Astrochronology of the Anisian stage (Middle Triassic) at the Guandao reference section, South China

Astrochronology of the Anisian stage (Middle Triassic) at the Guandao reference section, South China

Earth and Planetary Science Letters 482 (2018) 591–606 Contents lists available at ScienceDirect Earth and Planetary Science Letters

5MB Sizes 0 Downloads 14 Views

Earth and Planetary Science Letters 482 (2018) 591–606

Contents lists available at ScienceDirect

Earth and Planetary Science Letters

Astrochronology of the Anisian stage (Middle Triassic) at the Guandao reference section, South China Mingsong Li a,b,c , Chunju Huang a,∗ , Linda Hinnov b , Weizhe Chen a , James Ogg d,a , Wei Tian a a

State Key Laboratory of Biogeology and Environmental Geology, School of Earth Sciences, China University of Geosciences (Wuhan), Wuhan 430074, China Department of Atmospheric, Oceanic, and Earth Sciences, George Mason University, Fairfax, VA 22030, USA c Department of Geosciences, Pennsylvania State University, University Park, PA 16802, USA d Department of Earth, Atmospheric and Planetary Sciences, Purdue University, 550 Stadium Mall Drive, West Lafayette, IN 47907-2051, USA b

a r t i c l e

i n f o

Article history: Received 2 July 2017 Received in revised form 11 November 2017 Accepted 17 November 2017 Editor: M. Frank Keywords: cyclostratigraphy paleoclimate change magnetostratigraphy timescale Early Triassic Great Bank of Guizhou

a b s t r a c t A high-precision global timescale for the Early and Middle Triassic is the key to understanding the nature, pattern and rates of biotic recovery following the end-Permian mass extinction. The Guandao section of Guizhou Province of South China is an important reference section for the magnetic polarity pattern, conodont datums, geochemical anomalies and interpreted temperature history through the Anisian (Middle Triassic). We analyzed the high-resolution gamma-ray and magnetic susceptibility series from the complete Anisian stage. Intensity variations are indicative of fluctuating terrestrial clay influxes showing strong signals that match predicted astronomical solutions for eccentricity and precession. Astronomical tuning of these series to interpreted 405-kyr long-eccentricity cycles yields a 5.3 Myr duration for the Anisian at Guandao. When combined with the astrochronology of the Early Triassic, then the projected age of the Anisian–Ladinian boundary relative to the base-Triassic date of 251.9 Ma is 241.5 ± 0.1 Ma. This provides a 10-Myr reference timescale for other key geological events, including conodont zones, geomagnetic polarity chrons, rates of marine carbon- and oxygen isotope excursions and global sea-level changes, that were associated with the repeated biotic crises and recovery episodes after the end-Permian mass extinction. The middle Anisian humid phase in ca. 244–244.5 Ma was probably a global event, which may have been linked to the middle Anisian warming event and sea-level change. Sea-level fluctuations at Guandao generally correlate with those in western Tethyan and Boreal regions in time, confirming sea-level changes during the Anisian were of eustatic origin. © 2017 Elsevier B.V. All rights reserved.

1. Introduction In the aftermath of the end-Permian mass extinction, Early Triassic environments experienced a succession of extreme temperature swings (Sun et al., 2012; Trotter et al., 2015), anoxic episodes (e.g., Song et al., 2014), global carbon perturbations (Payne et al., 2004) and other episodes that were associated with crises in both continental and marine ecosystems (e.g., Wignall, 2015), which delayed the rise of the modern marine ecosystems with metazoan reefs and efficient nutrient upwelling until the


Corresponding author at: State Key Laboratory of Biogeology and Environmental Geology, School of Earth Sciences, China University of Geosciences, Wuhan 430074, Hubei, China. E-mail addresses: [email protected] (M. Li), [email protected] (C. Huang), [email protected] (L. Hinnov), [email protected] (W. Chen), [email protected]u (J. Ogg), [email protected] (W. Tian). 0012-821X/© 2017 Elsevier B.V. All rights reserved.

middle Anisian of the Middle Triassic (e.g., Benton et al., 2013; Chen and Benton, 2012; Grasby et al., 2016). The timing and rates of these Early Triassic through Anisian crises requires a precise global timescale, but published radioisotopic dating provides only a few sporadic constraints for stage boundaries and biozones during this interval, e.g., the Permian– Triassic system or Changhsingian–Induan stage boundary (Baresel et al., 2017; Burgess et al., 2014), the beginning of the Olenekian stage (Galfetti et al., 2007a), the beginning of the Anisian stage (e.g., Lehrmann et al., 2015; Ovtcharova et al., 2015), and the Anisian–Ladinian stage boundary (Brack et al., 2005; Mundil et al., 2010; Wotzlaw et al., 2017). By contrast, this interval has wellestablished carbon-isotope chemostratigraphy and magnetostratigraphy that enable global correlations (e.g., Hounslow and Muttoni, 2010; Payne et al., 2004; Sun et al., 2012; Szurlies, 2007; Wignall, 2015). The sedimentary record of Milankovitch cycles of precession, obliquity, and orbital eccentricity can be used to generate


M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606

a high-resolution continuous astronomical timescale framework for these geochemical proxies and magnetostratigraphic scales (Hinnov and Hilgen, 2012; Strasser et al., 2006). An astronomicallytuned magnetostratigraphy in South China and Germany now provides an integrated timescale for the Early Triassic (e.g., Li et al., 2016a, 2016b; Szurlies, 2007) and for most of the Late Triassic (e.g., Li et al., 2017; Kent et al., 2017; Olsen et al., 2011; Zhang et al., 2015). However, an astronomically-tuned magnetic time scale for the majority of the Middle Triassic has not yet been established, which hinders global correlation and a precise time frame for the environmental changes and biotic recovery. Establishing an astrochronology for the Middle Triassic has encountered many difficulties. For example, the Latemar atoll-like platform in the Italian Dolomites displays over 500 meter-scale depositional cycles spanning the late Anisian-early Ladinian, which were interpreted as a record of 20-kyr precession cycles modulated by short-term (100-kyr) eccentricity cycles based on stacking patterns and spectral analysis (Hinnov and Goldhammer, 1991). However, these were re-interpreted as millennial-scale cycles based on constraints from U–Pb ages from coeval tuffs, magnetostratigraphy, and quantitative analysis of sedimentary cycles (e.g., Kent et al., 2004; Meyers, 2008; Mundil et al., 1996; Wotzlaw et al., 2017). The deep-sea bedded chert sequence of Japan enables a ∼70 Myr long astronomical time scale for the Triassic–Jurassic (Ikeda and Tada, 2014), nevertheless, a reliable correlation of the Japanese sections to other locations will be possible only once magnetostratigraphy and/or global correlative bio-zones can be defined. Only a partial astronomical tuning of the earliest Anisian magnetic polarity zones has been accomplished in the Germanic Basin (Szurlies, 2007) and South China (Li et al., 2016b). The Nanpanjiang Basin of South China received a nearly continuous deposition of conodont-bearing marine sediments throughout the Early and Middle Triassic. The Guandao road-cut sections on the outer depositional apron of the Great Bank of Guizhou in this Nanpanjiang Basin are some of the best exposed and intensely studied Anisian outcrops in the world. The array of biostratigraphy, magnetostratigraphy, stable-isotope stratigraphy, cyclostratigraphy (for the late Olenekian and the earliest Anisian) and radio-isotope geochronology enable the Guandao sections to serve as a global reference for the Anisian stage (e.g., Lehrmann et al., 1998, 2006, 2015; Li et al., 2016b; Payne et al., 2004; Wang et al., 2006). Here, we present the astrochronology of the Anisian portion of the Guandao section to calibrate its conodont biostratigraphy, magnetic polarity zones, sea-level changes and stable-isotope trends. We also estimate the durations of the Anisian substages based on both their proposed correlations to conodont datums and magnetic polarity zones. When merged with the Early Triassic timescale (Li et al., 2016b), this provides a continuous 10-Myr astrochronology for biotic recovery and climate change after the end-Permian great dying. 2. Geological setting, stratigraphy and Anisian substages of the Guandao section 2.1. Geological setting The South China block was characterized by shallow-water carbonate deposits during much of the late Proterozoic through the Middle Triassic, and terrestrial strata during the Late Triassic. The depositional environment evolved from marine to terrestrial conditions as a result of collision between the South China and North China blocks during the Middle–Late Triassic (Dong et al., 1997; Feng et al., 1997). During the Late Permian–Early Triassic South

