Orbital Forcing and Sea-Level Changes in the Earliest Triassic of the Meishan Section, South China

Home , Niobrara Formation

Journal of Earth Science, Vol. 25, No. 1, p. 64–73, February 2014 ISSN 1674-487X Printed in China DOI: 10.1007/s12583-014-0400-3

Shifeng Tian*1, 2, 3, Zhong-Qiang Chen4, Chunju Huang5 1. Key Laboratory of Tectonics and Petroleum Resources of Ministry of Education, China University of Geosciences, Wuhan 430074, China 2. Faculty of Earth Resources, China University of Geosciences, Wuhan 430074, China 3. State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum, Beijing 102249, China 4. State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Wuhan 430074, China 5. Faculty of Earth Sciences, China University of Geosciences, Wuhan 430074, China

ABSTRACT: The Earliest Triassic Yinkeng Formation is exposed at the well-known Meishan Section, South China, which contains the Global Stratotype of Section and Point (GSSP) for the Permian- Triassic boundary (PTB). It records centimeter-scale rhythmic alternations comprised mainly by marlstone and limestone. Seven types of couplet embedded in five types of bundles were recognized based on occurrence and thickness of the lithologic units, suggesting that their formation was controlled by cyclic processes. The various orders of cycles observed correlate well with other Early Triassic counterparts recorded in South China. Here, we present new cyclostratigraphic results based on lithologic thickness and relative carbonate content of the Yinkeng Formation. Power spectra of carbonate content show that the ratio of major wavelengths recognized throughout the formation is similar to that of the 100 kyr short eccentricity, 33 kyr obliquity, and 21 kyr precession cycles, indicating that astronomical signals are recorded in the Earliest Triassic rhythmic succession. Consistence between pronounced lithologic rhythmicity and sea-level changes obtained from Fischer plots indicates that high-frequency climatic cycles may have driven sea-level changes immediately after the PTB mass extinction. Fur- thermore, the 4th-order sea-level changes interpreted from the sedimentary record match well with 100 kyr short eccentricity component of carbonate content, reflecting that the 100 kyr short eccentricity- induced climate changes may have likely controlled the deposition of 4th-order sequences recognized from rhythmic successions. KEY WORDS: orbital cycle, limestone-marl alternation, carbonate content, Fischer plot, Yinkeng Formation.

1 INTRODUCTION extreme climatic and oceanic conditions prevailed throughout As a consequence of the most devastated biotic crisis of the Early Triassic (Chen and Benton, 2012; Sun et al., 2012; earth life at the end of the Permian, the Earliest Triassic marine Algeo et al., 2011). faunas are very rare worldwide (Chen and Benton, 2012; Erwin, Detecting cyclic sedimentation patterns and temporal pe- 1994). The depauperate nature of marine faunas suggests a riodicities offer geoscientists better understanding the origin of rather low bioturbation intensity in marine sediment, which sedimentary sequences and their driving mechanisms enables the excellent preservation of very fine rhythmic cyclic- (Paulissen and Luthi, 2011). Milankovitch signals have been ity in the Early Triassic succession. The well-preserved, detected in Early Triassic marine records (Wu et al., 2012; centimeter-scale rhythmic sediments are characteristic of the Huang et al., 2011; Guo et al., 2008; Li et al., 2007; Tong et al., Early Triassic successions in South China (Wu et al., 2012; 2007; Yang and Lehrmann, 2003; Hansen et al., 2000; Rampino Guo et al., 2008; Tong et al., 2007; Yang and Lehrmann, 2003; et al., 2000). At the GSSP Meishan Section, the centimeter- Lehrmann et al., 2001). The Early Triassic was a critical period scale, rhythmic alternation is pronounced in the Early Triassic witnessing the much delayed recovery of ecosystems following succession (Chen et al., 2007, 2002; Yin et al., 2001). Moreover, the End-Permian mass extinction. Many studies show that the biostratigraphy, chronostratigraphy, event stratigraphy and ecostratigraphy of the Latest Permian to Early Triassic of *Corresponding author: [email protected] Meishan have been well studied, and thus offer high-resolution ©China University of Geosciences and Springer-Verlag Berlin time constraints for further high-resolution cyclostratigraphic Heidelberg 2014 analysis (Chen et al., 2010; Zhang et al., 2007; Yin et al., 2001). Whereas high-resolution cyclostratigraphy and sequence strati- Manuscript received April 22, 2013. graphy have received far less attention. Tong (1997) attempted Manuscript accepted July 11, 2013. to recognize Milankovitch cyclicity in the Early Triassic of

Tian, S. F., Chen, Z. Q., Huang, C. J., 2014. Orbital Forcing and Sea-Level Changes in the Earliest Triassic of the Meishan Section, South China. Journal of Earth Science, 25(1): 64–73, doi:10.1007/s12583-014-0400-3 Orbital Forcing and Sea-Level Changes in the Earliest Triassic of the Meishan Section, South China 65

