Previous   |    Next

doi:10.14379/iodp.pr.351.2015

Background and setting

Regional setting

The western part of the Philippine Sea plate (PSP), where Site U1438 is located, is complex both topographically and geologically; the following features are highlighted in Figure F2: (1) the Early Cretaceous (131–119 Ma) Huatung Basin (Deschamps et al., 2000); (2) Gagua Ridge; (3) conjugate plateaus of possible hotspot origin Benham Rise and Urdaneta Plateau, now separated by (4) oceanic crust comprising the West Philippine Basin (WPB); (5) Mesozoic remnant arcs (Amami Plateau and Daito and Oki-Daito Ridges); (6) the possible “ocean island” or “hot-spot” related Oki-Daito Rise (Ishizuka et al., 2013); and (7) the Eocene–Oligocene KPR. Basins occur between the ridges in the northwest Philippine Sea, including (1) the Minami-Daito Basin, between the Amami Plateau and Daito Ridge; (2) the Kita-Daito Basin, between the Daito and Oki-Daito Ridges; (3) the ASB; and (4) east of the KPR, the Shikoku Basin.

Generalities

Regional considerations

The PSP, Earth’s eleventh largest lithospheric plate, containing the largest back-arc basin, is surrounded by subduction zones and transform faults and has a complex tectonic and magmatic history. The most comprehensive and recent plate tectonic reconstructions (Deschamps and Lallemand, 2002; Whittaker et al., 2007; Seton et al., 2012) adopt subduction inception at ~50 Ma in the proto-Izu-Bonin arc (KPR), concurrent with a change in Pacific plate motion, cannibalizing former northwest-southeast–trending transform faults associated with the Izanagi-Pacific Ridge (Fig. F3). In essence, this is a revised version of the original hypothesis for subduction initiation specifically proposed for the IBM system by Uyeda and Ben-Avraham (1972), of convergence across a transform fault following plate reorganization. Izanagi-Pacific Ridge subduction was the last stage in the development through the Mesozoic of closure between the Australian plate to the south and Asia to the north, resulting in subduction along the Asian continental margin of a series of formerly linked spreading ridges from the Indian through to the Pacific Ocean (Seton et al., 2012). Whittaker et al. (2007) propose subduction of the Izanagi-Pacific Ridge along much of eastern Asia at ~60 Ma initiated the reorganization of plate motions that culminated in the change in motion of the Pacific plate at 50 Ma relative to the Eurasian plate. Of course, the original extent of the PSP is not known, as much has been subducted along the Nankai Trough and the Ryukyu and Philippine Trenches. But the areal extent of the PSP has also been augmented by creation of new back-arc basin oceanic crust generated through seafloor spreading in the West Philippine, Shikoku, and Parece Vela Basins and the Mariana Trough. An additional aspect of reconstruction is that the PSP has undergone northward migration mostly between 50 and 25 Ma, accompanied by clockwise rotation approaching 90°, since inception (Yamazaki et al., 2010).

At KPR inception, a Cretaceous-age island arc system existed, presently preserved as the Amami Plateau–Daito–Oki Daito Ridges; there may have been an arc conjugate on the other flank of the WPB spreading center, now accreted to Luzon (Deschamps and Lallemand, 2002). Lewis et al. (1982) had previously proposed the WPB is a back-arc basin developed north of the East Mindanao-Samar arc. The relationship of the Izanagi-Pacific Ridge to this arc system is unclear but critical in terms of understanding the nature of the basement of the KPR in the vicinity of the Amami Plateau–Daito–Oki Daito Ridges and the history of Site U1438. Deschamps and Lallemand (2002) invoke northward subduction of oceanic lithosphere north of Australia beneath a late Mesozoic–Cenozoic east-west–striking arc. Back-arc spreading north of this arc formed the initial stages of the WPB, split the Amami Plateau–Daito–Oki Daito Ridges from the arc front in the south, and left these ridges as a remnant arc. Coincidentally and possibly critical in initiation of spreading in the WPB was ocean island basalt (OIB)-like magmatism, preserved as the Benham Rise-Urdaneta Plateau conjugates and the Oki Daito Rise. The source of this compositional type of magmatism may have been the Manus Basin hotspot (Macpherson and Hall, 2001; Ishizuka et al., 2013). Whittaker et al. (2007), on the other hand, proposed development of a southeastward ridge jump as the Izanagi-Pacific Ridge was consumed; the linking transform bypassed the Amami Plateau–Daito Ridge to join the WPB spreading center and subsequently itself became cannibalized to form the proto-IBM arc.

