2024 Volume 119 Issue 1 Article ID: 231222
The present study reports the geochemical characteristics and new precise zircon U-Pb age of alkali basalt to trachyte that form the basement of the reef limestones of the Akiyoshi Terrane, Southwest Japan. The rocks of the Yamano Group (part of the Permian Maizuru Terrane) in the Yakawa area, eastern Hiroshima Prefecture are unconformably overlain by the Inakura Formation of the Cretaceous Kanmon Group. The latter contains limestone blocks and is accompanied by underlying volcanic rocks that are extremely vesicle-rich and highly altered. We carried out detailed petrographic studies on one trachytic rock, and three basaltic rock samples from the Yakawa area. Apart from that, we also have studied two basaltic samples from the Akiyoshi-dai area, one from the Atetsu-dai area, one from the Oga-dai area, and one associated with the Hina Limestone. Geochemically, except for one trachyte-phonolite sample, all other samples from Yakawa, Hina Limestone, Oga-dai, and Atetsu-dai areas are alkali basalt, while samples from Akiyoshi-dai are basalt. The U-Pb zircon age of the trachytic sample from the Yakawa area yielded a crystallization age of 335.8 ± 1.6 Ma. The tectonic discrimination diagram indicates that the samples of the Yakawa area and Hina Limestone are of French-Polynesia-type superplume origin, while the rest are of normal hotspot region affinity. Zircon geochemistry also corroborates the fact that their origin was in the hotspot-type oceanic island.
The rocks of oceanic crustal origin that occur on the present-day Earth’s surface are mostly younger than ∼ 200 Ma (Seton et al., 2020). The oceanic crust before that was mostly lost due to plate subduction. Hence, research on understanding the geochemical characterization and origin of oceanic crust >∼ 200 Ma becomes restricted. Only the limited occurrences of ophiolites, i.e., the obducted oceanic plate rocks on the preexisting continental margin, and oceanic crustal rocks present in accretionary complexes become the important key study materials.
The Japanese island arc has been formed close to the plate subduction boundary for at least the last ∼ 500 million years. As a result, ophiolitic rocks and greenstones of accretionary complexes from the Paleozoic to the Cenozoic ages occur in Japan. Their places of formation even in the oceanic crust are also varied. The ophiolites and greenstones found in the Japanese island arc are therefore of particular importance in elucidating the origin and tectonic history of the Phanerozoic oceanic crust.
The Inner Zone of SW Japan (Fig. 1a) preserves Late Paleozoic Yakuno ophiolite in Maizuru Terrane and early Paleozoic Oeyama ophiolite in the Sangun-Renge Belt (Ishiwatari, 1985; Kurokawa, 1985; Suda et al., 2014; Kimura and Hayasaka, 2019). Due to the improvement of microanalytical instruments and techniques helping high precision measurement of U-Pb isotopic ratios for minerals like zircon in recent decades, precise geochronological data of Japanese Islands have been reported from not only zircon-abundant rocks but also zircon-poor ophiolites and associated rocks (Tsujimori et al., 2005; Tsutsumi et al., 2010; Suda et al., 2014; Kimura and Hayasaka, 2019; Kimura et al., 2021; Sawada et al., 2022; Kitano et al., 2023). However, available data on the greenstones in the accretionary complexes are still meager which hinders the understanding of the formation mechanism and geological setting of the protoliths of such greenstones.
The Inner Zone of SW Japan has elongated belts (Fig. 1a) of accretionary complexes in the time range between the Permian and Jurassic periods (Kojima et al., 2016). A lot of greenstones occur in such accretionary complexes e.g., Akiyoshi Terrane, Ultra-Tamba Terrane, and Tamba-Mino-Ashio Terrane (Tatsumi et al., 2000; Ichiyama et al., 2008). However, their ages are mostly estimated from fossils occurring in the overlying sedimentary rocks (e.g., Okimura, 1966; Ishiga, 1982). As a result, in most cases, this only indicates the younger age bounds from the stratigraphic relationship between the greenstone units and fossil-bearing sedimentary units. So far, only two Sm-Nd radiometric age values have been reported (Sano and Tazaki, 1989; Tazaki et al., 1989). The data from the mafic rocks of the Jurassic accretionary complex (Tamba Terrane) have a high error range (13-26%), and data from some of the tholeiitic and alkali basalts of Akiyoshi Terrane even show a 50% error in the age value. Hence, high-precision age determination, along with petrochemical characterization of greenstones, has become essential.
The Permian accretionary complex of Akiyoshi Terrane is known for its characteristically large occurrence of reef limestone associated with the greenstones. These greenstones have a high potential to record the geochemical character of pre-Jurassic oceanic crust leading to the characterization of the magma generation process, timing, and tectonic settings during the formation of these oceanic crusts (e.g., Sano et al., 2000). As far as the age of formation of such oceanic crust is concerned, most of the rocks being metamorphosed greenstone, precise isotopic age is difficult to acquire. The geochemical proxies are also altered by late thermal or hydrothermal alteration, reflecting uncertainty in age determination.