Meishan but facies homogeneity at outcrop exposure in the tion differentiated by lithology. Two to three units can usually lower Yangtze region prevents a high-resolution sedimen- be grouped into a couplet. These couplets are interpreted as the tological analysis (Tong and Yin, 1998; Zhang et al., 1997; Yin depositional response to one cycle of sea-level change. The and Tong, 1996). This means that it is difficult to verify orbital couplets stack into lower-order cycles, here termed bundle, forcing mechanism without quantitative analysis. Whether or- based on the pattern of upward facies change. In general, black bital forcing was a major climatic driver during this critical shale, claystone, greenish gray mudstone, and calcareous mud- period is crucial for the understanding of the controlling factors stone characterize the lower part of cycles whereas greenish and depositional mechanisms during the period of biotic recov- gray mudstone, muddy limestone/marlstone, and limestone the ery following the End-Permian environmental crisis. upper part. Hence, we interpret the couplets and bundles to Here, we present a case study of the Early Triassic represent shallowing-upward sequences on the basis of greater Yinkeng Formation exposed at the GSSP Meishan Section, argillaceous content and thinner beds in their lower parts, and South China. We aim to (1) distinguish the different orders of progressively increasing bedding thickness and decrease in embedded cycles and verify the origin of the high-frequency argillaceous content in the upper parts of the units. Of the iden- cyclicity; (2) employ a modified version of Fischer plots to tified 332 units, seven basic couplet types, 5 bundle types and define the high-frequency sea-level changes and compare them 158 couplets have been recognized. with the global eustatic sea-level changes; and (3) explore the potential causes of sea-level changes in the Earliest Triassic. 3.1.1 Basic couplets Couplets 1-3 (BC1-3) are all thickening-upward couplets 2 GEOLOGICAL SETTING which consist of two portions. The lower part is black shale, The GSSP Meishan Section is located at Meishan Town of capped by pale muddy limestone or gray marlstone, gray lime- Changxing County, Zhejiang Province, South China. The stone, and gray calcareous mudstone as the upper part, respec- Meishan area is situated in the eastern part of the South China tively. The shale is interpreted to be deposited in a quiet Block, which was a giant island in the eastern Paleo-Tethys deep-water environment with thickness ranging from 1–5 cm, Ocean near the tropics during the Permian-Triassic transition while the three lithologic types of the upper part of couplet (Ziegler et al., 1998). Unlike the rather heightened paleo- indicate relative shallow-water environments with thickness geographic contoural difference prior to the PTB biocrisis, the ranging 2–18, 4–11, and 3–24 cm, respectively. Among these Earliest Triassic paleogeography in South China was relatively three basic couplets, BC1 is very common across the entire uniform: massive carbonate ramp bounded with numerous plat- Yinkeng Formation while BC2 is rare and only develops in the forms and shelf basins mainly in the western part of the South uppermost of Yinkeng Formation. BC3 occurs mainly in the China Block (Zhang et al., 2007; Yin and Tong, 1996). The Early Triassic succession comprises the Yinkeng and Helongshan formations. The former conformably overlies above the platform carbonate rocks of the Latest Permian Changxing Formation. This formation (Bed 25 to Bed 60) is characterized by interbedded black shale, greenish gray mudstone, gray calcareous mudstones, gray marlstone, pale muddy limestone, and limestone (Figs. 1, 2). The overlying Helong- shan Formation is dominated by gray thin-bedded limestone. The Yinkeng Formation and the lower Helongshan Formation are of Griesbachian age (Chen et al., 2007). The Yinkeng For- mation is much better exposed than the Helongshan Formation and thus is studied in detail from the PTB to Bed 60 herein. Centimeter to decimeter-scale limestone-marl alternations are recognized throughout the Yinkeng Formation (Figs. 1b, 1c, 2). A total of 13.34 m stratigraphic thickness which is composed of 33 beds (332 smaller units) was examined vertically. The data collected consist of the lithological description and thickness of each unit.

3 DATA AND METHODS 3.1 Cycle Recognition Assigning high-frequency cyclicity in a hierarchical man- Figure 1. (a) Permian-Triassic succession exposed at the ner helps understanding the origin and driving mechanisms Meishan Section (Quarry B) showing the Permian-Triassic (Paulissen and Luthi, 2011). Based on the occurrence and ar- boundary (PTB), End-Permian mass extinction (EPME) rangement of lithotypes, two orders of cycles can be recognized horizon, and the topmost Yinkeng Formation; (b) and (c) throughout the Yinkeng Formation. The terminology used here mudstone and limestone alternations in the Upper Yinkeng for the sedimentary cycles, in ascending order, is unit, couplet, Formation. The pen is 11 cm long. and bundle. The unit is the smallest, centimeter-scale stratifica-

66 Shifeng Tian, Zhong-Qiang Chen and Chunju Huang

Sta- Fm.Bed Litho- Bundle Haq et al. Zhang et This study Fourth- Sta-Fm.Bed Litho- Bundle Haq et al. Zhang et This study Fourth- ge logy (1987) al. (1997) order ge logy (1987) al. (1997) order sequence sequence 62 Fall Rise Fall Rise HST FOS8 50 150 m 46 Bc3 TST Fm. 61

BC3 Helongshan 45 BB2 60 FOS7 BC2 44 BC7 59 BB5

FOS13 43 FOS6 58 BC4 BC7 57 BB4 42 BC3 56 FOS12 BC3 FOS5 60 BB2 55 40 41 54 20

Yinkeng Fm. Yinkeng

Griesbachian

Yinkeng Fm. Yinkeng

Griesbachian 0 cm

FOS4 40 BB4 FOS11 53 39 BB7 BS4 38 60 FOS3 37 40 52 36 20 FOS10 35 0 cm BC7 34 BC3 HST 33 FOS2 BB3 51 32 31 FOS9 BC7 30 50 B1C BC3 29 BB3 28 49 27 B1C FOS1 B1B BC3 26 PTB 48 25 FOS8

. . EPME 47 BC3 24 Fall Rise50 150 m Fall Rise BB2

Ch hs.

Claystone Black shale Calcareous Marlstone Muddy Limestone Siltstone mudstone limestone

Figure 2. Lower Triassic lithostratigraphy and third- and fourth-order sea-level changes from Haq et al. (1987), Zhang et al. (1997), and this study showing high-frequency cyclicities and the differences among the sea-level changes. The fourth-order sequences are also shown based on the dominant frequencies of orbital controls. Fm.. Formation; Ch.hs.. Changhsing; PTB. Permian-Triassic boundary; EPME. End-Permian mass extinction; BC. basic couplet; BB. basic bundle; FOS. fourth-order sequence; TST. transgressive system tract; HST. highstand system tract. middle Yinkeng Formation. muddy limestone/marlstone, and gray limestone in ascending Basic couplet 4 (BC4) consists of gray calcareous mud- order. It is rare and occurs occasionally in the uppermost stone at its lower part and pale muddy limestone or gray marl- Yinkeng Formation. stone at its upper part. Calcareous mudstone is 2–13 cm thick, Basic couplet 7 (BC7) comprises black shale, greenish gray and marlstone/muddy limestone 5–20 cm thick. These are in- or gray calcareous mudstone, and gray muddy limestone/ terpreted to represent a thickening- and shallowing-upward marlstone up the section. They are interpreted to represent a sequence and mainly occur in the upper Yinkeng Formation. thickening-up sequence. The sequence is 12 cm thick in average, Basic couplet 5 (BC5) is similar to basic couplet 4. They and largely develops in the middle Yinkeng Formation. both have the same lithology in the lower part. BC5, however, has gray limestone at its upper portion. This couplet type is 3.1.2 Basic Bundles only recognized in the upper Yinkeng Formation. Basic couplets are stacked vertically to form five different Basic couplet 6 (BC6) is formed by calcareous mudstone, types of basic bundles, which are interpreted to represent long

