The Floating Astronomical Time Scale for the Terrestrial Late Cretaceous

Earth and Planetary Science Letters 278 (2009) 308–323

Contents lists available at ScienceDirect

Earth and Planetary Science Letters

journal homepage: www.elsevier.com/locate/epsl

The ﬂoating astronomical time scale for the terrestrial Late Cretaceous Qingshankou Formation from the Songliao Basin of Northeast China and its stratigraphic and paleoclimate implications

Huaichun Wu a,b,⁎, Shihong Zhang a, Ganqing Jiang c, Qinghua Huang d a State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Beijing, 100083, China b School of Marine Science, China University of Geosciences, Beijing, 100083, China c Department of Geoscience, University of Nevada, Las Vegas, NV 89154-4010, U.S.A. d Exploration and Development Research Institute of Daqing Oil Field Corporation Ltd., Daqing, Heilongjiang, 163712, China article info abstract

Article history: The Upper Cretaceous Qingshankou Formation (K2qn) in the Songliao Basin (SLB) of Northeast China consists Received 19 May 2008 of up to 550 m thick, lacustrine mudstone and shale that constitute one of the most important source rocks of Received in revised form 14 October 2008 the Daqing oil ﬁeld. A high-resolution cyclostratigraphic analysis of the natural gamma-ray logging from 10 Accepted 8 December 2008 wells of the Qingshankou Formation (K qn) reveals orbital cycles of precession (20 ka), obliquity (40 ka) and Available online 21 January 2009 2 eccentricity (100 ka and 405 ka), providing strong evidence for astronomically driven climate changes in the Editor: P. DeMenocal Late Cretaceous terrestrial environments. Floating astronomical time scales (ATS) are established for all

sections, which demonstrate variable durations of K2qn across the basin (1.09 Ma–5.20 Ma) and strong Keywords: diachroneity of the lacustrine strata. Four periods of high depositional rates can be identiﬁed in the central Late Cretaceous parts of the basin, possibly recording deposition during times of sustained wet climate and high chemical Milankovitch cycles weathering. An ATS established from well M206 in the central depression zone of the basin, where the most astronomical time scale (ATS) complete and stable Milankovitch cycles are present, suggests that the maximum duration of the K2qn is lacustrine anoxic event 1 (LAE1) 5.20 Ma (from 94.27 Ma to 89.07 Ma; Late Cenomanian to Early Coniacian). The lacustrine anoxic event 1 Songliao Basin (LAE1) at the Cenomanian–Turonian boundary lasted ~210–310 ka, during which the most proliﬁc petroleum source rocks in SLB were deposited. The onset (~94.21–94.18 Ma) and duration (~210–310 Ka) of LAE1 in SLB are comparable to those of the oceanic anoxic event 2 (OAE2; onset at 94.21 Ma and duration of ~320–

900 ka), suggesting that the same trigger mechanism, such as increased atmospheric CO2 from large-scale igneous activity, may have initiated high primary productivity and organic carbon burial in both marine and terrestrial systems. © 2009 Elsevier B.V. All rights reserved.

1. Introduction graphic events including OAEs (e.g. Fiet et al., 2006; Li et al., 2008; Locklair and Sageman, 2008; Mitchell et al., 2008). With the The Cretaceous represents one of the warmest periods of the progresses made in the last decade, it is anticipated that an ATS Phanerozoic eon and a time with signiﬁcant ocean anoxic events covering the entire Cretaceous Period will be completed in a few years (OAEs). Establishing a high-precision chronostratigraphic framework (Hinnov and Ogg, 2007). The existing ATS, however, is restricted to is fundamental for a better understanding of the global climate change marine strata. There is an immediate need to establish the Cretaceous under the Cretaceous supergreenhouse conditions. Owing to the ATS from terrestrial sedimentary basins so that the marine and limitations of radiometric ages and paleontological data in general, the terrestrial records can be compared and a better understanding of astronomical time scale (ATS), established by tuning astronomical the Cretaceous Earth system change can be achieved. forcing signals recorded in sedimentary strata, plays critical roles for Continental rift basins are unique repositories for long-term deﬁning and correlating the Cretaceous paleoclimatic/paleoceano- palaeoclimate records (Olsen and Kent, 1999). Lacustrine mudstone and shales deposited in continental rift basins, due to their sensitivity to changes in precipitation–evaporation ratios and/or water-level changes, are particularly suitable for high-resolution cyclostratigraphic studies ⁎ Corresponding author. School of Marine Science, China University of Geosciences, (e.g., Olsen and Kent, 1999; Prokoph and Agterberg, 2000) and for Xueyuan Road 29, Haidian District, Beijing 100083, China. Tel.: +86 10 82334699; fax: +86 10 82320065. establishing astronomical time scales with resolution potentially down E-mail addresses: [email protected], [email protected] (H. Wu). to 0.02 to 0.40 Ma (e.g., Hinnov, 2004; Hinnov and Ogg, 2007).