Specifics

Huatung Basin–Gagua Ridge

Amphibole separated from gabbros dredged from the western flank of the Gagua Ridge give ⁴⁰Ar/³⁹Ar ages of 116–121 Ma; in combination with interpretation of magnetic anomalies, Deschamps et al. (2000) conclude the Huatung Basin formed between 131 and 127 Ma and suggest the oceanic crust underlying a sediment cover formed at a Cretaceous spreading ridge or back-arc basin. They further suggest the Gagua Ridge is a transpressional ridge between relatively old oceanic crust of the Huatung Basin and the younger WPB.

Mesozoic remnant arcs (Daito Ridge group) and hotspot-related volcanic rocks (Daito Rise, Urdaneta Plateau, and Benham Rise)

The Daito Ridge group is a complex array of ridges and basins. The group comprises three remnant arcs: the Amami Plateau, the Daito and Oki-Daito Ridges, and two basins neighboring these ridges (the Kita-Daito and Minami-Daito Basins) (Fig. F2). Granites and arc volcanics of Cretaceous age (e.g., Matsuda et al., 1975; Hickey-Vargas, 2005) are exposed on the Amami Plateau, which has crustal thickness up to 19 km (Nishizawa et al., 2014). Geochemical characteristics of the volcanic rocks are consistent with formation of the plateau by Cretaceous-aged subduction zone magmatism (Hickey-Vargas, 2005). These remnant arcs predate the inception of IBM arc magmatism at 52 Ma (Ishizuka et al., 2006a, 2011a).

The Daito Ridge is generally east-west–trending and intersects the KPR at its eastern end (Fig. F2). Low-grade metamorphic rock, sedimentary rock, and some volcanic rock were recovered by dredging, apparently from beneath Eocene sedimentary rock (Mizuno et al., 1975, 1978; Yuasa and Watanabe, 1977), whereas recent shallow drilling recovered fresh volcanic rock from the eastern part of the Daito Ridge. Two of these drilled samples (andesites) with the distinctive trace element characteristics of arc magmas yielded ⁴⁰Ar/³⁹Ar ages of 116.9 and 118.9 Ma (Ishizuka et al., 2011b). Both of these ages are significantly older than the KPR volcanism. These results revealed that the Daito Ridge comprises Mesozoic arc rock overlain by middle Eocene sedimentary rock (e.g., Mizuno et al., 1978).

The Oki-Daito Ridge is WNW–ESE trending and ~600 km long. This ridge is characterized by crust ranging from 20 to 25 km in thickness, based on its seismic velocity structure (Nishizawa et al., 2014). A wide bathymetric high west of the Oki-Daito Ridge is the Oki-Daito Rise, which occupies an area of ~200 km²; dredged rocks have intraplate ocean island geochemical characteristics and were probably formed through the same hotspot magmatism responsible for the Urdaneta Plateau–Benham Rise (Ishizuka et al., 2013). All of these samples have to varying extents Pb isotopic characteristics of the “enriched mantle 2” (EM-2)-type source. The eastern margin of the rise appears to overlap the western part of the Oki-Daito Ridge. The rise is characterized by much thinner crust (10–15 km) compared to the neighboring Oki-Daito Ridge. Ishizuka et al. (2013) suggest the age-progressive volcanism from ~50 to 35 Ma, becoming younger toward the WPB from the Minami-Daito Basin, is consistent with a hotspot origin.

The Kita-Daito Basin separates the Amami Plateau and the Daito Ridge and contains irregularly shaped seamounts and ridges. The Minami-Daito Basin is located between the Daito and Oki-Daito Ridges. This basin has many more bathymetric highs compared to the Kita-Daito Basin, including conical seamounts. Drilling at Deep Sea Drilling Project (DSDP) Site 446 in the western part of the basin recovered basalt sills with ⁴⁰Ar/³⁹Ar plateau ages of 51.3 and 42.8 Ma (Hickey-Vargas, 1998a). These basalts are geochemically varied with tholeiitic and alkaline compositions and clearly have OIB-like geochemical characteristics (Hickey-Vargas, 1998a).