Orbital Forcing and Sea-Level Changes in the Earliest Triassic of the Meishan Section, South China 67 period, shallowing-up cycles (Fig. 2). Spectral analysis is utilized to investigate and evaluate the Two BC1s, 15–20 cm thick, characterize basic bundle 1 cyclicity in carbonate content data series. Prior to spectral (BB1), which develops in the basal Yinkeng Formation. analysis, we removed a long-term trend by using 15% weighted Basic bundle 2 (BB2) is composed of two BC3s, 5–45 cm average from carbonate content (Fig. 3). As to the cumulative thick with average thickness of 12 cm, which is thinnest among depature thickness of couplets which is used to construct the all basic bundles and present in the lower and middle parts of the Yinkeng Formation. Basic bundle 3 (BB3) comprises repetitive alternation of 0.8 (a) 0.6 BC3 and BC7 and is commonly present in the middle Yinkeng 0.4 Formation. The thickness varies between 12 and 27 cm. content Carbonate 0.2 Basic bundle 4 (BB4) consists of BC7 and BC4 at the Precession lower and upper parts, respectively. BB4, 12–23 cm thick, de- 0.1 velops mainly in the upper Yinkeng Formation. 0.0

Basic bundle 5 (BB5) is characterized by alternations of -0.1 out Filter BC7 and BC2, with an average thickness of 16 cm. It only oc- 0.1 Obliquity curs in the uppermost Yinkeng Formation. 0.0

3.2 Relative Carbonate Content out Filter -0.1 Eccentricity Although the exposure of the Early Triassic succession is 0.05 spectacular, it has received far less attention than the 0.00

well-studied PTB interval (Chen et al., 2007). Due to the cyclic -0.05 out Filter sediments in the Early Triassic, several proxies can be used for 0.2 cyclostratigraphic study. Of these, carbonate content is consid- 0.0 ered as numerical proxy herein. -0.2 The carbonate-rich sediments such as limestone and marls -0.4

After remove After were developed during the relative low sea-level stage. In con- average weighted 0 200 400 600 800 1 000 1 200 Stratigraphic thickness (cm, from PTB) trast, high sea-levels reduce the imported carbonate and con- 0.10 (b) tribute to the formation of mudstone/shale and calcareous mud- 102 cm 33.4 22.1 stone. Therefore, sediments from shallow-water are rich in 0.08 20.8 99% 95% carbonate and those from abyssal depths are absence/lack of 0.06 2 015.6 90% carbonate, meaning that carbonate content can reflect the M changes of water depth of depositional environments, which is Power 0.04 11.2 cm sensitive to the climate oscillations induced by Earth’s orbital 0.02 parameters. Consequently, when analyzed with high-resolution 0.00 carbonate contents can constitute a powerful tool for detecting 0.00 0.05 0.10 0.15 0.20 Frequency (cycles/cm) astronomically driven cyclicities reflecting carbonate productivity cycles and/or clay input cycles (Weedon, 2003; Hinnov, Figure 3. Carbonate content and its interpreted orbital 2000). components. (a) Top to bottom, curve 1. the calculated car- We calculated relative carbonate contents of the Yinkeng bonate content on a 4-cm regular spaced sampling, the dot- Formation on a 4-cm regular spaced sampling interval assuming ted line represents 15% weighted average of carbonate that the carbonate content of limestone, muddy limestone, marl- content; curves 2–4. band-pass-filtered 100, 33, and ~21 kyr stone, calcareous mudstone, shale, and clay is 90%, 65%, 50%, interpreted short eccentricity, obliquity, and precession 35%, 10%, and 5%, respectively. In Meishan, the carbonate con- signals of the carbonate content, Gaussian band-pass filters tent shows composite cyclicities quite prominently (Fig. 3). were centered at frequency of 0.009 8±0.003 cycles/cm (100

kyr), 0.03±0.007 cycles/cm (33 kyr), and 0.049±0.013 cycles/ 3.3 Spectral Analysis cm (~21 kyr); curve 5. the carbonate content series after Radiometric age dating at Meishan does not have the re- removing 15% weighted average (in Synergy Software’s quired resolution and accuracy in the Early Triassic in order to Kaleida Graph). (b) The 2π MTM (multitaper method) establish the potential imprint of driving force on cyclic sedi- power spectrum of the carbonate content (resample at step mentary record. Therefore, cyclostratigraphic analysis is un- of 2 cm before spectral analysis) in the depth domain, the dertaken based on quantitative description of high-frequency power spectrum estimated using the SSA-MTM Toolkit cycles recorded in the GSSP Meishan section. We selected (Ghil et al., 2002). Curve M is the median smoothed, fitted carbonate content as the numerical proxy to undertake cyc- red noise spectrum; the upper 90%, 95% and 99% confi- lostratigraphic analysis which can exhibit orbital cyclicity in dence levels are also shown. Significant peaks are labeled in sedimentary successions (Cleaveland et al., 2002; Prokoph and centimeters. The power spectrum of the carbonate content Thurow, 2000). Carbonate content has been widely applied to record shows that the ~102, 33.4, 22.1, 20.8, 20, 15.6, and detect Milankovitch cycles in strata records (Cleaveland et al., 11.2 cm peaks are statistically significant. 2002; Prokoph and Thurow, 2000; Hays et al., 1976).

68 Shifeng Tian, Zhong-Qiang Chen and Chunju Huang

nents is chosen and accomplished by AnalySeries 2.0.4 120 (Paillard et al., 1996). 100 80 3.4 Fischer Plot 60 Fischer plots in which cumulative cycle thickness cor- 40 15 20 rected for linear subsidence is plotted against time were origi- 10 0 nally used in the study of Triassic peritidal sequences to define -20 5 short-duration departures in relative sea level in the Milank-

Cumulative departure (cm) departure Cumulative 0 average ovitch range (Read and Goldhammer, 1988; Fischer, 1964). -5 Later, Fischer plots were developed to extract long-term

(a) weighted remove After sea-level changes from carbonate platform facies successions 0 200 400 600 800 1 000 1 200 (Soreghan, 1994; Read et al., 1991). Sadler et al. (1993) also Stratigraphic thickness (cm, from PTB) suggested that the horizontal axis of the plots should be labeled ~2 m 89 by cycle number instead of a time scale in order to avoid the 200 78 99% poor absolute time control on the stratigraphic record (Husinec 98 cm 95% 150 90% et al., 2008). Day (1997) plotted cumulative stratigraphic 49 M thickness as the horizontal axis rather than cycle number, which 100

