tokyo, is located atthe southern part of the Izu collision zone (Fig. 1). The collision of the

Izu–Bonin arc with the Honshu arc began at ~15 Ma (Taira et al., 1989).Since then, blocks of juvenile Izu–Bonin arc upper crust have beensuccessively accreted onto the Honshu arc, resulting in a sequentialsouthward migration of the collision boundary. Amano (1991)identified four accreted blocks; the southernmost of these, the Izublock, corresponds to the Izu Peninsula (Fig. 1). The Izu block is boundedon the north by a major thrust, the Kannawa Fault (Fig. 1). Abiostratigraphic study of the trough-fill deposit genetically associatedwith development of the Kannawa Fault shows that the Izu block beganto collidewith theHonshuarc from~1 Ma (Huchon andKitazato, 1984).Accreted blocks other than the Izu block are associated with syncolli-sional silicic plutonic rocks (Saito et al., 2007; Tani et al., 2010) thatrange from tonalite to granite (Fig. 1). The existence of silicic plutonsbeneath the Izu Peninsula is also inferred from tonalitic xenolithswithinthe volcaniclastic deposits of theQuaternary volcanoes and small dioriteintrusive bodies found in the tunnels of goldmines in thewesternpart ofthe peninsula (Sakamoto et al., 1999).

The northern half of the Izu Peninsula is largely mantled byQuaternary polygenetic volcanoes active from 1.8 Ma to 0.2 Ma,followed by numerous Quaternary monogenetic volcanoes, called theHigashi–Izu monogenetic volcanic group active from 0.3 Ma andcontinues to present (Fig. 2; Hasebe et al., 2001). These youngvolcanoes are underlain by widespread volcaniclastic deposits plusassociated lava flows and intrusive bodies called the Yugashima andShirahama Groups.

The Yugashima Group, as mapped by previous workers (Sawa-mura, 1955; Sawamura et al., 1970), consists of a wide variety of lavaflows, coeval intrusive bodies, and associated submarine volcaniclas-tic deposits ranging in composition from basalt to dacite (Koyama,1986; Kurasawa and Michino, 1976). Outcrops of the YugashimaGroup are limited, and are mostly found in the northern Izu Peninsula,where they form a basal complex underlying the Quaternarypolygenetic volcanoes (Fig. 2). The most conspicuous characteristicof the Yugashima Group is that almost all units display pervasivehydrothermal alteration (Sawamura, 1955; Sawamura et al., 1970).On the basis of nannofossil and foraminifera ages, the YugashimaGroup has been interpreted to be of Middle Miocene age (Koyama,1986; Okada, 1987; Saito, 1963). However, the limestone outcropsthat include these nannofossil and foraminifera are limited andgenerally occur as blocks or lenses in sedimentary rocks, makingtheir stratigraphic relationship with the surrounding volcaniclastics

Fig. 1. (A) General tectonic setting of the study area. The Izu–Bonin arc is outlined by the 3000 m bathymetric contour. Thick dotted lines represent locations of rear-arc seamountchains. (B) Geologic map of Izu collision zone (modified after Saito et al., 2007). IB, Izu block; ISTL, Itoigawa–Shizuoka tectonic line; KB, Koma block; KF, Kannawa fault; MB, Misakablock; SAT, Sagami trough; SRT, Suruga trough; TB, Tanzawa block; TATL, Tonoki–Aikawa tectonic line.

226 K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 3

Author's personal copy

uncertain. In the absence of confirming radiometric ages, the alterednature of these rocks alone led previous workers to assume theYugashima Group constitutes a Middle Miocene basement complexunderlying all of the Izu Peninsula and nearby areas. For example,several boreholes within the caldera and on the southeast flank of theQuaternary Hakone volcano, ~20 km to the northeast (Fig. 1),encountered altered volcanic rocks and on this characteristic alonethese units were assumed to be basement correlated with theYugashima Group (Hirata et al., 2001; Matsumura and Fujimoto,2008).

The southern part of the Izu Peninsula also exposes assemblages oflava flows, coeval intrusive bodies, and associated volcaniclasticdeposits mostly ranging from basalt to andesite, with subordinateamounts of dacite and rhyolite (Kano, 1989; Matsumoto et al., 1985;Tamura, 1994), but in contrast to the Yugashima Group, theseassemblages are generally fresh. These deposits, collectively known asthe Shirahama Group (Fig. 2), have yielded biostratigraphic ages of LateMiocene to Pliocene (Ibaraki, 1976, 1981; Koyama, 1986), and havebeen the focus of process-oriented studies of submarine volcanicprocesses, such as retention of heat in freshly erupted submarinebreccias (Kano, 1989; Tamura et al., 1991) and the distinctive pumice-lithic clast assemblages formed by fallout of tephra through the watercolumn accompanying submarine silicic eruptions (Cashman and Fiske,1991). Previous workers (e.g. Koyama, 1986; Sawamura et al., 1970)assumed that the Shirahama Group unconformably overlay the more

altered Yugashima basement but generally acknowledged that bothgroupswere products of submarine eruptions, even though the tectonicsetting where these eruptions took place remained unresolved.

2.2. Altered upper crustal units in the northern Izu–Bonin arc

The volcanic front of the northern Izu–Bonin arc is marked by a N–Strending line of islands, most of which are Quaternary basalticstratovolcanoes (Fig. 3). Another feature of this part of the arc is thepresence of Quaternary rear-arc volcanoes lying to thewest, behind thevolcanic front (Fig. 3). Some of these rear-arc volcanoes have grownabove sea level to form islands (e.g. Niijima and Kozushima), chieflycomposed ofmonogenetic rhyolite lava domes that are built upon aNE–SW trending bathymetric high called the Zenisu Ridge. The Izu–Boninrear-arc volcanism is characterized by the presence of NE–SW trendingseamount chains behind the volcanic front, termed rear-arc seamountchains (Fig. 1). The Zenisu Ridge has been interpreted to be thenorthernmost of these rear-arc seamount chains (Ishizuka et al., 2006).

Zenisu Rocks, a feature located ~50 km southwest of Kozushima(Fig. 3), is a small reef on the Zenisu Ridge consisting of hornblende-bearing andesite to dacite displaying strong hydrothermal alteration.Isshiki (1980) correlated these altered volcanic rocks with theYugashima Group in the Izu Peninsula and concluded that they werepart of a widespread Miocene basement underlying the northern Izu–Bonin arc. Similar correlations were also inferred from an altered dacite

Fig. 2. Geologic map of Izu Peninsula (modified from seamless digital geological map of Japan 1: 200,000, Geological Survey of Japan, 2010) with newly obtained zircon U–Pb ages(numbers in parentheses are mean square weighted deviations). Ages from volcanic breccias and tuffs are shown in italic. Black triangles represent the inferred volcanic centers ofQuaternary polygenetic volcanoes. Locations of monogenetic cones in the Higashi–Izu monogenetic volcanic group (gray triangles) are from Hasebe et al. (2001).

227K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 4

Author's personal copy

lava flow making up the oldest stratigraphic unit in the northern partof Kozushima (Taniguchi, 1977). Abundant xenoliths of altered dacitecontained in pyroclasitc deposits in southern Niijima were similarlycorrelated with what was believed to be a widespread Middle Miocenebasement, corresponding to theYugashimaGroup (Isobe andNakashima,2001; Isshiki, 1987).

Altered rocks underlying more distant Quaternary volcanoes alongthe northern Izu–Bonin volcanic front also have been correlated withthe Yugashima Group in the Izu Peninsula. These include the alteredandesitic to dacitic xenoliths contained in Quaternary volcaniclasticdeposits on Izu–Oshima (Isshiki, 1984; Kuno, 1959), located ~20 kmto the north where Zenisu Ridge intersects the volcanic front, and onMiyakejima (Isshiki, 1960), ~70 km south of Izu–Oshima (Fig. 3).Further south, altered andesitic to basaltic rocks and associatedturbidite deposits encountered in the lowermost unit of a 1500 mborehole on the southern flank of Higashiyama volcano on the islandof Hachijyojima (Fig. 3), have been similarly correlated with the

Yugashima Group (Hirata et al., 1997). Collectively, in the absence ofdetailed stratigraphic and geochronologic information, the wide-spread occurrence of altered volcanic rocks, and their gross similarityto units within the Yugashima Group, have led to the widely acceptedbelief that the northern Izu–Bonin arc is underlain by, thus built upon,an extensive basement composed chiefly of altered volcanic andintrusive rocks of Middle Miocene age (e.g. Sakamoto et al., 1999 andreferences therein).

3. Sample descriptions

Samples chosen for zircon U–Pb geochronology were collectedfrom volcaniclastic and intrusive rocks in the central and southern IzuPeninsula, outcrops previously mapped as the Yugashima or Shir-ahama Groups. Altered volcanic rocks from Zenisu Rocks, Kozushima,and Niijima, previously considered to represent the Middle Miocenebasement of the northern Izu–Bonin arc, were selected and analyzed

Fig. 3. Bathymetric map of northern Izu–Bonin arc showing newly obtained zircon U–Pb ages (numbers in parentheses are meanweighted deviations); bathymetric contour interval200 m. Thick dotted line indicates the volcanic front. The volcanic front is extended to the Quaternary Asama volcano, northwest of Hakone volcano. The isodepth contours (km) ofPacific Plate (solid line) and Philippine Sea Plate (dotted line) are estimates of Nakajima et al. (2009). Open squares show locations of Ar–Ar and K–Ar ages (ages in italic) fromunaltered volcanic rocks of northern Izu–Bonin rear-arc in Ishizuka et al. (2006).

228 K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 5

Author's personal copy

for comparison. Locations of the analyzed samples are shown in Figs. 2and 3 for Izu Peninsula and northern Izu–Bonin arc, respectively;precise sampling location coordinates are presented in Table 1.

3.1. Yugashima Group

Three samples were selected for zircon U–Pb dating from outcropsconsidered to be representative of the Yugashima Group (Fig. 2). Thefirst, an altered andesitic tuff breccia (sample: 07102405), was collectedat the type locality of the YugashimaGroup (Sawamura, 1955), near thetown of Yugashima. The second, an andesitic tuff breccia fromNumanokawa (sample: 06020412), in the central Izu Peninsula, camefrom the Lower Yugashima Group (Sawamura et al., 1970), and thethird, a quartz-bearing andesitic volcanic breccia (sample: 99031615)from the Ikeshiro River valley in the southern part of the peninsula, wasclassified by Sawamura et al. (1970) as coming from the UpperYugashima Group. All three samples, greenish in appearance, arestrongly altered; phenocrysts in volcanic clasts are replaced by clayminerals, chlorite, and epidote, typical petrographic features of theYugashimaGroup (Sawamura, 1955).Most vesicles in volcanic clasts arefilled with zeolites.