The present study aims to focus on the altered and metamorphosed mafic igneous rocks, i.e., greenstones affiliated to the Akiyoshi Terrane in the Yakawa area in eastern Hiroshima Prefecture, SW Japan. We found a zircon-bearing greenstone from the Yakawa area, even though zircon grains are rather rare in similar greenstones, which is useful to derive the precise zircon U-Pb isotopic ages of crystallization in these greenstones, and the tectonic setting of magma generation. Hence, we analyzed zircon grains from the presently studied greenstones of the Yakawa area for precise U-Pb age and chemical discrimination for the tectonic setting of this oceanic crust. We also investigate the geochemical affinities of greenstones from the Yakawa area, in addition to those from the Akiyoshi, Atetsu, and Oga bodies. All these data help to understand the formation of pre-existing oceanic crust during the Paleozoic and its eventual accretion process at the East Asian continental margin.
The Japanese island arc preserves the rock records of several accretionary complexes of the Paleozoic to Mesozoic Era. They are widely distributed on the present-day surface and are accompanied by large amounts of oceanic crust-origin greenstones (e.g., Tatsumi et al., 2000). The Akiyoshi Terrane is one of the Permian accretionary complexes exposed as a part of the Inner Zone of Southwest Japan, mainly distributed from Chugoku district to northern Kyushu with a small distribution in the Itoigawa area (Fig. 1a). The Akiyoshi Terrane is characterized by large limestone plateaus such as the Akiyoshi-dai in Yamaguchi Prefecture (Fig. 1b), the Hirao-dai in Fukuoka Prefecture, the Taishaku-dai in Hiroshima Prefecture, the Atetsu-dai and Oga-dai in Okayama Prefecture (Fig. 1c), and the Omi Limestone in the Itoigawa area (Fig. 1a). The Permian strata of Akiyoshi Terrane are characterized by Carboniferous to Permian reef limestones with underlying greenstones of the possible contemporaneous age (Kanmera et al., 1990). Apart from the reef limestones and underlying greenstones, there are some reported units of chert, mudstone, and sandstone (Sano and Kanmera, 1988; Kojima et al., 2016).
Origin of most of the greenstones underlying the Akiyoshi Limestone Group is considered to be of oceanic islands and seamounts derived from intraplate igneous activities, based on their geochemical compositions and occurrence, and the lithology of associated reef limestone (Sano and Kanmera, 1988). The bulk-rock geochemical and Nd isotopic compositions of the alkali basalts at Taishaku-dai suggest that they are of superplume origin (Sano et al., 2000; Tatsumi et al., 2000). The Sm-Nd radiometric age of greenstones in the Akiyoshi Terrane has been reported to be ∼ 370 Ma (Tazaki et al., 1989), which has a very large error range and can only be treated as a reference value. The ages of fossil content in reef limestones imply their growth between the Early Carboniferous to Middle Permian Periods (Sada, 1965; Okimura, 1966; Hase et al., 1974; Nakazawa, 1997; Kojima et al., 2016). Most of the ages from several kinds of fossils (conodont, fusulinacean, coral, brachiopod, and bryozoan) present in the reef limestones of the Akiyoshi Terrane can be traced back to the Visean age (Kojima et al., 2016). Therefore, the underlying greenstones were considered to be the same or older than that. Furthermore, fossil ages of the Tournaisian age have also been reported from Akiyoshi-dai and Hina Limestone (Yanagida, 1973; Kobayashi and Hamada, 1978; Haikawa, 1986, 1988; Niko, 2006, 2007). Hence, the exact age of the basement greenstones is yet to be specified.
Oga-dai, that is distributed in the western Okayama Prefecture, is composed of the Nakamura Limestone, the Koyama Limestone, and associated with smaller limestone blocks, e.g., Hina Limestone (Fig. 1c). The Permian Yamano Group of the Maizuru Terrane is widely exposed to the south of the Oga-dai (Fig. 1c). The Yamano Group is unconformably covered by the Late Cretaceous ignimbrites at its northern margin, and these boundaries are characterized by the isolated limestone masses of the Akiyoshi Terrane origin, which are collectively referred to as the Joe-Yakawa limestones (Fig. 1c).