Power 33 cm allows the Fischer plots to be drawn directly alongside strati- 50 (b) graphic columns to plot accommodation change (Husinec et al., 2008). In a word, the Fischer plot approach has been widely 0 0 0.01 0.02 0.03 0.04 0.05 0.06 used to analyze the high-frequency cyclicity from the Protero- Frequency (cycles/cm) zoic to the Holocene (Martín-Chivelet et al., 2000; Boss and Rasmussen, 1995; Sadler et al., 1993; Goldhammer et al., Figure 4. (a) The upper line represents modified Fischer 1990). Following Day (1997), the modified Fischer plot was plot of the Yinkeng Formation. Fischer diagram plot cu- obtained for the Yinkeng Formation of Meishan Section, in mulative departure from mean couplet thickness (vertical which cumulative deviation is plotted against the real, meas- axis) as a function of stratigraphic thickness (horizontal ured, cumulative thickness of the stratigraphic section (Fig. 4a). axis). We divided the research section into three parts by two points which are 4.72 and 9.67 m from PTB, to remove 4 RESULTS AND DISCUSSION 30% weighted average separately. These two points are 4.1 Origin of Cyclicity labeled as black stars. The lower line represents cumulative Major wavelength bands recognized in the Yinkeng For- departure thickness after removing 30% weighted average mation shows a hierarchy of cycles. Notable wavelengths with (in Synergy Software’s Kaleida Graph). (b) Spectral analy- spectral peaks exceeding the 99% confidence level of the mod- sis of the cumulative departure thickness in the SSA-MTM eled noise include ~102, 33.4, 22.1, 20.8, 20, 15.6, and 11.2 cm Toolkit (Ghil et al., 2002). Curve M and the upper 90%, (Fig. 3b). Among these wavelengths, the ratio of ~102, 33.4, 95%, 99% confidence levels are the same meanings as in and 20.8 cm is about 100 : 33 : 21, which is correlative to Early Fig. 3. Significant peak is labeled in centimeters. The power Triassic Milankovitch cycles of 100 kyr : 33 kyr : 20.7 kyr spectrum of the cumulative departure thickness shows that (short eccentricity, obliquity, and precession), and similar to the the ~200, 98, 89, 78, 49, and 33 cm peaks are statistically most recent model for the Earth’s orbital parameters (Laskar et significant. al., 2011). We thus consider that the sedimentary cyclicity in the Yinkeng Formation at Meishan Section was likely to be Fischer plot (Fig. 4a), the 30% weighted average was removed formed by orbital forcing. To explore the origin of high- separately from three divided sections of the Yinkeng Forma- frequency cyclicities, we cannot rule out some other factors. tion (Fig. 4a). Initial spectral analysis of the carbonate content A repetition of seven types of basic couplets embedded in using a simple Fast Fourier Transform (FFT) technique sug- five types of bundles characterizes the Yinkeng Formation suc- gests the possibility of cycle periods at the Milankovitch band. cession. Each basic couplet or bundle shows the superposition To improve the resolution of the spectral estimates, we apply of two or three distinct facies, which imply a deepening- several other Fourier techniques to the high-resolution carbon- shallowing character with argillaceous content decreasing and ate content data series, including Blackman-Tukey method bed thickness increasing upward (Li et al., 2007). Controls on (BTM) and more sophisticated techniques such as the Multita- the deepening-shallowing cycle formation generally include per method (MTM) of Thomson (Thomson, 1982). Both meth- tectonic activities, sedimentation processes, and/or eustatic ods are applied using the SSA-MTM Toolkit (Ghil et al., 2002) sea-level fluctuations. Two orders of cycles in the Yinkeng and the estimated spectra are tested against robust red noise Formation are laterally very extensive, and can be traced within models (Mann and Lees, 1996). We found that the results did 2 km in the field. Previous studies have distinguished almost not differ significantly between two methods. Therefore, the the same types of cycles in Lower Triassic in Chaohu, Anhui spectral analysis results presented here are based on the MTM. Province, South China (Guo et al., 2008; Li et al., 2007). The To aid in the recognition of orbital cycles, the Gaussian widespread, well ordered organization of the cycles and their band-pass filter used to extract the dominant spectral compo- hierarchy preclude an interpretation based on prevailing auto-

Orbital Forcing and Sea-Level Changes in the Earliest Triassic of the Meishan Section, South China 69 cyclic retrogradation-progration. Other genetic mechanisms, highstand systems tract (HST), respectively. The former is such subsidence pulsation of tectonic origin may also be ex- dominated by calcareous mudstone and argillaceous sediments. cluded in this case, because the Early Triassic is a time during Number of the shale- or mudstone-dominated couplets and which the subsidence was in its late stage within a passive con- thickness of shale or mudstone units increase up the section and tinental margin, characterized by very constant subsidence thus exhibits an upward-deepening trend where, the basic bun- rates. dles pass from BB1 gradually upwards to BB2 and BB3. This is Although the forming mechanisms of these cycles are still also accompanied by a marked upward decrease in carbonate not thoroughly understood, certain dominant frequencies have contents (Fig. 3) and increase in fossil abundance and diversity been recognized by previous work, which may be attributed to (Chen et al., 2007). These cycles occur mainly on the rising Milankovitch cycles (Tong, 1997). Reports on the PTB interval part of the modified Fischer plot curve (Fig. 2). As interpreted also raise the possibility that the Early Triassic Yinkeng Forma- in other TSTs, the above characteristics overall indicate that tion is astronomically forced stratigraphy (Wu et al., 2012; accommodation volume increases faster than carbonate produc- Huang et al., 2011; Algeo et al., 2010; Yin et al., 2001; Hansen tion and that retrogradation occurs. The end of TST occurs et al., 2000; Rampino et al., 2000). In order to probe into the when the rate of accommodation falls to match sediment pro- possible orbital-driven mechanism, we carried out spectral duction and supply, and most likely coincides with the onset of analysis of high-resolution time series of carbonate content to progradation. While the turnaround from retrogradational to determine the dominant frequency components and test progradational stacking pattern that marks the maximum whether Milankovitch orbital signals are recorded in the rhyth- flooding surface is not abrupt in this section. Instead, it is char- mic successions (Fig. 3). Comparing the relative ratio of the acterized by an interval of aggradational stacking. Conse- observed cycles in the sedimentary records with that of the quently, we recognized the maximum flooding surface (mfs) in Milankovitch cycles is the most commonly used methods to the middle of Bed 46 on the Fischer plot, where the inflection determine the potential astronomical forcing (Strasser et al., point of positive and negative slope is. The overall HST is char- 2006; Hinnov, 2000). In addition to the exclusion of autocyclic acterized by stacking patterns ranging from aggradation to gra- retrogradation-progration and tectonic activities, the strong dation. High argillaceous sediments like black shale, grayish similarity between the spectral analysis of carbonate content green mudstone and calcareous mudstone are still recognizable data series and the most recently calculated astronomical solu- in the aggradational part, where the succession mainly com- tion reveal that the high-frequency cyclicities recognized in the posed of BB2s and BB3s (Fig. 2). The bundles passing upwards Yinkeng Formation were astronomically driven. The major to BB4s with BB5 occurring occasionally in the uppermost cycle periods ~102, 33.4, and 20.8 cm were formed by short Yinkeng Formation shows that number of shale- or mudstone- eccentricity (100 kyr), obliquity (33 kyr), and precession (20.7 dominated bundles becomes less and thickness of the shale- or kyr), respectively. As to the other wavelengths corresponding to mudstone-unit in the bundles are thinner than TST (Fig. 2). the peak frequencies in Fig. 3b, we think they are formed by Moreover, the occurrence of relative shallow water limestone, noises, unrelated to orbital perturbations, or may represent spa- increasing carbonate content, and rare bioturbation collectively tial shifts in orbitally induced stratigraphic rhythms due to dominate the higher HST gradational part. sedimentation rate changes (Locklair and Sageman, 2008). Sea-level rise/fall is usually expressed as positive or negative departures in the Fischer plot (Martín-Chivelet et al., 2000; 4.2 Sequence Stratigraphy and Sea-Level Changes Day, 1997; Read and Goldhammer, 1988). Thus, the different The Fischer plot begins and ends at the same elevation but segments of positive and negative slope can reveal higher-order has the form of an irregular train of asymmetric waves, which sea-level changes. In order to express the higher-order sea-level rise gradually and fall steeply. Before interpreting the signifi- changes prominently, we adopted Boulila et al.’s proposal to cance of rises and falls we ought to know the basic mechanism construct the 4th-order eustatic sequence hierarchy based on the of the Fischer plot. If there is a long-term rise in sea level, rela- dominant frequencies of orbital controls (Boulila et al., 2011). tively thick cycles will develop because accommodation space The power spectrum shows that ~2 m, 98 cm, 89 cm, 78 cm, is increased due to linear subsidence. If there is a long-term fall, 48.6 cm, and 33 cm are statistically significant. The ratio of 98 thin cycles will develop because the falling sea-level decreases and 33 cm is about 100 : 33, which is also similar to Early Tri- the accommodation space. As Fischer plots rise where assic Milankovitch cycles of 100 kyr : 33 kyr (short eccentricity thicker-than-average cycles are grouped together, and they fall and obliquity). We thus extracted the interpreted 100 kyr short where there are runs of thinner cycles (Sadler et al., 1993), so, eccentricity component from the modified Fischer plot with the Fischer plot is able to define relative sea-level changes. Gaussian band-pass filters in Analyseries2.0.4.2 freeware cen- The overall shape of the wave derived from Meishan (Fig. tered at frequency of 0.01 cycles/m with bandwidth 0.003 (Pail- 2) is categorized as part of third-order sea-level change of Vail lard et al., 1996). Thirteen 4th-order sea-level changes were et al. (1984) in view of its duration inferred based on radiomet- recognized and numbered FOS1 (4th-order sequence) to FOS13 ric ages from Meishan (Mundil et al., 2004; Bowring et al., ascendingly (Fig. 2). 1998). Neither distinct erosional surfaces marking sequence boundaries nor pronounced transgressive surface is recogniz- 4.3 Comparison with Global Sea-Level Changes able from the outcrop. However, retrogradational and prograda- Effect of eustasy and tectonism on accommodation is dif- tional stacking patterns of high-frequency cycles are rather ficult to be discriminated and quantified. Eustatic changes distinct and characterize transgressive systems tracts (TST) and curve therefore is difficult to be constructed if only based on

