2025 Volume 120 Issue 1 Article ID: 250214
To constrain the timing of 24-22 Ma silicic volcanism prior to the opening of the Sea of Japan in 21-15 Ma, zircon U-Pb dating was conducted on rhyolitic welded tuffs from the Nishitani Formation and the Goroku-Kamiwazumi-Matsunagi formations around the Toyama Basin. We obtained weighted mean 206Pb/238U dates of 21.8 ± 0.2 and 22.6 ± 0.2 Ma of from the Nishitani Formation and the Goroku-Kamiwazumi-Matsunagi formations, respectively. This suggests that 24-22 Ma silicic volcanism occurred or that rhyolitic pyroclastic density currents reached the Noto Peninsula in addition to the Toyama Basin. Additionally, spatial variation in zircon U-Pb dates reported in this and previous studies suggest the possibility that eruptive vents shifted from northeast to southwest during the 24-22 Ma interval. The results of this study constrain both the timing and temporospatial evolution of the 24-22 Ma silicic volcanism in and around the Toyama Basin.
The Toyama Basin is one of the Miocene to Quaternary sedimentary basins in the back-arc side of the Japan arc. Large-scale silicic volcanism, caused by crustal assimilation of basaltic to andesitic magmas, occurred in the Toyama Basin during the opening of the Sea of Japan (Yamada et al., 2023). Prior to the rapid spreading of the Sea of Japan in 21-15 Ma, rhyolitic welded tuffs (known as ‘moonstone rhyolite’) were widely deposited in the Toyama Basin during 24-22 Ma (Yamada and Takahashi, 2021; Fig. 1). The moonstone rhyolite contains quartz and alkali feldspar grains, sometimes reflecting blue or white light (moonstone or hecatolite) (Ishida et al., 1998). Silicic tuffs dated to 24-22 Ma ages are also scattered across many areas of the northeast and southwest Japan arcs (Hataya and Otsuki, 1991; Sato et al., 2009; Shinjoe et al., 2018; Haji and Yamaji, 2019; Hosoi et al., 2019). Therefore, understanding of large-scale silicic volcanism during 24-22 Ma is key to elucidating early-stage magmatic processes predating the opening of the Sea of Japan and subsequent tectonic processes in the circum-back-arc regions. To reveal the formation timing and temporospatial variation of 24-22 Ma rhyolitic welded tuffs, we conducted zircon U-Pb dating of the Lower Miocene rhyolitic welded tuffs in the Hokuriku Sedimentary Province. In this paper, we present the results and discuss their formation ages and the temporospatial shifts.
Lower Miocene (partly upper Oligocene) rhyolitic welded tuffs in the Hokuriku Sedimentary Province are intercalated between pre-Cenozoic basement rocks and Lower Miocene andesites formed during 21-17 Ma, with unconformities (Yamada and Takahashi, 2021). They are mainly exposed as the Johana, Wassogatake, and Nishitani formations along the margin of the Toyama Basin (Nakae et al., 2013; Nagata et al., 2023; Yamada, 2025; Fig. 1). These formations are composed of rhyolitic welded tuffs with conglomerate and are correlated to the Nanto Group (Yamada, 2025). The Johana Formation is a newly defined formation name, replacing the Tori Formation, by Yamada (2025). Zircon U-Pb dates of 22.5 ± 0.5 and 22.8 ± 0.2 Ma have been reported from the Johana Formation by Ota et al. (2019) and Yamada (2024). The Wassogatake Formation is also a newly defined formation name by Nagata et al. (2023), who reported zircon U-Pb dates of 21.75 ± 0.50, 21.79 ± 0.49, and 21.91 ± 0.50 Ma. No zircon U-Pb dates have been reported from the Nishitani Formation. Although the Uryu Formation mainly consists of andesites, rhyolitic welded tuff, possibly correlated with the 24-22 Ma rhyolitic welded tuffs, is intercalated in its lower part (Sumi et al., 1989). Contemporaneous volcanic rocks are also distributed in the Noto Peninsula, on the back-arc side of the Hokuriku region, as the Goroku, Kamiwazumi and Matsunagi formations (Yoshikawa et al., 2002; Ozaki, 2010). These formations are mainly composed of andesitic to dacitic lava and pyroclastic rocks (Yoshikawa et al., 2002; Ozaki, 2010). Momma et al. (2016) reported a zircon U-Pb date of 23.5 ± 0.7 Ma from the Goroku Formation.