China was an isolated carbonate platform in the equatorial eastern Tethys (ca. 6–9◦ N) apart from the Pangaea supercontinent (e.g., Sun et al., 2009; Zhang et al., 2015). The South China block drifted northward during the Triassic and reached ca. 30◦ N in Late Triassic time (Li et al., 2017). The Guandao section for our Anisian study is situated 2 km south of Bianyang town, Luodian County, Guizhou Province, South China. During the latest Permian through the earliest Late Triassic (early Carnian), the Guandao section was situated on the northern flank of the Great Bank of Guizhou in the Nanpanjiang Basin of South China (Lehrmann et al., 2015). Drowning of the southern margin of the Yangtze platform formed a deep-marine embayment, the Nanpanjiang Basin in the Permian–Triassic transition, during which initial accumulation of the Great Bank of Guizhou occurred (Lehrmann et al., 1998). After a low-relief bank stage during the Early Triassic, the Great Bank of Guizhou came to be characterized by a progressively steepening bank to reef-rimmed architecture during the Middle Triassic (Lehrmann et al., 1998) making it the oldest-known platform-margin reef complex of the Mesozoic Era (Payne et al., 2006) (Fig. 1). The Great Bank of Guizhou was terminated at the beginning of the Late Triassic as sea level increased over the platform top and the platform was buried by marine shales and siliciclastic turbidites (Lehrmann et al., 1998). This platform was later uplifted and tilted, and the Guandao road sections are a continuous exposure through its outer depositional apron (Fig. 1D). 2.2. Stratigraphy of the Guandao section The Guandao section has more than 770 m from the Late Permian to the Carnian of the Late Triassic (Lehrmann et al., 2015). The Late Permian Wuchiaping Formation is characterized by cherty bioclastic limestone and overlain by the latest Permian Talung (Dalong) Formation. The Talung Formation mainly consists of black chert and shale and represents the drowning of the Yangtze Platform (Dong et al., 1997; Lehrmann et al., 2015). During the Early and Middle Triassic the section was deposited in a dysaerobic, quiet, deep marine environment. The Early Triassic through earliest Anisian Luolou Formation consists of thin-bedded pelagic micritic to wackestone limestone with thin shale interbeds and a few carbonate packstone–grainstone turbidite deposits and debrisflow breccia beds (Lehrmann et al., 1998). Grains of the problematic reef-forming taxa Tubiphytes occur below the first occurrence (FO) of the Olenekian–Anisian boundary conodont marker Chiosella timorensis (Payne et al., 2006). Several beds of volcanic ash layers occur regionally near this Olenekian–Anisian boundary, and some of these have been dated at Guandao (e.g., Lehrmann et al., 2015). The Anisian–Ladinian portion of the Guandao section consists of marine slope deposits of pelagic micrite-rich limestones punctuated during the Anisian by carbonate packstone–grainstone turbidites and debris flow breccia beds (Lehrmann et al., 1998, 2015). Tubiphytes reefs developed rapidly at the platform margin in the early Anisian (Payne et al., 2006). A gradual thickening of the breccia beds at Guandao during the Anisian reflects the steepening of the basin-margin slope (Lehrmann et al., 1998, 2015). Beneath these breccia beds, the pelagic micritic limestone intervals display some soft-sediment deformation. A notable shift in facies to grainstone breccia with Tubiphytes boundstone clasts and fragmented Tubiphytes debris occurred during the late Ladinian (Payne et al., 2006). Our study interval at Guandao is overlain by the Carnian Bianyang Formation, which is dominated by shallow-marine to terrestrial siliciclastic sediments interbedded with breccia at the base (Dong et al., 1997).

M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606


Fig. 1. Paleogeographic context, study location and the Great Bank of Guizhou. A. The global paleogeography of the Early-Middle Triassic, from B. Triassic paleogeography of South China modified from Feng et al. (1997) and Li et al. (2016b) indicates the location of the Nanpanjiang Basin. C. Simplified cross-section of the Great Bank of Guizhou showing the location of the Guandao section as red line (A to B ; Lehrmann et al., 2015; Li et al., 2016b). The cross-section line a–a is indicated in B. D. Google Earth map of the Guandao section (hillslope outcrop A from A to A and road-cut outcrop B from B to B ). (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

2.3. Anisian chronostratigraphy of the Guandao section Conodont biostratigraphy provides the main constraints for the depositional age of the Guandao section (Lehrmann et al., 2007, 2015; Payne et al., 2004; Wang et al., 2006). For example, Wang et al. (2006) proposed five conodont zones for the Anisian (Chiosella timorensis, Nicoraella germanica-Ni. kockeli, Paragondolella bulgarica, Neogondolella constricta, N g. constricta cornuda) and three conodont zones for the Ladinian (Budurovignathus truempyi, B v. hungaricus, Bv. mungoensis). No systematic study of ammonoids at Guandao has been reported. The Olenekian–Anisian stage (Spathian–Aegean substage) boundary is placed at the first occurrence (FO) of the conodont Chiosella timorensis, which is within the upper part of a thick reversed-polarity zone “GD2r” (Lehrmann et al., 2007, 2015). A brief normal-polarity magnetozone MT1n within the coeval reversed-polarity zone in other global reference sections has been

proposed as a global marker for the Olenekian–Anisian boundary (Hounslow and Muttoni, 2010). A positive carbon-isotope excursion at the base-Anisian in the Guandao section represents an additional marker for the boundary (Lehrmann et al., 2015; Payne et al., 2004). The substages of the Anisian are defined by ammonoid zones, and the correlation of conodont markers to these ammonoid zones is uncertain. Orchard (in Lehrmann et al., 2015) placed the Aegean–Bithynian boundary at the FO of Ni. germanica at Guandao, the Bithynian–Pelsonian boundary roughly between the last occurrence (LO) of N g. ex gr. regalis and the FO of Pg. bulgarica, and the Pelsonian–Illyrian boundary at the FO of Paragondolella ex gr. excela (Lehrmann et al., 2015) (see Fig. 7 below for definitions of the Aegean–Bithynian–Pelsonian–Illyrian). However, some of these conodont datums may be diachronous among regions. In their composite of Anisian magnetostratigraphy sections that have biostratigraphy from ammonoids and conodonts,


M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606

Fig. 2. Comparison of the pre-2013 Middle Triassic reference scale to the Guandao section. Left panel: Age model of Timescale Creator 2012 chart after Ogg (2012) scaled the magnetic polarity pattern of Hounslow and Muttoni (2010) according to their placement relative to marine macrofossil zones. Global sequences after Jacquin and Vail (in Hardenbol et al., 1998) for these biozones. Right panel: Chrono-, litho-, magneto-, and bio- stratigraphy of the Guandao section. Anisian–Ladinian boundary placed either at the first occurrence (FO) of conodont Bv. truempyi (Option ➀) or at the top of GD8n normal zone if it represents MT8n of Hounslow and Muttoni (2010) (Option ➁). Conodont zones are from Lehrmann et al. (2015) and Wang et al. (2006), magnetostratigraphy is from Lehrmann et al. (2015). Sequence stratigraphy and gamma-ray data are from this study. cpm: counts per minutes.

Hounslow and Muttoni (2010) concluded that the bases of these substages should be placed as follows: the Aegean–Bithynian substage boundary at approximately the base of reversed-polarity chron MT3r (in their chron notation; with correlation uncertainty spanning ca. 0.2 Myr in the age model of Ogg et al., 2016), the Bithynian–Pelsonian boundary within the upper normal-polarity subchron of MT4n (∼0.15 Myr correlation uncertainty), and the Pelsonian–Illyrian boundary fairly precisely in the lowermost part of reversed-polarity chron MT5r. The magnetostratigraphy of Guandao (Lehrmann et al., 2015) can be correlated to these magnetic polarity chrons (Figs. 2 and 3), especially after compensating for distortions from breccia beds. We display both the conodont-based and the magnetostratigraphy-based substage assignments in Table 1, and the magnetostratigraphy-based version by Hounslow and Muttoni (2010) is more consistent with the relative number

of ammonoid zones within each substage (e.g., Jenks et al., 2015, as diagrammed in Ogg et al., 2016). The Anisian–Ladinian stage (Illyrian–Fassanian substage) boundary has been formally defined at Bagolino in northern Italy where it coincides with the first occurrence of the ammonoid Eoprotrachyceras curionii (Brack et al., 2005). Wang et al. (2006) and Lehrmann et al. (2015) assigned the base of the Ladinian to the base of the Budurovignathus truempyi conodont zone, which is 2.5 m higher than the top of the GD8n at Guandao after removing breccia layers (Fig. 3). A secondary global boundary marker for the latest Anisian is the top of a brief normal polarity zone “SC2r.1n” at the Seceda section in the Southern Alps (chron MT8n in the scheme of Hounslow and Muttoni, 2010, Fig. 2), and we interpret this as the top of the normal-polarity zone that is 2.5 m lower than the FO of Bv. truempyi at Guandao. This 2.5-m offset between these two markers for the Anisian–Ladinian boundary at Guandao

M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606


Fig. 3. Stratigraphy of the Anisian Guandao section, before and after removing breccia and ash layers. Lithology and beds follow Wang et al. (2006). Gamma-ray (GR) and magnetic susceptibility (MS) values (solid lines) and their 50-m ‘loess’ trend (dashed lines) are shown for the breccia-adjusted column. For spectral analysis, the gamma-ray series were interpolated to a median 0.2-m sample rate, and the magnetic series were interpolated to a median 0.1-m sample rate. See Fig. 2 for detailed explanations of ➀ and ➁, and references for biozones and magnetic polarity patterns.