3.2. Shirahama Group

Six samples from areas mapped as Shirahama Group were selectedfor zircon U–Pb dating (Fig. 2). Three samples of hornblende-bearingporphyritic dacite (samples: No4, 06112909, and 06062403) weresampled from vertical dikes exposed near the towns of Ugusu and Arari.These dikes are considered to represent the magmatic activityaccompanying deposition of the Shirahama Group, and they intrudebasaltic to andesitic volcaniclastic deposits of the Yugashima and olderShirahama Groups exposed along the west coast of the Izu Peninsula(Sawamura et al., 1970). One sample of altered pumiceous tuff wasobtained from Kumomizaki (sample: KUMO) located on the south-western coast of the peninsula. An orthopyroxene-bearing daciticpumiceous breccia (sample: ISK2-3) was collected near the town ofMera from anoutcrop assigned to the Isshiki Formation, a subunit of theMiddle Shirahama Group (Yamada and Sakaguchi, 1987). A quartz-bearing porphyritic rhyolite sample (sample: 08012405) was collectedfrom a lava flow exposed at Onigasaki on the eastern coast of thepeninsula, ~1 km south from the town of Kawazu. Shirahama rocks areless altered than in the Yugashima Group because their plagioclasephenocrysts are better preserved, even though their mafic phenocrystsare replaced by chlorite, clay minerals, and Fe–Ti oxides. However, theKumomizaki tuff sample (sample: KUMO) has experienced stronghydrothermal alteration, showing complete replacement of original

minerals to various clayminerals and chlorite, associatedwith abundantsecondary pyrite.

3.3. Zenisu Rocks

Two samples of altered andesite (samples: R42237-6 andR42240-1) were analyzed from the Zenisu Rocks (Fig. 3), bothcollected by N. Isshiki and archived at Geological Survey of Japan.Zenisu Rocks consist of two small clusters of outcrops, ~2.5 km apart,protruding a few meters above sea level. Both samples are from thesouthwestern cluster, collected from different exposures. SampleR42237-6 (NI 60080401 in Isshiki, 1980) is an altered andesitic clastfrom a volcanic breccia, whereas sample R42240-1 (NI60080404 inIsshiki, 1980) is from an altered andesite lava flow. Both samples arestrongly altered and contain abundant secondary clay minerals,chlorite, Fe–Ti oxides, and pyrite that replace all the originalphenocrysts and groundmass. Relict textures suggest that theoriginal mafic phenocrysts were orthopyroxene and hornblende(Isshiki, 1980).

3.4. Kozushima

One strongly altered dacite lava flow sample R45321 (NI6071803in Isshiki, 1982) was collected by N. Isshiki from Kaesuhama, on thenorthern coast of Kozushima island (Fig. 3), and archived atGeological Survey of Japan. The sample displays strong hydrothermalalteration; all original phenocrysts of orthopyroxene and hornblendeare now aggregates of clay minerals and Fe–Ti oxides; feldsparphenocrysts have been replaced by clay minerals and quartz. Thesubaerial portion of Kozushima is composed of more than 16 rhyoliticmonogenetic volcanoes and associated volcaniclastic deposits; theanalyzed Kaesuhama lava flow underlies these volcanoes, and hasbeen regarded as the Tertiary basal unit for these volcanoes (Isshiki,1982).

3.5. Niijima

One strongly altered porphyritic dacite xenolith (sample: NJ01),~5 cm in diameter, contained in a pyroclastic deposit of Mukaiyamavolcano, was collected at the southwestern end of Niijima island(Fig. 3). All original mafic and some feldspar phenocrysts have beenreplaced by clay minerals and chlorite. A wide variety of rocks appearas xenoliths in this deposit, including volcanic and granitic clasts(Isobe and Nakashima, 2001; Isshiki, 1987). Niijima is mantled bymore than 15 monogenetic Quaternary volcanoes, mostly rhyolitic incomposition, but one basaltic volcano lies on the northern part of theisland (Isshiki, 1987). The Mukaiyama volcano is the youngest among

Table 1Location of samples analyzed in this study.

Area Locality Sample name Sampling position (WGS84) Rock type

Yugashima Group, Izu Peninsula Ikeshiro 99031615 34° 45.25′N 138° 50.82′E Quartz-bearing andesitic volcanic brecciaYugashima 07102405 34° 53.22′N 138° 55.44′E Altered andesitic tuff brecciaNumanokawa 06020412 34° 48.72′N 138° 54.23′E Andesitic tuff breccia

Shirahama Group, Izu Peninsula Ugusu (Ugusu beach) No4 34° 50.59′N 138° 45.91′E Hornblende-bearing porphyritic dacite dikeArari (Shin-Koganezaki tunnel) 06112909 34° 50.23′N 138° 45.91′E Hornblende-bearing porphyritic dacite dikeArari (Arari beach) 06062403 34° 49.86′N 138° 46.07′E Hornblende-bearing porphyritic dacite dikeOnigasaki 08012405 34° 44.14′N 138° 00.01′E Quartz-bearing prophyritic rhyolite lavaMera ISK2-3 34° 39.60′N 138° 47.26′E Orthopyroxene-bearing dacitic pumicious brecciaKumomizaki KUMO 34° 43.70′N 138° 44.53′E Altered pumiceous tuff

Northern Izu–Bonin arc Mamashitaura, Niijima NJ01 34° 21.54′N 139° 14.75′E Altered porphyritic dacite lava (xenolith inQuaternary pyroclastic deposit)

Kaesuhama, Kozushima R45321 34° 14.37′N 139° 09.11′E Orthopyroxene and hornblende-bearing dacitic lavaZenisu Rocks R42240-1 33° 56.55′N 138° 49.08′E Orthopyroxene and hornblende-bearing andesite lavaZenisu Rocks R42237-6 33° 56.79′N 138° 49.00′E Orthopyroxene and hornblende-bearing andesite lava

(lava clast in breccia)

229K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 6

Author's personal copy

the rhyolitic monogenetic volcanoes and has yielded a 14C eruptionage of 1120±75 y. B.P. (Isshiki, 1987).

4. Sample preparation and analytical methods

4.1. Sample preparation

Zircons were separated from hand specimens (~1 kg) by coarsecrushing (to ~300 μm) in a tungsten-carbide rotary mill. For smallsamples (b100 g), original rock fragments were hand-crushed in atungsten-carbide mortar. The heavy minerals were concentrated bypanning and further processed with a hand magnet, and theremaining fractions were purified using heavy liquid (diiodo-methane) separation. Zircon grains were then handpicked under astereo microscope. These grains were ~40 to several hundred μm longand had variable euhedral to subhedral grain shapes with dominantforms ranging from prisms to dipyramids (Fig. 4). However, all of thezircon grains separated from pumicious tuff from Kumomizaki(sample: KUMO), Shirahama Group, showed abraded crystalmorphologies (Fig. 4).

4.2. Zircon U–Pb analyses

Zircon U–Pb ages were analyzed over several sessions using theSensitive High-Resolution Ion Microprobe (SHRIMP)-II at the

National Institute of Polar Research (NIPR), Japan. The zircon grainswere mounted in epoxy, polished down to grain centers, cleaned andthen coated with ~10 nm of high-purity gold. Internal zoning ofzircons was observed in backscattered electron and cathodolumi-nescence (CL) images using the JEOL JSM7300 scanning electronmicroprobe at NIPR. A primary O2

− ion beam with a surface currentranging from−1.7 to−11 nA between sessions was used to produce20×15 to 30×25 μm flat-floored oval craters. Abundance of U wascalibrated against zircon standard SL13 (238 ppm), provided byAustralian National University. Reduction of raw data for standardsand samples was performed using the Microsoft Excel 2003 add-insSQUID v.1.12a (Ludwig, 2001) and Isoplot v.3.71 (Ludwig, 2003).Sample U–Pb measurements were calibrated against zircon standardTEMORA-2 (417 Ma; Black et al., 2004) using a calibration exponentof 2. Scatter on 207Pb-corrected (Pb/U)/(UO/U2) ratios of TEMORA-2in each sessions was typically less than 1% (2 sigma, 2σ), withexternal spot-to-spot errors of less than 3% (2σ); the latter errors areincluded in the errors of weighted mean ages for each sample. Zirconisotopic compositions and ages are listed in Table 2, with Tera–Wasserburg concordia plots in Fig. 5. Initial 238U–230Th disequilib-riumwas corrected using the method described in Parrish and Noble(2003). The Th/U ratio of the magma at the time of zirconcrystallization, required for disequilibrium correction, was assumedas (Th/U)magma=2.94±1.45 (2 standard deviation, n=31), asestimated from an average of whole-rock analyses of basaltic to

Fig. 4. Cathodoluminescence images of representative analyzed zircons. Areas of spot analyses are marked with ellipses and labeled according to grain number and spot number, aslisted in Table 2.

230 K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 7

Author's personal copy

dacitic rocks collected from the northern Izu–Bonin rear-arc volca-noes (Ishizuka et al., 2003a, 2006; Machida and Ishii, 2003).Corrections for common Pb on U/Pb values and ages for sampleswere performed using the 207Pb-correction method described inIreland and Williams (2003), which assumes concordance between206Pb/238U and 207Pb/235U ages, and the common Pb model of Staceyand Kramers (1975) for the approximate age of each analysis. Agesreported in the main text and figures are weighted means of 207Pb-230Th-corrected 206Pb/238U ages, quoted at the 95% confidence level,from 4 to 14 analyses per sample (Table 2). Outliers excluded frommean age calculations are indicated in Table 2 and Fig. 5, and areselected from statistical variation only.

5. Results

All of the analyzed zircon grains exhibit magmatic oscillatory and/or sector zoning (Fig. 4) and lack textual evidence for secondarymodification, suggesting that ages obtained in each sample representthe timing of magmatic crystallization of the zircon. However, itshould be noted that ages obtained from breccias and tuffs will onlyprovide a maximum estimate of deposition age, even if the obtainedage shows a well-constrained single age population.