West Philippine Basin

The WPB is an ocean basin occupying the western half of the PSP (Fig. F2). The Ryukyu and Philippine Trenches mostly bound the western margin of the basin. Between these trenches, the Gagua Ridge separates the WPB from the Cretaceous-age Huatung Basin. Prominent bathymetric features in the WPB include the broad highs of the Benham Rise and the Urdaneta Plateau. These features are located approximately equidistant from the so-called Central Basin Fault, which is an extinct spreading center. The Benham Rise was drilled on its southeastern flank at DSDP Site 292. A thick plagioclase-porphyritic basalt layer was recovered from beneath Eocene–early Oligocene sediment (Karig, Ingle, et al., 1975). Hickey-Vargas (1998b) reported ⁴⁰Ar/³⁹Ar ages of 35.6 and 36.2 Ma for this basalt and OIB-like geochemical characteristics. The Urdaneta Plateau has dimensions of about 300 km × 200 km and consists of two bathymetric highs, with seafloor fabrics similar to overlapping spreading centers or dueling propagators (Deschamps et al., 2008) and ⁴⁰Ar/³⁹Ar ages between 35.87 and 39.79 Ma (Ishizuka et al., 2013). These oceanic plateaus within and north of the WPB form age-progressive volcanic chains that are hypothesized to have been produced by a mantle plume that remained fixed at the spreading center for ~10 My (Ishizuka et al., 2013).

The origin of the WPB has long been debated. Hilde and Lee (1984) published magnetic lineation data for this area (Fig. F4A). Based on these data, they propose that spreading from the Central Basin Fault formed the WPB. A spreading direction for the WPB was determined to be northeast–southwest between 60 and 45 Ma at a rate of 44 mm/y. After 45 Ma, the spreading direction changed to a more north–south direction associated with a reconfiguration of the Central Basin spreading center, and the spreading rate decreased to 18 mm/y. The major phase of spreading is inferred to have finished at ~35 Ma. Other suggestions for the origin of the WPB include that of Lewis et al. (1982), who proposes the basin formed as a back-arc basin behind the east Mindanao-Samar arc, and Seno and Maruyama (1984), who suggest the WPB formed behind the KPR. The hypothesis that the WPB is of back-arc origin has been further developed by recent studies (Fig. F4B–F4C) (e.g., Fujioka, et al., 1999; Deschamps and Lallemand, 2002, 2003; Okino and Fujioka, 2003) based on new detailed bathymetric and geomagnetic data mainly acquired around the Central Basin Fault and the northern part of the basin. Combined with the notion that subduction initiation of the IBM arc was contemporaneous with or preceded the opening of the WPB, Hall et al. (1995), Hall (2002), and Deschamps and Lallemand (2002) propose models assuming the WPB opened between the two subduction zones of the East Philippine and IBM arcs, followed by clockwise rotation of the latter. Deschamps and Lallemand (2002) further propose rifting started at 55 Ma and spreading ended at 33–30 Ma. The spreading axis was parallel to the East Philippine arc but became inactive when a new spreading ridge propagated from the eastern part of the basin. Spreading of the basin occurred mainly from this new axis, which rotated counterclockwise during its existence.

An old hypothesis for the origin of the WPB is the so-called “trapped basin” model (Fig. F4D) (e.g., Uyeda and Ben-Avraham, 1972). Le Pichon et al. (1985) propose the extinct spreading center of the WPB was a remnant of the North New Guinea-Kula spreading ridge that was captured at 43 Ma. Jolivet et al. (1989) present a similar but modified model, arguing the WPB is a remnant of the Pacific-North New Guinea spreading ridge captured at 43 Ma by inception of a new subduction zone (i.e., the IBM) along a transform fault. More recent globally constrained plate tectonic reconstructions have established the Izanagi-Pacific spreading ridge was regionally important and do not confirm a role for the putative North New Guinea-Kula Ridge (Whittaker et al., 2007; Seton et al., 2012). Nevertheless, these more recent plate configurations also assume a trapped basin model in that the lithosphere north of the spreading center of the WPB was formerly part of the Izanagi plate.

Kyushu-Palau Ridge

The KPR forms the eastern margin of the WPB. It is now a remnant arc separating the WPB from the Shikoku and Parece Vela back-arc basins (Fig. F2). The KPR was an active arc in the Eocene and Oligocene (e.g., Mizuno et al., 1977; Shibata et al., 1977; Ishizuka et al., 2011b) and became isolated from the volcanic front of the IBM arc at ~25 Ma through the initiation of rifting and seafloor spreading in the Shikoku and Parece Vela Basins (Ishizuka et al., 2011b). The extinct spreading center of the WPB (Central Basin Fault) is truncated by the KPR at ~15°N. Radiometric ages for samples collected from the northern to central KPR range in age between 43 and 25 Ma but are mostly between 27 and 25 Ma, indicating arc volcanism ended on the KPR at about this time (Fig. F5). In other words, back-arc rifting (or spreading) of the Shikoku and Parece Vela Basins initiated at ~25 Ma (Ishizuka et al., 2011b). This interpretation is generally consistent with the estimated timing of rifting and spreading of the Shikoku Basin based on magnetic anomaly data and seafloor fabric observations. Okino et al. (1994) identifies a magnetic lineation corresponding to Anomaly 7 in the westernmost (oldest) margin of the basin and suggests spreading started at 26 Ma.