M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606

has a minor effect on our computed cycle-tuned duration for the Anisian stage. 3. Data and methods 3.1. Gamma-ray and magnetic susceptibility logs Natural gamma-ray intensity (GR) of the sedimentary successions is dominated by potassium (K), uranium (U) and thorium (Th) (Schnyder et al., 2006). K is concentrated in common minerals including clays, feldspar, mica, and chloride salts. U and Th are common in many sedimentary host minerals such as clays, heavy minerals, feldspar, and phosphate, and U is often concentrated in organic matter (Schnyder et al., 2006). Therefore GR variations at studied section are generally linked with both terrestrial runoff that is governed by precipitation in the hinterland and productivity (e.g., Li et al., 2016a, 2016b; ten Veen and Postmas, 1996; Zhang et al., 2015). Magnetic susceptibility (MS) is a measure of the degree of magnetization of a material when subjected to an external magnetic field (Kodama and Hinnov, 2015). Variations in MS relate to the relative detrital mineral influx into the ocean, which is linked to the intensity of continental weathering (Lehrmann et al., 2015; Kodama and Hinnov, 2015). Carbonate accumulation rate that is partly controlled by relative sea-level changes may also have contributed to GR and MS variations (e.g., Westphal et al., 2010). A conceptual model is that during a higher eccentricity phase that amplified the precession cycles, hotter summers during the closer approaches to the Sun would have resulted in an intensified monsoonal climate. Increased summer-monsoonal rainfall and runoff would lead to more terrestrial input into the marine environment, and result in higher gamma-ray and magnetic susceptibility values in the sediment (cf. Li et al., 2016b). Therefore, gamma-ray logs and magnetic susceptibility are appropriate for use in analysis of astronomically forced sedimentary cycles (Kodama and Hinnov, 2015; Zhang et al., 2015). Because the Great Bank of Guizhou is an isolated carbonate platform, re-distribution of oceanic currents (e.g., Strasser et al., 2015) may also have played a role in clay influx and thus variations in GR and MS. The time lag between the astronomical signal and the sedimentary proxy response is unclear (e.g., Hinnov and Hilgen, 2012), however, it only has a small effect on estimated durations of the Anisian events because our proxy series are not tuned to a specific astronomical target curve. Gamma-ray was measured using an RS-230 gamma-ray detector with a 60-second recording time. Measurements were taken at 0.2-m intervals for most of the section, but were spaced at up to 0.5-m intervals for thick-bedded limestone and for breccia intervals. We used a handheld KM-7 magnetic susceptibility meter with a sensitivity of 10−6 SI to record magnetic susceptibility at 0.05–0.2 m intervals (median of 0.1 m) through the Guandao section. Each susceptibility value is an average of 2–5 measurements for the same horizon. The 1071 spectral gamma-ray measurements and 2061 magnetic susceptibility data from the 259-m-thick Guandao section are tabulated in the Supplemental Material. 3.2. Time series methods The identification of potential astronomical signals within the logs followed a typical procedure (cf. Li et al., 2017). The gammaray and magnetic susceptibility series were interpolated using MATLAB function interp1.m with ‘cubic’ method. The series were pre-whitened using function smooth.m and subtracting a 50-m ‘loess’ long-term trend. Evolutionary fast Fourier transform (FFT) spectrograms for inspecting stratigraphic frequencies and trends of the time series were computed using MATLAB script evofft.m (Kodama and Hinnov, 2015). Taner–Hilbert bandpass filtering was

applied using tanerfilter.m to aid in the isolation of potential astronomical parameters (Kodama and Hinnov, 2015). The gamma-ray series was tuned using depthtotime.m (Kodama and Hinnov, 2015) based on identification of Earth’s 405-kyr long orbital eccentricity cycles in the proxy series as the age model. Power spectra of the tuned time series were examined for frequency peaks corresponding to the predicted orbital eccentricity, obliquity, and precession index of the La2010d solution (Laskar et al., 2011). The obliquity and precession index for the La2010d solution are calculated using the procedure provided in Appendix A of Wu et al. (2013). Both the untuned and tuned series were analyzed with the multitaper method (MTM) spectral estimator (Thomson, 1982); estimated spectra were tested against conventional red noise models at the 90%, 95% and 99% confidence levels using MATLAB scripts by Husson (2012). The Average Spectral Misfit (ASM) method provides a test for rejecting the null hypothesis (H0 ) (i.e. no astronomical signal) and objective estimation of the optimal sedimentation rate for a stratigraphic interval that was influenced by orbital forcing (Meyers and Sageman, 2007). Evolutionary ASM was calculated using “eASM” in the R package “astrochron” (Meyers, 2014). The gamma-ray series after removing breccia and ash layers was interpolated to a median 0.2-m sampling rate. 3π evolutionary harmonic analysis of the GR series was applied with a sliding window of 40 m and an estimation of frequencies from 0 to 1 cycles/m. The estimation step is 1 m, and zero-padding of the series in each window is 5000. Target frequencies were calculated from the 2π MTM power spectrum of astronomical solution of La2010d ETP (eccentricity, obliquity and precession) from 240 to 245 Ma (Laskar et al., 2011). The target frequencies are 1/403.32, 1/125.66, 1/96.18, 1/34.57, 1/21.38, 1/20.30 and 1/17.51 cycles/kyr. The H0 significance level test with a significance level above 80% for the GR series is applied for the evolutionary ASM in each 40 m window, for sedimentation rates spanning 2 to 20 cm/kyr. 4. Results 4.1. General observation The typical lithology of Guandao is characterized by cycles between thick-bedded limestone sets and intervals of thin-bedded carbonate with clayey interbeds (e.g., Fig. 4). Higher values of gamma-ray and magnetic susceptibility occur in the thin-bedded lithologies, which have increased clay and organic content, and lower values occur within thick-bedded carbonate. However, volcanic ash can also lead to high gamma-ray and irregular magnetic susceptibility (cf. Li et al., 2016b) and low values occur in redeposited shallow-water breccia intervals. Breccia beds and volcanic ash beds were deposited almost instantly. The time-series analysis is sensitive to thickness of the sedimentary record (Kodama and Hinnov, 2015). Therefore, we made an effort to remove all of these ash-associated peaks and breccia-induced lows using field observations of tuffs and breccias from the section log before time series analysis (Fig. 3). 4.2. Average Spectral Misfit and spectral analysis Evolutionary ASM results for the GR series show that the optimal sedimentation rate is relatively stable, ranging from 5.5 to ∼7 cm/kyr (Fig. 5). The interpretation of these sedimentation rates for the early Anisian is less straightforward. Different sedimentation rates at 5.5, 7.4, and 8.3 cm/kyr are possible. Sedimentation rates may also have been variable during the early Anisian, with switches between all three indicated sedimentation rate values

M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606


Fig. 4. Field photo of the Guandao section from 109 m to 130 m (cf. Fig. 2, right panel). The 5–6-m-thick cycles (orange arcs) are interpreted as corresponding to 100-kyr short-eccentricity cycles. Five meter-scale bundles (yellow arcs) within the rightmost 6-m cycle are probably precession-related bedding. Depth values in white rectangles are original depth in meters as measured in the outcrop. For scale, the height of the dog in the lower right corner is 0.6 m. (For interpretation of the colors in this figure, the reader is referred to the web version of this article.)

possible within the eASM window. These differences in interpretations for this interval may be caused by a higher degree of weathering and by the occurrence of several thin ash beds near the Olenekian–Anisian boundary. Power spectra of the gamma-ray and the magnetic susceptibility series at Guandao show similar characteristics. Both power spectra show significant periods of 12–23 m, 5–8 m, ∼3 m and 1.4–2 m wavelengths (cycle thicknesses) (Fig. 6). Based on estimated average sedimentation rates of 5.5–7 cm/kyr (Fig. 5), the 12 m to 23 m wavelengths probably represent the 405-kyr long orbital eccentricity cycles. This wide range of wavelengths may be explainable by differential compaction of sedimentary rocks, variable sedimentation rates within each limestone–marl couplet, and potential hiatuses (see Section 4.4). With this assumption, the 6–8 m, ∼3 m, and 1.4–2 m cycles would represent ∼100-kyr short-eccentricity, 33-kyr obliquity, and 20-kyr precession cycles. For example, the 5–6 m cycles in Fig. 4 are interpreted as 100-kyr eccentricity cycles; each of the interpreted eccentricity cycles consists of five meter-scale bundles that are interpreted as precessionrelated beddings. Each bundle contains several decimeter-scale bedding couplets of marl and limestone, which may represent subMilankovitch paleoclimate oscillations.

success or failure of the tuning to realign other astronomical parameters. The power spectra of the 405-kyr-tuned gamma-ray and magnetic susceptibility series of Guandao have significant peaks at 405 (long orbital eccentricity), 135–129, 100 (short orbital eccentricity), 42–35 (obliquity), 22–20 and 18–15 kyr (precession) cycles (Fig. 8). This result is compatible with theoretical Middle Triassic astronomical terms (Fig. 8; Laskar et al., 2011). The interpreted long orbital eccentricity cycles (Fig. 6) are labeled “E” with a continuation of the Early Triassic “E” numbering by Li et al. (2016a, 2016b) that had included the Olenekian–Anisian boundary interval at Guandao. The base of each “E” cycle is the center of the relatively low-clay phase in the suite of sections in South China, and the base-Triassic (FO of conodont Hindeodus parvus) at the mid-point of Cycle E0 at the Meishan GSSP is assigned the radio-isotopic age of 251.90 Ma (Burgess et al., 2014). Therefore, the ages of all of events calibrated to these “E” cycles in South China can be computed as a multiple and relative placement of the corresponding long-eccentricity 405-kyr cycle (Table 1). For example, the FO of Ch. timorensis in nearby Guando-3 (Mingtang) section is 0.053 Myr younger than the start minimum of the E13 405-kyr cycle (see Fig. 6 in Li et al., 2016b), which suggests the base of the Anisian is 13% up in Cycle E13.