Most of the analyzed samples show a tight cluster of ages, exceptfor the two samples from the Shirahama Group; pumicious tuff fromKumomizaki (sample: KUMO) and a pumicious breccia from Mera(sample: ISK2-3), both of which yielded multiple ages within a singlesample (Table 2 and Fig. 5). In the case of the Kumomizaki sample, outof 17 analyzed grains, 12 analyses on 12 grains define a weightedmean age of 1.711±0.057 Ma (mean square weighted deviation,MSWD=1.43). Furthermore, 4 analyses on 3 grains of higher Ucontent define a mean age of 2.227±0.073 Ma (MSWD=0.58).Another 2 analyses on 2 grains gave 8.3 and 2.9 Ma. For the Merasample, 8 analyses on 8 grains define a weighted mean age of 5.48±0.19 Ma (MSWD=0.80), with one older grain of 11.0 Ma. Because ofthe igneous morphology of the zircon, we interpret that the youngestand well-constrained ages of 1.711±0.057 Ma for Kumomizaki and5.48±0.19 Ma for Mera best represent the maximum estimates of thedeposition ages for these volcaniclastic samples, and that the olderages can be attributed to xenocrystic zircon incorporated duringeruption and/or deposition.

The ages of the Yugashima Group (7.4–2.7 Ma) and the ShirahamaGroup (5.5–1.7 Ma) obtained in this study broadly overlap (Fig. 6),indicating that these Groups are not successive stratigraphic units butformed as a result of broadly contemporaneous volcanic activity. Mostages indicate a Pliocene volcanic activity and are significantly youngerthan the previously accepted Middle Miocene age for the YugashimaGroup. The altered tuff breccia from Yugashima (sample: 07102405),the type locality for the Yugashima Group, yielded a well-constrainedPliocene age of 4.53±0.21 Ma (MSWD=0.95), requiring redefinitionof stratigraphic nomenclature in that area. The youngest age fromShirahama Group (1.71 Ma) obtained from the Kumomizaki tuff,extends the age of the Group into the Pleistocene.

On the other hand, two altered dacitic lava samples from ZenisuRocks (samples: R42237-6 and R42240-1) yielded well-constrainedPleistocene ages of 2.54±0.34 Ma (MSWD=0.71) and 2.14±0.17 Ma (MSWD=1.10), respectively. These ages are significantlyyounger than the previously assumed Middle Miocene age of theZenisu Ridge basement.

The altered dacite lava flow from Kozushima (sample: R45321)yielded an even young age of 0.934±0.026 Ma (MSWD=1.34), andthe altered dacitic xenolith from Niijima (sample: NJ01) showed asimilar young age of 0.881±0.079 Ma (MSWD=0.83). Our new agedata thus show that these rocks are not representative of awidespread Miocene basement claimed to underlie these Quaternaryvolcanoes; instead, they probably represent slightly older units withinthe edifice that happen to be more altered.

6. Discussion

The new zircon U–Pb ages constrain the duration of volcanicactivity in the Izu Peninsula, providing insights to the paleo-tectonicsetting of the magmatism that fed these volcanoes and the tectonicframework of the Izu collision zone. Furthermore, the zircon agesobtained from the basal units of the Quaternary volcanoes in thenorthern Izu–Bonin arc temporally constrain the nature of oldersequences that underlies these volcanoes.

6.1. No evidence for an altered Miocene basement: Quaternaryhydrothermal alteration of the Yugashima Group

The rocks of the Yugashima and Shirahama Groups are litholog-ically similar (Sawamura, 1955; Sawamura et al., 1970), but the rocksof Yugashima Group display extensive hydrothermal alteration. Themapping of previousworkers shows the distribution of the YugashimaGroup to be largely restricted to areas near the Quaternarypolygenetic volcanoes (Fig. 2). K–Ar dating of sericite and aluniteformed by hydrothermal alteration of Yugashima Group rocks nearUgusu, yielded Quaternary ages of 1.57–1.42 Ma (Hamasaki, 2000),and they related this alteration to the contemporaneous volcanicactivity at the nearby Quaternary polygenetic Tanaba volcano (Fig. 2).Similar evidence for Quaternary hydrothermal alteration of Yuga-shima Group has been obtained by K–Ar dating of alunite fromNishina, ~10 km southeast of Ugusu (Ando and Tsutsumi, 2005),suggesting that the Yugashima Group was altered in the Quaternary.

The overlapping zircon ages presented here for the alteredYugashima and the comparatively unaltered Shirahama Groupssuggest that both units are products of contemporaneous magmaticactivity. We therefore suggest that units previously defined asbelonging to the Yugashima Group are hydrothermally alteredequivalents of the Shirahama Group, and these units exposed in thesouthern and central Izu Peninsula can be collectively regarded as asequence formed through precollisional submarine volcanic activityfrom Late Miocene to Pleistocene.

6.2. Tectonic setting of Late Miocene to Pleistocene magmatic activity onthe Izu Peninsula

Even though the southern Izu Peninsula, especially the ShirahamaGroup, has been an ideal site to study the volcanic processesassociated with submarine pyroclastic eruptions (e.g. Cashman andFiske, 1991; Kano, 1989; Tamura et al., 1991), little attention has beenpaid to the paleo-tectonic setting of the volcanic activity; specificallywhether these volcanoes grew at the volcanic front, or in the rear-arcenvironment. Such tectonic setting affects factors that controlvolcanic processes, such as rate of magma production, magmacomposition, and water depth of the eruptions. Previous studiesregarded the volcanic deposits in the Izu Peninsula to have formed inthe volcanic front environment (e.g. Koyama and Umino, 1991),without clear justification. However, the fact that the current locationof the Izu Peninsula is ~10 to ~50 km west of the present volcanicfront (Fig. 3), and the lack of evidence for major tectonic offset aftercollision, suggests that the Late Miocene to Pleistocene volcanicsequences exposed on the Izu Peninsula originated in a rear-arcsetting before the onset of Izu Block collision at ~1 Ma (Huchon andKitazato, 1984).

This interpretation is supported by our new zircon ages from theIzu Peninsula and recent Ar–Ar and K–Ar ages from the rear-arcvolcanoes in the northern Izu–Bonin arc. The submarine rear-arcvolcanoes (from 33°N, near Hachijyojima, to 34.5°N, immediatelysouth of the Izu Peninsula) yielded Ar–Ar and K–Ar ages of 7.8 to0.12 Ma (Fig. 3, Ishizuka et al., 2006), contemporaneous with theduration of submarine volcanic activity (7.4 to 1.7 Ma) in the IzuPeninsula reported here (Fig. 6). Ishizuka et al. (2006) interpreted

231K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 8

Author's personal copyTa

ble2

SHRIMPzircon

U–Pb

analyses

ofrock

sfrom

theIzuPe

nins

ulaan

dno

rthe

rnIzu–

Boninarc.

Spot

206 Pb c

(%)

U (ppm

)Th (ppm

)

232 Th/

238 U

238 U

/206Pb

±%

207Pb

/206Pb

±%

206Pb

*(p

pm)

206Pb

/238U

age(M

a)(1

)

1sigm

aerr

206 Pb/

238 U

Meanage

(Ma)

(1)

(95%

c)

MSW

D23

8 Pb/

206 U

(Thc

)±%

207 Pb/

206 Pb

(Thc

)±%

206 Pb/

238 U

age(M

a)(Thc

)

1sigm

aerr

206 Pb/

238 U

Meanage

(Ma)

(Thc

)(95%

c)

MSW

D

Yuga

shim

aGroup

:Ikeshiro

volcan

icbrecciasample:

9903

1615

(stderr=

0.41

%)7.34

±0.34

1.10

7.44

±0.34

1.08

9903

1615

-1.1

6.8

164

104

0.65

823

4.9

0.10

013

0.17

17.29

0.38

(n=

7)81

44.8

0.09

9013

7.38

0.38

(n=

7)99

0316

15-2.1

9.8

119

510.44

912

6.2

0.12

421

0.11

26.37

0.46

901

6.1

0.12

221

6.47

0.46

9903

1615

-3.1

4.3

114

600.55

777

5.7

0.08

0022

0.12

67.93

0.49

769

5.7

0.07

9122

8.03

0.49

9903

1615

-4.1

6.0

165

980.61

802

4.8

0.09

3318

0.17

77.55

0.40

794

4.8

0.09

2318

7.64

0.40

9903

1615

-5.1

8.1

100

370.38

812

6.1

0.11

015

0.10

67.29

0.48

802

6.1

0.10

915

7.39

0.48

9903

1615

-6.1

8.6

7835

0.46

809

7.7

0.11

431

0.08

287.28

0.67

800

7.6

0.11

332

7.38

0.67

9903

1615

-7.1

3.7

202

119

0.61

822

4.3

0.07

5224

0.21

27.55

0.37

813

4.2

0.07

4425

7.64

0.37

Yuga

shim

aGroup

:Yug

ashimatuffbrecciasample:07

1024

05(stderr=

0.37

%)4.44

±0.21

0.99

4.53

±0.21

0.95

0710

2405

-2.1

4.9

8441

0.51

1330

5.4

0.08

4922

0.05

414.61

0.27

(n=

10)

1306

5.3

0.08

3322

4.70

0.27

(n=

10)

0710

2405

-3.1

7.7

4720

0.43

1219

8.2

0.10

736

0.03

354.88

0.48

1197

8.0

0.10

537

4.98

0.48

0710

2405

-4.1

8.3

303

387

1.3

1253

2.9

0.11

130

0.20

74.72

0.26

1238

2.8

0.11

030

4.78

0.26

0710

2405

-5.1

6.1

7839

0.51

1495

7.3

0.09

4525

0.04

494.05

0.32

1465

7.1

0.09

2626

4.14

0.32

0710

2405

-6.2

2846

170.38

1155

110.26

524

0.03

444.03

0.62

1136

110.26

024

4.13

0.63

0710

2405

-7.1

1245

170.39

1254

120.13

940

0.03

064.53

0.66

1232

120.13

640

4.63

0.66

0710

2405

-8.1

3.7

7341

0.58

1256

6.2

0.07

5426

0.05

004.94

0.33

1235

6.1

0.07

4127

5.03

0.33

0710

2405

-9.1

4.9

9049

0.57

1504

6.3

0.08

4722

0.05

154.07

0.28

1474

6.2

0.08

3023

4.17

0.28

0710

2405

-10.1

1065

360.57

1380

9.4

0.12

724

0.04

054.20

0.43

1354

9.2

0.12

424

4.29

0.43

0710

2405

-11.1

3.7

114

860.78

1431

5.2

0.07

5321

0.06

834.34

0.24

1406

5.1

0.07

4021

4.42

0.24

Yuga

shim

aGroup

:Num

anok

awatuffbrecciasample:

0602

0412

(stderr=

0.41

%)2.58

2±

0.09

30.22

2.66

2±

0.09

40.24

0602

0412

-1.1

1868

880

31.2

2012

3.6

0.19

17.5

0.29

42.61

50.11

(n=

6)19

733.5

0.18

77.6

2.68

30.11

(n=

6)06

0204

12-2.1

4.9

938

1327

1.5

2378

3.3

0.08

529.0

0.33

92.57

60.08

823

313.2

0.08

359.2

2.63

40.08

906

0204

12-3.1

6.4

643

420

0.68

2298

4.0

0.09

6811

0.24

02.62

50.11

2231

3.9

0.09

4011

2.71

40.11

0602

0412

-4.1

7.7

811

391

0.50

2330

3.8

0.10

710

0.29

92.55

30.10

2256

3.7

0.10

49.8

2.64

90.10

0602

0412

-5.1

1833

227

10.84

2162

5.6

0.18

712

0.13

22.44

80.16

2107

5.5

0.18

213

2.53

10.17

0602

0412

-6.1

6.5

568

284

0.52

2298

4.4

0.09

7414

0.21

22.62

30.13

2226

4.3

0.09

4415

2.71

80.13

Shirah

amaGroup

:Ugu

suintrus

ivesample:

No4

(stderr=

0.31

%)4.73

±0.17

1.86

4.82

±0.17

1.79

N04

-3.2

6379

290.38

584

3.3

0.54

14.5

0.11

74.12

0.39

(n=

9)57

93.3

0.53

74.6

4.22

0.39

(n=

9)N04

-7.1

7465

300.47

423

3.1

0.63

44.0

0.13

33.89

0.55

420

3.1

0.63

14.1

3.99

0.56

N04

-8.1

2110

040

0.41

1111

3.8

0.21

27.6

0.07

704.58

0.21

1094

3.8

0.20

87.7

4.68

0.21

N04

-9.1

5110

943

0.41

634

2.9

0.44

74.2

0.14

75.00

0.30

628

2.9

0.44

34.2

5.10

0.30

N04

-10.1

5810

948

0.45

631

2.9

0.50

44.7

0.14

94.29

0.35

625

2.9

0.50

04.8

4.39

0.35

N04

-11.1

4811

545

0.41

765

3.0

0.42

54.4

0.13

04.38

0.25

756

3.0

0.42

04.5

4.48

0.25

N04

-12.1

4610

445

0.45

761

3.1

0.41

04.6

0.11

84.56

0.26

753

3.1

0.40

64.7

4.66

0.26

N04

-13.1

3135

323

70.69

890

2.1

0.29

13.0

0.34

15.00

0.14

880

2.0

0.28

83.0

5.09

0.14

N04

-14.1

4612

770

0.57

698

2.8

0.40

94.0

0.15

64.99

0.25

691

2.7

0.40

54.0

5.09

0.25

Shirah

amaGroup

:Arariintrus

ivesample:

0611

2909

(stderr=

0.31

%)4.91

±0.12

1.06

4.98

±0.12

0.97

0611

2909

-1.1

7777

290.40

417

4.1

0.65

35.8

0.15

93.58

0.79

(n=

9)41

44.0

0.64

95.9

3.68

0.80

(n=

9)06

1129

09-2.1

3093

370.41

959

3.8

0.28

26.8

0.08

364.71

0.25

946

3.8

0.27

86.9

4.81

0.25

0611

2909

-3.1

4910

647

0.46

715

3.1

0.43

15.3

0.12

74.62

0.31

708

3.1

0.42

75.4

4.72

0.31

0611

2909

-4.1

1580

397

01.25

1001

2.7

0.16

73.3

0.68

95.45

0.15

992

2.7

0.16

53.3

5.52

0.16

0611

2909

-5.1

3723

414

70.65

830

2.5

0.34

17.4

0.24

24.86

0.28

821

2.5

0.33

87.5

4.95

0.29

0611

2909

-6.1

5412

562

0.51

651

2.8

0.47

63.9

0.16

54.51

0.29

646

2.8

0.47

14.0

4.61

0.29

0611

2909

-7.1

4119

515

10.80

733

2.5

0.37

04.9

0.22

85.18

0.25

727

2.5

0.36

75.0

5.27

0.25

0611

2909

-8.1

1479

688

81.2

1101

1.8

0.16

02.9

0.62

15.01

0.10

1089

1.7

0.15

82.9

5.08

0.10

0611

2909

-9.1

1399

614

471.5

1143

1.7

0.14

82.7

0.74

84.91

0.08

911

331.7

0.14

62.8

4.97

0.09

006

1129

09-10.1

3916

063

0.41

810

2.8

0.35

45.2

0.16

94.85

0.24

801

2.8

0.35

05.2

4.95

0.24

Shirah

amaGroup

:Arariintrus

ivesample:

0606

2403

(stderr=

0.31

%)4.87

±0.19

1.12

4.96

±0.20

1.08

0606

2403

-1.1

4610

030

0.31

692

3.2

0.41

34.8

0.12

44.98

0.30

(n=

9)68

53.2

0.40

94.9

5.08

0.30

(n=

9)06

0624

03-2.1

4517

911

90.69

700

2.5

0.40

03.6

0.21

95.08

0.22

694

2.5

0.39

63.6

5.17

0.23

0606

2403

-3.1

6459

210.37

487

3.6

0.55

45.0

0.10

54.72

0.52

484

3.5

0.55

05.1

4.82

0.53

0606

2403

-4.1

4589

290.34

761

3.6

0.40

55.6

0.10

04.62

0.30

752

3.5

0.40

15.7

4.72

0.31

232 K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 9

Author's personal copy

0606

2403

-5.1

4951

220.43

795

5.0

0.43

08.0

0.05

554.16

0.42

786

4.9

0.42

58.1

4.26

0.42

0606

2403

-6.1

2112

463

0.52

1055

3.7

0.21

17.2

0.10

14.84

0.22

1040

3.7

0.20

87.3

4.93

0.22

0606

2403

-7.1

4456

220.41

850

5.0

0.39

38.2

0.05

624.25

0.38

840

5.0

0.38

88.3

4.35

0.39

0606

2403

-8.1

2116

812

30.75

1221

3.3

0.21

06.4

0.11

84.18

0.17

1202

3.3

0.20

76.5

4.27

0.17

0606

2403

-9.1

3578

270.36

840

6.9

0.32

29.4

0.08

025.00

0.45

829

6.8

0.31

89.5

5.10

0.46

0606

2403

-10.1

2313

759

0.45

975

3.3

0.22

46.2

0.12

15.12

0.21

962

3.3

0.22

16.3

5.22

0.21

Shirah

amaGroup

:Oniga

saki

lava

sample:

0801

2405

(stderr=

0.18

%)3.41

±0.12

0.98

3.49

±0.12

0.95

0801

2405

-1.1

1294

560.62

1474

7.9

0.14

019

0.05

453.85

0.34

(n=

14)

1446

7.7

0.13

720

3.95

0.34

(n=

14)

0801

2405

-2.1

2.9

1969

2549

1.3

1853

2.2

0.06

886.0

0.91

33.38

0.07

618

222.2

0.06

776.2

3.44

0.07

708

0124

05-3.1

2344

170.40

1429

110.22

423

0.02

663.49

0.47

1399

100.21

923

3.59

0.48

0801

2405

-4.1

1347

200.43

1620

110.15

133

0.02

503.45

0.47

1583

110.14

834

3.55

0.47

0801

2405

-5.1

1862

320.53

1544

9.1

0.18

521

0.03

433.44

0.37

1512

8.9

0.18

121

3.53

0.38

0801

2405

-6.1

2041

170.43

1495

120.20

031

0.02

343.47

0.53

1464

110.19

632

3.57

0.53

0801

2405

-7.1

9.5

8142

0.53

1614

8.6

0.12

126

0.04

333.61

0.35

1579

8.4

0.11

826

3.71

0.35

0801

2405

-8.1

3652

280.56

1687

110.32

722

0.02

662.46

0.45

1649

110.32

023

2.55

0.46

0801

2405

-9.1

1674

360.51

1542

8.2

0.17

019

0.04

123.52

0.34

1509

8.1

0.16

719

3.62

0.34

0801

2405

-11.1

4.5

347

332

0.99

1700

4.2

0.08

1812

0.17

53.62

0.16

1668

4.1

0.08

0312

3.70

0.16

0801

2405

-12.1

2653

220.42

1615

110.25

127

0.02

842.96

0.48

1578

110.24

528

3.06

0.48

0801

2405

-13.1

3056

220.41

1667

110.28

021

0.02

892.72

0.41

1628

100.27

421

2.82

0.42

0801

2405

-14.1

1011

865

0.57

1636

6.7

0.12

818

0.06

173.53

0.26

1601

6.6

0.12

518

3.63

0.26

0801

2405

-15.1

1595

460.50

1621

7.5

0.16

519

0.05

063.38

0.30

1585

7.4

0.16

219

3.47

0.30

Shirah

amaGroup

:Merabrecciasample:

ISK2-3(stderr=

0.18

%)5.40

±0.19

0.85

5.48

±0.19

0.80

ISK2-3-1.1

2054

270.52

1036

8.1

0.20

219

0.04

504.99

0.50

(n=

8)10

217.9

0.19

919

5.09

0.51

(n=

8)ISK2-3-2.1

1839

130.33

886

8.8

0.18

620

0.03

835.99

0.63

874

8.7

0.18

420

6.09

0.63

ISK2-3-3.1

1481

410.52

1164

7.0

0.15

918

0.05

944.74

0.39

1146

6.9

0.15

718

4.84

0.39

ISK2-3-4.1

1159

220.39

1060

7.7

0.13

622

0.04

765.39

0.48

1044

7.6

0.13

422

5.49

0.48

ISK2-3-6.1

1964

280.45

1024

7.3

0.19

716

0.05

415.09

0.45

1009

7.2

0.19

416

5.19

0.45

ISK2-3-8.1

4.7

128

710.58

562

3.9

0.08

3111

0.19

610

.90.45

558

3.9

0.08

2411

11.0

0.45

ISK2-3-9.1

2.8

377

231

0.63

1125

4.6

0.06

8610

0.28

75.56

0.26

1109

4.5

0.06

7610

5.65

0.26

ISK2-3-10

.12.0

1265

1363

1.1

1153

2.2

0.06

206.1

0.94

35.48

0.12

1139

2.1

0.06

136.2

5.55

0.12

ISK2-3-11

.19.1

108

520.50

1120

6.0

0.11

818

0.08

295.23

0.35

1103

5.9

0.11

618

5.33

0.35

Shirah

amaGroup

:Kum

omizak

ituffsample:

Kum

o(stderr=

0.35

%)1.61

7±

0.05

61.49

1.71

1±

0.05

71.43

KUMO-1.1

1424

913

10.54

2545

2.5

0.15

48.7

0.08

422.18

60.07

0mea

nag

eof

analyses:

2.1,

3.1,

4.1,

5.1,

6.1,

7.1,

12.1,1

3.1,

14.1,

15.1,1

6.1,

17.1

(n=

12)

2458

2.4

0.14

99.0

2.28

00.07

1mea

nag

eof

analyses:

2.1,

3.1,

4.1,

5.1,

6.1,

7.1,

12.1,1

3.1,

14.1,

15.1,1

6.1,

17.1

(n=

12)

KUMO-1.2

1514

764

0.45

2670

3.3

0.16

37.5

0.04

712.05

60.07

725

723.2

0.15

77.7

2.15

40.07

8KUMO-2.1

2615

469

0.46

2791

3.1

0.25

06.0

0.04

731.71

20.07

026

843.0

0.24

16.3

1.80

90.07

2KUMO-3.1

2713

075

0.60

2821

3.3

0.25

96.3

0.03

971.66

90.07

327

173.2

0.24

96.5

1.76

10.07

5KUMO-4.1

2410

457

0.56

3421

4.1

0.23

48.7

0.02

621.43

40.07

732

683.9

0.22

49.1

1.52

80.08

0KUMO-5.1

2112

260

0.51

3215

3.8

0.21

47.9

0.03

251.57

80.07

430

763.6

0.20

58.3

1.67

30.07

6KUMO-6.1

3092

410.46

2789

4.0

0.28

09.2

0.02

831.62

80.10

2682

3.8

0.26

99.6

1.72

50.10

KUMO-7.1

1917

111

20.68

3071

3.3

0.19

88.5

0.04

781.69

40.07

12.13

3±

0.07

20.62

2953

3.1

0.19

18.9

1.78

30.07

32.22

7±

0.07

30.58

KUMO-8.1

1319

211

70.63

2667

2.8

0.14

67.1

0.06

192.11

20.06

7mea

nag

eof

analyses:

1.1,

1.2,

8.1,

9.1(n

=4)

2576

2.7

0.14

17.3

2.20

30.06

8mea

nag

eof

analyses:

1.1,

1.2,

8.1,

9.1(n

=4)

KUMO-9.1

1618

394

0.53

2511

2.7

0.17

06.1

0.06

252.16

40.06

924

262.7

0.16

46.3

2.25

90.07

0KUMO-10.1

1.3

128

470.38

776

1.9

0.05

677.5

0.14

28.19

20.17

767

1.9

0.05

617.6

8.29

20.17

KUMO-11.1

1.7

1378

339

0.25

2310

1.5

0.05

924.4

0.51

32.74

40.04

322

301.5

0.05

724.6

2.84

90.04

3KUMO-12.1

2314

489

0.64

2712

100.22

616

0.04

561.83

50.22

2618

100.21

816

1.92

60.22

KUMO-13.1

3012

777

0.63

2967

5.2

0.28

010

0.03

671.52

80.11

2854

5.0

0.27

011

1.61

90.12

KUMO-14.1

2218

411

70.65

3438

5.0

0.21

911

0.04

611.46

30.09

432

894.8

0.21

012

1.55

30.10

KUMO-15.1

3389

370.42

2495

6.0

0.30

912

0.03

081.72

20.16

2408

5.8

0.29

812

1.82

10.16

KUMO-16.1

4373

280.39

2334

6.5

0.38

215

0.02

701.58

70.22

2256

6.2

0.36

915

1.68

70.23

KUMO-17.1

3791

430.48

2177

5.5

0.33

810

0.03

601.86

40.16

2112

5.3

0.32

810

1.96

10.17

NorthernIzu–

Boninarc:

Niijim

alava

xeno

lithsample:

NJ01(stderr=

0.41

%)0.77

2±

0.07

60.90

0.88

1±

0.07

90.83

NJ01-1.1

4987

260.31

3339

8.8

0.43

615

0.02

230.97

60.18

(n=

7)31

788.4

0.41

516

1.08

0.19

(n=

7)NJ01-1.2

1361

180.31

8059

150.14

948

0.00

648

0.69

50.13

7182

140.13

354

0.79

90.14

NJ01-2.1

1662

190.31

5953

130.17

134

0.00

902

0.91

10.14

5460

110.15

737

1.01

0.14

NJ01-2.2

4352

100.21

5388

130.38

924

0.00

833

0.67

70.17

4966

120.35

926

0.78

40.18

NJ01-3.1

1411

130

0.28

6977

120.16

044

0.01

360.79

10.12

6303

110.14

549

0.89

50.13

NJ01-3.2

0.11

327

330.11

8139

6.2

0.04

6940

0.03

450.79

10.05

371

845.5

0.04

1446

0.90

30.05

4NJ01-3.3

3.3

118

330.29

1024

116

0.07

2189

0.00

989

0.60

90.11

8856

140.06

2310

30.71

30.11

(con

tinu

edon

next

page)

233K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 10

Author's personal copy

Table2(con

tinu

ed)

Spot

206 Pb c

(%)

U (ppm

)Th (ppm

)

232 Th/

238 U

238 U

/206Pb

±%

207Pb

/206Pb

±%

206Pb

*(p

pm)

206Pb

/238U

age(M

a)(1

)

1sigm

aerr

206 Pb/

238 U

Meanage

(Ma)

(1)

(95%

c)

MSW

D23

8 Pb/

206 U

(Thc

)±%

207 Pb/

206 Pb

(Thc

)±%

206 Pb/

238 U

age(M

a)(Thc

)

1sigm

aerr

206 Pb/

238 U

Meanage

(Ma)

(Thc

)(95%

c)

MSW

D

NorthernIzu–

Boninarc:

Koz

ushimalava

sample:

R453

21(stderr=

0.37

%)0.84

9±

0.04

22.36

0.93

4±

0.02

61.77

R453

21-1.1

1.6

1322

1882

1.5

7083

2.7

0.05

8810

0.16

00.89

50.02

5(n

=10

)66

832.9

0.05

5511

0.95

30.02

8(n

=10

)R4

5321

-3.1

3933

621

00.65

4265

3.6

0.35

47.5

0.06

770.92

10.06

240

383.4

0.33

57.9

1.01

0.06

5R4

5321

-4.1

4.7

200

129

0.67

7702

7.1

0.08

2826

0.02

230.79

80.06

169

976.5

0.07

5229

0.88

70.06

3R4

5321

-5.1

7.7

364

165

0.47

6821

4.4

0.10

713

0.04

580.87

20.04

262

174.0

0.09

7414

0.96

90.04

3R4

5321

-6.1

3.5

797

352

0.46

7728

3.3

0.07

3412

0.08

860.80

50.02

869

593.0

0.06

6113

0.90

30.02

9R4

5321

-7.1

5.1

573

322

0.58

7931

3.8

0.08

6013

0.06

210.77

20.03

271

593.5

0.07

7614

0.86

40.03

3R4

5321

-8.1

1.0

241

141

0.60

6648

6.1

0.05

4236

0.03

120.96

00.06

461

025.7

0.04

9739

1.05

0.06

5R4

5321

-9.1

7.7

298

125

0.43

7563

5.1

0.10

616

0.03

380.78

70.04

468

184.6

0.09

6018

0.88

60.04

6R4

5321

-10.1

5.8

397

190

0.49

6712

4.1

0.09

1912

0.05

080.90

50.04

061

323.8

0.08

3913

1.00

0.04

1R4

5321

-11.1

7.1

340

203

0.62

7017

4.6

0.10

215

0.04

170.85

30.04

364

144.3

0.09

3416

0.94

50.04

5NorthernIzu–

Boninarc:

Zenisu

lava

sample:

R422

40-1

(stderr=

0.46

%)2.04

±0.17

1.14

2.14

±0.18

1.10

R422

40-1.1

2476

350.48

2730

100.23

825

0.02

391.79

0.26

(n=

9)26

2810

0.22

926

1.88

0.27

(n=

9)R4

2240

-2.1

1246

160.36

2814

130.14

134

0.01

392.02

0.30

2701

120.13

536

2.12

0.30

R422

40-3.1

5.6

7131

0.45

2763

100.09

0748

0.02

212.20

0.26

2658

100.08

7249

2.30

0.27

R422

40-4.1

6.8

6329

0.48

2265

9.8

0.10

032

0.02

382.65

0.28

2194

9.5

0.09

7033

2.75

0.29

R422

40-5.1

5.6

6928

0.41

2964

110.09

0044

0.02

002.05

0.25

2841

100.08

6346

2.15

0.25

R422

40-6.1

3349

220.47

2746

120.31

026

0.01

531.56

0.30

2643

120.29

827

1.66

0.31

R422

40-8.1

1.5

7029

0.43

2922

100.05

8253

0.02

072.17

0.24

2804

100.05

5955

2.27

0.24

R422

40-9.1

1073

300.42

2901

9.9

0.12

930

0.02

161.99

0.22

2783

9.5

0.12

431

2.09

0.23

R422

40-10.1

1669

290.43

2826

9.8

0.17

326

0.02

111.91

0.23

2715

9.4

0.16

727

2.01

0.23

NorthernIzu–

Boninarc:

Zenisu

lava

clastin

brecciaR4

2237

-6(stderr=

0.46

%)2.45

±0.34

0.71

2.54

±0.34

0.71

R422

37-1.1

8.0

9154

0.61

2626

8.4

0.10

929

0.02

992.26

0.21

(n=

4)25

378.1

0.10

630

2.35

0.22

(n=

4)R4

2237

-1.2

1469

280.42

1885

200.15

841

0.03

162.93

0.65

1834

190.15

442

3.03

0.65

R422

37-2.1

1855

270.52

2043

260.18

757

0.02

302.59

0.79

1987

250.18

258

2.69

0.80

R422

37-2.2

2.1

5728

0.50

2313

110.06

2662

0.02

132.73

0.32

2241

100.06

0664

2.82

0.33

Errors

arequ

oted

at1-sigm

a,ex

cept

formea

nsampleag

es,w

hich

are95

%co

nfide

nceintervals.Pb

can

dPb

*indicate

theco

mmon

andradiog

enic

portions

,respe

ctively.

(stderr)

Errorin

Stan

dard

calib

ration

(%sigm

a,no

tinclud

edin

abov

eerrors

butrequ

ired

whe

ncalculatingmea

nag

esor

combining

data

withthosefrom

differen

tsessions

ormetho

ds).

(1)Co

mmon

Pbco

rrectedby

assu

ming

206Pb

/238U–207Pb

/235U

age-co

ncorda

nce.