In this study, we collected two rhyolitic welded tuff samples from the Nishitani Formation in the Yarasudake area (YD-1) and from the Goroku-Kamiwazumi-Matsunagi formations in the Tsurugiji area (AK-1) (Table 1; Supplementary Fig. S1; Figs. S1-S2 are available online from https://doi.org/10.2465/jmps.250214). These sampling sites are situated on the margins of the 24-20 Ma rhyolitic welded tuffs and are suitable for discussing their distribution area. YD-1 is a reddish, brown-colored rhyolitic welded tuff containing quartz, alkali feldspar, altered amphibole, and opaque minerals (Fig. 2a). AK-1 is a white-colored rhyolitic welded tuff, containing quartz, alkali feldspar, and opaque minerals, along with volcanic glass shards and pumice (Fig. 2b). Zircon occurs as an accessory mineral in the matrix, mafic minerals such as amphibole, and opaque minerals (Supplementary Fig. S2).
Sample ID | Latitude (N) | Longitude (E) | Mineral assemblage | 238U-206Pb weighted mean date (Ma; ±2σ) |
Number | MSWD |
YD-1 | 35°50′10.62′′ | 136°4′42.7′′ | Qz, Afs, Amp, Opq, Zrc | 21.81 ± 0.16 | 22 | 1.08 |
AK-1 | 37°15′11.09′′ | 136°43′3.36′′ | Qz, Afs, Opq, Zrc | 22.58 ± 0.21 | 13 | 1.00 |
Afs, alkali feldspar; Amp, amphibole; Opq, opaque minerals; Qz, quartz; Zrc, zircon.
Zircon grains in YD-1 and AK-1 were separated via rock crushing, hydraulic elutriation, and panning without heavy liquid separation at Niigata University, Japan. The separated zircon grains were mounted in photocurable resin by hand-picking. Cathodoluminescence (CL) imaging was performed using Chroma CL2 (Gatan Inc., USA) with a JSM-IT500HR (JEOL Inc., Japan) at Fukui Prefectural Dinosaur Museum, Japan.
Zircon U-Pb dating was conducted using a Nu Plasma II (Nu instruments, UK) multi-collector-inductively coupled plasma-mass spectrometer (MC-ICP-MS) coupled to a multiple-spot femtosecond laser-ablation system (CARBIDE, Light Conversion, USA; Makino et al., 2019) at the University of Tokyo, Japan. Measured isotopic ratios used for date calculations (207Pb/206Pb, 207Pb/235U) were corrected using repeated measurements of two primary reference materials: NIST SRM612 glass (207Pb/206Pb = 0.90726; Jochum et al., 2005) and GJ-1 zircon (206Pb/238U = 0.17948; Jackson et al., 2004). During analytical sessions, we measured OD-3 zircon (33.0 ± 0.1 Ma; Iwano et al., 2013) as a secondary reference. The results of OD-3 zircon provide a weighted mean 206Pb/238U date of 32.87 ± 0.28 Ma (±2σ; n = 9; MSWD = 0.54), consistent with the previous reference value. In this study, analytical results with 206Pb/238U and 207Pb/235U dates coinciding within 2σ uncertainties are regarded as concordant data. The U-Pb isotope data were visualized using Isoplot 4.15 (Ludwig, 2012).