4.3. Astrochronology and age-model tuned to 405-kyr long orbital eccentricity cycles

4.4. Uncertainty of the cyclostratigraphy at Guandao

Because of chaotic behavior of the Solar system, reliable complete astronomical solutions are not available for the Mesozoic (Laskar et al., 2011). However, Earth’s 405-kyr orbital eccentricity cycle is induced by motions of the orbital perihelia of Jupiter and Venus and is relatively stable for the past 250 Myr due to the large mass of Jupiter. Therefore, the 405-kyr cycle has been designated as an astronomical “metronome” for Mesozoic astrochronology (Hinnov and Hilgen, 2012; Laskar et al., 2011). We develop a floating astronomical time scale for the conodont zones and magnetic polarity pattern at Guandao based on tuning of the filtered 12 m to 23 m sedimentary cycles at Guandao to with a 405-kyr periodicity (Fig. 7). Tuning of a time series to a single astronomical frequency can also be used to assess the

There are three assumptions in our cyclostratigraphic analysis. First, while sedimentation rate within each limestone–marl couplet is likely to have varied (e.g., Tucker et al., 2009), our study assumes that the sedimentation rate within each running window of the eASM analysis described above is constant. Second, the assumption is made that differential compaction plays a minor role in the proxy variations. Differential compaction can lead to distortion in bed thicknesses; decompaction factors of 2.5–3 for mudstone and marls and 1.5–2 for packstones and wackestones were adopted in the study of a Jurassic carbonate platform in Swiss Jura Mountains (Strasser et al., 2012). Differential compaction may contribute to slightly lower “sedimentation rates” in the clay-rich, lower part of the Guandao section and slightly higher rates in the carbonaterich, upper part. Third, for the purpose of time series analysis, we


M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606

Fig. 5. Evolutionary ASM result for the GR series and interpreted average sedimentation rates at Guandao. GR: gamma-ray series after removing breccia and ash layers and removing a 50-m ‘loess’ long-term trend. EHA: evolutionary harmonic analysis F-test confidence level. eASM: Evolutionary ASM plot, displaying H0 significance level. Linear sedimentation rate: sedimentation rates interpreted from the evolutionary ASM results. See Section 3.2 for ASM input parameters.

assume that there are no major hiatuses in the studied section; this is partly based on field observation and the previous conclusion that “there is no evidence of significant erosion beneath the debris flow units” at Guandao (Lehrmann et al., 2015). However, brief hiatuses related to scouring at the base of breccia beds and/or formation of hardgrounds cannot be ruled out (e.g., Strasser et al., 2012). The recognition of three of the 405-kyr cycles (Cycles E15, E19 and E22) is tenuous because of outcrop conditions and uncertainties in sedimentological interpretation. (1) It was necessary to shift from hillslope section A –A to road-cut section B –B at the 42 m level, which required estimating their overlap across a 60-m covered interval (Fig. 1D). Therefore, Cycle E15 is likely distorted because of the uncertain overlap between the two sections or a potential loss of data coverage. (2) There may be scouring at the bases of thick breccia beds in section B –B , even though the

field studies (Lehrmann et al., 2015) concluded there was no evidence of significant erosion. If there was a major scour with the arrival of the 26-m-thick breccia bed at the 120 m level (Fig. 6), then the peak assigned as Cycle E19 is potentially a juxtaposition of the relatively clay-rich portions of two cycles, one above and one below that breccia bed at this level. This possibility of a hiatus is also suggested by the coincidence of this debris-flow episode with the boundary between polarity zones GD4n and GD4r and a conodont zone boundary. If so, then an additional 405-kyr cycle would need to be added to our scale. (3) The existence of breccia beds at 167 m may explain an apparent out-of-phase shift between the peaks of Cycle E22 in the gamma-ray and Cycle E22 in the magnetic susceptibility logs. In sum, according to the evidence at Guandao presented above, the Anisian most likely contains thirteen 405-kyr long orbital ec-

M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606


Fig. 6. Cyclostratigraphy of the Guandao section. Chrono-, magneto-, bio- and litho-stratigraphy at Guandao are detailed in Fig. 2. Detrended gamma-ray (GR) and magnetic susceptibility (MS) are from Fig. 3 showing filtered ∼20-m cycles (red, passband is 0.050 ± 0.035 cycles/m). Evolutionary power spectra (color areas) are calculated using a 40-m window. 2π MTM power spectra are shown with mean, 90%, 95%, and 99% confidence levels. See Fig. 2 for detailed explanations of ➀ and ➁, and references for biozones and magnetic polarity patterns. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)


M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606

Fig. 7. Astronomically tuned time series for physical logs, conodont zones and magnetic polarity patterns at Guandao. The tuned GR and MS logs are plotted with filtered 405-kyr (red, passband is 0.00248 ± 0.0005 cycles/kyr) and 100-kyr cycles (dashed blue, passband is 0.01 ± 0.002 cycles/kyr). Substage definitions are based on conodont zones (Substage #c) and magnetic polarity zones (Substage #m). See Fig. 2 for detailed explanations of ➀ and ➁, and references for biozones and magnetic polarity patterns. ➂: The FO of Ch. timorensis at Guandao. ➃ The FO of Ch. timorensis in nearby Guando-3 (Mingtang) section (Li et al., 2016b). (For interpretation of the colors in this figure, the reader is referred to the web version of this article.)

native definitions for placing the base-Anisian and base-Ladinian at Guandao (Section 2 and Table 1; cf. Li et al., 2016b). If a full 405-kyr cycle was removed by a scour at the base of the breccia flow in the middle of Cycle E19, then an additional 0.405 Myr would have to be added to this duration. 4.5. Sea-level change

Fig. 8. 3π MTM power spectra of the Laskar 2010d astronomical solution from 242 to 247 Ma compared to 405-kyr tuned GR and MS time series at Guandao.

centricity cycles (13% up in Cycle E13 to the base of Cycle E26), and therefore, a duration of 5.3 Myr. The uncertainty of approximately 0.1 Myr includes uncertainty of 405-kyr tuning and alter-

The Guandao section was characterized by pelagic limestone deposited in a quiet, deep marine environment punctuated by debris flows and turbidity currents (Lehrmann et al., 1998). Triggering of debris-flow beds at Guandao may have been complex, involving earthquakes, storms and an overload of unconsolidated sediments on the Great Bank of Guizhou. Exposure of the Great Bank of Guizhou during low relative sea-level stands could also be responsible for increased frequency of rock falls and debrisflows at the studied section. High GR values associated with increased concentrations of clay minerals and organic matter may be linked to the maximum flooding (e.g., Catuneanu, 2006). Therefore, the breccia-dominated intervals at Guandao could be indicators of low stands, and high GR values thus would be related to maximum flooding surfaces. Based on these inferences, relative sealevel changes at Guandao section are proposed in the right panel of the Fig. 2. Astrochronology at Guandao provides a high-resolution time frame for relative sea-level variations in the Nanpanjiang Basin of South China. The sequence boundary (SB) Ol4 occurred at 246.9 ± 0.1 Ma, just below the Olenekian–Anisian boundary. SB Ol4 correlates with a basal brecciated (karstic) limestone of the Dongma’anshan Formation overlain by terrestrial clastics at Chaohu section, South China (Li et al., 2016b). The subsequent SB An1 occurs close to the minimum between Cycle E16 and Cycle E17. SB An1 is at the base of the reversed magnetic polarity zone GD3r, with

M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606


Table 1 Astrochronology for conodont datums and magnetic polarity zones at the Guandao reference section for the Anisian. Placements of events within 405-kyr long-eccentricity “E” cycles are according to relative positions in Lehrmann et al. (2015), with stratigraphic ties to the section measured during gamma-ray and magnetic susceptibility logging, and then adjusted by removing major debris flow events and volcanic ash beds. Magnetic zones at Guandao (“GD” nomenclature of Lehrmann et al. (2015) is shown with interpreted assignment to the Middle Triassic magnetic polarity reference scale of Hounslow and Muttoni (2010)). Anisian substages, which are defined by ammonoid zones in the Mediterranean region, are placed according to interpreted conodont proxies (Lehrmann et al., 2015) and magnetic polarity calibrations (Hounslow and Muttoni, 2010). Age model is based on composite “E” scale calibrated to radio-isotopic dated (251.9 Ma) base of Triassic at Meishan GSSP (see text for details).

an age of 245.3 ± 0.1 Ma. SB An2 occurs at the beginning of Cycle E20, at the base of reversed magnetic polarity zone GD4r with an age of 244.1 ± 0.1 Ma. SB An3 is near the minimum between Cycle E22 and Cycle E23, in the lower part of the reversed magnetozone GD5r with an age of 243.2 ± 0.1 Ma. SB An4, the topmost sequence boundary in the Anisian, occurs in the maximum of the Cycle E25, at the middle of the GD7r with an age of 241.8 ± 0.1 Ma.