(Thc

)230Th

correctedby

assu

ming(Th/U) m

agma=

2.94

.Sa

mplean

alysis

initalic:ex

clud

edas

anou

tlierwhe

ncalculatingmea

nag

es.

234 K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 11

Author's personal copy

that Miocene to Pliocene rear-arc volcanism in the northern Izu–Bonin arc was active over a wider area extending 120 km west of thecurrent volcanic front, whereas Quaternary rear-arc volcanism islimited to within ~40 km of the volcanic front, and attributed thistemporal shift to the steepening of the subducting Pacific Plate. Thenewly obtained zircon ages exhibit similar spatial distribution, LateMiocene to Pleistocene ages were obtained from a broad zonecovering the Izu Peninsula and Zenisu Rock, and located ~30–70 kmwest of the volcanic front. In contrast, younger (b1 Ma) ages arerestricted to the basal units on the islands of Niijima and Kozushima,situated within the 40 km of the present volcanic front (Fig. 6).

The whole-rock compositions of the volcanic rocks from IzuPeninsula also indicate a rear-arc origin. Published whole-rockcompositions of the volcanic rocks from Shirahama and YugashimaGroup (Kurasawa, 1984; Kurasawa and Michino, 1976; Matsumotoet al., 1985; Tamura, 1994; Tamura and Tatsumi, 2002) are enriched inlarge ion lithophile (LIL) and high field strength (HFS) elements, suchas K, Rb, and Zr compared to rocks from the volcanic front (Fig. 7).Most of the Izu Peninsula rocks, including those from the northernQuaternary polygenetic volcanoes, fall in the medium-K field in SiO2–

K2O plot, whereas rocks from the Quaternary volcanic front showdistinctive low-K trends (Fig. 7). Such enrichment of LIL and HFSelements, as observed in Izu Peninsula rocks, is a typical geochemicalcharacteristic of Miocene to Pliocene Izu–Bonin rear-arc volcanoes(Hochstaedter et al., 2000). The compiled whole-rock compositions ofthe submarine Miocene to Pliocene Izu–Bonin rear-arc volcanoes thusagree well with those exposed subaerially on the Izu Peninsula(Fig. 7).

Petrograhic features of Izu Peninsula rocks also support the aboveinference. The evolved (~dacite) volcanic and intrusive rocks of IzuPeninsula often contain hornblende as a phenocryst or as a relict inaltered rocks (e.g. Matsumoto et al., 1985; Sawamura, 1955), and forexample, porphyritic dacite intrusive rocks from Ugusu and Ararianalyzed in this study (No4, 06112909, and 06062403) containabundant hornblende phenocrysts. Such hydrous minerals (mostlyamphiboles, rarely biotite) only occur as phenocryst in the rear-arcmagmas whereas silicic magmas erupting at volcanic front havedistinctive anhydrous mafic phenocryst assemblage, mostly pyrox-enes (e.g. Isshiki, 1982, 1987; Ishizuka et al., 2002, 2006; Tamura &Tatsumi, 2002; Tani et al., 2008).

Despite the general geochemical and petrographycal similaritiesbetween the rocks from Izu Peninsula and those from the submarineIzu–Bonin rear-arc volcanoes to the south, there are noteworthycompositional differences among them. In the northern Izu–Boninrear-arc, the Miocene volcanism is dominated by andesite, whereasPliocene to Quaternary volcanic products are characterized bybimodal compositions of basalt and rhyolite (Hochstaedter et al.,2001; Ishizuka et al., 2006). This compositional change occurred at 3–2 Ma in the rear-arc region (Hochstaedter et al., 2001) as well as alongthe volcanic front (Bryant et al., 2003). In contrast, even though thesubmarine volcanic deposits in the Izu Peninsula are mostly basalt toandesite, considerable volumes of dacite to rhyolite can also be found(Koyama, 1986; Kurasawa and Michino, 1976; Matsumoto et al.,1985; Tamura, 1994). Our zircon ages from the Shirahama Group(5.5–1.7 Ma) were obtained from dacitic to rhyolitic rocks, suggestingthat silicic volcanism existed at least from Late Miocene in the IzuPeninsula. Hochstaedter et al. (2001) attributed the inception of back-arc rifting and associated basaltic magmatism, which began at ~3 Ma,to the compositional shift from andesitic volcanism to bimodal basaltand rhyolite volcanism in the rear-arc seamount chains. This isunlikely mechanism for the prolonged silicic volcanism in the IzuPeninsula, as this region, including the northern Izu–Bonin arc, ischaracterized by the absence of back-arc rifting (Ishizuka et al., 2006).One possibility is that as Izu Peninsula, being above sea level, providesa more complete record of submarine rear-arc volcanism than thatobtained from the submarine rear-arc volcanoes. In short, rocks of the

Izu Peninsulamight display the full compositional diversity that existsin the Miocene rear-arc seamount chains to the south. This inferenceis supported by the sporadic occurrences of Miocene (7–5 Ma) daciticto rhyolitic rocks in the rear-arc seamount chains in central Izu–Boninarc (Ishizuka et al., 2003b).

6.3. Across-arc geochemical variation of the Quaternary volcanoes in theIzu collision zone

Fuji and Hakone volcanoes, located ~20–40 km north of the IzuPeninsula, are two large Quaternary active volcanoes within the Izucollision zone (Fig. 1). Despite the fact that historical explosiveeruptions of these volcanoes, especially of Fuji volcano, pose severevolcanic hazards to the surrounding region, tectonic interpretations ofmagma genesis for these volcanoes have remained unresolved, chieflybecause of tectonic complexities within the collision zone. Theprevious inference that submarine volcanic sequences of the IzuPeninsula originated in the volcanic front environment (e.g. Koyamaand Umino, 1991) added to the confusion. An unexplained largetectonic offset would be required to displace volcanic front rocks tothe current position of the Izu Peninsula, which is ~10 to ~50 kmwestof the present volcanic front. However, with the new insightssuggesting that Late Miocene to Pleistocene Izu Peninsula rocks arerear-arc origin, allow us to reconsider the tectonic framework of theQuaternary volcanism in the Izu collision zone.

Recent estimation of the isodepth of the subducting slabs of PacificPlate and Philippine Sea Plate in the northern Izu–Bonin arc and Izucollision zone (Nakajima et al., 2009) show that volcanic front of thenorthern Izu–Bonin arc, from Izu–Oshima to Hachijyojima and furthersouth, lie above the 120 km isodepth of the Pacific Plate (Fig. 3). Northof Izu–Oshima volcano, and into the Izu collision zone, the volcanicfront shifts to the west and toward a deeper isodepth of ~140 km forHakone volcano, corresponding to the deformation of the overridingPhilippine Sea Plate as a result of the northwestward collision of thebuoyant Izu–Bonin arc crust against the Honshu arc (Fig. 3). Hakonevolcano, even though it is situated above a deeper isodepth,dominantly erupts low-K tholeiitic basalt similar to those of theIzu–Bonin volcanic front volcanoes, along with low-K to medium-Kdacite and rhyolite (Fig. 7). In contrast, the whole rock compositionsof ejecta from Fuji volcano, situated ~30 km northwest of the Hakonevolcano, are distinctively enriched in LIL and HFS elements similar tothose of the Izu–Bonin rear-arc volcanoes, showing even higherenrichment of these elements than the Izu Peninsula rocks (Fig. 7).Watanabe et al. (2006) showed that mantle source of the Fuji volcanois isotopically identical to those of the Izu–Bonin rear-arc volcanoes,with a minimum amount of crustal assimilation from the Honshu arc.

The above inferences suggest that Hakone and Fuji volcanoescorrespond to the volcanic front and rear-arc volcanoes of the Izu–Bonin arc, respectively. The rear-arc origin of the Izu Peninsula,indicating the lack of major tectonic offset of the region after thecollision, supports this inference. Thus the geochemical variationobserved within Hakone and Fuji volcanoes likely to represent theIzu–Bonin across-arc variations, still undisturbed at the earlier phaseof collision.

6.4. What is the age of the basement underlying the QuaternaryIzu–Bonin volcanoes?

It has beenwidely believed that theQuaternary Izu–Bonin volcanoesdirectly lie upon a widespread and older Miocene basement (e.g.Sakamoto et al., 1999). However, our newly obtained zircon U–Pb agesfrom the Izu Peninsula and basal units of the Quaternary volcanoes inthe northern Izu–Bonin arc require reconsideration of the nature andorigin of the rocks underlying these volcanoes.

Firstly, the altered volcanic rocks and associated volcaniclasticdeposits of the Yugashima Group exposed in the Izu Peninsula do not

235K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 12

Author's personal copy

236 K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 13

Author's personal copy

constitute a widespread Middle Miocene basement. We show that thevolcanic sequences exposed in the Izu Peninsula represent material ofsubmarine rear-arc volcanoes active from LateMiocene to Pleistocene.This volcanic assemblage was later uplifted as a result of the Izu–Honshu collision at ~1 Ma (Huchon and Kitazato, 1984). The samplesfrom Zenisu Rocks yielded Pleistocene ages of 2.54–2.14 Ma, whichfall within the periods of petrographically similar rear-arcmagmatismin the northern Izu–Bonin arc from 7.8 to 0.12 Ma (Ishizuka et al.,2006), suggesting that the Zenisu Rocks, although altered, representone of these rear-arc volcanoes rather than altered Middle Miocenebasement as proposed by Isshiki (1980).

Secondly, the highly altered lava flow forming the basal unit atKozushima, and the lithologically similar xenolith collected from thepyroclastic deposit on Niijima, yield ages (b1 Ma) significantlyyounger than either the Late Miocene to Pleistocene volcanicsequences exposed in Izu Peninsula or the Pleistocene Zenisu Rocks.This suggests that the conspicuous island volcanoes of the northernIzu–Bonin arc are underlain by products of only slightly older phasesof volcanism, rather than older unrelated basement. A recent single-channel seismic survey conducted at the Quaternary Sumisu volcano,a submarine caldera located at the Izu–Bonin volcanic front ~200 kmsouth of Hachijyojima, supports this inference (Tani et al., 2008). Theseismic profiles show that Sumisu volcano is a compound edificeconstructed of its own thick volcaniclastic strata that rest upon apronsof at least two adjacent submarine volcanoes, and lack a reflector thatwould suggest the presence of an uncomformable older basement.

It is possible that much older (NMiocene) basal units underlie theQuaternary Izu–Bonin volcanoes at a greater depth, but the findingspresented here indicate that these conspicuous island volcanoes wereconstructed upon submarine volcanic complexes that formed duringearlier phases of the same magmatic episode.