The lack of systematic age variations of volcanic rock along the KPR indicates that rifting was initiated almost concurrently along the entire ridge between 30°N and 11°N. By extension, this means that initiation of the Shikoku and Parece Vela Basins and isolation of the KPR as a remnant arc occurred at about the same time. Even though the KPR is a remnant part of the IBM arc, the geochemical characteristics of the KPR magmatic lithologies are distinct compared to those of Quaternary age at the IBM arc volcanic front. Whereas the Quaternary IBM arc has clear along-arc geochemical variations (e.g., more enriched isotopic and trace element signatures) (Fig. F6) both toward Honshu and the “alkalic volcano province,” typified by the Iwo-To volcano at the Izu-Bonin/Mariana intersection in the volcanic front and the rear arc (Bloomer et al., 1989; Ishizuka et al., 2003, 2006b, 2007), the KPR does not show analogous systematic along-arc variations (Ishizuka et al., 2011b). This observation suggests the magmatic record obtained at Site U1438 is representative of adjacent along-strike KPR arc magmatism and provides us with representative magmatic evolution of the Paleogene stage of the IBM system.

However, there are locations where distinct geochemical signatures have been obtained along the KPR. One of these is where the Daito Ridge intersects the KPR. High-K andesite of Eocene age occurs in this area. These Eocene volcanic rocks from the KPR/Daito Ridge intersection have a distinctly enriched trace element and isotopic character relative to the surrounding KPR samples. In particular, they have higher ²⁰⁶Pb/²⁰⁴Pb and elevated light versus heavy rare earth element ratios in combination with small deviations of ²⁰⁸Pb/²⁰⁴Pb from the Northern Hemisphere Reference Line (i.e., Δ²⁰⁸Pb/²⁰⁴Pb) relative to the KPR (Fig. F7). Arc magmatism at the KPR/Daito Ridge intersection is thought to have been established on preexisting Daito Ridge crust, which is a Mesozoic remnant arc. Therefore, it is possible the distinct geochemical characteristics of the KPR/Daito Ridge intersection are related to the involvement of sub-Daito Ridge lithospheric mantle or subarc crust, which was metasomatized by a Cretaceous-aged subduction event. This critically important hypothesis can be tested through the geochemical character of the arc basement, part of which was drilled during Expedition 351 at Site U1438. Similarly, such geochemical variations in KPR volcanic products will facilitate determination of the provenance of Site U1438 volcaniclastic sediments.

Amami Sankaku Basin

The initial products of the IBM system are preserved today in two longitudinal belts: (1) one forming the eastern margin of the WPB and abandoned as a remnant arc (the KPR) (Fig. F2) when the Parece Vela-Shikoku Basin opened; and (2) a second belt preserved in the IBM fore arc that is mostly submarine but emerges sporadically as islands, such as Chichijima and Guam. The ASB is bordered to the west across a major north-south–striking fault scarp (Minami-Amami Escarpment [MAE]) by the Amami Plateau, to the south by the Daito Ridge group, and to the northeast by the KPR. It is important to note the KPR is not oriented parallel to either the Amami Plateau-Daito Ridge or the MAE. Instead, the strike of the KPR is at a high angle to the trends of these features, apparently constructed independently of any preexisting basement fabric. If the MAE represents a fossil transform fault adjacent to an ASB basement formed through seafloor spreading, it appears the initiation of the KPR was independent of preexisting transform fault control, inconsistent with the majority of models that have been proposed for subduction inception for this arc system.

The eastern part of the Daito Ridge south of the ASB has a ⁴⁰Ar/³⁹Ar date of 118 Ma (Ishizuka et al., 2011b). Prior to Expedition 351, it appeared possible that the early ASB sediment and basement might be Early Cretaceous or older (i.e., Neotethyan). The Seton et al. (2012) plate reconstruction estimates Pacific plate basement with an age ranging from ~100 Ma in the vicinity of the ASB to ~140 Ma in the southern Marianas.