5. Discussion 5.1. Timescale for the Anisian The astrochronology of the Anisian stage is anchored to the base of the Triassic in the middle of Cycle E0 at Meishan GSSP (Li et al., 2016b), which has an interpolated radio-isotopic date of


M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606

251.902 ± 0.024 Ma (Burgess et al., 2014), although a new Bayesian analysis of dating at Meishan and two other sections by Baresel et al. (2017) suggests this age might be ca. 0.05 Myr older. Based on the Meishan base-Triassic GSSP date of 251.9 Ma and uncertainty in the placement of stage boundaries relative to the E-cycles, the Olenekian–Anisian boundary is 246.8 Ma and the Anisian–Ladinian boundary is 241.5 Ma, with an uncertainty on each age of ∼0.1 Myr (choice of boundary markers and uncertainty in each placement within the E cycle). In addition, there is a “potential systematic offset” of up to ca. 0.08 Myr from the divergent opinions on the radio-isotopic dating for the base-Triassic GSSP level plus their analytical uncertainty of ca. 0.03 Myr, and an uncertainty of ca. 0.05 Myr in the exact placement of the GSSP level within Cycle E0. 5.2. Comparison to radio-isotopic dating of the Anisian At first glance, the 405 kyr cycle-derived age of 246.8 ± 0.1 Ma for the Olenekian–Anisian boundary is about 0.5 Myr younger than the 247.28 ± 0.12 Ma based linear extrapolation of bracketing constraints from radio-isotopic dating of zircons from volcanic ash layers near the boundary at Guandao (Lehrmann et al., 2015). However, the individual zircons in the ash beds in this Guizhou region have a considerable spread in analytical ages. A very detailed dating of the Olenekian–Anisian boundary interval at the Monggan (Wantou) section, about 150 km south of Guandao by Ovtcharova et al. (2015) indicated a considerable scatter in zircon dates, and statistics on mean dates for a succession of ash beds violated the stratigraphic order because of assumed potential lead loss and unknown crystallization history. In that Monggan study, a statistical model suggested an approximate age of 247.305 ± 0.040 Ma for the base of the conodont Ch. timorensis boundary marker (Ovtcharova et al., 2015), but excluded the majority of the youngest and oldest individual zircon dates. An ash bed located at 17.25 m above the Olenekian–Anisian boundary at Guandao yielded an interpreted U–Pb date of 246.50 ± 0.11 Ma (Lehrmann et al., 2015). This ash bed at 25% up in Cycle E14 in our scale would project to an age of 246.33 Ma relative to the base-Triassic, i.e., close to that date. Radio-isotope dating for the Anisian–Ladinian boundary is currently constrained in the Monte San Giorgio region of southern Switzerland to between (1) a volcanic ash in the Grenzbitumen horizon within the lower Nevadites secedensis Ammonoid Zone of the latest Anisian (Mundil et al., 2010) that yielded 242.2 ± 0.6 Ma date by U–Pb CA-TIMS (chemical abrasion, thermal ionization mass spectrometry) analysis of zircons, or a 239.5 ± 0.5 Ma date by Ar–Ar analysis of sanadines, recalibrated as 241.16 ± 0.73 using current Ar–Ar monitor standards (Schmitz, 2012); and (2) a tuff in the P. gredleri Ammonoid Zone of the early Ladinian, which yielded an interpreted date of 241.07 ± 0.13 Ma based on U–Pb ID-TIMS dating (Stockar et al., 2012). Therefore, timescale compilations suggested a ca. 242.0 Ma (Mundil et al., 2010) or 241.5 ± 1 Ma (Ogg et al., 2016) age for the Anisian–Ladinian boundary. A latest U–Pb CA-TIMS dating of the Buchenstein Formation at Passo Feudo and Seceda in the Dolomites of Italy improved the calibration of the Anisian–Ladinian stage boundary (Wotzlaw et al., 2017). The U–Pb based age model at Seceda projected the stage boundary to 241.43 ± 0.15/0.17/0.31 Ma after excluding one sample of SEC-C. The SEC-C is just below the Anisian–Ladinian boundary with an age of 241.98 ± 0.11/0.13/0.27 Ma, which has been considered to be out of stratigraphic order. Constrained by cyclostratigraphy, the age of the Anisian–Ladinian boundary at Seceda was estimated at 241.464 ± 0.064/0.097/0.28 Ma (Wotzlaw et al., 2017). The 405 kyr cycle-derived age of the Anisian–Ladinian boundary from Guandao is 241.5 ± 0.1 Ma (241.51 Ma using top of

GD8r.1n (MT8n) as a boundary marker at Guandao or 241.45 Ma using FO of conodont Bv. truempyi). This age is indistinguishable from the independent survey of the stage boundary age of the Buchenstein Formation in the Dolomites by Wotzlaw et al. (2017). Therefore, our cycle-derived 5.3-Myr time span for the Anisian and its ages relative to the base of the Triassic are compatible with the constraints from U–Pb and Ar–Ar dating, and are comparable with the ca. 5.2-Myr duration estimated by Mundil et al. (2010) and used in the International Chronostratigraphic Chart (v2017/02,, and to an estimated 5.3-Myr duration used in GTS2016 (Ogg et al., 2016). 5.3. Global sea-level changes during the Anisian Global eustatic sea-level change has been one of the key controls on the sedimentary record in tune with a broad temporal band (e.g., Haq et al., 1987; Ogg, 2012; Li et al., 2016a). Haq et al. (1987) recognized two global sea-level falls in the Anisian. Subsequently, larger datasets incorporated in global and regional sealevel database provided clues for high-resolution sea-level changes. For example, Gianolla and Jacquin (1998) recognized four major sea-level falls during the Anisian in the Southern and Northern Alps (western Tethys). A compilation of sequences in the Timescale Creator 2012 chart provides an ideal target for the global Triassic sea-level changes after Ogg (2012). The compilation scaled the magnetic polarity pattern of Hounslow and Muttoni (2010) and western Tethyan and Boreal sequences after Hardenbol et al. (1998) and Snedden and Liu (2010) according to their placement relative to marine macrofossil zones. However, ages of the sequences in these compilations remain un-calibrated and in dispute. Aided with magnetostratigraphy and conodont biostratigraphy, the cyclostratigraphy at Guandao provides a high-resolution time scale for the global sea-level changes during the Anisian. The first major sea-level fall indicated by SB An1 occurred near the base of a reversed magnetic polarity zone (GD3r/MT3r) both at Guandao and in the global compilation in Fig. 2. The third global sea-level fall near SB An3 occurred in the lower part of the reversed magnetozone GD5r at Guandao. This correlates with the SB An3 in the global compilation, in the contemporaneous magnetozone MT5r (Ogg, 2012). The times of the SB An2 and SB An4 at Guandao do not match those in the global compilation (Fig. 2). For example, the second sea-level fall near SB An2 occurred in the middle of the normal magnetic zone MT4n, but near the top of same normal magnetozone GD4n. This discrepancy may be caused by uncertainties in the recognition of sea-level changes at Guandao and/or in the global compilation and potential interference from regional tectonics (e.g., Gianolla and Jacquin, 1998; Hardenbol et al., 1998; Ogg, 2012). In sum, sea-level changes at Guandao in the eastern termination of the Tethyan region generally correlate with sea-level fluctuations in western Tethyan and Boreal regions. These suggest the sea-level changes during the Anisian were of eustatic origin. 5.4. Paleotemperature change and the middle Anisian humid phase The Triassic world was characterized by multiple “greenhouse crises” of unusually warm and wet conditions (e.g., Retallack, 2013; Stefani et al., 2010; Trotter et al., 2015). The Early Triassic has long been recognized as a lethally hot greenhouse (e.g., Chen and Benton, 2012; Retallack et al., 1996; Sun et al., 2012). During the Olenekian–Anisian transition interval (246.8 Ma), a distinct cooling event led to a more favorable climate for tropical life and the onset of more efficient recycling of nutrients and greater oceanic productivity (Sun et al., 2012; Trotter et al., 2015). A cooler climate ensued during the early–middle Anisian, as interpreted from subsequent stable and relatively heavy δ 18 O data in central–western Tethys (Italy and Turkey) and eastern Tethys (South China) (Fig. 9).

M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606


Fig. 9. Astronomically (405 kyr cycle)-tuned timescale for biotic recovery following the end-Permian dying. 405 kyr cycles (“E”, numbered from end-Permian) and magnetic polarity patterns for the Early Triassic are from Li et al. (2016b) using magnetozone nomenclature of Szurlies (2007) and Hounslow and Muttoni (2010), and for the Anisian are from this paper with magnetozone pattern from Lehrmann et al. (2015) plus “MT” nomenclature from Hounslow and Muttoni (2010). Placement of Anisian substages is using estimated placement relative to both conodont zones (left, Lehrmann et al., 2015) and magnetic polarity chrons (right, Hounslow and Muttoni, 2010). Carbon isotope data is from Payne et al. (2004). Oxygen isotope data of conodont apatite is from Sun et al. (2012) and Trotter et al. (2015). Humid phases are from Galfetti et al. (2007b), Stefani et al. (2010) and Trotter et al. (2015). Reef records are from Payne et al. (2006) and Chen and Benton (2012). Timing of biotic events are from Li et al. (2016b) (Early Triassic) and this paper (Anisian).

