Global perturbations of carbon cycle during the Triassic–Jurassic transition recorded in the mid-Panthalassa

Earth and Planetary Science Letters 500 (2018) 105–116

Contents lists available at ScienceDirect

Earth and Planetary Science Letters www.elsevier.com/locate/epsl

Global perturbations of carbon cycle during the Triassic–Jurassic transition recorded in the mid-Panthalassa Wataru Fujisaki a,∗ , Yohei Matsui a,b,c , Hisashi Asanuma d , Yusuke Sawaki e , Katsuhiko Suzuki b , Shigenori Maruyama f a

Research and Development (R&D) Center for Submarine Resources, Japan Agency for Marine–Earth Science and Technology (JAMSTEC), 2-15, Natsushima-cho, Yokosuka-city, Kanagawa, 237-0061, Japan b Project Team for Development of New-Generation Research Protocol for Submarine Resources, Japan Agency for Marine–Earth Science and Technology (JAMSTEC), 2-15, Natsushima-cho, Yokosuka-city, Kanagawa, 237-0061, Japan c Department of Subsurface Geobiological Analysis and Research (D-SUGAR), Japan Agency for Marine–Earth Science and Technology (JAMSTEC), 2-15, Natsushima-cho, Yokosuka-city, Kanagawa, 237-0061, Japan d Geochemical Research Center, The University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 113-0033, Japan e Department of Earth Science and Astronomy, The University of Tokyo, 3-8-1 Komaba, Meguro, Tokyo 153-8902, Japan f Earth-Life Science Institute, Tokyo Institute of Technology, 2-12-1 O-okayama, Meguro, Tokyo, 152-8550, Japan

a r t i c l e

i n f o

Article history: Received 22 March 2018 Received in revised form 17 July 2018 Accepted 19 July 2018 Available online xxxx Editor: D. Vance Keywords: Triassic–Jurassic boundary Central Atlantic Magmatic Provinces (CAMP) organic carbon isotopes redox-sensitive elements bedded cherts Japan

a b s t r a c t To examine environmental changes in the biosphere during the Triassic–Jurassic transition, with a particular focus on the global carbon cycle related to Central Atlantic Magmatic Provinces (CAMP) volcanism in the mid-Panthalassa, we established stratigraphic δ 13 Corg variations using Rhaetian (Late Triassic) to Hettangian (Early Jurassic) shales interbedded within deep-sea cherts in the Katsuyama section in the Mino-Tanba belt, SW Japan. High-resolution record of Rhaetian to Hettangian δ 13 Corg values in the mid-Panthalassa contain three distinct negative carbon isotopic excursions (NCIEs) before and across the Triassic–Jurassic boundary (TJB): the Rhaetian NCIE1 and NCIE2 show a deviation of 5.0h from ca. −24.0h to ca. −29.0h, whereas NCIE3 across the TJB shows a 3.5h deviation from ca. −23.5h to ca. −27.0h. Our newly obtained NCIEs in the deep mid-Panthalassa can be correlated with the δ 13 Corg records in the shallow-marine Tethyan regions (i.e., precursor, initial, and main CIEs), suggesting that three NCIEs in the Tethys and mid-Panthalassa likely reflected the global perturbations of the carbon cycle. Three NCIEs before and across the TJB can be interpreted as the consequence of the multiple CAMP volcanic episodes; i.e., the release of thermogenic methane from organic-rich sediments by CAMP intrusive rocks for NCIE1 and large-scale volcanically derived carbon species for NCIE2 and NCIE3. In addition, progressive increase of atmospheric pCO2 throughout three NCIEs was possibly attributed to accumulation of volcanically derived CO2 from multiple CAMP eruptions, which resulted in the development of ocean acidification across the TJB. On the other hand, in view of the oxic conditions in the deep mid-Panthalassa during three NCIEs, the development of coeval oceanic anoxic– euxinic conditions was restricted solely to shallow-marine regions. Therefore, ocean acidification together with localized shallow-marine anoxia acted as environmental stresses on the biosphere, which eventually resulted in the severe biotic crisis at the end of the Triassic. © 2018 Elsevier B.V. All rights reserved.

1. Introduction The biodiversity crisis across the Triassic–Jurassic boundary (TJB; ca. 201 Ma) has been regarded as the one of the biggest mass extinctions in the Phanerozoic history of life (e.g., Sepkoski, 1997; Alroy, 2010). The emplacement of the Central Atlantic Mag-

*

Corresponding author. E-mail address: [email protected] (W. Fujisaki).

https://doi.org/10.1016/j.epsl.2018.07.026 0012-821X/© 2018 Elsevier B.V. All rights reserved.

matic Province (CAMP), which was associated with the breakup of Pangea, has been considered as a main trigger for the extinctionrelated environmental changes; i.e., an increase in atmospheric pCO2 (McElwain et al., 1999; Bonis et al., 2010a; Steinthorsdottir et al., 2011), an Oceanic Anoxic Event (OAE) in the shallowmarine regions (Tethys—Bonis et al., 2010b; Richoz et al., 2012; Jaraula et al., 2013; Eastern Panthalassa—Kasprak et al., 2015), and ocean acidification in the shallow-marine Tethys (Hautmann, 2004) and deep mid-Panthalassa (Abrajevitch et al., 2013; Ikeda et al., 2015). Of these environmental changes, the development of ocean

106

W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116

cycle related to CAMP volcanism during the Triassic–Jurassic transition. Deep-sea bedded cherts in the Mino-Tanba belt in SW Japan are appropriate for obtaining paleo-environmental information in the deep mid-Panthalassa because of the successive occurrence of the extinction-related intervals in the Middle Permian to Early Jurassic (Matsuda and Isozaki, 1991). To clarify extinction-related environmental changes in the deep mid-Panthalassa during the Triassic– Jurassic transition, Kuroda et al. (2010) determined organic carbon isotope (δ 13 Corg ) values and Os isotopic compositions from deepsea cherts in the Kurusu section in SW Japan (Fig. 1C). They detected a signature of CAMP volcanism during the Rhaetian based on the negative 187 Os/188 Os excursion. The existing δ 13 Corg profile, however, may not trace the carbon cycle around the Triassic– Jurassic transition in detail, possibly due to the low sampling resolution. Therefore, more information is needed about carbon cycle variations, with high resolution, to identify the extinction-related perturbations in the mid-Panthalassa that are relevant to environmental changes at the end of the Triassic, and especially to understand the linkage with CAMP volcanism. To examine possible links between CAMP volcanism and perturbations of the carbon cycle in the mid-Panthalassa during the Triassic–Jurassic transition, we established δ 13 Corg variations in Triassic to Jurassic shales (siliceous claystones) interbedded within deep-sea cherts in the Katsuyama section in the Inuyama area, Mino-Tanba belt. Shales have an advantage over cherts in terms of monitoring fluctuations in the long-term carbon cycle because of their low sedimentation rates (Hori et al., 1993). In addition, we measured the abundances of major, trace, and rare earth elements (REEs) to constrain redox conditions in the deep mid-Panthalassa. Based on the newly obtained δ 13 Corg values, trace element (i.e., molybdenum and uranium) abundances, and REE patterns (i.e., cerium anomalies), we discuss the multiple carbon isotopic fluctuations, their possible association with CAMP volcanism, in combination with the redox conditions in the mid-Panthalassa during the Triassic–Jurassic transition. 2. Geological setting Fig. 1. (A) Latest Rhaetian paleogeography (modified after Abrajevitch et al., 2013) with ancient location of deep-sea bedded cherts exposed in the Katsuyama section. CAMP volcanism is illustrated in red color. (B) A simplified geological map of the Japanese Islands and the location of the Inuyama area. The Jurassic accretionary complexes are denoted by shaded areas. (C) Geological map of the Inuyama area in the Mino-Tanba belt, which is one part of the imbricated Jurassic accretionary complexes (modified after Fujisaki et al., 2016). The stars represent the locations of the Katsuyama and Kurusu sections. White areas show post-Jurassic cover sediments. (For interpretation of the colors in the figure(s), the reader is referred to the web version of this article.)