The present study area is located in the Yakawa area, eastern Hiroshima Prefecture, where the rocks of the Yamano Group is unconformably overlain by the Inakura Formation of the Cretaceous Kanmon Group, which is in turn overlain by the Late Cretaceous rhyolitic lava and ignimbrite (Fig. 2). The Inakura Formation consists of limestone conglomerate to limestone clast-bearing mudstone and sandstone, including massive limestone blocks over several tens of meters in length (Figs. 3a-3d). These limestone blocks are accompanied by underlying volcanic rocks that are extremely vesicle-rich and highly altered (Hayasaka et al., 2000). Such volcanic rocks have traditionally been considered as alkali basalt on the top of the Akiyoshi seamounts, however, Hayasaka et al. (2000) referred to it as scoria-tuff associated with the Cretaceous Inakura Formation. Visean to Middle Permian fossil ages were reported from the Joe-Yakawa limestones and the Kawai Limestone (Fig. 1c) (Hase, 1963; Ehiro et al., 2012). Furthermore, Visean fossil ages were also reported from the lowest strata of the Hina Limestone, Koyama Limestone, and Shimodani Limestone around the Oga area (Hase and Yokoyama, 1975; Fujimoto et al., 1995; Ishida et al., 2013).
Samples were collected from four studied locations in the Yakawa area. Samples IWY-Q2 (GPS coordinates: 34°39′23.6′′N, 133°21′09.1′′E), IWY-BS1 (34°39′25.0′′N, 133°21′06.9′′E), IWY-BS2A and 2B (34°39′23.3′′N, 133°21′09.4′′E), and IWY-391 (34°39′17.6′′N, 133°21′21.2′′E) were taken from a thin patch of greenstone occurring between a massive limestone and Inakura Formation (Fig. 2). Out of these five samples, IWY-BS2A and 2B were taken just at the basal part of the limestone. The boundary is sharp (Fig. 3e) with comparatively fresh limestone on top and a weathered reddish basaltic unit at the bottom. Whereas, the IWY-Q2 sample was collected from nearly 5 m below this sharp contact. Sample IWY-391 was collected from the southeastern part of the studied area as shown in Figure 2. Additionally, two samples from Akiyoshi-dai (AKY-BS2A and B; 34°16′45.3′′N, 131°20′29.3′′E), one sample from Atetsu-dai (ATT-BS1; 34°56′24.0′′N, 133°30′40.0′′E), one sample from Oga-dai (OGD-BS1; 34°43′57.1′′N, 133°25′52.6′′E), and one sample from Hina Limestone body (YSI-BS1; 34°41′56.6′′N, 133°23′33.9′′E) were collected for geochemical analysis.
The studied trachytic rock (IWY-Q2) contains phenocrysts of plagioclase in a devitrified glassy matrix (Fig. 4a). Plagioclase grains are elongate (500 µm to 1 mm) showing lamellar twinning. In places, small-grained plagioclase is found in the matrix. Circular vesicles are filled up by secondary chlorite. Part of the matrix is replaced by chlorite and glauconite. Calcite veins are common. Minor minerals include zircon and apatite (Fig. 4a). A few specs of deep brown biotite with included quartz and iron oxide along its cleavage are also observed in the thin section. Overall texture can be described as hyaloophitic. Inclusions in zircon grains are mostly fluor-apatite, and chloritized biotite. Some altered melt-like inclusions are also present.
In the Yakawa area, other associated volcanic rocks include nearly aphyric basaltic rock (IWY-BS1, IWY-BS2A, IWY-BS2B; Figs. 4b-4d) with near absence of any phenocrysts (at places minor altered plagioclase are present, though). The devitrified glassy matrix contains numerous vesicles which are filled up by chlorite, and calcite. The matrix is replaced by fine-grained clay and calcite.
On the other hand, rocks of the nearby Akiyoshi-dai (AKY-BS2A and B) show euhedral olivine phenocrysts (which are later skeletally replaced by chlorite and calcite) (Fig. 4e). Some plagioclase phenocrysts are also observed. The groundmass is made up of plagioclase microlites and devitrified glassy materials in the overall hyaloophitic texture of the rock. Vesicles in these rocks are filled up by calcite and chlorite.
The rock of Atetsu-dai (ATT-BS1) contains fine-grained plagioclase (<200 µm), chlorite, and opaque minerals showing an intersertal texture, and aphyric (Fig. 4f). The rocks of Oga-dai (OGD-BS1; Fig. 4g) and basalt associated with Hina Limestone (YSI-BS1; Fig. 4h) are nearly similar in mineralogy and texture as that of the above-mentioned Atetsu-dai rock. However, a few olivine skeletal crystals and vesicles are also observed in the two latter samples.
The collected samples were cleaned and dried using an ultrasonic cleaner. The dried samples were then pulverized using a hydraulic press and vibration mill made of agate mortar and used for whole-rock chemical analyses and zircon dating.
Mineral chemistry was measured using an electron probe microanalyzer (EPMA; JEOL JXA 8200 Superprobe) installed at the Natural Science Center for Basic Research and Development (N-BARD), Hiroshima University. The operation condition includes 15 kV of accelerating voltage, 12-20 nA beam current, and 4-5 µm beam diameter. Synthetic and natural standards supplied by JEOL Datum were used during the analysis.