δ 18 O records indicate that warming events of low magnitude punctuate the Bithynian–Pelsonian and the transition of the Anisian– Ladinian (Trotter et al., 2015). At least five Triassic humid phases have been inferred from the synthesis of sedimentological, paleobotanical, and geochemical proxies in the Dolomites of the western Tethys, dated to early Olenekian, middle Anisian, late Ladinian, middle Carnian, and late Norian times (e.g., Galfetti et al., 2007b; Stefani et al., 2010; Trotter et al., 2015; Zhang et al., 2015). The middle Anisian (Bithynian–Pelsonian) humid phase (Stefani et al., 2010) coincided with the middle Anisian warming event within long period of dry

and cool conditions during the Anisian (Trotter et al., 2015). At Guandao, the GR series was generally linked with precipitation and productivity (section 3.1). High GR values in the late Bithynian– early Pelsonian may have derived from high precipitation in South China old-land, higher nutrient input, and increased productivity in the marine environment. The evidence for more humid conditions recorded in the western Tethys may also apply to Guandao in the eastern Tethys. Therefore, the middle Anisian humid phase appears to have been a global event. The age of this humid phase is ca. 244.5–244 Ma on our time scale. This humid phase and the middle Anisian warming event appear to have coincided with a


M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606

maximum flooding surface suggesting a complex climate response worth exploring (Fig. 9 and Trotter et al., 2015). 5.5. Timeline for biotic recovery in the aftermath of the great dying 5.5.1. Early Triassic The end-Permian dying at 251.9 Ma spanned only 60 ± 48 kyr based on U–Pb dating of volcanic ash beds across the Permian– Triassic boundary at the Meishan GSSP section (Burgess et al., 2014) and less than 40 kyr based on cycles from the same section (Li et al., 2016b). Biotic recovery after the end-Permian great dying was intermittent over the following 4 million years because of repeated Early Triassic climate upheavals, recurrent global warming, ocean acidification, ocean anoxia and inefficient nutrient recycling as indicated by major perturbations in the global carbon cycle, oxygen–isotope analyses of conodonts and other proxies of sulfides and nitrogen isotopes (Grasby et al., 2016; Payne et al., 2004; Sun et al., 2012; Trotter et al., 2015; Zhang et al., 2017). Astronomical cycles recorded in South China and the Germanic Basin provide a timescale for the Early Triassic (Li et al., 2016b), which provides a time frame for the main episodes of partial biotic recovery and relapses (Fig. 9). The latest Smithian event (248.2 Ma) at 3.7 Myr after the endPermian extinction was the most severe biocrisis after the endPermian dying. Preceded by a thermal maximum that may have been lethal for vertebrates living on tropical landmasses (Sun et al., 2012) and accompanied by ocean euxinia and a major perturbation in the carbon cycle (Galfetti et al., 2007b; Payne et al., 2004), oceanic and terrestrial life both experienced a major turnover across the Smithian–Spathian boundary (Galfetti et al., 2007b; Hermann et al., 2011; Song et al., 2011). This event is characterized by a climate shift from humid and hot conditions in the Smithian to drier climate in the Spathian (Galfetti et al., 2007b; Stefani et al., 2010). The latest Smithian event occurred suddenly and lasted only ∼50 kyr (Li et al., 2016b), suggesting a cataclysm (e.g., Romano et al., 2013). 5.5.2. Early–Middle Triassic transition Following the end-Permian extinction, an initial recovery of marine reptiles did not begin until the middle to late Spathian. The fossil beds in the Chaohu section in South China that have yielded nearly 100 skeletons of diverse marine reptiles are 405 kyr cycledated at 247.10 Ma to 247.25 Ma (Li et al., 2016b). Reefs have been a central focus of many studies on extinctions and recoveries because reef ecosystems have a greater susceptibility to environmental stresses (Erwin, 2006). The end-Permian mass extinction led to a 5-Myr reef gap, which is the longest documented gap for any extinction (Erwin, 2006). The recovery of reef ecosystems coincided with the cooler climate of the Anisian (Trotter et al., 2015). During the Anisian, Tubiphytes reefs developed rapidly at the platform margin of the Great Bank of Guizhou (Payne et al., 2006). In the slope deposits at Guandao, the first Triassic Tubiphytes appeared ca. 100 kyr earlier than the first occurrence of the conodont Chiosella timorensis that marks the Olenekian–Anisian boundary (Payne et al., 2006), and therefore has an age of 246.9 Ma (Fig. 9). 5.5.3. Anisian The Luoping biota of South China occured in the conodont Nicoraella kockeli Zone of the mid–late Anisian (Zhang et al., 2009). The Luoping assemblage records a complete tropical structure from primary producers and consumers through meso-consumers to top predators of marine reptiles and predatory fishes, and is considered to mark the final recovery of a complex marine ecosystem after the end-Permian extinction (Benton et al., 2013;

Chen and Benton, 2012). The late Anisian has the oldest dinosauromorph suggesting that dinosaurs, or their immediate precursors may have existed at the same time; however, the record of true dinosaurs did not begin until the late Carnian (Benton et al., 2014; Furin et al., 2006; Nesbitt et al., 2010; Olsen et al., 2011). The end-Permian extinction included the extinction of wetland plants, major contributors to coals globally (e.g., Retallack, 2013). The extinction of wetland plants ushered in a multiple millionyear hiatus in peat formation leading to an Early Triassic “coal gap” (Chen and Benton, 2012; Retallack et al., 1996; Retallack, 2013). The coal gap ended during the Anisian; only in the Late Triassic did Permian levels of plant diversity and peat thickness reappear (Retallack, 2013). In general, the recovery of the terrestrial biosphere after the end-Permian extinction appears to have been in pace with (or even later than) the marine ecosystem (Benton et al., 2013; Chen and Benton, 2012; Payne et al., 2006). 6. Conclusions The Guandao section of Guizhou Province from the former Nanpanjiang Basin of South China contains a complete record of conodont datums, stable isotope trends and magnetostratigraphy through the entire Anisian stage of the Middle Triassic (Lehrmann et al., 2015). Astronomical tuning of gamma-ray and magnetic susceptibility data from the Guandao section based on interpreted 405-kyr long orbital eccentricity cycles provides a high-resolution astronomical time and indicates a 5.3 Myr duration for the Anisian stage. If there was a complete long orbital eccentricity cycle removed by scour at the base of the major debris flow near the Bithynian–Pelsonian substage boundary, then an additional 0.405-Myr would need to be added to this duration, but field observations do not indicate any evidence for major erosion at the base of this debris flow bed (Lehrmann et al., 2015). This Anisian 405 kyr cycle-tuned astronomical scale is merged with that from the Early Triassic (Li et al., 2016b), which includes cyclostratigraphy across the 251.9 Ma radio-isotopic dated basal Triassic at the Meishan GSSP. The composite astrochronology projects the Olenekian–Anisian stage boundary at 246.8 ± 0.1 Ma and the Anisian–Ladinian stage boundary at 241.5 ± 0.1 Ma. This high-resolution astrochronology provides age constraints and rates for carbon- and oxygen–isotope events, magnetic polarity chrons, sea-level and biozones for the Early Triassic and the Middle Triassic Anisian stage. This 10-Myr timescale allows a new assessment of the convoluted history of biotic recovery and paleoclimate change following the end-Permian extinction. The middle Anisian humid phase appears to have been a global event that lasted from 244.5 Ma to 244 Ma. This humid phase and the middle Anisian warming apparently coincided with a maximum flooding surface at Guandao. The sea-level changes at Guandao in eastern Tethys generally correlate with sea-level fluctuations in western Tethyan and Boreal regions, supporting their eustatic origin. Acknowledgements We thank Haishui Jiang, Kunyuan Ma and Xiaokun Huang for field assistance. This study was supported by the National Natural Science Foundation of China (No. 41772029, 41322013). Natural Science Foundation for Distinguished Young Scholars of Hubei Province of China (2016CFA051) and the 111 Project (No. B14031 and B08030). Prof. Martin Frank, Prof. André Strasser and one anonymous reviewer are acknowledged for their helpful comments. M. Li acknowledges the China Scholarship Council for Ph.D. work at Johns Hopkins University, USA. Support for J. Ogg was from a visiting professorship at the State Key Laboratory of Geobiology and Environmental Geology funded by the Overseas Top Scholar Fellowship program of the Ministry of Education of the People’s