acidification and anoxia was proposed as a possible cause for the mass killing at the end of the Triassic. In addition, these environmental changes occurred alongside significant carbon-cycle perturbations, as shown by a complex pattern of positive and negative carbon isotope excursions during the Triassic–Jurassic transition in fossiliferous shallow-marine strata deposited along the continental margins of Tethys (Pálfy et al., 2001, 2007; Hesselbo et al., 2002; Kürschner et al., 2007; Ruhl and Kürschner, 2011) and Eastern Panthalassa (Ward et al., 2001, 2004; Williford et al., 2007). These investigations during the Triassic–Jurassic transition emphasized the causal relationships between CAMP volcanism and perturbations of the carbon cycle (e.g., Hesselbo et al., 2002; Ruhl and Kürschner, 2011), but the number and duration of perturbations still remain unclear. In addition to these studies on continental shelf sediments, information from the mid-Panthalassa, which accounted for a major portion of the Triassic to Jurassic global ocean (Fig. 1A), is indispensable for reconstructing changes on the global carbon-

2.1. Jurassic accretionary complex in the Mino-Tanba belt The Mino-Tanba belt is a large part of a Jurassic accretionary complex that was developed along the East Asia margin from the Late Triassic to the earliest Cretaceous, as constrained by detailed macrofossil records (Fig. 1B; e.g., Matsuda and Isozaki, 1991). The Katsuyama section in the Inuyama area is characterized by Early Triassic to Middle Jurassic bedded cherts and Middle to Late Jurassic mudstones and sandstones repeated in thrust sheets (Yao et al., 1980). The metamorphic grade in the MinoTanba belt is pumpellyite–actinolite to prehnite–pumpellyite facies or lower, constrained by the mineral assemblage of basaltic rocks (Hashimoto and Saito, 1970). 2.2. Katsuyama section in the Inuyama area The bedded cherts in the Katsuyama section have been thoroughly investigated through geological, structural (Fujisaki et al., 2016), and paleontological studies (Fig. 1C; Hori, 1992; Carter and Hori, 2005). In the Katsuyama section, rhythmically-bedded cherts that exhibit a cyclic pattern corresponding to Milankovitch cycles (Ikeda and Tada, 2014) were deposited from the Middle Triassic to the Early Jurassic without a major hiatus. The bedded cherts strike mostly N–S with a vertical or sub-vertical dip to the west, and show several colors (i.e., black, gray, white, green, red, and purple). Triassic to Jurassic red and purple bedded cherts in the study interval have a total thickness of 3.5 m (Fig. 2A). The thickness of

W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116

107

Fig. 2. Lithological column and outcrop photos of the Rhaetian to Hettangian bedded cherts in the Katsuyama section. (A) Lithological columns of the Katsuyama section. The depositional ages of bedded cherts were determined using radiolarian and conodont records (Hori, 1992; Carter and Hori, 2005). Ikeda and Tada (2014) calculated the age of the end-Triassic radiolarian extinction (i.e., LAD of the G. tozeri assemblage) as 201.4 ± 0.2 Ma. (B) Thin red bedded cherts in the Hettangian in the upper part of the study section. (C) Red and purple bedded cherts across the TJB. (D) Rhaetian red bedded cherts in the middle to lower part of the study section. RZ, Radiolarian zone; CZ, Conodont zone.

the individual chert beds gradually decreases upwards in the study interval (Figs. 2B and 2D). A distinct purple chert and thick shale interbed (ca. 6 cm) occurs in the middle part of the study interval (Fig. 2C). 2.3. Microfossil ages Radiolarian and conodont dating studies constrain the depositional ages of red to purple bedded cherts across the TJB in the Katsuyama section (Fig. 2A; Hori, 1992; Carter and Hori, 2005). Previous paleontological studies recognized the following three radiolarian assemblage zones in the study interval (total thickness of 3.5 m) from bottom to top: (1) Haeckelicyrtium breviora, (2) Globolaxtorum tozeri, and (3) the Pantanellium tanuense (Fig. 2A; Hori, 1992; Carter and Hori, 2005). The last appearance datum (LAD) of Triassic radiolarians (i.e., the G. tozeri assemblage) and conodont (i.e., Misikella posthernsteini) were found in the distinct purple chert (Fig. 2A). The first appearance datum (FAD) of Jurassic radiolarians (including P. tanuense) was confirmed just above the distinct purple chert (Fig. 2A). The depositional age of the LAD of Triassic radiolarians was estimated as 201.4 ± 0.2 Ma (Ikeda and Tada, 2014). In view of the rate of deposition of the Rhaetian bedded cherts (ca. 16 kyr; Ikeda and Tada, 2014), the time lag between the LAD of Triassic radiolarians and the FAD of Jurassic radiolarians is negligible. Thus, we place the TJB just above the distinct purple chert

and infer a 201.4 ± 0.2 Ma for the TJB in the Katsuyama section (Fig. 2A). 3. Analytical methods We collected 64 samples from shales interbedded within cherts to measure δ 13 Corg ratios and total organic carbon (TOC) contents from the 3.5 m thick studied section. We sampled shales from each layer across the TJB to conduct a high-resolution reconstruction. 3.1. Total organic carbon contents and isotopic analyses For our carbon isotopic study, we handpicked fresh shale fragments without veins or nodules before crushing the samples into a fine powder. We ultrasonically washed the fresh fragments three times with Milli-Q water and then heated them to achieve complete dryness. We powdered the 64 shale samples (∼1.0 g) using an alumina mortar and then carefully weighed them and digested them overnight with 10 mL of 6 M HCl at 70 ◦ C to completely dissolve carbonate minerals. We then extracted HCl and rinsed the remaining residues five times with Milli-Q water to avoid contamination with carbon from carbonate minerals. The residues were then dried overnight at 80 ◦ C and wrapped (10 to 50 mg) in tin foil. The carbon isotopic composition and TOC concentration were measured with an elemental analyzer-isotope ratio mass spectrometer (EA-IRMS; a Thermo Finnigan DELTA plus Advantage mass