Whole-rock analyses were conducted using fused glass beads and X-ray fluorescence spectrometry (XRF) equipped with an Rh-W dual anode tube (Rigaku ZSX-101e) at the Department of Earth and Planetary Systems Science (DEPSS), Hiroshima University, Japan. The analytical precision is 0.001-0.14 wt% for major elements and 0.68-2.39 ppm for trace elements (Kanazawa et al., 2001).
Precise trace element concentrations for only one sample, IWY-Q2, were measured using whole rock powder fused by 1:1 of Spectroflux100B. Trace elements were measured using fused glass beads and a 213 nm Nd-YAG Laser (New Wave Research UP-213) with an inductively coupled plasma ionization mass spectrometer (ICP-MS: Agilent 7500) at the DEPSS. NIST SRM 610 was used as the standard. Three sample spots and four standard spots were measured. Sr and Zr concentrations from XRF analysis were used as internal standards.
Zircon grains were separated from rock powder by the panning method using water. The internal structure of zircon grains was observed using CL images from the scanning electron microscope (SEM) at the DEPSS to determine the analysis spots. Zircon FC1 standard (207Pb/206Pb age of 1099.0 ± 0.6 Ma; Paces and Miller, 1993) was used for correction of the U-Pb ratio, and glass standard NIST SRM 610 was used for correction of Th/U ratio. Zircon U-Pb isotope analysis was performed using a 213 nm Nd-YAG Laser (New Wave Research UP-213) attached with an ICP-MS (LA-ICP-MS) installed at the DEPSS. The laser spot diameter of 25 µm and the repetition of 4 Hz were chosen during analysis. The other details of the method are described in Katsube et al. (2012).
Raw data were processed using the data reduction program Pepi-AGE (Dunkl et al., 2008), and final statistical plotting using Isoplot/Ex (Version 3; Ludwig, 2003). The isotopic ratio and age values are quoted at 2σ. In this study, concordant data were treated as those intersecting the concordia curve and having a calculated discordance of less than 10%.
Before the analysis of the trace elements in zircon using LA-ICP-MS, Hf concentration was measured to normalize the trace element concentrations using an EPMA (JEOL JXA 8200 Superprobe) at the N-BARD, Hiroshima University. Sc metal standard provided by JEOL Datum is used for the analysis of Sc, while a zircon standard of JEOL was used for the analysis of Hf (HfO2 = 1.876%). The operating conditions of EPMA for measurements of zircon were 15 kV accelerating voltage, 200 nA beam current, and 6 µm beam diameter. Following that, the LA-ICP-MS analysis was carried out for precise evaluation of trace elements in the same zircon spots with 15 µm of laser ablation beam diameter.
Mineral chemistry was analyzed for the trachytic sample of IWY-Q2. Plagioclase feldspar phenocrysts are albitic with an An-content of 1-3 mol%. Lamellar twinned plagioclase composition is albitic, possibly due to the later albitization of the sample. Biotite flakes present in a minor amount (with intergrown quartz and Fe-oxide) are anomalous in chemistry. They are Ti-rich (TiO2 = 7.3-8.9 wt%) with XMg ranges between 0.46-0.56. All the mineral chemical data of sample IWY-Q2 are presented in the Supplementary Tables S1 and S2 (Supplementary Tables S1-S3 are available online from https://doi.org/10.2465/jmps.231222).
Whole-rock CompositionA total of nine samples including four samples from the Yakawa area were analyzed (Table 1). The rocks of the Yakawa area contain a total SiO2 of 26.9-56.4 wt%. The lower values of SiO2 content are from aphyric basaltic host rock (same mineralogy and texture as others) having calcite veins and calcite-filled vesicles. LOI (estimated from the difference between 100% and the total of major elements) ranges from 6.4 to 22.0 wt%. The other rocks of Akiyoshi-dai (2 samples) and one each of three other samples (Oga-dai, Atetsu-dai, and basalt associated with Hina Limestone) have SiO2 contents ranging between 42.6-47.6 wt%. LOI ranges between 1 and 7 wt%. Global MORB-normalized trace element distribution pattern shows general LILE enrichment. The Yakawa sample shows negative anomalies of Sr, K, and Ba (Fig. 5a). Other samples mostly show negative Sr anomalies (except IWY-Q2). IWY-Q2 sample is unique with negative anomalies at P, Ti, V, and Sc and positive Zr anomaly with respect to the other samples of the Yakawa area (Fig. 5a). The chondrite-normalized REE diagram for this sample shows a LREE-enriched pattern with an overall 10-350 times enrichment for all elements (Fig. 5b). Negative Eu anomaly with a value of Eu/Eu* = 0.56 is prominent.