Reconnaissance of the subseafloor crust along the western margin of the ASB was begun on a Shinkai 6500 dive (337) conducted at the MAE (Fig. F8) in 1996. The dive started at the foot of a 1 km high steep cliff and ascended to the top of the escarpment. Lithologies identified on the dive transect from shallower to deeper parts along the submersible track line were ash turbidite with burrows, altered tuffs, calcareous chalk, scoria, and basalt breccia with calcareous matrixes, all covered with pelagic mud and manganese sediments. Occasional pumice blocks were scattered on the sediment surface. Intact basaltic basement (pillows or sheet flows) was not observed. Sediment samples obtained during this dive were predominantly pelagic brown mud, indicating deposition below the carbonate compensation depth (CCD). However, calcareous chalk is consistent with a shallower depositional environment for the older lithologies. The topography of the escarpment is a combination of gentler sedimented slopes with steep to occasional overhanging cliffs. A notable slump scar, erosional gulley, and slope failure–induced debris flows and turbidites were seen everywhere along the dive track. These phenomena strongly suggest the occurrence of past slope failure in relation to likely fault movement along the MAE.

Seismic studies/site survey data

The Japan Oil, Gas, and Metals National Corporation (JOGMEC) acquired extensive multichannel seismic (MCS) reflection data in the northern part of the PSP (Higuchi et al., 2007). These surveys covered a wide area of the IBM-KPR, as well as the Amami Plateau and Daito Ridge regions; the major objective of the surveys was to obtain detailed images of the sedimentary and deeper crustal structures. These data provide important information for drilling into the sedimentary and igneous sections, particularly in the ASB. The interpretation of these profiles (Fig. F9) coupled with information from DSDP holes (e.g., 296, 445, and 448) and Ocean Drilling Program (ODP) Site 1201, have resulted in the description of two major layers in the ASB, the igneous basement and overlying sedimentary section (Fig. F10). On the basis of comparison of MCS reflection profiles in the northwestern Philippine Sea, including some that traverse the ASB, Higuchi et al. (2007) suggest the oldest sediments in the ASB are late to middle Eocene in age.

According to Higuchi et al. (2007), the sediments of this region of the ASB contain five stratigraphic units. The top unit (~110 m) is estimated to comprise Pliocene–Pleistocene pelagic sediment. The second unit (~160 m) is suggested to be upper Miocene turbidites, which may have come from the KPR but are proposed to be more likely pelagite given the termination of eruptive activity on the KPR by this time. The third unit (~310 mbsf) is suggested to be lower Miocene turbidites derived from the magmatically extinct KPR. The fourth unit (490 m) is estimated to be Oligocene and Eocene volcaniclastic turbidites from the KPR. The thickness of this unit increases eastward toward the KPR with a maximum exceeding 1 km, which is consistent with a predominant source on the ridge. The final sedimentary unit (230 m) is suggested to be pelagic sediment of Eocene or older age; the distribution of this layer is discontinuous across the ASB but present at Site U1438.

Site U1438 was selected at the intersection of two MCS profiles (Lines D98-A and D98-8) obtained by JOGMEC (Figs. F9, F10) located ~50–80 km southwest of the nearest part of the KPR.

It is also useful to examine the seismic and lithologic structures of another drilled site in the WPB because of the shedding of pyroclastic debris and ash from the KPR at Site U1438. One of the objectives of ODP Leg 195 was coring and casing a hole (Site 1201) in the WPB (Figs. F2, F11) for the installation of a broadband seismometer as part of the International Ocean Network of seismometers (Shipboard Scientific Party, 2002). Site 1201 lies ~100 km west of the KPR on putative 49 Ma crust (near Chron 21) formed at the Central Basin spreading center of the WPB. The basement at Site 1201 presumably formed just after IBM arc initiation and so cannot be used to satisfy the specific objectives for Expedition 351. Nevertheless, some aspects of the sedimentological processes at the former location are contextually important. As Site 1201 drifted away from the Central Basin spreading center, volcanism ceased and about 0.5 km of sediment was deposited in three stages: (1) quiescent marine sedimentation in deep water into the late Eocene; (2) pelagic sedimentation mixed with, and finally overwhelmed by, volcaniclastic turbidites from the KPR from the late Eocene through the early Oligocene; and (3) waning turbidite deposition, followed by deep-sea pelagic sedimentation below the CCD from the early Oligocene to the early Pliocene, when sedimentation ceased altogether, as reported by Salisbury et al. (2006).

Subsequent to ODP drilling at Site 1201, a new MCS line (D99-2) (Fig. F11) was run northwest–southeast through the site and across the KPR into the Parece Vela Basin. One feature of this seismic line is the thickening of the upper part of the sedimentary package toward the prominent topographic high of the KPR and the relative constancy in thickness of the lower parts.

Top of page   |    Previous   |    Next