108

W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116

spectrometer coupled with an EA 1112 Series FLASH Elemental Analyzer) at the Japan Agency for Marine–Earth Science and Technology. The δ 13 Corg data are reported using delta notation relative as per mil (h) deviations from Vienna Pee Dee Belemnite (VPDB) standard. The analytical uncertainty in δ 13 Corg and TOC values, determined by replicate analyses using in-house standard materials, was better than ±0.11h and 2.7%, respectively. 3.2. Analyses of major and rare earth elements Fujisaki et al. (2016) reported the concentrations of major, trace and rare earth elements from 26 samples of Triassic to Jurassic shales. This study also includes abundances of these elements from 20 newly obtained Triassic to Jurassic shale samples, determined by ALS Minerals (Methods PREP-31 and CCP-PKG01) and listed in Tables S1 and S2. See Fujisaki et al. (2016) for a detailed description of the analytical procedure and detection limits. We calculated enrichment factors (EFs) for the redox-sensitive elements (Mo and U) in the shale samples to evaluate the degree of authigenic enrichment, as previously reported by Algeo and Tribovillard (2009). We present the EFs as TMEF = (TM/Alsample )/ (TM/AlPAAS ), in which the Al-normalized concentration of a given trace metal (TM) in the sample is divided by that of Post-Archean Australian Shale (PAAS; Taylor and McLennan, 1985). 4. Results 4.1. Organic carbon isotopic composition Table 1 lists all the measurements of δ 13 Corg values and TOC contents of 64 samples from the study section. Fig. 3 shows stratigraphic profiles of δ 13 Corg values and TOC contents. The δ 13 Corg values range from −29.07h to −21.10h and TOC contents vary from 0.003 wt.% to 0.085 wt.% (Table 1). Analysis of correlation between δ 13 Corg values and TOC contents show no clear correlations in the analyzed shale samples (r 2 = 0.18): therefore, the δ 13 Corg values were not significantly disturbed by secondary alteration and/or diagenetic process. Although previous workers demonstrated that alteration of organic matter during regional metamorphism increases δ 13 Corg ratios, this process is insignificant below greenschist facies (e.g., Hayes, 1983). In view of the metamorphic facies in this section (up to pumpellyite–actinolite), we consider that changes in the δ 13 Corg ratio in the shales reflect secular variation in the global carbon-cycle. Stratigraphic δ 13 Corg changes showed three distinct negative excursions before and across the TJB in the Katsuyama section (Table 1; Fig. 3). Our results show two abrupt drops in the Rhaetian from ca. −24.0h to ca. −29.0h at heights from −1.83 m to −1.70 m and from −1.23 m to −0.83 m, respectively. Another negative δ 13 Corg excursion from ca. −23.5h to ca. −27.0h occurs across the TJB from −0.25 m to 0.21 m in the section. We named these negative δ 13 Corg excursions NCIE1, NCIE2, and NCIE3, in ascending order. The astrochronologic calibration of the bedded cherts in the section (Ikeda and Tada, 2014), in which a chert-shale couple corresponds to ca. 16 kyr for the Rhaetian and ca. 21 kyr for the Hettangian, indicates durations of ca. 50, ca. 100, and ca. 200 kyr for NCIE1, NCIE2, and NCIE3, respectively. 4.2. Major and trace element compositions Tables S1 shows the abundances of major and trace elements from 46 Triassic to Jurassic shale interbeds from the Katsuyama section. The main components of the rock samples in our study are SiO2 (46.3–78.1 wt.%), Al2 O3 (5.7–22.3 wt.%), and Fe2 O3 (3.5–15.4 wt.%), with smaller amounts of K2 O (1.68–6.40 wt.%) and MgO (1.30–3.39 wt.%). Concentrations of TiO2 , MnO, CaO, Na2 O,

P2 O5 , SrO and BaO are 0.30–1.66, 0.37–6.17, 0.32–0.72, 0.09–1.54, 0.07–0.43, 0.01–0.09, and 0.04–1.37 wt.%, respectively. Cr2 O3 content of the rock samples is 0.01–0.02 wt.% or less. Fig. 3 illustrates stratigraphic profiles of EFs and concentrations of Mo and U, and of Ba/Al and Ba/Ti ratios. Mo and U contents vary from <1.0 to 3.0 ppm and from 1.2 to 5.7 ppm, respectively, corresponding to variations in EFs of <1.0 to 9.9 and 0.7 to 2.1. Ba/Al ratios range from 0.004 to 0.133, and Ba/Al ratios vary from 0.076 to 0.722, except for one outlier at 2.353. Stratigraphic variations in Ba/Al and Ba/Ti ratios show no clear correlation with δ 13 Corg values. 4.3. Rare earth element (REE) abundances Table S2 lists the REE concentrations determined in this study. The total REE concentration (REE) positively correlates with Fe2 O3 (r 2 = 0.55) and P2 O5 contents (r 2 = 0.81), indicating that minerals such as phosphate and iron-oxide may be the main REE carriers. We found a negative correlation between SiO2 content and REE, indicating that the REE is diluted with biogenic and/or detrital silica. We defined Ce anomalies as follows: Ce/Ce* = 2 × (Ce/CePAAS )/ (La/LaPAAS + Pr/PrPAAS ). Fig. 3 shows the stratigraphic profile of Ce anomalies. The Ce anomalies vary from 1.27 to 1.70 throughout the studied section (Table S2). They range from 1.27 to 1.65 during the Rhaetian, and increase up to 1.70 toward the TJB. Afterwards, the Ce anomalies decrease to 1.41 and fluctuate between 1.27 and 1.51 during the Hettangian. 5. Discussion 5.1. Chemostratigraphic correlation 5.1.1. Comparison with mid-Panthalassic sites The new δ 13 Corg isotopic data presented here for the Katsuyama section demonstrate three distinct negative excursions before and across the TJB in the mid-Panthalassa (Fig. 3). NCIE1 and NCIE2 are characterized by a 5.0h excursion from ca. −24.0h to ca. −29.0h along with relatively elevated TOC contents (0.010 to 0.085 wt.%), whereas NCIE3 shows a 3.5h excursion from ca. −23.5h to ca. −27.0h along with low TOC contents (0.004 to 0.008 wt.%) (Table 1; Fig. 3). Radiolarian biostratigraphic data in the Katsuyama section (Hori, 1992; Carter and Hori, 2005) show that the three NCIEs occur in different radiolarian zones; i.e., NCIE1 in the H. breviora Zone (i.e., late Rhaetian age), NCIE2 spans the boundary between the H. breviora and G. tozeri Zones (i.e., late Rhaetian age), and NCIE3 spans the TJB between the G. tozeri and P. tanuense Zones (i.e., latest Rhaetian to earliest Hettangian ages) (Figs. 3 and 4). Kuroda et al. (2010) established the first stratigraphic δ 13 Corg profile from deep-sea cherts in the Kurusu section, located ca. 600 m west of the Katsuyama section (Fig. 1C). They reported an interval marked by more negative δ 13 Corg values from the H. breviora to P. tanuense Zones (Fig. 4). The radiolarian biostratigraphic comparison demonstrates that a negative δ 13 Corg excursion in the H. breviora Zone corresponds to NCIE1 and that the predominance of more negative δ 13 Corg values in the transitional interval between the G. tozeri and P. tanuense Zones is comparable to NCIE3 (Fig. 4). However, no negative δ 13 Corg excursion was recognized in the transitional interval between the H. breviora and G. tozeri Zones in the Kurusu section. The lack of NCIE2 in the Kurusu section may be attributable to either post-depositional processes or the low sampling resolution. No negative correlation was detected between TOC and δ 13 Corg values of cherts in the Kurusu section, indicating that post-depositional processes caused negligible selective loss

W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116

109

Table 1 Isotopic compositions and concentrations of total organic carbon in the Rhaetian (Late Triassic) to the Hettangian (Early Jurassic) shales. Bed number

Sample ID

Lithology

Age

Height (m)

δ 13 Corg vs. VPDB (h)

TOC (wt.%)

Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed Bed

IYF53E-27 IYF53E-26 IYF53E-25 IYF53E-24 IYF53E-23 IYF53E-22 IYF53E-21 IYF53E-20 IYF53E-19 IYF53E-17 IYF53E-16 IYF53E-15 IYF53E-14 IYF53E-13 IYF53E-12 IYF53E-11 IYF53E-10 IYF53E-9 IYF53E-8 IYF53E-6 IYF53E-5 IYF53E-4 IYF53E-3 IYF53E-2 IYF53E-1 IYF53E-0 IYF53D-4 IYF53D-3 IYF53D-2 IYF53D-1 IYF53D-0 IYF53B-0 IYF53A-2 IYF53A-1 IYF53A-0 IYF52-10 IYF52-9 IYF52-8 IYF52-7 IYF52-6 IYF52-5 IYF52-4 IYF52-3 IYF52-2 IYF52-0 IYF51-8 IYF51-7 IYF51-6 IYF51-3 IYF51-2 IYF51-1 IYF50-12b IYF50-12a IYF50-11 IYF50-10b IYF50-10a IYF50-9 IYF50-8 IYF50-7 IYF50-5 IYF50-4 IYF50-1 IYF50-0 IYF49-11

red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale purple shale purple shale purple shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale red shale

Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Hettagian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian Rhaetian

1.20 1.10 0.99 0.95 0.90 0.88 0.85 0.83 0.80 0.75 0.72 0.70 0.68 0.65 0.62 0.60 0.56 0.49 0.45 0.41 0.37 0.35 0.33 0.31 0.29 0.21 0.14 0.12 0.70 0.20 −0.01 −0.06 −0.12 −0.16 −0.18 −0.25 −0.35 −0.43 −0.49 −0.60 −0.70 −0.77 −0.83 −0.93 −1.20 −1.23 −1.28 −1.34 −1.58 −1.64 −1.70 −1.80 −1.83 −1.87 −1.99 −2.07 −2.11 −2.16 −2.22 −2.31 −2.36 −2.60 −2.59 −2.67

−24.83 −24.58 −23.78 −24.23 −24.71 −24.20 −24.03 −24.07 −24.95 −24.89 −24.86 −23.32 −24.64 −24.58 −23.74 −22.57 −22.48 −24.29 −23.68 −21.10 −24.26 −25.22 −24.81 −24.70 −24.58 −27.58 −25.59 −27.50 −26.95 −27.19 −27.46 −26.22 −26.64 −24.70 −27.31 −26.33 −24.90 −22.70 −23.73 −24.01 −24.62 −25.70 −27.64 −28.88 −28.84 −27.33 −24.57 −23.94 −24.16 −23.58 −27.51 −26.58 −29.07 −25.07 −24.08 −23.72 −24.13 −23.10 −22.63 −23.58 −21.82 −22.21 −22.60 −24.42

0.009 0.007 0.008 0.006 0.006 0.011 0.009 0.015 0.007 0.006 0.010 0.006 0.008 0.011 0.006 0.004 0.006 0.007 0.006 0.006 0.006 0.022 0.021 0.016 0.019 0.005 0.008 0.006 0.006 0.004 0.005 0.006 0.006 0.006 0.006 0.008 0.008 0.004 0.003 0.005 0.007 0.008 0.023 0.085 0.032 0.026 0.007 0.005 0.008 0.018 0.032 0.010 0.058 0.028 0.015 0.010 0.017 0.011 0.013 0.025 0.027 0.012 0.012 0.021

64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1

of 12 C. Thus, we consider that the non-detection of NCIE2 in the Kurusu section is likely the result of the low sampling resolution in Kuroda et al. (2010). Despite the lack of the negative δ 13 Corg excursion across the H. breviora and G. tozeri Zones in the Kurusu section, our newly obtained three NCIEs were probably common isotopic signatures in the mid-Panthalassa.

5.1.2. Comparisons with the shallow-marine regions Many researchers have analyzed the δ 13 Corg and δ 13 Ccarb records around the TJB in the shallow-marine regions of Tethys and Eastern Panthalassa. Lindström et al. (2017) recently compiled the existing δ 13 Corg records in shallow-marine Tethys using palynological records; i.e., P. polymicroforatus abundant interval for

110

W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116

Fig. 3. Stratigraphic profiles of δ 13 Corg , TOC, Ba/Al, Ba/Ti, Mo, U, and Ce/Ce* (Ce anomalies) in the Katsuyama section. The circles and triangles show the data from this study and Fujisaki et al. (2016), respectively. The location of the TJB is represented by the dotted red line. The purple, blue, and light-gray horizontal bars represent three NCIEs in the mid-Panthalassa, respectively. The durations of the three NCIEs were constrained using astrochronologic calibration by Ikeda and Tada (2014). RZ, Radiolarian zone; CZ, Conodont zone.

the Rhaetian, and the FAD of Pinuspollenites minimus for the Hettangian. It is, however, difficult to compare δ 13 Corg profiles based on the palynological assemblage with those based on radiolarian biostratigraphy. Instead, to compare the δ 13 Corg profile of the midPanthalassa with those of shallow-marine regions, we compiled a selection of the existing δ 13 Corg and δ 13 Ccarb data around the TJB using on ammonite-based biostratigraphy (Figs. 4 and 5). These newly complied carbon isotopic profiles allow the investigation of global-scale perturbations of the carbon cycle during the Triassic– Jurassic transition. We note that the positive and negative δ 13 Corg and δ 13 Ccarb shifts around the TJB from Yager et al. (2017) are slightly different from other studies (Fig. 5). The negative carbon isotope excursion across the TJB has been widely recognized in shallow-marine Tethys, such as in Røbdy-1 well (Denmark—Lindström et al., 2017) (Fig. 5). Guex et al. (2004) and Ward et al. (2007) also reported a negative δ 13 Corg excursion across the TJB in Nevada (North America) in the Eastern Panthalassa, but detailed ammonite biostratigraphy in Nevada (Bartolini et al., 2012) revealed that this negative δ 13 Corg excursion occurred in the Early Hettangian, not at the TJB as initially assigned. On the other hand, other negative carbon isotope excursions have been reported from Rhaetian continental shelf sediments in the Tethys; ˝ i.e., Csovár (Hungary—Pálfy et al., 2001, 2007), and Eastern Panthalassa; i.e., Kennecott Point (Canada—Ward et al., 2001, 2004; Williford et al., 2007). Hesselbo et al. (2002) named two negative

δ 13 Corg excursions in St. Audrie’s Bay in England as follows: a short “initial” negative δ 13 Corg excursion (i.e., initial CIE) of ca. 5h in the Rhaetian and a longer “main” negative δ 13 Corg excursion (i.e., main CIE) of ca. 3h across the TJB (Fig. 4). These two negative δ 13 Corg excursions, which correspond to the initial and main CIEs, have been discovered in the Tethys; i.e., Lavernock Point (England—Korte et al., 2009), Schandelah and Mingolsheim (Germany—Quan et al., 2008; Lindström et al., 2017) (Fig. 5). Ruhl and Kürschner (2011) recently found a “precursor” negative δ 13 Corg excursion (i.e., precursor CIE) of ca. 3h in Tiefengraben (Austria), that preceded the initial CIE in the Tethys. They further mentioned the existence of the precursor CIE in St. Audrie’s Bay in England (Fig. 4). In addition to Tiefengraben and St. Audrie’s Bay, a ca. 3h deviation occurred before the initial CIE in Kuhjoch (Ruhl et al., 2009, 2011), which has been officially designated as the base Jurassic Global Stratotype Section and Point (GSSP). Thus, we regard the deviation as the precursor CIE in Kuhjoch (Fig. 5). Kuroda et al. (2010) suggested that the more negative δ 13 Corg values across the TJB in the Kurusu section are comparable to those identified as the initial CIE in the Tethys (Fig. 4). However, recent biostratigraphic comparison does not support this idea, because the detailed ammonite biostratigraphy in Kuhjoch demonstrates that the initial CIE occurred in the Rhaetian, clearly before the TJB (Ruhl et al., 2009). Instead, the main CIE in the Tethys is equivalent to the mid-Panthalassic NCIE3, because both CIEs occurred

W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116

111

Fig. 4. Comparison of δ 13 C data in combination with atmospheric pCO2 and 187 Os/188 Os values across the TJB in different regions. The blue squares, black circles, and a bold black line represent 187 Os/188 Os values, δ 13 Corg values, and atmospheric pCO2 , respectively. The purple, blue, and light-gray horizontal bars represent correspondences between mid-Panthalassic NCIEs and shallow-marine CIEs. The stratigraphic position of the TJB is shown by the dotted red line. We define the TJB in the Astartekløft and St. Audrie’s Bay sections as the FAD of Jurassic-type leaf fossils (i.e., Thaumatopteris biozones) and ammonites (i.e., Psiloceras planorbis), respectively. Taking into account of a minor hiatus immediately below the TJB (Hesselbo et al., 2002), we have assigned the precursor, initial, and main CIEs in the Astartekløft section. This definition is consistent with previous works of Hesselbo et al. (2002) and Steinthorsdottir et al. (2011). MZ, Microfloral zone; AZ, Ammonite zone; RZ, Radiolarian zone; CZ, Conodont zone; G. t., Globolaxtorum tozeri; P. tanuense, Pantanellium tanuense.