| Sample name | IWY-Q2* | IWY-BS2A | IWY-BS2B | IWY-391 | AKY-BS2A | AKY-BS2B | ATT-BS1 | OGD-BS1 | YSI-BS1 |
| Rock type | Phonolite | Alkali basalt | Alkali basalt | Alkali basalt | Basalt | Basalt | Alkali basalt | Alkali basalt | Alkali basalt |
| SiO2 (wt%) | 56.41 | 37.47 | 41.61 | 26.86 | 46.63 | 46.47 | 47.64 | 45.66 | 42.61 |
| TiO2 | 0.39 | 2.53 | 2.77 | 1.63 | 2.30 | 2.22 | 4.22 | 2.55 | 2.14 |
| Al2O3 | 15.33 | 11.92 | 11.60 | 7.79 | 14.93 | 14.75 | 14.19 | 15.48 | 14.53 |
| Fe2O3 | 5.58 | 8.47 | 8.29 | 6.74 | 13.48 | 13.13 | 15.51 | 12.69 | 13.04 |
| MnO | 0.10 | 0.11 | 0.16 | 0.23 | 0.14 | 0.14 | 0.22 | 0.09 | 0.11 |
| MgO | 4.39 | 3.29 | 8.60 | 3.29 | 6.36 | 5.84 | 4.77 | 5.52 | 7.81 |
| CaO | 4.39 | 17.52 | 11.49 | 30.47 | 11.08 | 11.59 | 5.08 | 4.40 | 7.99 |
| Na2O | 6.71 | 0.02 | 0.75 | 0.02 | 3.45 | 3.39 | 4.85 | 1.23 | 3.51 |
| K2O | 0.22 | 1.15 | 0.54 | 0.51 | 0.49 | 0.56 | 0.29 | 6.01 | 0.74 |
| P2O5 | 0.13 | 0.79 | 0.76 | 0.45 | 0.11 | 0.10 | 0.75 | 0.53 | 0.57 |
| Total | 93.65 | 83.26 | 86.56 | 77.97 | 98.96 | 98.17 | 97.51 | 94.15 | 93.03 |
| Estimated LOI | 6.35 | 16.74 | 13.44 | 22.03 | 1.04 | 1.83 | 2.49 | 5.85 | 6.97 |
| Sc (ppm) | 1.2 | 25.7 | 25.0 | 20.6 | 39.5 | 38.5 | 31.5 | 33.9 | 24.8 |
| V | 39.2 | 216.5 | 211.6 | 103.6 | 353.5 | 349.2 | 338.8 | 339.3 | 219.1 |
| Cr | 10.8 | 325.7 | 283.8 | 281.0 | 415.4 | 496.1 | 5.1 | 822.4 | 403.2 |
| Co | 11.0 | 26.7 | 49.5 | 37.8 | 57.7 | 60.9 | 38.6 | 44.6 | 41.2 |
| Ni | 35.2 | 161.9 | 246.7 | 164.9 | 205.6 | 265.4 | 8.3 | 273.1 | 192.2 |
| Cu | 2.9 | 8.3 | 50.8 | 35.7 | 73.0 | 27.8 | 17.8 | 36.7 | 56.7 |
| Zn | 59.9 | 61.0 | 100.7 | 64.0 | 88.4 | 86.9 | 177.8 | 78.8 | 74.5 |
| Ga | 21.5 | 9.8 | 15.0 | 10.1 | 17.5 | 17.2 | 24.7 | 19.1 | 18.2 |
| Rb | 10.4 | 40.0 | 17.4 | 21.7 | 10.0 | 11.2 | 7.1 | 97.8 | 11.6 |
| Sr | 217.5 | 106.7 | 166.7 | 213.4 | 368.6 | 382.0 | 295.9 | 377.1 | 487.4 |
| Y | 27.4 | 30.3 | 29.0 | 19.2 | 22.1 | 22.3 | 52.7 | 31.0 | 26.2 |
| Zr | 779.1 | 333.2 | 379.3 | 207.1 | 87.7 | 84.8 | 315.2 | 243.5 | 254.3 |
| Nb | 102.3 | 69.0 | 73.7 | 42.6 | 6.8 | 6.7 | 44.9 | 38.3 | 42.0 |
| Cs | 9.6 | 12.6 | 10.3 | 5.1 | 0.1 | 0.7 | <LLD | 13.8 | 13.2 |
| Ba | 182.8 | 132.6 | 154.7 | 88.5 | 76.5 | 76.4 | 159.2 | 989.4 | 380.9 |
| La | 84.8 | 79.3 | 82.4 | 49.6 | 10.8 | 8.5 | 45.4 | 36.6 | 36.9 |
| Ce | 144.5 | 136.6 | 144.5 | 89.1 | 18.3 | 19.6 | 95.5 | 73.9 | 75.4 |
| Pr | 15.1 | ||||||||
| Nd | 54.5 | 50.9 | 57.7 | 33.0 | 14.1 | 13.4 | 50.0 | 35.8 | 33.2 |
| Sm | 9.3 | ||||||||
| Eu | 1.4 | ||||||||
| Gd | 6.6 | ||||||||
| Tb | 0.9 | ||||||||
| Dy | 5.4 | ||||||||
| Ho | 1.0 | ||||||||
| Er | 2.9 | ||||||||
| Tm | 0.4 | ||||||||
| Yb | 2.8 | 2.6 | 1.2 | 1.8 | 2.6 | 1.5 | 4.0 | 2.2 | 1.7 |
| Lu | 0.4 | ||||||||
| Hf | 17.3 | 3.7 | 8.2 | <LLD | 0.1 | <LLD | 5.2 | 4.4 | 6.3 |
| Pb | 3.9 | 7.6 | 11.4 | 6.2 | 1.4 | 1.0 | 2.8 | 2.1 | 3.6 |
| Th | 31.8 | 14.0 | 14.3 | 10.1 | 1.6 | 3.0 | 4.9 | 5.8 | 6.3 |
| U | 2.2 | 2.4 | 2.1 | 0.8 | <LLD | 0.8 | 0.2 | 3.9 | 0.3 |
*IWY-Q2 sample is separately analyzed for REE and some trace elements using LA-ICP-MS.