70 Shifeng Tian, Zhong-Qiang Chen and Chunju Huang lithology and architecture of stratigraphic record. When the prove the time-resolution of the pattern of biotic recovery, but interplay between tectonism and eustasy is emphasized, marine also model and predict the future climate changes. accommodation changes are referred to as relative sea-level The 4th-order sea-level changes obtained from Fischer changes. Here, we inferred the relative sea-level changes curve plot shows characteristic of high-frequency oscillations, which from the Fischer plots (Fig. 4a). It combines the effects of gives way to demonstrate whether it's in tune with the ap- eustasy and tectonism, without attempting to discriminate their proximately 100 kyr short eccentricity or even higher- relative importance. frequency orbital signals. If sea-level changes were indeed Several global eustatic cycle charts of sea-level changes orbitally driven in the Milankovitch band, then power spectra have been published over the past thirty years, but the most of the calculated sea-level changes curve should also show the cited one was proposed by Haq et al. (1988). It has been cor- significant periodicities. We have already demonstrated the roborated by many independent studies with various ap- possible existence of orbital signals previously and extracted proaches (Stoll and Schrag, 2000; Pekar and Miller, 1996). Haq the interpreted 100 kyr short eccentricity. The 4th-order et al. (1988) indicated a major persisting long-term sea-level sea-level changes curve has the same peak numbers and can rise from Early to Middle Triassic. However, the coeval compares remarkably close with the 100 kyr short eccentricity long-term sea-level change curve derived from Meishan is modulation cycles (Fig. 5). Thus, we concluded that 100 kyr slightly from the global standard (Fig. 2). Like the global curve, short eccentricity induced climate changes translated into the local sea-level change curve from Meishan also shows a sea-level oscillations, which in turn controlled the formation of rise in sea level during the earliest Triassic, but the different lies 4th-order sequences with different lithological expressions. in that the transgression reached its maximum in the middle This observation also supports the view that fourth and higher Early Triassic followed by a regression and sea level falling. orders sea-level sequences are controlled by precession- This difference may attribute to not considering the relative eccentricity in Mesozoic greenhouse (Strasser et al., 2006, importance of tectonism. However, even if the tectonism is 1999). Finally, it’s important to note that the simple relationship stable, the sea-level curve we obtained may or may not corre- between 4th-order sea-level changes and astronomical cycles late with the global sea-level changes curve (Haq et al., 1988, does not rule out other stochastic processes such as plate-scale 1987) because the curve of Haq et al. (1988, 1987) is based on collisional events causing slow sea-level change. Although the data from West Europe, North America, Russia and Gondwana, which is clearly different from the case of the Eurasian Tethys 0.4 (Yin and Tong, 1996). In particular, the relative sea-level 0.2 0.1 changes curve in the Yangtze region from the Late Permian to Early Triassic is different from the global eustasy curve which 0.0 may be caused by the regional or local tectonic subsidence. -0.2 0.0 -0.4 4.4 Potential Causes of Sea-Level Changes out Filter