7. Conclusions

1) Newly obtained zircon U–Pb ages show that widespread subma-rine volcanic and intrusive sequences on the Izu Peninsula wereproducts of continuous magmatic activity extending from the LateMiocene to Pleistocene. Previous workers subdivided these

sequences into the Middle Miocene Yugashima Group and theUpper Miocene to Pliocene Shirahama Group, based chiefly ondegree of alteration, but our new ages show that the magmaticactivity associated with both groups was contemporaneous. Rocks

Fig. 7. SiO2–K2O, –Rb, and –Zr plots for the volcanic rocks from the Izu collision zone,Izu–Bonin rear-arc, and Quaternary Izu–Bonin volcanic front. IB-RA, Izu–Bonin rear-arc;IB-VF, Izu–Bonin volcanic-front. Data for Izu Peninsula from Kurasawa and Michino(1976), Kurasawa (1984), Matsumoto et al. (1985), Tamura (1994), and Tamura andTatsumi (2002). Data for Hakone and Fuji volcanoes from Watanabe et al. (2006) andYamashita and Kasama (2008), respectively. Data for rear-arc seamount chains fromHochstaedter et al. (2000), Ishizuka et al. (2002, 2003a, 2006), and Machida and Ishii(2003). Data for Quaternary volcanic front from Hochstaedter et al. (2000), Shukunoet al. (2006), and Taylor and Nesbitt (1998). Rock classification in SiO2–K2O plot is fromLe Maitre (1989) and Rickwood (1989).

Fig. 6. Spatial distribution of ages in the Izu Peninsula and northern Izu–Bonin rear-arcregion. Ar–Ar and K–Ar ages from northern Izu–Bonin rear-arc volcanoes of Ishizukaet al. (2006) are plotted for comparison. Error bar represents 95% confidence intervalfor zircon Pb/U ages and 2σ for Ar–Ar and K–Ar ages.

Fig. 5. 230Th-corrected Tera–Wasserburg concordia plots of SHRIMP-analyzed samples. Error ellipses represent 68.3% confidence levels. Quoted ages areweightedmeans of 207Pb and230Th corrected 206Pb/238U spot ages (n analyses, errors at 95% confidence intervals). Solid ellipses, analyses used in weighted mean age calculation; dotted ellipses, excludedanalyses. Labels on individual ellipses correspond to analysis spot numbers in Table 2. Dashed lines represent linear projections from weighted mean ages to common leadcompositions estimated using the two-stage common-Pb model of Stacey and Kramers (1975).

237K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 14

Author's personal copy

of the Yugashima Group are more altered because they are locatedin areas adjacent to hydrothermal systems associated with thenumerous nearby Quaternary volcanoes.

2) The submarine volcanoes and associated intrusive rocks exposedon the Izu Peninsula were contemporaneous with the LateMiocene to Pleistocene rear-arc volcanoes in the northern Izu–Bonin arc. Moreover, previously reported geochemical data ofvolcanic rocks from the Izu Peninsula are virtually identical tothose of Izu–Bonin rear-arc volcanoes to the south. These resultscollectively suggest the Shirahama Group (and its alteredequivalent, the Yugashima Group) on the Izu Peninsula are partof an accreted block of Izu–Bonin rear-arc volcanoes.

3) Zircon ages of basal units underlying Kozushima and NiijimaQuaternary volcanoes in the northern Izu–Bonin arc are onlyslightly older (b1 Ma) than those from the main body of theseedifices that rest upon them. These underlying units weretherefore erupted during earlier phases of edifice growth ratherthan being parts of an older, unrelated basement.

Our findings show that Izu Peninsula is an uplifted upper crustalassemblage of submarine volcanic edifices formed in the Izu–Boninrear-arc environment. The resulting subaerial exposures provide anunusual opportunity to investigate in detail the growth andarchitecture of a rear-arc region in an active intra-oceanic arc. Animportant concept emerging from the present study is that theexposed LateMiocene to Pleistocene assemblages of lava flows, coevalintrusives, and associated volcaniclastic deposits in the Izu Peninsulaare remnants of numerous overlapping and roughly coeval submarineIzu–Bonin rear-arc volcanoes, rather than a widespread, layercake-like succession of progressively older stratigraphic units.

Acknowledgements

We are indebted to K. Shiraishi and K. Misawa for use of the NIPRSHRIMP. We thank N. Isshiki and T. Nakajima for permitting access tosamples from Kozushima and Zenisu Rocks archived in GSJ/AIST, andT. Ooi and Y.U. Kim for providing the Ugusu and Arari samples. Y.Tamura assisted us during the Izu Peninsula sampling. A.R.L. Nichols,M. Arima, and M. Takahashi provided us with valuable comments andsuggestions. We are grateful to two anonymous reviewers forconstructive reviews and T.M. Harrison for editorial handling.

References

Amano, K., 1991. Multiple collision tectonics of the South Fossa Magna in central Japan.Mod. Geol. 15, 315–329.

Ando, Y., Tsutsumi, S., 2005. K–Ar ages of acid alteration in the western Izu Peninsula. J.Mag. Min. Petrol. Sci. 34, 89–95 (in Japanese with English abstract).

Black, L.P., Kamo, S.L., Allen, C.M., Davis, D.W., Aleinikoff, J.N., Valley, J.W., Mundil, R.,Campbell, I.H., Korsch, R.J., Williams, I.S., Foudoulis, C., 2004. Improved 260Pb/238Umicroprobe geochronology by the monitoring of a trace-element-related matrixeffect; SHRIMP, ID-TIMS, ELA-ICP-MS and oxygen isotope documentation for aseries of zircon standards. Chem. Geol. 205, 115–140.

Bryant, C.J., Arculus, R.J., Eggins, S.M., 2003. The geochemical evolution of the Izu–Boninarc system: a perspective from tephras recovered by deep-sea drilling. Geochem.Geophys. Geosyst. 4, 1094. doi:10.1029/2002GC000427.

Cashman, K.V., Fiske, R.S., 1991. Fallout of pyroclastic debris from submarine volcanic-eruptions. Science 253, 275–280.

Geological Survey of Japan, AIST (Ed.), 2010. Seamless digital geological map of Japan 1:200, 000, Feb 1, 2010 version. Research Information Database DB084. Geol. Surv.Jpn. AIST.

Gill, J.B., Kodaira, S., Tamura, Y., 2009. Rear arc contributions to continental crust. EOSTrans. AGU 90 (52) Fall Meet. Suppl., Abstract T31D-07.

Hamasaki, S., 2000. K–Ar ages of the advanced argillic alteration in the Ugusu area,western Izu Peninsula, Japan. Bull. Geol. Surv. Jpn. 51, 369–372 (in Japanese withEnglish abstract).

Hasebe, N., Fukutani, A., Sudo, M., Tagami, T., 2001. Transition of eruptive style in anarc–arc collision zone: K–Ar dating of Quaternary monogenetic and polygeneticvolcanoes in the Higashi–Izu region, Izu peninsula, Japan. Bull. Volcanol. 63,377–386.

Hirata, D., Yamashita, H., Taniguchi, H., Nishikawa, T., Aoike, K., Suzui, M., Saito, N., Suga,K., Arima, M., 1997. Evolution of a volcanic edifice in the Izu–Ogasawara (Bonin)

arc: lithological characteristics of a 1500 meter core sample in Hachijojima. Bull.Kanagawa Prefect. Mus. Nat. Hist. 26, 57–88 (in English).

Hirata, D., Yamashita, H., Mannen, K., Taniguchi, H., 2001. Lithological characteristics ofthe basement of Hakone Volcano: using of the deep drilling core sample. J. Geogr.110, 420–426 (in Japanese with English abstract).

Hochstaedter, A.G., Gill, J.B., Taylor, B., Ishizuka, O., Yuasa, M., Morita, S., 2000. Across-arc geochemical trends in the Izu–Bonin arc: Constraints on source compositionand mantle melting. J. Geophys. Res. 105, 495–512.

Hochstaedter, A., Gill, J., Peters, R., Broughton, P., Holden, P., Taylor, B., 2001. Across-arcgeochemical trends in the Izu–Bonin arc: contributions from the subducting slab.Geochem. Geophys. Geosyst. 2 (2000GC000105).

Huchon, P., Kitazato, H., 1984. Collision of the Izu block with central Japan during theQuaternary and geological evolution of the Ashigara area. Tectonophysics 110,201–210.

Ibaraki, M., 1976. Notes on planktonic foraminifera from the Harada Formation,Shirahama Group, the Southern Izu Peninsula. Geosci. Rep. Shizuoka Univ. 2, 1–7(in Japanese with English abstract).

Ibaraki, M., 1981. Geologic ages of “Lepidocyclina”, Miogypsina horizons in IzuPeninsula as determined by planktonic foraminifera. J. Geol. Soc. Jpn. 87,417–420 (in Japanese).

Ireland, T.R., Williams, I.S., 2003. Considerations in zircon geochronology by SIMS. In:Hanchar, J.M., Hoskin, P.W.O. (Eds.), Zirco. : Reviews in Mineralogy andGeochemistry, 53. Mineralogical Society of America, Washington D.C., pp. 215–241.

Ishizuka, O., Yuasa, M., Uto, K., 2002. Evidence of porphyry copper-type hydrothermalactivity from a submerged remnant back-arc volcano of the Izu–Bonin arc —implications for the volcanotectonic history of back-arc seamounts. Earth Planet.Sci. Lett. 198, 381–399.

Ishizuka, O., Taylor, R.N., Milton, J.A., Nesbitt, R.W., 2003a. Fluid-mantle interaction inan intra-oceanic arc: constraints from high-precision Pb isotopes. Earth Planet. Sci.Lett. 211, 221–236.

Ishizuka, O., Uto, K., Yuasa, M., 2003b. Volcanic history of the back-arc region of the Izu–Bonin (Ogasawara) arc. In: Larter, R.D., Leat, P.T. (Eds.), Intra-oceanic SubductionSystems: tectonic and Magmatic Processes: Geological Society, London, Spec. Pub.,219, pp. 187–205.

Ishizuka, O., Taylor, R.N., Milton, J.A., Nesbitt, R.W., Yuasa, M., Sakamoto, I., 2006.Variation in the mantle sources of the northern Izu arc with time and space —

constraints from high-precision Pb isotopes. J. Volcanol. Geoth. Res. 156, 266–290.Isobe, I., Nakashima, R., 2001. Molluscan fossil found from pyroclastics of the Niijima

volcano and its basement rocks. Bull. Geol. Surv. Jpn 52, 41–48 (in Japanese withEnglish abstract).