across the TJB (Figs. 4 and 5). The age of the ammonite-based definition of the TJB in the Tethys (201.36 ± 0.17 Ma; Schoene et al., 2010; Wotzlaw et al., 2014) is almost identical to the age of the radiolarian-based definition of the TJB in the mid-Panthalassa (201.4 ± 0.2 Ma; Ikeda and Tada, 2014), supporting the correspondence between the NCIE3 and the main CIE (Fig. 5). Although it is difficult to precisely compare other CIEs due to the lack of criteria for integrating radiolarian and ammonite biostratigraphies, the precursor and initial CIEs in the Tethys likely correspond to NCIE1 and NCIE2 in the mid-Panthalassa, respectively. Therefore, we conclude that three NCIEs in the Tethys and mid-Panthalassa likely reflected global perturbations of the carbon cycle, rather than local phenomena. 5.2. CAMP volcanism as the driver for global perturbations of carbon cycle Three possible mechanisms are considered for NCIEs that we have identified in the mid-Panthalassa; i.e., (1) changes in primary

productivity, (2) changes in the relative contribution of terrestrial plants, and (3) 13 C-depleted carbon inputs associated with CAMP volcanism. Lower biological productivity associated with microfossil (radiolarian and conodont) turnovers can account for NCIE2 and NCIE3, occurring alongside the disappearance of microfossils before and across the TJB, respectively (Fig. 3). However, based on an estimation by Ikeda and Tada (2014), the durations of NCIE2 (ca. 100 kyr) and NCIE3 (ca. 200 kyr) were longer than the timescale of radiolarian turnovers (∼20 kyr; Hori, 1992; Carter and Hori, 2005). Moreover, the stratigraphic profiles of Ba/Al and Ba/Ti ratios, which are indices of primary productivity, show no clear correlation with δ 13 Corg values (Fig. 3). This geochemical evidence indicates that lower biological productivity is unlikely as the main trigger for NCIE2 and NCIE3. Another possibility, suggested by van de Schootbrugge et al. (2008), is that the initial CIE in Mingolsheim (Germany) is likely explained by changes in the relative contributions of a few groups of marine algae and terrestrial plants. Terrestrial material input, however, potentially had a

112 W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116 Fig. 5. Comparison of δ 13 C data across the TJB in different regions. A continental shelf sequence in the Kennecott Point entirely covers the Rhaetian δ 13 Corg records. The green, blue, and black symbols/lines show the data from Eastern Panthalassa, Tethys, and mid-Panthalassa, respectively. The purple, blue, and light-gray horizontal bars illustrate three NCIEs in the mid-Panthalassa, which correspond to the precursor, initial, and main CIEs in the Tethys, respectively. A red and gray dashed line represents the location of the TJB and Norian–Rhaetian boundary (NRB), respectively. The age of the NRB was referred to Wotzlaw et al. (2014). We defined the TJB as the FAD of the Psiloceras ammonite zone in the Eastern Panthalassic and Tethyan regions. The age of the ammonite-based TJB was constrained to 201.36 ± 0.17 Ma (Schoene et al., 2010; Wotzlaw et al., 2014), whereas that of the radiolarian-based TJB in the Katsuyama section was constrained to 201.4 ± 0.2 Ma (Ikeda and Tada, 2014). Nor., Norian; AZ, Ammonite zone; RZ, Radiolarian zone; C. marshi, Choristoceras marshi; C. crickmayi, Choristoceras crickmayi; P. spelae, Psiloceras spelae; P. tilmanni, Psiloceras tilmanni; P. pacificum, Psiloceras pacificum; P. calliphyllum, Psiloceras calliphyllum; P. planorbis, Psiloceras planorbis; P. hagenowi, Psiloceras hagenowi; A. li., Alsatites liasicus; C. j., Caloceras johnstoni; A. laqueus, Alsatites laqueus; S. angulata, Schlotheimia angulata; H. breviora, Haeckelicyrtium breviora; G. tozeri, Globolaxtorum tozeri; P. tanuense, Pantanellium tanuense.

W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116

negligible effect on three NCIEs, because the bedded cherts in the Katsuyama section deposited far from continents (Fig. 1A). Davies et al. (2017) suggested, on the basis of zircon and baddeleyite U–Pb dating of intrusive rocks, that the degassing of the organic-rich sediments by CAMP intrusive rocks may have played an important role in atmospheric disturbance during the Triassic– Jurassic transition. Enrichments of Hg in marine and terrestrial sections provide further evidence that CAMP volcanism significantly influenced the biosphere across the TJB (Thibodeau et al., 2016; Percival et al., 2017). Cohen and Coe (2002) and Kuroda et al. (2010) demonstrated that negative Os isotopic excursions around the TJB coincided with CAMP volcanism (Fig. 4). In the Kurusu section, 187 Os/188 Os values decreased to ca. 0.2 in the Rhaetian, and increased to ca. 0.6 across the TJB. Three NCIEs correspond to the interval showing an upswing in Os isotopic values. On the other hand, in the St. Audrie’s Bay section, 187 Os/188 Os values decreased to ca. 0.1 toward the TJB. The precursor and initial CIEs correspond to the interval featuring this decrease. A high 187 Os/188 Os ratio at the TJB, meanwhile, was attributed to transient enhanced continental weathering (Cohen and Coe, 2002). Kuroda et al. (2010) attributed these discrepancies in both the absolute values and the timing of the Os isotopic records to the proximity of the St. Audrie’s Bay section to CAMP volcanism (Fig. 1A). A large amount of unradiogenic Os would have been supplied to the European marginal seas, including the St. Audrie’s Bay section, by CAMP basalt (Cohen and Coe, 2002). This would have been removed from seawater immediately before being mixed into the open ocean. Therefore, the discrepancies in the Os isotopic record can be attributed to the short residence time of Os and the spatial relationship between the St. Audrie’s Bay and the Kurusu sections. An abrupt increase in 187 Os/188 Os value at the TJB could alternatively be explained as a local signature in the St. Audrie’s Bay section. In spite of these small differences, δ 13 Corg variations show a clear link with Os isotopic excursions. We therefore suggest that the three NCIEs were likewise related to CAMP volcanism. The negative Os isotope excursion precedes three NCIEs in the mid-Panthalassa (Fig. 4). The gap in timing between negative 187 Os/188 Os shift and the first NCIE (i.e., NCIE1), which spans 7 chert-shale couplets (ca. 110 kyr), likely reflected the difference in oceanic residence time between osmium (∼10 kyr; e.g., Oxburgh, 1998) and carbon reservoirs (∼100 kyr). On the other hand, in the shallow-marine Tethys, Cohen and Coe (2007) judged the negative 187 Os/188 Os shift, except for a positive 187 Os/188 Os excursion at the TJB, to coincide with the main CIE. They do not, however, link the initial and precursor CIEs and 187 Os/188 Os values (Fig. 4). The precursor, initial, and main CIEs in the Tethys are considered to have resulted from CAMP volcanism; i.e., carbon (possibly as thermogenic methane) released from subsurface organic-rich sediments by extensive Late Triassic dike and sill intrusion for the precursor CIE (Ruhl and Kürschner, 2011), outgassing of volcanically derived CO2 (Ruhl and Kürschner, 2011) or rapid release of methane hydrate (e.g., Pálfy et al., 2001; Beerling and Berner, 2002; Ruhl et al., 2011) for the initial CIE, and long-lasting injection of the volcanically derived CO2 for the main CIE (e.g., Hesselbo et al., 2002; Ruhl and Kürschner, 2011). The linkage between the initial CIE and the emplacement of CAMP volcanism is further supported by the detection of Hg enrichments in the interval that corresponds to the initial CIE in Kuhjoch (Austria) by Percival et al. (2017). Therefore, three NCIEs detected prior to and across the TJB can be interpreted as the consequence of the multiple emplacements of CAMP volcanism; i.e., the release of thermogenic methane from organic-rich sediments by CAMP intrusive rocks for NCIE1, and large-scale volcanically derived carbon species for NCIE2 and NCIE3.