Sea-level changes at a variety of scales have prompted (%) content Carbonate -0.6 numerous researches for their potential causes. The widely accepted opinions are that first- and second-order sequences are -0.1 attributed to tectono-eustatic changes while the fourth and 3 higher orders depositional sequences are linked to climato- 15 2 eustatic changes driven by Earth’s orbital parameters (Boulila 1 10 et al., 2011). Third-order sequences were often related to both 0

climatic and tectonic influences (Vail et al., 1991), but many 5 -1 out Filter recent studies have proposed orbitally forced climate is the -2 0 major driving mechanism of third-order sea-level changes be- -3 cause tectonics fail to explain global sea-level changes (Boulila -5 -4 et al., 2011; Strasser et al., 2000). Sea-level change is a domi- -10

nant control on sedimentary processes and sedimentation pat- changes sea-level Fourth-order 0 200 400 600 800 1 000 1 200 terns and has the capacity to produce rhythmic stratigraphic Stratigraphic thickness (cm, from PTB) sequences. Especially, sea-level changes resulting from varia- Figure 5. Relationship between fourth-order sea-level tions of polar ice volume is potentially one of the biggest cause changes defined by extracting the interpreted 100 kyr short for future climate change (Andrews et al., 2007). Although eccentricity components of the modified Fischer plot and previous studies have proposed that astronomical signals are the carbonatecontent depth record. The fourth-order registered in the Lower Triassic strata at different sections in sea-level changes and interpreted 100 kyr short eccentricity South China (Wu et al., 2012; Huang et al., 2011; Guo et al., components were extracted with Gaussian band-pass filters 2008; Li et al., 2007; Yang and Lehrmann, 2003; Tong, 1997), in Analyseries2.0.4.2 freeware centered at frequency of 0.01 the relationship between astronomical forcing and sea-level cycles/cm with bandwidth 0.003 and frequency of 0.009 8 changes in Early Triassic has not been taken into account. To cycles/m with bandwidth 0.004 5, respectively (Paillard et further clarify the origin and decipher the effects of astronomi- al., 1996). cal forcing on sea-level changes perhaps can help not only im-

Orbital Forcing and Sea-Level Changes in the Earliest Triassic of the Meishan Section, South China 71 orbitally induced eustatic changes are global, their lithological Bowring, S. A., Erwin, D. H., Jin, Y. G., et al., 1998. U/Pb Zircon expression depends strongly on regional and local factors be- Geochronology and Tempo of the End-Permian Mass Ex- cause the sequences may be enhanced or attenuated by the re- tinction. Science, 280(5366): 1039–1045, doi: gional subsidence pattern. 10.1126/science.280.5366.1039 Chen, Z. Q., Benton, M. J., 2012. The Timing and Pattern of Bi- 5 CONCLUSIONS otic Recovery Following the End-Permian Mass Extinction. High-frequency centimeter-scale cycles characterize the Nature Geoscience, 5(6): 375–383, doi:10.1038/NGEO1475 lowest Triassic Yinkeng Formation of the GSSP Meishan Sec- Chen, Z. Q., Shi, G. R., Kaiho, K., 2002. A New Genus of Rhyn- tion, South China. Total seven types of couplets and five types chonellid Brachiopod from the Lower Triassic of South Chi- of bundles are recognizable from the Earliest Triassic succes- na and Implications for Timing the Recovery of Brachiopoda sion. Power spectral analysis of carbonate contents shows evi- after the End-Permian Mass Extinction. Palaeontology, 45(1): dence of the Earth’s orbital variations, with short eccentricity, 149–164, doi:10.1111/1475-4983.00231 obliquity, and precession as the main frequencies. Comparison Chen, Z. Q., Tong, J. N., Kaiho, K., et al., 2007. Onset of Biotic between third-order sea-level changes obtained from Fischer and Environmental Recovery from the End-Permian Mass plot and the global eustatic changes exists some differences, Extinction within 1–2 Million Years: A Case Study of the which means that the eustasy is not exactly worldwide syn- Lower Triassic of the Meishan Section, South China. Pa- chronous and the differences may be caused by the local tec- laeogeography, Palaeoclimatology, Palaeoecology, tonic subsidence effects. Finally, the remarkable relationship 252(1–2): 176–187, doi:10.1016/j.palaeo.2006.11.042 between the peaks of 4th-order sea-level changes and inter- Chen, Z. Q., Tong, J. N., Liao, Z. T., et al., 2010. Structural preted 100 kyr short eccentricity prove that orbitally induced Changes of Marine Communities over the Permian-Triassic climate changes translated into 4th-order sea-level oscillations Transition: Ecologically Assessing the End-Permian Mass in Meishan area in the Earliest Triassic, which will improve the Extinction and Its Aftermath. Global and Planetary Change, relative time-resolution of stratigraphic intervals. 73(1–2): 123–140, doi:10.1016/j.gloplacha.2010.03.011 Cleaveland, L. C., Jensen, J. L., Goese, S., et al., 2002. Cyc- ACKNOWLEDGMENTS lostratigraphic Analysis of Pelagic Carbonates at Monte dei This study was jointly supported by the National Program Corvi (Ancona, Italy) and Astronomical Correlation of the on Key Basic Research Project (No. 2012CB822000), the Na- Serravallian-Tortonian Boundary. Geology, 30(10): 931–934, tional Natural Science Foundation of China (No. 41302113), doi:10.1130/0091-7613(2002)030<0931:CAOPCA>2.0.CO; China Postdoctoral Science Foundation (No. 2013M531765), 2 the Foundation of State Key Laboratory of Petroleum Re- Day, P. I., 1997. The Fischer Diagram in the Depth Domain: A sources and Prospecting, China University of Petroleum, Bei- Tool for Sequence Stratigraphy. Journal of Sedimentary Re- jing (No. PRP/open-1301), the Project of Key Laboratory of search, 67(5): 982–984, doi:10.2110/jsr.67.982 Tectonics and Petroleum Resources of Ministry of Education, Erwin, D. H., 1994. The Permo-Triassic Extinction. Nature, 367: China University of Geosciences (No. TPR-2012-06), and the 231–236, doi:10.1038/367231a0 Fundamental Research Funds for the Central Universities of Fischer, A. G., 1964. The Lofer Cyclothems of the Alpine Triassic. China (No. 14CX02034A). Kansas Geological Survey Bulletin, 38(1): 107–149 Ghil, M., Allen, M. R., Dettinger, M. D., et al., 2002. Advanced REFERENCES CITED Spectral Methods for Climatic Time Series. Reviews of Geo- Algeo, T. J., Hinnov, L., Moser, J., et al., 2010. Changes in Pro- physics, 40(1): 1–41, doi:10.1029/2001rg000092 ductivity and Redox Conditions in the Panthalassic Ocean Goldhammer, R. K., Dunn, P. A., Hardie, L. A., 1990. Deposi- during the Latest Permian. Geology, 38(2): 187–190, doi: tional Cycles, Composite Sea-Level Changes, Cycle Stack- 10.1130/g30483.1 ing Patterns, and the Hierarchy of Stratigraphic Forcing: Algeo, T. J., Kuwahara, K., Sano, H., et al., 2011. Spatial Variation Examples from Alpine Triassic Platform Carbonates. Geo- in Sediment Fluxes, Redox Conditions, and Productivity in logical Society of America Bulletin, 102(5): 535–562, doi: the Permian-Triassic Panthalassic Ocean. Palaeogeography, 10.1130/0016-7606(1990)102<0535:dccslc>2.3.co;2 Palaeoclimatology, Palaeoecology, 308(1–2): 65–83, doi: Guo, G., Tong, J. N., Zhang, S. X., et al., 2008. Cyclostratigraphy 10.1016/j.palaeo.2010.07.007 of the Induan (Early Triassic) in West Pingdingshan Section, Andrews, J. E., Portman, C., Rowe, P. J., et al., 2007. Sub-Orbital Chaohu, Anhui Province. Science in China Series D: Earth Sea-Level Change in Early MIS 5e: New Evidence from the Sciences, 51(1): 22–29, doi:10.1007/s11430-007-0156-z Gulf of Corinth, Greece. Earth and Planetary Science Letters, Hansen, H. J., Lojen, S., Toft, P., et al., 2000. Magnetic Suscepti- 259(3–4): 457–468, doi:10.1016/j.epsl.2007.05.005 bility and Organic Carbon Isotopes of Sediments across Boss, S. K., Rasmussen, K. A., 1995. Misuse of Fischer Plots as Some Marine and Terrestrial Permo-Triassic Boundaries. In: Sea-Level Curves. Geology, 23(3): 221–224, doi: Yin, H. F., Dickins, J. M., Shi, G. R., et al., eds., Permian- 10.1130/0091-7613(1995)023<0221:MOFPAS>2.3.CO;2 Triassic Evolution of Tethys and Western-Circum Pacific. Boulila, S., Galbrun, B., Miller, K. G., et al., 2011. On the Origin Elsevier, Amsterdam. 271–289 of Cenozoic and Mesozoic “Third-Order” Eustatic Se- Haq, B. U., Hardenbol, J., Vail, P. R., 1987. Chronology of Fluc- quences. Earth-Science Reviews, 109(3–4): 94–112, doi: tuating Sea Levels since the Triassic. Science, 235(4793): 10.1016/j.earscirev.2011.09.003 1156–1167, doi:10.1126/science.235.4793.1156