Isshiki, N., 1960. Geology of the Miyakejima district, Quadrangle Series, scale 1:50, 000.Geol. Surv. Jpn. (in Japanese with English abstract).

Isshiki, N., 1980. Geology of the Mikurajima, Inambajima and Zenisu districts, QuadrangleSeries, scale 1:50, 000. Geol. Surv. Jpn. (in Japanese with English abstract).

Isshiki, N., 1982. Geology of the Kozushima district, Quadrangle Series, scale 1:50, 000.Geol. Surv. Jpn. (in Japanese with English abstract).

Isshiki, N., 1984. Geology of the O-shima district, Quadrangle Series, scale 1:50, 000.Geol. Surv. Jpn. (in Japanese with English abstract).

Isshiki, N., 1987. Geology of the Niijima district, Quadrangle Series, scale 1:50, 000. Geol.Surv. Jpn. (in Japanese with English abstract).

Kano, K., 1989. Interactions between andesitic magma and poorly consolidatedsediments: examples in the Neogene Shirahama Group, south Izu, Japan. J.Volcanol. Geoth. Res. 37, 59–75.

Kawate, S., Arima, M., 1998. Petrogenesis of the Tanzawa plutonic complex, central Japan:Exposed felsic middle crust of the Izu–Bonin–Mariana arc. Island Arc 7, 342–358.

Kitamura, K., Ishikawa, M., Arima, M., 2003. Petrogical model of the northern Izu–Bonin–Mariana arc crust: constraints from high-pressure measurements of elastic wavevelocities of the Tanzawa plutonic rocks, central Japan. Tectonophysics 371, 213–221.

Kodaira, S., Sato, T., Takahashi, N., Miura, S., Tamura, Y., Tatsumi, Y., Kaneda, Y., 2007.New seismological constraints on growth of continental crust in the Izu–Boninintra-oceanic arc. Geology 35, 1031–1034.

Koyama, M., 1986. Geologic history of the Izu Peninsula and Quaternary deformationsof the Ashigara and Oiso areas. Earth Mon. 8, 743–752 (in Japanese).

Koyama, M., Umino, S., 1991. Why does the Higashi–Izu monogenetic volcano groupexist in the Izu Peninsula?: relationships between Late Quaternary volcanism andtectonics in the northern tip of the Izu–Bonin arc. J. Phys. Earth 39, 391–420.

Kuno, H., 1959. Geology and petrology of O-shima volcano. Int. Geol. Rev. 1, 48–59.Kurasawa, H., 1984. Strontium isotopic consequence of the volcanic rocks from Fuji,

Hakone, and Izu areas. Bull. Geol. Surv. Jpn 35, 637–659.Kurasawa, H., Michino, K., 1976. Petrology and chemistry of the volcanic rocks from the

western and southern part of Izu Peninsula, Central Japan. Bull. Volcanol. Soc. Jpn.Ser. 2 (21), 11–19 (in Japanese with English abstract).

Le Maitre, R.W. (Ed.), 1989. A Classification of Igneous Rocks and Glossary of Terms.Blackwell, Oxford, p. 193.

Ludwig, K.R., 2001. SQUID: A User's Manual. Berkeley Geochronology Center Spec. Pub. 2.Ludwig, K.R., 2003. User's Manual for Isoplot 3.00: A Geochronological Toolkit for

Microsoft Excel. Berkeley Geochronology Center, Spec. Pub. 4.Machida, S., Ishii, T., 2003. Backarc volcanism along the en echelon seamounts: the Enpo

seamount chain in the northern Izu–Ogasawara arc. Geochem. Geophys. Geosyst. 4,9006. doi:10.1029/2003GC000554.

Matsumoto, R., Katayama, T., Iijima, A., 1985. Geology, igneous activity, andhydrothermal alteration in the Shimoda district, southern part of Izu Peninsula,central Japan. J. Geol. Soc. Jpn 91, 43–63 (in Japanese with English abstract).

Matsumura, M., Fujimoto, K., 2008. Hydrothermal alteration of the basement inside ofthe caldera rim of the Hakone volcano. Bull. Tokyo Gakugei Univ. Nat. Sci. 60,111–119 (in Japanese with English abstract).

238 K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239

Page 15

Author's personal copy

McLennan, S.M., Taylor, S.R., 1982. Geochemical constraints on the growth of thecontinental crust. J. Geol. 90, 347–361.

Nakajima, K., Arima, M., 1998. Melting experiments on hydrous low-K tholeiite:implications for the genesis of tonalitic crust in the Izu–Bonin–Mariana arc. IslandArc 7, 359–373.

Nakajima, J., Hirose, F., Hasegawa, A., 2009. Seismotectonics beneath the Tokyometropolitan area, Japan: effect of slab–slab contact and overlap on seismicity. J.Geophys. Res. 114, B08309.

Okada, H., 1987. Calcareous nannofossil biostratigraphy and paleoenvironmentalanalysis of marine formations exposed in the South Fossa-Magna region. Fossils43, 5–8 (in Japanese).

Parrish, R.R., Noble, N.R., 2003. Zircon U–Th–Pb geochronology by isotope dilution —

thermal ionization mass spectrometry (ID-TIMS). In: Hanchar, J.M., Hoskin, P.W.O.(Eds.), Zircon. : Reviews inMineralogy and Geochemistry, 53. Mineralogical Societyof America, Washington D.C., pp. 182–213.

Rickwood, P.C., 1989. Boundary lines within petrologic diagrams which use oxides ofmajor and minor elements. Lithos 22, 247–263.

Saito, T., 1963. Miocene planktonic Foraminifera from Honshu Japan. Sci. Rep. TohokuUniv. Ser. 2 (35), 123–209.

Saito, S., Arima, M., Nakajima, T., Misawa, K., Kimura, J.I., 2007. Formation of distinctgranitic magma batches by partial melting of hybrid lower crust in the Izu arcCollision Zone, central Japan. J. Petrol. 48, 1761–1791.

Sakamoto, I., Hirata, D., Fujioka, K., 1999. Description of basement rocks from the Izu–Ogasawara arc. Res. Rep. Kanagawa Prefect. Mus. Nat. His. 9, 21–39 (in Japanesewith English abstract).

Sawamura, K., 1955. Geological map of Japan, scale 1:50, 000, Shuzenji. Geol. Surv. Jpn.(in Japanese with English abstract).

Sawamura, K., Sumi, K., Ono, K., Moritani, T., 1970. Geology of the Shimoda district,Quadrangle Series, scale 1:50, 000. Geol. Surv. Jpn. (in Japanesewith English abstract).

Shukuno, H., Tamura, Y., Tani, K., Chang, Q., Suzuki, T., Fiske, R.S., 2006. Origin of silicicmagmas and the compositional gap at Sumisu submarine caldera, Izu–Bonin arc,Japan. J. Volcanol. Geoth. Res. 156, 187–216.

Stacey, J.S., Kramers, J.D., 1975. Approximation of terrestrial lead isotope evolution by atwo-stage model. Earth Planet. Sci. Lett. 26, 207–221.

Taira, A., Tokuyama, H., Soh, W., 1989. Accretion tectonics and evolution of the PacificOcean Margins. In: Ben-Avraham, Z. (Ed.), The Evolution of the Pacific OceanMargins. Oxford University Press, New York, pp. 100–123.

Takahashi, N., Kodaira, S., Tatsumi, Y., Yamashita, M., Sato, T., Kaiho, Y., Miura, S., No, T.,Takizawa, K., Kaneda, Y., 2009. Structural variations of arc crusts and rifted marginsin the southern Izu–Ogasawara arc–back arc system. Geochem. Geophys. Geosyst.10, Q09X08. doi:10.1029/2008GC002146.

Tamura, Y., 1994. Genesis of island arc magmas by mantle-derived bimodalmagmatism: evidence from the Shirahama Group, Japan. J. Petrol. 35, 619–645.

Tamura, Y., Tatsumi, Y., 2002. Remelting of an andesitic crust as a possible origin forrhyolitic magma in oceanic arcs: an example from the Izu–Bonin-arc. J. Petrol. 43,1029–1047.

Tamura, Y., Koyama, M., Fiske, R.S., 1991. Paleomagnetic evidence for hot pyroclasticdebris flow in the shallow submarine Shirahama Group (Upper Miocene–Pliocene),Japan. J. Geophys. Res. 96 (B13), 21779–21787.

Tani, K., Fiske, R.S., Tamura, Y., Kido, Y., Naka, J., Shukuno, H., Takeuchi, R., 2008. Sumisuvolcano, Izu–Bonin arc, Japan: site of a silicic caldera-forming eruption from a smallopen-ocean island. Bull. Volcanol. 70, 547–562.

Tani, K., Dunkley, D.J., Kimura, J.-I., Wysoczanski, R.J., Yamada, K., Tatsumi, Y., 2010.Syncollisional rapid granitic magma formation in an arc–arc collision zone:evidence from the Tanzawa plutonic complex, Japan. Geology 38, 215–218.

Taniguchi, H., 1977. Volcanic geology of Kozu-shima, Japan. Bull. Volcanol. Soc. Jpn. 22,133–147 (in Japanese with English abstract).

Tatsumi, Y., Shukuno, H., Tani, K., Takahashi, N., Kodaira, S., Kogiso, T., 2008. Structureand growth of the Izu–Bonin–Mariana arc crust: 2. Role of crust–mantletransformation and the transparent Moho in arc crust evolution. J. Geophys. Res.113, B02203. doi:10.1029/2007JB005121.

Taylor, R.N., Nesbitt, R.W., 1998. Isotopic characteristics of subduction fluids in an intra-oceanic setting, Izu–Bonin arc, Japan. Earth Planet. Sci. Lett. 164, 79–98.

Watanabe, S., Widom, E., Ui, T., Miyaji, N., Roberts, A.M., 2006. The evolution of achemically zoned magma chamber: the 1707 eruption of Fuji volcano, Japan. J.Volcanol. Geoth. Res. 152, 1–19.

Yamada, E., Sakaguchi, K., 1987. Stratigraphy and geological structure of the Neogeneformations, southeastern part of the Izu Peninsula. Japan. Bull. Geol. Surv. Jpn. 38,357–383 (in Japanese with English abstract).

Yamashita, H., Kasama, T., 2008. Database of whole rock compositions for eruptiveejecta and basement rocks of Hakone volcano. Res. Rep. Kanagawa Prefect. Mus.Nat. His. 13, 211–218 (in Japanese with English abstract).

239K. Tani et al. / Earth and Planetary Science Letters 303 (2011) 225–239