M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606

Republic of China (MS2013ZGDZ[WH]028). This study would not have been possible without the advice of Daniel Lehrmann, Paul Montgomery, Zhong-Qiang Chen and Mike Orchard about the importance of the Guandao section and their extensive documentation on its stratigraphy. Appendix A. Supplementary material Supplementary material related to this article can be found online at References Baresel, B., d’Abzac, F.-X., Bucher, H., Schaltegger, U., 2017. High-precision time– space correlation through coupled apatite and zircon tephrochronology: an example from the Permian–Triassic boundary in South China. Geology 45, 83–86. Benton, M.J., Forth, J., Langer, M.C., 2014. Models for the rise of the dinosaurs. Curr. Biol. 24. R87–95. Benton, M.J., Zhang, Q., Hu, S., Chen, Z.-Q., Wen, W., Liu, J., Huang, J., Zhou, C., Xie, T., Tong, J., Choo, B., 2013. Exceptional vertebrate biotas from the Triassic of China, and the expansion of marine ecosystems after the Permo-Triassic mass extinction. Earth-Sci. Rev. 125, 199–243. 2013.05.014. Brack, P., Rieber, H., Nicora, A., Mundil, R., 2005. The global boundary stratotype section and point (GSSP) of the Ladinian Stage (Middle Triassic) at Bagolino (Southern Alps, Northern Italy) and its implications for the Triassic time scale. Episodes 28, 233–244. Burgess, S.D., Bowring, S., Shen, S.-z., 2014. High-precision timeline for Earth’s most severe extinction. Proc. Natl. Acad. Sci. USA 111, 3316–3321. 1073/pnas.1317692111. Catuneanu, O., 2006. Principles of Sequence Stratigraphy. Elsevier, p. 375. Chen, Z.-Q., Benton, M.J., 2012. The timing and pattern of biotic recovery following the end-Permian mass extinction. Nat. Geosci. 5, 375–383. 1038/ngeo1475. Dong, W., Lin, S., Chen, Y., Yin, G., Jiao, H., Wang, K., Wang, H., Gan, X., Zhang, J., Xu, H., 1997. Lithostratigraphy of Guizhou Province. China University of Geosciences (Wuhan) Publishing House, p. 306. Erwin, D.H., 2006. Extinction: How Life on Earth Nearly Ended 250 Million Years Ago. Princeton Univ. Press, p. 320. Feng, Z., Bao, Z., Li, S., 1997. Lithofacies Paleogeography of the Middle and Lower Triassic of South China. Petroleum Industry Press, Beijing, pp. 1–222. Furin, S., Preto, N., Rigo, M., Roghi, G., Gianolla, P., Crowley, J.L., Bowring, S.A., 2006. High-precision U–Pb zircon age from the Triassic of Italy: implications for the Triassic time scale and the Carnian origin of calcareous nannoplankton and dinosaurs. Geology 34, 1009–1012. Galfetti, T., Bucher, H., Ovtcharova, M., Schaltegger, U., Brayard, A., Brühwiler, T., Goudemand, N., Weissert, H., Hochuli, P.A., Cordey, F., Guodun, K., 2007a. Timing of the Early Triassic carbon cycle perturbations inferred from new U–Pb ages and ammonoid biochronozones. Earth Planet. Sci. Lett. 258, 593–604. https:// Galfetti, T., Hochuli, P.A., Brayard, A., Bucher, H., Weissert, H., Vigran, J.O., 2007b. Smithian–Spathian boundary event: evidence for global climatic change in the wake of the end-Permian biotic crisis. Geology 35, 291–294. 1130/G23117A. Gianolla, P., Jacquin, T., 1998. Triassic sequence stratigraphic framework of western European basins. In: Mesozoic and Cenozoic Sequence Stratigraphy of European Basins. In: Soc. Sediment. Geol. Spec. Publ., pp. 643–650. Grasby, S.E., Beauchamp, B., Knies, J., 2016. Early Triassic productivity crises delayed recovery from world’s worst mass extinction. Geology 44, 779–782. https://doi. org/10.1130/g38141.1. Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctuating sea levels since the Triassic. Science 235, 1156–1167. Hardenbol, J., Thierry, J., Farley, M.B., Jacquin, T., Graciansky, P.-C.D., Vail, P.R., 1998. Mesozoic and Cenozoic sequence chronostratigraphic framework of European basins. In: Mesozoic and Cenozoic Sequence Stratigraphy of European Basins. In: Soc. Sediment. Geol. Spec. Publ., pp. 3–13. Hermann, E., Hochuli, P.A., Bucher, H., Brühwiler, T., Hautmann, M., Ware, D., Roohi, G., 2011. Terrestrial ecosystems on North Gondwana following the endPermian mass extinction. Gondwana Res. 20, 630–637. gr.2011.01.008. Hinnov, L.A., Goldhammer, R.K., 1991. Spectral-analysis of the Middle Triassic Latemar Limestone. J. Sediment. Petrol. 61, 1173–1193. Hinnov, L., Hilgen, F.J., 2012. Cyclostratigraphy and astrochronology. In: Gradstein, F., Ogg, J., Schmitz, M., Ogg, G. (Eds.), The Geologic Time Scale 2012. Elsevier, pp. 63–83. Hounslow, M.W., Muttoni, G., 2010. The geomagnetic polarity timescale for the Triassic: linkage to stage boundary definitions. In: Lucas, S.G. (Ed.), The Triassic Timescale. In: Geol. Soc. London Spec. Publ., pp. 61–102.


Husson, D., 2012. RedNoise_ConfidenceLevels. matlabcentral/fileexchange/45539-rednoise-confidencelevels?focused= 3817728&tab=function (October, 2017). Ikeda, M., Tada, R., 2014. A 70 million year astronomical time scale for the deep-sea bedded chert sequence (Inuyama, Japan): implications for Triassic– Jurassic geochronology. Earth Planet. Sci. Lett. 399, 30–43. 1016/j.epsl.2014.04.031. Jenks, J.F., Monnet, C., Balini, M., Brayard, A., Meier, M., 2015. Biostratigraphy of Triassic ammonoids. In: Klug, C., Korn, D., De Baets, K., Kruta, I., Mapes, R.H. (Eds.), Ammonoid Paleobiology: from Macroevolution to Paleogeography. Springer, pp. 329–388. Kent, D.V., Muttoni, G., Brack, P., 2004. Magnetostratigraphic confirmation of a much faster tempo for sea-level change for the Middle Triassic Latemar platform carbonates. Earth Planet. Sci. Lett. 228, 369–377. 2004.10.017. Kent, D.V., Olsen, P.E., Muttoni, G., 2017. Astrochronology polarity time scale (APTS) for the Late Triassic and Early Jurassic from continental sediments and correlation with standard marine stages. Earth-Sci. Rev. 166, 153–180. 10.1016/j.earscirev.2016.12.014. Kodama, K.P., Hinnov, L., 2015. Rock Magnetic Cyclostratigraphy. Wiley–Blackwell. Laskar, J., Fienga, A., Gastineau, M., Manche, H., 2011. La2010: a new orbital solution for the long-term motion of the Earth. Astron. Astrophys. 532. 1051/0004-6361/201116836. Lehrmann, D.J., Ramezani, J., Bowring, S.A., Martin, M.W., Montgomery, P., Enos, P., Payne, J.L., Orchard, M.J., Hongmei, W., Jiayong, W., 2006. Timing of recovery from the end-Permian extinction: geochronologic and biostratigraphic constraints from south China. Geology 34, 1053–1056. G22827A.1. Lehrmann, D.J., Ramezani, J., Bowring, S.A., Martin, M.W., Montgomery, P., Enos, P., Payne, J.L., Orchard, M.J., Hongmei, W., Jiayong, W., 2007. Timing of recovery from the end-Permian extinction: geochronologic and biostratigraphic constraints from south China: comment and reply. Geology 35, e136–e137. https:// Lehrmann, D.J., Stepchinski, L., Altiner, D., Orchard, M.J., Montgomery, P., Enos, P., Ellwood, B.B., Bowring, S.A., Ramezani, J., Wang, H., Wei, J., Yu, M., Griffiths, J.D., Minzoni, M., Schaal, E.K., Li, X., Meyer, K.M., Payne, J.L., 2015. An integrated biostratigraphy (conodonts and foraminifers) and chronostratigraphy (paleomagnetic reversals, magnetic susceptibility, elemental chemistry, carbon isotopes and geochronology) for the Permian–Upper Triassic strata of Guandao section, Nanpanjiang Basin, south China. J. Asian Earth Sci. 108, 117–135. 10.1016/j.jseaes.2015.04.030. Lehrmann, D.J., Wei, J., Enos, P., 1998. Controls on facies architecture of a large Triassic carbonate platform: the Great Bank of Guizhou, Nanpanjiang Basin, South China. J. Sediment. Res. 68, 311–326. Li, M., Huang, C., Hinnov, L., Ogg, J., Chen, Z.-Q., Zhang, Y., 2016a. Obliquity-forced climate during the Early Triassic hothouse in China. Geology 44, 623–626. Li, M., Ogg, J., Zhang, Y., Huang, C., Hinnov, L., Chen, Z.-Q., Zou, Z., 2016b. Astronomical-cycle scaling of the end-Permian extinction and the Early Triassic Epoch of South China and Germany. Earth Planet. Sci. Lett. 441, 10–25. https:// Li, M., Zhang, Y., Huang, C., Ogg, J., Hinnov, L., Wang, Y., Zou, Z., Li, L., 2017. Astronomical tuning and magnetostratigraphy of the Upper Triassic Xujiahe Formation of South China and Newark Supergroup of North America: implications for the Late Triassic time scale. Earth Planet. Sci. Lett. 475, 207–223. 10.1016/j.epsl.2017.07.015. Meyers, S.R., 2008. Resolving Milankovitchian controversies: the Triassic Latemar Limestone and the Eocene Green River Formation. Geology 36, 319. https://doi. org/10.1130/g24423a.1. Meyers, S., 2014. Astrochron: an R package for astrochronology. Available at: Meyers, S.R., Sageman, B.B., 2007. Quantification of deep-time orbital forcing by average spectral misfit. Am. J. Sci. 307, 773–792. 05.2007.01. Mundil, R., Brack, P., Meier, M., Rieber, H., Oberli, F., 1996. High resolution U–Pb dating of Middle Triassic volcaniclastics: time-scale calibration and verification of tuning parameters for carbonate sedimentation. Earth Planet. Sci. Lett. 141, 137–151. Mundil, R., Palfy, J., Renne, P.R., Brack, P., 2010. The Triassic timescale: new constraints and a review of geochronological data. In: Lucas, S.G. (Ed.), The Triassic Timescale. In: Geol. Soc. London Spec. Publ., pp. 41–60. Nesbitt, S.J., Sidor, C.A., Irmis, R.B., Angielczyk, K.D., Smith, R.M., Tsuji, L.A., 2010. Ecologically distinct dinosaurian sister group shows early diversification of Ornithodira. Nature 464, 95–98. Ogg, J.G., 2012. Triassic. In: Gradstein, F., Ogg, J., Schmitz, M., Ogg, G. (Eds.), The Geological Time Scale 2012. Elsevier, pp. 681–730, Chapter 25. Ogg, J.G., Ogg, G.M., Gradstein, F.M., 2016. The Concise Geologic Time Scale 2016. Elsevier. 234 pp. Olsen, P.E., Kent, P.E., Whiteside, J.H., 2011. Implications of the Newark Supergroupbased astrochronology and geomagnetic polarity time scale (Newark-APTS) for