72 Shifeng Tian, Zhong-Qiang Chen and Chunju Huang

Haq, B. U., Hardenbol, J., Vail, P. R., 1988. Mesozoic and Ceno- Prokoph, A., Thurow, J., 2000. Diachronous Pattern of Milank- zoic Chronostratigraphy and Cycles of Sea-Level Change. ovitch Cyclicity in Late Albian Pelagic Marlstones of the SEPM Special Publication, 42: 71–108 North German Basin. Sedimentary Geology, 134(3): Hays, J. D., Imbrie, J., Shackleton, N. J., 1976. Variations in the 287–303, doi:10.1016/S0037-0738(00)00050-6 Earth’s Orbit: Pacemarker of the Ice Age. Science, 194(4270): Rampino, M. R., Prokoph, A., Adler, A., 2000. Tempo of the 1121–1132, doi:10.1126/science.194.4270.1121 End-Permian Event: High-Resolution Cyclostratigraphy at Hinnov, L. A., 2000. New Perspectives on Orbitally Forced Stra- the Permian-Triassic Boundary. Geology, 28(7): 643–646, tigraphy. Annual Review of Earth and Planetary Science, doi:10.1130/0091-7613(2000)28<643:TOTEEH>2.0.CO;2 28(1): 419–475, doi:10.1146/annurev.earth.28.1.419 Read, J. F., Goldhammer, R. K., 1988. Use of Fischer Plots to Huang, C. J., Tong, J. N., Hinnov, L., et al., 2011. Did the Great Define Third-Order Sea-Level Curves in Ordovician Peritidal Dying of Life Take 700 k.y.? Evidence from Global Astro- Cyclic Carbonates, Appalachians. Geology, 16(10): 895–899, nomical Correlation of the Permian-Triassic Boundary Inter- doi:10.1130/0091-7613(1988)016<0895:UOFPTD>2.3.CO;2 val. Geology, 39(8): 779–782, doi:10.1130/g32126.1 Read, J. F., Koerschner, W. F., Osleger, D. A., et al., 1991. Field Husinec, A., Basch, D., Rose, B., et al., 2008. FISCHERPLOTS: and Modelling Studies of Cambrian Carbonate Cycles, Vir- An Excel Spreadsheet for Computing Fischer Plots of Ac- ginia Appalachians: Reply. Journal of Sedimentary Petrology, commodation Change in Cyclic Carbonate Successions in 61(4): 647–652 both the Time and Depth Domains. Computers & Geo- Sadler, P. M., Osleger, D. A., Montañez, I. P., 1993. On the Label- sciences, 34(3): 269–277, doi:10.1016/j.cageo.2007.02.004 ing, Length, and Objective Basis of Fischer Plots. Journal of Laskar, J., Fienga, A., Gastineau, M., et al., 2011. La2010: A New Sedimentary Research, 63(3): 360–368, doi: Orbital Solution for the Long-Term Motion of the Earth. As- 10.1306/D4267AFF-2B26-11D7-8648000102C1865D tronomy & Astrophysics, 532: A89, Soreghan, G. S., 1994. Stratigraphic Responses to Geologic Proc- doi:10.1051/0004-6361/201116836 esses: Late Pennsylvanian Eustasy and Tectonics in the Pe- Lehrmann, D. J., Yang, W., Wei, J., et al., 2001. Lower Triassic dregosa and Orogrande Basins, Ancestral Ricky Mountains. Peritidal Cyclic Limestone: An Example of Anachronistic Geological Society of America Bulletin, 106(9): 1195–1211, Carbonate Facies from the Great Bank of Guizhou, Nanpan- doi:10.1130/0016-7606(1994)106<1195:srtgpl>2.3.co;2 jiang Basin, Guizhou Province, South China. Palaeogeogra- Stoll, H. M., Schrag, D. P., 2000. High-Resolution Stable Isotope phy, 173(3): 103–123, doi:10.1016/S0031-0182(01)00302-9 Records from the Upper Cretaceous Rocks of Italy and Spain: Li, S., Tong, J. N., Liu, K., et al., 2007. The Lower Triassic Cyclic Glacial Episodes in a Greenhouse Planet? Geological Society Deposition in Chaohu, Anhui Province, China. Palaeogeogra- of America Bulletin, 112(2): 308–319, doi: phy, 252(1): 188–199, doi:10.1016/j.palaeo.2006.11.043 10.1130/0016-7606(2000)112<308:hsirft>2.0.co;2 Locklair, R. E., Sageman, B. B., 2008. Cyclostratigraphy of the Strasser, A., Hilgen, F. J., Heckel, P. H., 2006. Cyclostratigraphy- Upper Cretaceous Niobrara Formation, Western Interior, Concepts, Definitions, and Applications. Newsletters on Stra- U.S.A.: A Coniacian–Santonian Orbital Timescale. Earth and tigraphy, 42(2): 75–114, doi: Planetary Science Letters, 269(3–4): 540–553, doi: 10.1127/0078-0421/2006/0042-0075 10.1016/j.epsl.2008.03.021 Strasser, A., Hillgärtner, H., Hug, W., et al., 2000. Third-Order Mann, M. E., Lees, J. M., 1996. Robust Estimation of Background Depositional Sequences Reflecting Milankovitch Cyclicity. Noise and Signal Detection in Climatic Time Series. Climatic Terra Nova, 12(6): 303–311, doi: Change, 33(3): 409–445, doi:10.1007/BF00142586 10.1046/j.1365-3121.2000.00315.x Martín-Chivelet, J., Osleger, D. A., Montañez, I. P., 2000. Modi- Strasser, A., Pittet, B., Hillgärtner, H., et al., 1999. Depositional fied Fischer Plots as Graphical Tools for Evaluating Thick- Sequences in Shallow Carbonate Dominated Sedimentary ness Patterns in Stratigraphic Successions. Journal of Geo- Systems: Concepts for a High-Resolution Analysis. Sedi- science Education, 48(2): 179–183 mentary Geology, 128(3): 201–221, doi: Mundil, R., Ludwig, K. R., Metcalfe, I., et al., 2004. Age and 10.1016/S0037-0738(99)00070-6 Timing of the Permian Mass Extinctions: U/Pb Dating of Sun, Y., Joachimski, M. M., Wignall, P. B., et al., 2012. Lethally Closed-System Zircons. Science, 305(5691): 1760–1763, doi: Hot Temperatures during the Early Triassic Green House. 10.1126/science.1101012 Science, 338(6105): 366–370, doi:10.1126/science.1224126 Paillard, D., Labeyrie, L. D., Yiou, P., 1996. Macintosh Program Thomson, D. J., 1982. Spectrum Estimation and Harmonic Analy- Performs Time-Series Analysis. EOS, Transactions American sis. Proceedings of the IEEE, 70(9): 1055–1096, doi: Geophysical Union, 77(39): 379, doi:10.1029/96EO00259 10.1109/PROC.1982.12433 Paulissen, W. E., Luthi, S. M., 2011. High-Frequency Cyclicity in Tong, J. N., 1997. A Study on the Griesbachian Cyclostratigraphy a Miocene Sequence of the Vienna Basin Established from of Meishan Section, Changxing, Zhejiang Province. Sympo- High-Resolution Logs and Robust Chronostratigraphic Tun- sium of the International Conference on Stratigraphy and ing. Palaeogeography, 307(1–4): 313–323, doi: Tectonic Evolution of Southeast Asia and the South Pacific. 10.1016/j.palaeo.2011.05.029 Bangkok, Thailand. 158–163 Pekar, S., Miller, K. G., 1996. New Jersey Oligocene “Icehouse” Tong, J. N., Yin, H. F., 1998. The Marine Triassic Sequence Sequences (ODP Leg 150X) Correlated with Global δ18O Stratigraphy of Lower Yangtze. Science in China Series D: and Exxon Eustatic Records. Geology, 24(6): 567–570, doi: Earth Sciences, 41(3): 255–261, doi:10.1007/BF02973113 10.1130/0091-7613(1996)024<0567:njoiso>2.3.co;2 Tong, J. N., Zhang, S. X., Zuo, J. X., et al., 2007. Events during