LLD, lower limit of detection
Relatively less mobile element-based classification diagram e.g., Nb/Y versus Zr/Ti diagram (Pearce, 1996) shows that the IWY-Q2 plot between trachyte to phonolite fields, samples from Akiyoshi-dai plot in basalt field, and other studied samples plot in the alkali basalt field, respectively (Fig. 5c). The tectonic discrimination diagram using Ti-Zr-Y values (Pearce and Cann, 1973) indicates a within-plate basalt affinity for all the studied samples with a variable spread in the field (Fig. 5d). While the other discrimination diagram using Nb-Zr-Y (Meschede, 1986) shows a similar within-plate alkali basalt (WPA) affinity (Fig. 5e). However, samples of Oga-dai and Atetsu-dai plot in within-plate basalt (WPB) field, while that of Akiyoshi-dai plot in the within-plate tholeiite (WPT) and volcanic-arc basalt (VAB) field (Fig. 5e). In Nb/Zr versus Nb/Y diagram (Tatsumi et al., 1998) used for discriminating the superplume-related basalts, the samples of Yakawa area and basalt associated with Hina Limestone plot in French-Polynesia-type superplume field, and other samples plot in the normal hotspot region (Fig. 5f).
Zircon U-Pb ageIn the trachytic sample of IWY-Q2, euhedral zircon grains are mostly 100-300 µm in size along the longer axis. These grains display clear oscillatory zoning in the SEM-CL images (Fig. 6a). In total, 49 spots were analyzed from 13 grains. Out of these, 45 concordant 206Pb/238U ages range from ∼ 321 to 347 Ma (Fig. 6b) and yielded a weighted average age of 335.8 ± 1.6 Ma (Fig. 6c, 2σ, MSWD = 1.3). Th/U ratio ranges from 0.55 to 1.33 except for one data of 2.04 (Supplementary Table S3).
Trace element concentrations of 10 points from 6 zircon grains from the sample IWY-Q2 are also measured using LA-ICP-MS (Table 2). Figure 7a shows the C1 chondrite-normalized REE diagram. IWY-Q2 showed an REE pattern with HREE enrichment, strong positive Ce anomaly (Ce/Ce* = 32.4-97.8), and negative Eu anomaly (Eu/Eu* = 0.07-0.24) (Fig. 7a and Table 2). One spot from the bright oscillatory center showed abundant REE content (Ce/Ce* = 3.6; Eu/Eu* = 0.32). In the Hf versus U/Yb diagram and Y versus U/Yb diagram (Grimes et al., 2007), IWY-Q2 zircon data were plotted in the field of continental zircon region (Figs. 7b and 7c). Furthermore, IWY-Q2 zircon data were mainly plotted in the field of oceanic island zircon chemistry in the Nb/Yb versus Sc/Yb, Nb/Yb versus U/Yb, and Sc/Yb versus U/Yb diagrams (Grimes et al., 2015) (Figs. 7d-7f).