113

5.3. Environmental changes in the biosphere during the Triassic–Jurassic transition On the basis of EFs of Mn, Fe, Mo, and U along with Ce anomalies in shales (Fujisaki et al., 2016) and iron speciation analysis on adjacent cherts (Sato et al., 2012) in the Katsuyama section, it has been proposed that the deep mid-Panthalassa was oxic across the TJB. Our newly obtained EFs of Mo and U together with Ce anomalies confirm that mid-Panthalassic NCIEs occurred under oxic conditions (Figs. 3 and 6A–C). In the Kurusu section, 187 Os/188 Os ratios decreased to ca. 0.2 before NCIE1 and increased up to ca. 0.45 throughout NCIE2 and NCIE3 (Fig. 4). The former indicates intensified weathering of CAMP basaltic rocks, whereas the latter is interpreted as the result of radiogenic Os from the continental crust overriding unradiogenic Os from CAMP basaltic rocks, even though the CAMP basalt’s weathering rate remained high (Kuroda et al., 2010). Based on the carbon isotope chemostratigraphy, oxic conditions continued in the aftermath of CAMP volcanism in the deep mid-Panthalassa. Several studies have reconstructed oceanic redox conditions from shallow-marine shelf sequences around the TJB. Bonis et al. (2010b) reported TOC contents and paleontological data from the Eiberg Basin sections; i.e., Kuhjoch, Hochalplgraben, and Tiefengraben in Austria (Fig. 1A), and suggested the appearance of oceanic anoxia during the initial CIE in the Tethys (Fig. 6B). Additionally, based on biomarker proxies, episodic euxinia in the photic zone across the TJB (main CIE) was recently indicated in the Tethyan regions; i.e., Mingolsheim (Germany—Richoz et al., 2012), Rosswinkel (Luxembourg—Richoz et al., 2012), and St. Audrie’s Bay (England—Jaraula et al., 2013), and Eastern Panthalassa; i.e., Kennecott Point (Canada—Kasprak et al., 2015) (Figs. 1A and 6C). In contrast, using inorganic geochemical proxies; i.e., U/Th ratios and Ce anomalies, Pálfy and Zajzon (2012) proposed no such anoxic conditions during the initial and main CIEs in the Kendlbachgraben in Austria (Figs. 1A, 6B and C). In view of oxic conditions in the deep mid-Panthalassa throughout three NCIEs, the development of reducing conditions during the initial and main CIEs appears to be restricted solely to some shallow-marine regions, and played an important role in the collapse of the shallow-marine ecological system. Cohen and Coe (2002) detected sharp 187 Os/188 Os increases (up to 0.74) during the precursor and main CIEs in England in the Tethys, but such high 187 Os/188 Os values are not recognized throughout three NCIEs in the mid-Panthalassa (Fig. 4; up to 0.52). This indicates that the enhancement of continental weathering rate, probably triggered by the input of a huge amount of CO2 to the atmosphere by CAMP volcanism, was limited solely to shallowmarine Tethys. The intensified continental weathering likely promoted primary productivity in shallow-marine settings, resulting in an increase of organic matter accumulation in the water column and local appearances of anoxic–euxinic conditions. Although 187 Os/188 Os records during the initial CIE are lacking (Fig. 4), we speculate that this mechanism worked during the initial CIE and locally developed anoxic–euxinic conditions in the shallow-marine Tethys (Fig. 6B). Steinthorsdottir et al. (2011) reconstructed atmospheric pCO2 during the Triassic–Jurassic transition, using the fossil leaf cuticle stomatal density and index in the Astartekløft, East Greenland. Additionally, they correlated atmospheric pCO2 with the δ 13 Corg data from fossil wood in the Astartekløft as follows; atmospheric pCO2 remained unchanged during the precursor CIE, increased during the initial CIE, and culminated during the main CIE (Fig. 4). Based on chemostratigraphic comparison, the increase in atmospheric pCO2 coincided with the shift in radiogenic 187 Os/188 Os values in the mid-Panthalassa. The progressive increase of atmospheric pCO2 throughout three NCIEs can be explained as the accumulation of volcanically derived CO2 from multiple CAMP eruptions, which is

114

W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116

δ 13 Ccarb records and its detection in the δ 13 Corg records (Fig. 5) would be consistent with direct assimilation of volcanically derived methane into organic matter without oxidation of the methane to CO2 , as this would leave atmospheric pCO2 and δ 13 Ccarb values unchanged. This idea is, however, inconsistent with redox conditions during the precursor CIE (Fig. 6A), as the oxic water mass would oxidize the methane. On the other hand, the sampling resolution of the fossil leaves on which atmospheric pCO2 has been reconstructed was low and plausibly insufficient to detect an event of such short duration as the precursor CIE. In any case, the change in atmospheric pCO2 during the precursor CIE may have been small compared with those during the initial and main CIEs. In this regard, it is noteworthy that mid-Panthalassic NCIE3 coincided with the end-Triassic radiolarian and conodont extinction. The long-lasting volcanic degassing during NCIE3 likely led to the accumulation of excess CO2 in the atmosphere, which resulted in the development of ocean acidification in the shallow-marine Tethys and the deep mid-Panthalassa (Fig. 6C; Hautmann, 2004; Abrajevitch et al., 2013; Ikeda et al., 2015). Therefore, we conclude that the multiple CAMP eruptions explain global perturbations of the carbon cycle, redox structure and ocean acidification of the whole ocean, and the stepwise increase of atmospheric pCO2 during the Triassic–Jurassic transition. In other words, ocean acidification, in combination with the recurring shallow-marine anoxia, likely placed environmental stresses on the biosphere, which eventually resulted in the ultimate biotic crisis at the end of the Triassic. 6. Conclusions To investigate environmental changes during the Triassic– Jurassic transition, with a particular focus on possible relationships between perturbations of the carbon cycle in the midPanthalassa and CAMP volcanism, we established organic carbon isotope records in the Katsuyama section in the Inuyama area, Mino-Tanba belt, SW Japan. On the basis of this study, we reached the following conclusions:

Fig. 6. Schematic diagrams showing the atmospheric pCO2 , perturbations of carbon cycle and redox conditions in the shallow-marine Tethys and deep mid-Panthalassa; (A) NCIE1 in the Rhaetian, (B) NCIE2 in the Rhaetian, and (C) NCIE3 across the TJB.

consistent with the interpretation that the continental weathering rate in the mid-Panthalassa was progressively enhanced by the atmospheric CO2 accumulation. The negligible pCO2 change during the precursor CIE may be attributable to either direct assimilation of volcanically derived methane or a low sampling resolution. Indeed, the lack of detection of the precursor CIE (NCIE1) in the

1. High-resolution Rhaetian (Late Triassic) to Hettangian (Early Jurassic) organic carbon isotopic records in the mid-Panthalassa contain three distinct negative excursions: i.e., the Rhaetian NCIE1 and NCIE2, and NCIE3 across the TJB. 2. Three distinct NCIEs in the mid-Panthalassa can be correlated with δ 13 Corg records (precursor, initial, and main CIEs) in the shallow-marine Tethyan regions such as Austria (Kuhjoch and Tiefengraben) and England (St. Audrie’s Bay). This suggests that the three NCIEs were probably global. 3. Three NCIEs in the mid-Panthalassa can be interpreted as the consequence of the multiple emplacements of CAMP volcanism; i.e., thermogenic methane emission from organic-rich sediments by CAMP intrusive rocks for NCIE1 and significant output of volcanic carbon species for NCIE2 and NCIE3. 4. In view of the oxic condition in the deep mid-Panthalassa during three NCIEs, the coeval oceanic anoxic–euxinic conditions that might be attributed to an increase in the continental weathering rate were restricted solely to shallow-marine regions. 5. Stepwise increase of atmospheric pCO2 level during the Triassic–Jurassic transition was likely caused by accumulation of volcanic CO2 released from multiple CAMP eruptions and associated with the development of ocean acidification. The appearance of ocean acidification together with local shallowmarine anoxia likely had a strong impact on the biosphere, which eventually resulted in mass extinction at the end of the Triassic.