M. Li et al. / Earth and Planetary Science Letters 482 (2018) 591–606

the tempo and mode of the early diversificiation of the Dinosauria. Earth Environ. Sci. Trans. R. Soc. Edinb. 101, 201–229. Ovtcharova, M., Goudemand, N., Hammer, Ø., Guodun, K., Cordey, F., Galfetti, T., Schaltegger, U., Bucher, H., 2015. Developing a strategy for accurate definition of a geological boundary through radio-isotopic and biochronological dating: the early–middle Triassic boundary (South China). Earth-Sci. Rev. 146, 65–76. Payne, J.L., Lehrmann, D.J., Christensen, S., Wei, J., Knoll, A.H., 2006. Environmental and biological controls on the initiation and growth of a Middle Triassic (Anisian) reef complex on the Great Bank of Guizhou, Guizhou Province, China. Palaios 21, 325–343. Payne, J.L., Lehrmann, D.J., Wei, J., Orchard, M.J., Schrag, D.P., Knoll, A.H., 2004. Large perturbations of the carbon cycle during recovery from the end-Permian extinction. Science 305, 506–509. Retallack, G.J., Veevers, J.J., Morante, R., 1996. Global coal gap between Permian– Triassic extinction and Middle Triassic recovery of peat-forming plants. Geol. Soc. Am. Bull. 108, 195–207. 108<0195:GCGBPT>2.3.CO;2. Retallack, G.J., 2013. Permian and Triassic greenhouse crises. Gondwana Res. 24, 90–103. Romano, C., Goudemand, N., Vennemann, T.W., Ware, D., Schneebeli-Hermann, E., Hochuli, P.A., Bruhwiler, T., Brinkmann, W., Bucher, H., 2013. Climatic and biotic upheavals following the end-Permian mass extinction. Nat. Geosci. 6, 57–60. Schmitz, M.D., 2012. Radiometric ages used in GTS2012. In: Gradstein, F.M., Ogg, J.G., Schmitz, M.D., Ogg, G.M. (Eds.), The Geologic Time Scale 2012. Elsevier Publ., Appendix 2. 1156 pp. Schnyder, J., Ruffell, A., Deconinck, J.-F., Baudin, F., 2006. Conjunctive use of spectral gamma-ray logs and clay mineralogy in defining late Jurassic–early Cretaceous palaeoclimate change (Dorset, UK). Palaeogeogr. Palaeoclimatol. Palaeoecol. 229, 303–320. Snedden, J., Liu, C., 2010. A compilation of Phanerozoic sea-level change, coastal onlaps and recommended sequence designations. Search Discovery Article 40594. snedden.pdf.html. Song, H., Wignall, P.B., Chen, Z.-Q., Tong, J., Bond, D.P.G., Lai, X., Zhao, X., Jiang, H., Yan, C., Niu, Z., Chen, J., Yang, H., Wang, Y., 2011. Recovery tempo and pattern of marine ecosystems after the end-Permian mass extinction. Geology 39, 739–742. Song, H., Wignall, P.B., Chu, D., Tong, J., Sun, Y., Song, H., He, W., Tian, L., 2014. Anoxia/high temperature double whammy during the Permian–Triassic marine crisis and its aftermath. Sci. Rep. 4, 4132. Stefani, M., Furin, S., Gianolla, P., 2010. The changing climate framework and depositional dynamics of Triassic carbonate platforms from the dolomites. Palaeogeogr. Palaeoclimatol. Palaeoecol. 290, 43–57. 2010.02.018. Stockar, R., Baumgartner, P., Condon, D., 2012. Integrated Ladinian biochronostratigraphy and geochronology of Monte San Giorgio (Southern Alps, Switzerland). Swiss J. Geosci. 105, 85–108. s00015-012-0093-5. Strasser, A., Hilgen, F.J., Heckel, P.H., 2006. Cyclostratigraphy – concepts, definitions, and applications. Newsl. Stratigr. 42, 75–114. 0078-0421/2006/0042-0075. Strasser, A., Pittet, B., Hug, W., 2015. Palaeogeography of a shallow carbonate platform: the case of the Middle to Late Oxfordian in the Swiss Jura Mountains. J. Palaeogeogr. 4, 251–268. Strasser, A., Védrine, S., Stienne, N., 2012. Rate and synchronicity of environmental changes on a shallow carbonate platform (Late Oxfordian, Swiss Jura Mountains). Sedimentology 59, 185–211.

Sun, Y., Joachimski, M.M., Wignall, P.B., Yan, C., Chen, Y., Jiang, H., Wang, L., Lai, X., 2012. Lethally hot temperatures during the Early Triassic greenhouse. Science 338, 366–370. Sun, Z., Hounslow, M.W., Pei, J., Zhao, L., Tong, J., Ogg, J.G., 2009. Magnetostratigraphy of the Lower Triassic beds from Chaohu (China) and its implications for the Induan–Olenekian stage boundary. Earth Planet. Sci. Lett. 279, 350–361. https:// Szurlies, M., 2007. Latest Permian to Middle Triassic cyclo-magnetostratigraphy from the Central European Basin, Germany: implications for the geomagnetic polarity timescale. Earth Planet. Sci. Lett. 261, 602–619. j.epsl.2007.07.018. ten Veen, J., Postmas, G., 1996. Astronomically forced variations in gamma-ray intensity: Late Miocene hemipelagic successions in the eastern Mediterranean basin as a test case. Geology 24, 15–18. (1996)024<0015:AFVIGR>2.3.CO;2. Thomson, D.J., 1982. Spectrum estimation and harmonic analysis. Proc. IEEE 70, 1055–1096. Trotter, J.A., Williams, I.S., Nicora, A., Mazza, M., Rigo, M., 2015. Long-term cycles of Triassic climate change: a new δ 18 O record from conodont apatite. Earth Planet. Sci. Lett. 415, 165–174. Tucker, M.E., Gallagher, J., Leng, M.J., 2009. Are beds in shelf carbonates millennialscale cycles? An example from the mid-Carboniferous of northern England. Sediment. Geol. 214, 19–34. Wang, H., Wang, X., Li, R., Wei, J., 2006. Triassic conodont succession and stage subdivision of the Guandao section, Bianyang, Luodian, Guizhou. Acta Palaeontol. Sin. 44, 611–626. Westphal, H., Hilgen, F., Munnecke, A., 2010. An assessment of the suitability of individual rhythmic carbonate successions for astrochronological application. EarthSci. Rev. 99, 19–30. Wignall, P.B., 2015. The Worst of Times: How Life on Earth Survived Eighty Million Years of Extinctions. Princeton University Press. 224 pp. Wotzlaw, J.-F., Brack, P., Storck, J.-C., 2017. High-resolution stratigraphy and zircon U–Pb geochronology of the Middle Triassic Buchenstein Formation (Dolomites, northern Italy): precession-forcing of hemipelagic carbonate sedimentation and calibration of the Anisian–Ladinian boundary interval. J. Geol. Soc. https://doi. org/10.1144/jgs2017-052. Wu, H., Zhang, S., Jiang, G., Hinnov, L., Yang, T., Li, H., Wan, X., Wang, C., 2013. Astrochronology of the Early Turonian–Early Campanian terrestrial succession in the Songliao Basin, northeastern China and its implication for long-period behavior of the Solar System. Palaeogeogr. Palaeoclimatol. Palaeoecol. 385, 55–70. Zhang, G., Zhang, X., Hu, D., Li, D., Algeo, T.J., Farquhar, J., Henderson, C.M., Qin, L., Shen, M., Shen, D., Schoepfer, S.D., Chen, K., Shen, Y., 2017. Redox chemistry changes in the Panthalassic Ocean linked to the end-Permian mass extinction and delayed Early Triassic biotic recovery. Proc. Natl. Acad. Sci. USA 114, 1806–1810. Zhang, Q., Zhou, C., Lu, T., Xie, T., Lou, X., Liu, W., Sun, Y., Huang, J., Zhao, L., 2009. A conodont-based Middle Triassic age assignment for the Luoping Biota of Yunnan, China. Sci. China, Ser. D 52, 1673–1678. 1007/s11430-009-0114-z. Zhang, Y., Li, M., Ogg, J.G., Montgomery, P., Huang, C., Chen, Z.-Q., Shi, Z., Enos, P., Lehrmann, D.J., 2015. Cycle-calibrated magnetostratigraphy of Middle Carnian from South China: implications for Late Triassic Time Scale and Termination of the Yangtze Platform. Palaeogeogr. Palaeoclimatol. Palaeoecol. 436, 135–166.