| 003IWY-01 | 004IWY-02 | 005IWY-04 | 006IWY-05 | 007IWY-06 | 010IWY-7 | 011IWY-8 | 012IWY-9 | 013IWY-10 | 014IWY-12 | |
| EPMA (mass%) | ||||||||||
| SiO2 | 32.85 | 32.68 | 32.61 | 32.90 | 33.02 | 32.74 | 31.92 | 32.80 | 33.14 | 32.84 |
| ZrO2 | 66.83 | 66.34 | 66.61 | 67.26 | 66.83 | 65.69 | 63.85 | 66.63 | 67.36 | 67.28 |
| Al2O3 | <LLD | <LLD | <LLD | <LLD | 0.0016 | <LLD | 0.0064 | <LLD | 0.0018 | <LLD |
| P2O5 | 0.0435 | 0.0186 | 0.1039 | 0.0193 | 0.0188 | 0.0232 | 0.0605 | 0.0323 | 0.0218 | 0.0178 |
| CaO | <LLD | <LLD | 0.0010 | <LLD | <LLD | 0.0069 | 0.0633 | 0.0011 | 0.0523 | 0.0058 |
| Sc2O3 | 0.0023 | 0.0031 | 0.0113 | 0.0011 | 0.0024 | 0.0044 | 0.0029 | 0.0028 | <LLD | <LLD |
| HfO2 | 0.9194 | 1.0180 | 1.1114 | 0.9405 | 0.9208 | 1.0169 | 0.8110 | 0.8986 | 0.9458 | 0.9572 |
| Total | 100.65 | 100.06 | 100.45 | 101.11 | 100.79 | 99.48 | 96.71 | 100.37 | 101.53 | 101.10 |
| EPMA (ppm) | ||||||||||
| Sc | 15.00 | 20.21 | 73.67 | 7.17 | 15.65 | 28.69 | 18.91 | 18.25 | <LLD | <LLD |
| Hf | 7796.3 | 8632.4 | 9424.4 | 7975.2 | 7808.2 | 8623.1 | 6877.1 | 7619.9 | 8020.2 | 8116.9 |
| LA-ICP-MS (ppm) | ||||||||||
| Y | >ULD | 620.93 | >ULD | 573.17 | 642.59 | 610.90 | >ULD | 1244.95 | 514.29 | 482.50 |
| Nb | 25.80 | 4.72 | 26.55 | 2.68 | 3.00 | 4.90 | 456.73 | 8.50 | 2.49 | 2.02 |
| La | 0.08 | 0.01 | 0.06 | 0.01 | 0.02 | 0.04 | 22.87 | 0.04 | <LLD | 0.00 |
| Ce | 33.05 | 10.12 | 60.96 | 7.40 | 6.96 | 11.90 | 493.28 | 15.39 | 7.02 | 6.36 |
| Pr | 0.45 | 0.15 | 0.52 | 0.15 | 0.14 | 0.10 | 50.54 | 0.30 | 0.13 | 0.07 |
| Nd | 6.44 | 3.27 | 8.89 | 2.03 | 2.30 | 1.19 | 469.83 | 4.66 | 2.16 | 1.54 |
| Sm | 10.40 | 4.76 | 13.76 | 3.47 | 3.91 | 4.49 | 357.54 | 6.96 | 3.99 | 2.65 |
| Eu | 0.60 | 0.20 | 2.31 | 0.35 | 0.44 | 0.22 | 45.49 | 0.61 | 0.36 | 0.27 |
| Gd | 49.94 | 15.81 | 62.85 | 14.18 | 15.49 | 15.67 | 519.37 | 28.47 | 12.93 | 11.93 |
| Tb | 19.83 | 5.56 | 20.69 | 4.86 | 5.20 | 5.42 | 120.19 | 10.23 | 4.27 | 4.08 |
| Dy | 240.58 | 62.17 | 240.15 | 52.59 | 58.66 | 60.77 | 980.17 | 123.44 | 53.30 | 49.48 |
| Ho | 91.18 | 22.69 | 82.23 | 17.93 | 22.47 | 23.13 | 269.16 | 42.87 | 17.05 | 16.44 |
| Er | 355.09 | 87.19 | 324.55 | 90.17 | 97.68 | 85.71 | 907.94 | 175.26 | 79.88 | 68.90 |
| Tm | 76.89 | 19.88 | 67.18 | 19.20 | 20.24 | 20.59 | 170.93 | 36.57 | 16.63 | 15.61 |
| Yb | 767.76 | 207.16 | 657.41 | 177.67 | 208.49 | 180.95 | 1508.67 | 362.99 | 171.68 | 172.22 |
| Lu | 97.82 | 28.60 | 90.02 | 26.86 | 36.18 | 27.16 | 200.08 | 58.11 | 26.14 | 24.57 |
| Pb | 118.04 | 35.40 | 204.50 | 20.47 | 17.20 | 39.14 | 199.84 | 68.67 | 19.89 | 16.60 |
| Th | 630.24 | 147.72 | 1879.83 | 91.48 | 59.61 | 150.54 | >ULD | 558.79 | 79.94 | 64.09 |
| U | 569.05 | 184.45 | 1054.01 | 108.81 | 91.95 | 227.51 | 1319.00 | 345.97 | 105.56 | 79.00 |
| Ce/Ce* | 41.62 | 89.12 | 83.85 | 39.90 | 32.42 | 48.17 | 3.56 | 34.46 | - | 97.79 |
| Eu/Eu* | 0.08 | 0.07 | 0.24 | 0.15 | 0.17 | 0.08 | 0.32 | 0.13 | 0.15 | 0.15 |
CeCN/CeCN* = CeCN/(LaCNPrCN)1/2
EuCN/EuCN* = EuCN/(SmCNGdCN)1/2
LLD, lower limit of detection
ULD, upper limit of detection
-, uncorrected data due to less than LLD
In previous studies, Visean fusulinacean, conodont, brachiopod, and coral fossils were reported from the lowest strata of the limestones that occurred in the Akiyoshi Terrane (Okimura, 1966; Igo, 1994; Nakazawa, 2001; Tazawa, 2018). Moreover, Tazaki et al. (1989) reported an Sm-Nd isotopic age data of 366 ± 185 Ma having a high error value. On the other hand, zircon grains from the trachytic rock sample of IWY-Q2 yielded a 206Pb/238U age of 335.8 ± 1.6 Ma (Fig. 6) in the present study. This is the first report of the precise zircon U-Pb age from the greenstone of Akiyoshi Terrane. These zircon grains displayed euhedral shape, oscillatory zoning, and high Th/U ratio. Therefore, this Carboniferous age is considered as the age of crystallization of this trachytic magma (Hoskin and Schaltegger, 2003). Furthermore, this volcanism can be the latest magmatic activity of the paleo-Yakawa seamount from the viewpoint of the stratigraphic position of studied rock (IWY-Q2; Fig. 3e) and highly differentiated whole-rock chemistry (Fig. 5).