W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116

Acknowledgements We appreciate reviews by S. Schoepfer and two anonymous reviewers, for their constructive comments that greatly improved the manuscript. We are grateful to D. Vance for his editorial handling. We wish to thank Drs. Shuichi Yanai and Yoshimitsu Akahori for their help in outsourcing the whole-rock geochemistry. This work was partly supported by JSPS grants (No. 23224012, 15J11148, 16K00534, 17J04985, and 18K13642) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. Appendix A. Supplementary material Supplementary material related to this article can be found online at https://doi.org/10.1016/j.epsl.2018.07.026. References Abrajevitch, A., Hori, R.S., Kodama, K., 2013. Rock magnetic record of the Triassic– Jurassic transition in pelagic bedded chert of the Inuyama section, Japan. Geology 41, 803–806. Algeo, T.J., Tribovillard, N., 2009. Environmental analysis of paleoceanographic systems based on molybdenum–uranium covariation. Chem. Geol. 268, 211–225. Alroy, J., 2010. The shifting balance of diversity among major marine animal groups. Science 329, 1191–1194. Bartolini, A., Guex, J., Spangenberg, J.E., Schoene, B., Taylor, D.G., Schaltegger, U., Atudorei, V., 2012. Disentangling the Hettangian carbon isotope record: implications for the aftermath of the end-Triassic mass extinction. Geochem. Geophys. Geosyst. 13, Q01007. Beerling, D.J., Berner, R.A., 2002. Biogeochemical constraints on the Triassic–Jurassic boundary carbon cycle event. Glob. Biogeochem. Cycles 16 (3), 1036. Bonis, N.R., Van Konijnenburg-Van Cittert, J.H.A., Kürschner, W.M., 2010a. Changing CO2 conditions during the end-Triassic inferred from stomatal frequency analysis on Lepidopteris ottonis (Goeppert) Schimper and Ginkgoites taeniatus (Braun) Harris. Palaeogeogr. Palaeoclimatol. Palaeoecol. 295, 146–161. Bonis, N.R., Ruhl, M., Kürschner, W.M., 2010b. Climate change driven black shale deposition during the end-Triassic in the western Tethys. Palaeogeogr. Palaeoclimatol. Palaeoecol. 290, 151–159. Carter, E.S., Hori, R.S., 2005. Global correlation of the radiolarian faunal change across the Triassic–Jurassic boundary. Can. J. Earth Sci. 42, 777–790. Cohen, A.S., Coe, A.L., 2002. New geochemical evidence for the onset of volcanism in the Central Atlantic magmatic province and environmental change at the Triassic–Jurassic boundary. Geology 30, 267–270. Cohen, A.S., Coe, A.L., 2007. The impact of the Central Atlantic Magmatic Province on climate and on the Sr- and Os-isotope evolution of seawater. Palaeogeogr. Palaeoclimatol. Palaeoecol. 244, 374–390. Davies, J.H.F.L., Marzoli, A., Bertrand, H., Youbi, N., Ernesto, M., Schaltegger, U., 2017. End-Triassic mass extinction started by intrusive CAMP activity. Nat. Commun. 8, 15596. Fujisaki, W., Sawaki, Y., Yamamoto, S., Sato, T., Nishizawa, M., Windley, B.F., Maruyama, S., 2016. Tracking the redox history and nitrogen cycle in the pelagic Panthalassic deep ocean in the Middle Triassic to Early Jurassic: insights from redox-sensitive elements and nitrogen isotopes. Palaeogeogr. Palaeoclimatol. Palaeoecol. 449, 397–420. Guex, J., Bartolini, A., Atudorei, V., Taylor, D., 2004. High-resolution ammonite and carbon isotope stratigraphy across the Triassic–Jurassic boundary at New York Canyon (Nevada). Earth Planet. Sci. Lett. 225, 29–41. Hashimoto, M., Saito, Y., 1970. Metamorphism of Paleozoic greenstones of the Tamba Plateau, Kyoto prefecture. J. Geol. Soc. Jpn. 76, 1–6. Hautmann, M., 2004. Effect of end-Triassic CO2 maximum on carbonate sedimentation and marine mass extinction. Facies 50, 257–261. Hayes, J.M., 1983. Geochemical evidence bearing on the origin of aerobiosis, a speculative hypothesis. In: Earth’s Earliest Biosphere: Its Origin and Evolution. Princeton University Press, Princeton, NJ, pp. 291–301. Hesselbo, S.P., Robinson, S.A., Surlyk, F., Piasecki, S., 2002. Terrestrial and marine extinction at the Triassic–Jurassic boundary synchronized with major carbon-cycle perturbation: a link to initiation of massive volcanism? Geology 30, 251–254. Hori, R.S., 1992. Radiolarian biostratigraphy at the Triassic/Jurassic period boundary in bedded cherts from the Inuyama area, Central Japan. J. Geosci., Osaka City Univ. 35, 53–65. Hori, R.S., Cho, C.-F., Umeda, H., 1993. Origin of cyclicity in Triassic–Jurassic radiolarian bedded cherts of the Mino accretionary complex from Japan. Isl. Arc 3, 170–180. Ikeda, M., Tada, R., 2014. A 70 million year astronomical time scale for the deepsea bedded chert sequence (Inuyama, Japan): implications for Triassic–Jurassic geochronology. Earth Planet. Sci. Lett. 399, 30–43.