We analyzed 30 spots (30 grains) for each sample (YD-1 and AK-1). The rim parts of zircon grains were selectively analyzed based on CL images. Isotopic ratios and dates of each analytical spot are shown in Supplementary Table S1 (Supplementary Table S1 is available online from https://doi.org/10.2465/jmps.250214). Zircon grains of YD-1 show very weak oscillatory zoning and have thin rims (<10 µm) in CL images (Fig. 3a). Meanwhile, zircon grains of AK-1 mainly show sector zoning with thick rim (>10 µm; Fig. 3b). In Tera-Wasserburg diagram, both YD-1 and AK-1 plot along mixing line between upper and lower intercepts, providing lower intercept dates of 21.81 ± 0.14 Ma (±2σ; n = 30; MSWD = 1.00) for YD-1 and 22.59 ± 0.15 Ma (±2σ; n = 30; MSWD = 1.11) for AK-1 (Figs. 3a and 3b). Concordant data provides weighted mean 206Pb/238U dates of 21.81 ± 0.16 Ma (±2σ; n = 22; MSWD = 1.08) for YD-1 and 22.58 ± 0.21 Ma (±2σ; n = 13; MSWD = 1.00) for AK-1 (Figs. 3c and 3d). In this paper, we use the weighted mean 206Pb/238U dates as zircon U-Pb ages.
Since YD-1 and AK-1 contain quartz and alkali feldspar (Fig. 2), they are considered as so-called ‘moonstone rhyolite’ distributed in and around the Toyama Basin (Ishida et al., 1998). From the Johana, Wassogatake, and Goroku formations, zircon U-Pb dates of 23.5-21.8 Ma have been reported by Momma et al. (2016), Ota et al. (2019), Nagata et al. (2023), and Yamada (2024). Although there is ∼ 1 Myr age gap between YD-1 and AK-1, their zircon U-Pb dates indicate that those rhyolitic welded tuffs were formed during the 24-22 Ma silicic volcanism stage. Accordingly, this study reveals that silicic volcanism occurred outside the Toyama Basin, or that pyroclastic density currents erupted from the Toyama Basin were also deposited in the northwest Noto Peninsula.
Zircon U-Pb dates from the 24-22 Ma rhyolitic welded tuffs vary beyond analytical uncertainties. Moreover, several flow units of pyroclastic density current (PDC) are distinguished within the Johana Formation (Ganzawa, 1983). These indicate that rhyolitic PDC eruptions occurred multiple times during 24-22 Ma. Additionally, 24-22 Ma rhyolitic welded tuffs appear to show a NE-SW distribution trend in Figure 1, oblique to the Toyama Basin. This suggests that 24-22 Ma silicic volcanism occurred prior to the formation of the Toyama Basin. Zircon U-Pb dates of 24-22 Ma rhyolitic welded tuffs seem to become younger toward the southwest from the Noto Peninsula. This possible temporospatial shift implies that the locations of the vents, possibly calderas, responsible for the PDC eruptions shifted southward during 24-22 Ma. However, no previous studies have addressed the stratigraphic relationships among these formations. Identifying the location of the calderas that have not yet been discovered and conducting detailed stratigraphic investigations are essential for discussing the temporospatial shift of the 24-22 Ma silicic volcanism in and around the Toyama Basin.
Yamada et al. (2023) discussed that rhyolitic magma during 24-22 Ma beneath the Toyama Basin was generated by crustal assimilation of basaltic to andesitic magma. Results and discussion in this study provide insights into the time scale of this magma genesis and implications for silicic volcanism prior to the rapid opening of the Sea of Japan.
We appreciate two anonymous reviewers and Dr. Tetsuo Kawakami, handling editor, for their constructive comments. We thank prefectural offices of Ishikawa for granting permission to collect samples in the Noto Hanto Quasi-National Park. Our sincere thanks go to Yuta Tsukiji of Fukui Prefectural Dinosaur Museum for helping with cathodoluminescence imaging.
Supplementary Figures S1-S2 and Table S1 are available online from https://doi.org/10.2465/jmps.250214.