Origin of the Yakawa seamountAll samples collected in the present study show the geochemical affinity to intraplate igneous activity based on discrimination diagrams using the less mobile elements (Fig. 5). The samples from the Yakawa area were within-plate alkali basalts (WPA). However, the sample IWY-Q2 which contained zircon, in particular, showed a significantly more differentiated composition than the other studied samples of the Yakawa area and also that of the normal WPA. The zircon trace element composition also shows a similar differentiated continental chemistry in terms of U/Yb ratios versus both Hf and Y contents (Figs. 7b and 7c). Moreover, the chemical characteristics of these zircon grains are similar to that of the ocean-island type tectonic setting (Figs. 7d-7f). These characters possibly suggest a hot spot origin for the rocks of the Yakawa area. The Nb/Zr versus Nb/Y diagram also suggests a superplume origin for the studied rocks of the Yakawa area (Fig. 5f). Tatsumi et al. (2000) suggested that some of the greenstones of Akiyoshi-dai and Taishaku-dai were of superplume origin. Sano et al. (2000) reported that the alkali basalt of Taishaku-dai originated from a superplume with EM-1 composition based on Nd isotopic data. The reef limestone-seamount top blocks of the Yakawa area are located to the southeast of the Taishaku-dai, Atetsu-dai, and Oga-dai limestone plateaus, suggesting that it was a product of a superplume origin similar to the alkali basalt of the Taishaku-dai. In other words, both the zircon chemistry and bulk-rock geochemical data of studied rocks of the Yakawa area strongly corroborate the earlier claims (Tatsumi et al., 2000; Sano et al., 2000) of superplume origin of these greenstones.
Comparison with the greenstones of the Inner Zone, SW JapanSm-Nd isotopic ages of 280-340 Ma had been reported from greenstones of the Tamba-Mino-Ashio Terrane, a Jurassic accretionary complex in the Inner Zone of Southwest Japan. The crystallization age of 335.8 ± 1.6 Ma for greenstone obtained in the present study is concordant with the oldest ages among them. In addition, Carboniferous coral fossils and fusulinids have been reported from limestones associated with greenstones of the Kozuki Formation of the Ultra-Tamba Terrane of the Permian accretionary complex (Igi, 1969; Goto and Yamagiwa, 1973) indicating its origin as an oceanic island. Many greenstones that occur in the accretionary complexes of the Inner Zone of Southwest Japan are of varying formation ages within Carboniferous Period. Moreover, they also spatially spread over different parts of the ocean plate. This suggests the possibility of plume activity on a spatially wide area in the erstwhile oceanic plate. More data on the U-Pb ages of these greenstones may help to elucidate the history and extent of mantle plume activity during the Carboniferous Period.
We thank Yasuhiro Shibata for his help with EPMA analyses, and Prof. Yoshio Takahashi for allowing us to use the ICP-MS facility during the zircon U-Pb dating and zircon trace element measurement. This study was partly supported by Grants-in-Aid for Scientific Research JP25400486 to Y. Hayasaka from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan. Critical comments of Dr. Yuji Ichiyama and an anonymous reviewer helped to revise the manuscript. Also, Dr. Kenta Kawaguchi helped us with several rounds of discussion. Handling editor, Dr. Yoshimitsu Suda provided us with many of his comments which also helped during revision.
Supplementary Tables S1-S3 are available online from https://doi.org/10.2465/jmps.231222.