115

Ikeda, M., Hori, R.S., Okada, Y., Nakada, R., 2015. Volcanism and deep-ocean acidification across the end-Triassic extinction event. Palaeogeogr. Palaeoclimatol. Palaeoecol. 440, 725–733. Jaraula, C.M.B., Grice, K., Twitchett, R.J., Böttcher, M.E., LeMetayer, P., Dastidar, A.G., Felipe Opazo, L., 2013. Elevated pCO2 leading to Late Triassic extinction, persistent photic zone euxinia, and rising sea levels. Geology 41, 955–958. Kasprak, A.H., Sepúlveda, J., Price-Waldman, R., Williford, K.H., Schoepfer, S.D., Haggart, J.W., Ward, P.D., Summons, R.E., Whiteside, J.H., 2015. Episodic photic zone euxinia in the northeastern Panthalassic Ocean during the end-Triassic extinction. Geology 43, 307–310. Korte, C., Hesselbo, S.P., Jenkyns, H.C., Rickaby, R.E.M., Spötl, C., 2009. Palaeoenvironmental significance of carbon- and oxygen-isotope stratigraphy of marine Triassic–Jurassic boundary sections in SW Britain. J. Geol. Soc. 166, 431–445. Kuroda, J., Hori, R.S., Suzuki, K., Gröcke, D.R., Ohkouchi, N., 2010. Marine osmium isotope record across the Triassic–Jurassic from a Pacific pelagic site. Geology 38, 1095–1098. Kürschner, W.M., Bonis, N.R., Krystyn, L., 2007. Carbon-isotope stratigraphy and palynostratigraphy of the Triassic–Jurassic transition in the Tiefengraben section—Northern Calcareous Alps (Austria). Palaeogeogr. Palaeoclimatol. Palaeoecol. 2007, 257–280. Lindström, S., van de Schootbrugge, B., Hansen, K.H., Pedersen, G.K., Alsen, P., Thibault, N., Dybkjær, K., Bjerrum, C.J., Nielsen, L.H., 2017. A new correlation of Triassic–Jurassic boundary successions in NW Europe, Nevada and Peru, and the Central Atlantic Magmatic Province: a time-line for the end-Triassic mass extinction. Palaeogeogr. Palaeoclimatol. Palaeoecol. 478, 80–102. Matsuda, T., Isozaki, Y., 1991. Well-documented travel history of Mesozoic pelagic chert in Japan: from remote ocean to subduction zone. Tectonics 10, 475–499. McElwain, J.C., Beerling, D.J., Woodward, F.I., 1999. Fossil plants and global warming at the Triassic–Jurassic boundary. Science 285, 1386–1390. Oxburgh, R., 1998. Variation in the osmium isotope composition of sea water over the past 200,000 years. Earth Planet. Sci. Lett. 159, 183–191. Pálfy, J., Zajzon, N., 2012. Environmental changes across the Triassic–Jurassic boundary and coeval volcanism inferred from elemental geochemistry and mineralogy in the Kendlbachgraben section (Northern Calcareous Alps, Austria). Earth Planet. Sci. Lett. 335–336, 121–134. ˝ I., 2001. Carbon isotope Pálfy, J., Demény, A., Haas, J., Hetényi, M., Orchard, M.J., Veto, anomaly and other geochemical changes at the Triassic–Jurassic boundary from a marine section in Hungary. Geology 29, 1047–1050. Pálfy, J., Demény, A., Haas, J., Carter, E.S., Görög, Á., Halász, D., Oravecz-Scheffer, ˝ I., Zajzon, N., 2007. A., Hetényi, M., Márton, E., Orchard, M.J., Ozsvárt, P., Veto, Triassic–Jurassic boundary events inferred from integrated stratigraphy of the ˝ Csovár section, Hungary. Palaeogeogr. Palaeoclimatol. Palaeoecol. 244, 11–33. Percival, L.M.E., Ruhl, M., Hesselbo, S.P., Jenkyns, H.C., Mather, T.A., Whiteside, J.H., 2017. Mercury evidence for pulsed volcanism during the end-Triassic mass extinction. Proc. Natl. Acad. Sci. USA 114, 7929–7934. Quan, T.M., van de Schootbrugge, B., Field, M.P., Rosenthal, Y., Falkowski, P.G., 2008. Nitrogen isotope and trace metal analyses from the Mingolsheim core (Germany): evidence for redox variations across the Triassic–Jurassic boundary. Glob. Biogeochem. Cycles 22, GB2014. Richoz, S., van de Schootbrugge, B., Pross, J., Püttmann, W., Quan, T.M., Lindström, S., Heunisch, C., Fiebig, J., Maquil, R., Schouten, S., Hauzenberger, C.A., Wignall, P.B., 2012. Hydrogen sulphide poisoning of shallow seas following the end-Triassic extinction. Nat. Geosci. 5, 662–667. Ruhl, M., Kürschner, W.M., 2011. Multiple phases of carbon cycle disturbance from large igneous province formation at the Triassic–Jurassic transition. Geology 39, 431–434. Ruhl, M., Kürschner, W.M., Krystyn, L., 2009. Triassic–Jurassic organic carbon isotope stratigraphy of key sections in the western Tethys realm (Austria). Earth Planet. Sci. Lett. 281, 169–187. Ruhl, M., Bonis, N.R., Reichart, G.-J., Sinninghe Damsté, J.S., Kürschner, W.M., 2011. Atmospheric carbon injection linked to end-Triassic mass extinction. Science 333, 430–434. Sato, T., Isozaki, Y., Shozugawa, K., Seimiya, K., Matsuo, M., 2012. 57 Fe Mössubauer analysis of the Upper Triassic–Lower Jurassic deep-sea chert; paleo-redox history across the Triassic–Jurassic boundary and the Toarcian oceanic anoxic event. Hyperfine Interact. 208, 95–98. Schoene, B., Guex, J., Bartolini, A., Schaltegger, U., Blackburn, T.J., 2010. Correlating the end-Triassic mass extinction and flood basalt volcanism at the 100 ka level. Geology 38, 387–390. Sepkoski Jr., J.J., 1997. Biodiversity: past, present and future. J. Paleontol. 71, 533–539. Steinthorsdottir, M., Jeram, A.J., McElwain, J.C., 2011. Extremely elevated CO2 concentrations at the Triassic/Jurassic boundary. Palaeogeogr. Palaeoclimatol. Palaeoecol. 308, 418–432. Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: Its Composition and Evolution. Blackwell, Oxford. Thibodeau, A.M., Ritterbush, K., Yager, J.A., West, J.A., Ibarra, Y., Bottjer, D.J., Berelson, W.M., Bergquist, B.A., Corsetti, F.A., 2016. Mercury anomalies and the timing of biotic recovery following the end-Triassic mass extinction. Nat. Commun. 7, 11147.

116

W. Fujisaki et al. / Earth and Planetary Science Letters 500 (2018) 105–116

van de Schootbrugge, B., Payne, J.L., Tomasovych, A., Pross, J., Fiebig, J., Benbrahim, M., Föllmi, K.B., Quan, T.M., 2008. Carbon cycle perturbation and stabilization in the wake of the Triassic–Jurassic boundary mass-extinction event. Geochem. Geophys. Geosyst. 9, Q04028. Ward, P.D., Haggart, J.W., Carter, E.S., Wilbur, D., Tipper, H.W., Evan, T., 2001. Sudden productivity collapse associated with the Triassic–Jurassic boundary mass extinction. Science 292, 1148–1151. Ward, P.D., Garrison, G.H., Haggart, J.W., Kring, D.A., Beattie, M.J., 2004. Isotopic evidence bearing on Late Triassic extinction events, Queen Charlotte Islands, British Columbia, and implications for the duration and cause of the Triassic/Jurassic mass extinction. Earth Planet. Sci. Lett. 224, 589–600. Ward, P.D., Garrison, G.H., Williford, K.H., Kring, D.A., Goodwin, D., Beattie, M.J., McRoberts, A., 2007. The organic carbon isotope and paleontological record across the Triassic–Jurassic boundary at the candidate GSSP section at Ferguson Hill, Muller Canyon, Nevada, USA. Palaeogeogr. Palaeoclimatol. Palaeoecol. 244, 281–289.

Williford, K.H., Ward, P.D., Garrison, G.H., Buick, R., 2007. An extended organic carbon-isotope record across the Triassic–Jurassic boundary in the Queen Charlotte Islands, British Columbia, Canada. Palaeogeogr. Palaeoclimatol. Palaeoecol. 244, 290–296. Wotzlaw, J.-F., Guex, J., Bartolini, A., Gallet, Y., Krystyn, L., McRoberts, C.A., Taylor, D., Schoene, B., Schaltegger, U., 2014. Towards accurate numerical calibration of the Late Triassic: high-precision U–Pb geochronology constraints on the duration of the Rhaetian. Geology 42, 571–574. Yager, J.A., West, J.A., Corsetti, F.A., Berelson, W.M., Rollians, N.E., Rosas, S., Bottjer, D.J., 2017. Duration of and decoupling between carbon isotope excursions during the end-Triassic mass extinction and Central Atlantic Magmatic Province emplacement. Earth Planet. Sci. Lett. 473, 227–236. Yao, A., Matsuda, T., Isozaki, Y., 1980. Triassic and Jurassic radiolarians from the Inuyama area, central Japan. J. Geosci., Osaka City Univ. 23, 135–155.