Orbital Forcing and Sea-Level Changes in the Earliest Triassic of the Meishan Section, South China 73

Early Triassic Recovery from the End-Permian Extinction. in Lower Triassic (Olenekian) Peritidal Carbonate Succes- Global and Planetary Change, 55(1): 66–80, doi: sions, Nanpanjiang Basin, South China. Palaeogeography, 10.1016/j.gloplacha.2006.06.015 201(3–4): 283–306, doi:10.1016/s0031-0182(03)00614-x Vail, P. R., Audemard, F., Bowman, S. A., et al., 1991. The Strati- Yin, H. F., Tong, J. N., 1996. Late Permian–Middle Triassic graphic Signatures of Tectonics, Eustasy and Sedimentology: Sea-Level Changes of Yangtze Platform. Journal of China An Overview. In: Einsele, G., Ricken, W., Seilacher, A., eds. University of Geosciences, 7(1): 101–104 Cycles and Events in Stratigraphy. Springer, Berlin. 617–659 Yin, H. F., Zhang, K. X., Tong, J. N., et al., 2001. The Global Vail, P. R., hardenbol, J., Todd, R. G., 1984. Jurassic Unconformi- Stratotype Section and Point (GSSP) of the Permian-Triassic ties, Chronostratigraphy, and Sea-Level Changes from Seis- Boundary. Episodes, 24(2): 102–114 mic Stratigraphy and Biostratigraphy. In: Schlee, J. S., ed., Zhang, K. X., Tong, J. N., Shi, G. R., et al., 2007. Early Triassic Interregional Unconformities and Hydrocarbon Accumula- Conodont-Palynological Biostratigraphy of the Meishan D tion. American Association of Petroleum Geologists, Tulsa. Section in Changxing, Zhejiang Province, South China. 129–144 Palaeogeography, 252(1–2): 4–23, Weedon, G. P., 2003. Time-Series Analysis and Cyclostratigraphy: doi:10.1016/j.palaeo.2006.11.031 Examining Stratigraphic Records of Environmental Cycles. Zhang, K. X., Tong, J. N., Yin, H. F., et al., 1997. Sequence Strati- Cambridge University Press, New York graphy of the Permian-Triassia Boundary Section of Wu, H., Zhang, S., Feng, Q., et al., 2012. Milankovitch and Changxing, Zhejiang, Southern China. Acta Geologica Sub-Milankovitch Cycles of the Early Triassic Daye Forma- Sinica (English Edition), 71(1): 90–103, doi: tion, South China and Their Geochronological and Paleocli- 10.1111/j.1755-6724.1997.tb00349.x matic Implications. Gondwana Research, 22(2): 748–759, Ziegler, A. M., Gibbs, M. T., Hulver, M. L., 1998. A Mini-Altas of doi:10.1016/j.gr.2011.12.003 Oceanic Water Masses in the Permian Period. Proceedings of Yang, W., Lehrmann, D. J., 2003. Milankovitch Climatic Signals Royal Society of Victoria, 110: 323–343