The Songliao Basin in northeastern China is one of the largest Daqing Oil ﬁeld (Li, 1995)(Figs. 1d and 2). Unconformably overlying

Cretaceous continental rift basins in the world (Fig. 1). Well-preserved the K2n, the Sifangtai (K2s) and Mingshui (K2m) Formations are Cretaceous lacustrine deposits in this basin provide a unique mainly composed of siliciclastic rocks deposited from alluvial and opportunity for the construction of terrestrial Cretaceous ATS. deltaic environments during basin inversion.

Particularly, the Upper Cretaceous Qingshankou Formation (K2qn) The Qingshankou Formation (K2qn) was deposited during the post- consists of up to 550 m thick, black mudstone, shale and oil shale that rift thermal subsidence stage. The thickness of this unit varies from 30 to have attracted considerable attention regarding their potential linkage 550 m across the basin (Fig. 2). It conformably overlies the Quantou with Milankovitch climate forcing (e.g., Wu et al., 2007; Cheng et al., Formation (K2q) and is unconformably or conformably overlain by the 2008). The basal interval of this unit is enriched in oil shale and was Yaojia Formation (K2y). On the basis of lithofacies changes, the K2qn can 1 considered as deposits during a lacustrine anoxic event (Huang et al., be divided into three members (Fig. 1d). Member 1 (K2qn ) is mainly 1998, 2007). This event, referred to as lacustrine anoxic event 1 composed of deep lacustrine black mudstone and shale. The oil shale at 1 (LAE1), has been suspected to be time equivalent to the oceanic anoxic the lower part of the K2qn is the most important marker for regional 2+3 event 2 (OAE2), but the lack of age controls prevents a chronostrati- stratigraphic correlation. Member 2 and 3 of the K2qn (K2qn )consist graphic correlation with the marine record. of interbedded black mudstone and gray- to celadon-colored siltstone. In this paper, we present a cyclostratigraphic study based on the In general, the Qingshankou Formation shows a shallowing—upward 1 high-resolution natural gamma-ray logging of the Qingshankou trend from deep lacustrine deposit in the lower part (K2qn )toshallow 2+3 Formation from 10 wells in an east–west transect across the Songliao laustrine, coastal and deltaic deposits in the upper part (K2qn )(Gao Basin (Fig. 2). The major objective of this study is to establish an et al., 1994; Li, 1995; Fig. 1b and c). astronomically calibrated time scale (ATS) for the Late Cretaceous lacustrine strata. With the new floating ATS, we discuss (1) the 2.2. The age of the Qingshankou Formation (K2qn) stratigraphic correlation across the basin and temporal variations in sedimentary rates, (2) the duration of the ostracod biozones Due to the lack of radiometric ages and difficulties of defining the constrained by the ATS that may be useful for stratigraphic correlation duration of terrestrial fossils, the age of the K2qn has been debated. between terrestrial and marine successions, and (3) the duration of Previous age assignments include (1) late Cenomanian-early Tur- the lacustrine anoxic event 1 (LAE1) and its correlation with the onian (Chen, 2000), (2) Cenomanian (Stratigraphy Committee of oceanic anoxic event 2 (OAE2). China, 2002; Wan et al., 2005; Sha, 2007), (3) Cenomanian–Turonian (Ye et al., 2002), and (4) late Cenomanian–early Coniacian (Wang 2. Geological setting et al., 2007). Recent radiometric ages from volcanic rocks of the Yingcheng

2.1. Tectonics and stratigraphy Formation (K1y) and paleontological data from Denglouku (K1d) and Quantou (K2q) Formations support a late Cenomanian age for the Geographically, the Songliao Basin (SLB) in northeastern China is Qingshankou (K1qn) Formation. The SHRIMP zircon U-Pb age of 111– surrounded by the Great Xing'an Mountains in the west, the Lessar 113 Ma (Zhang et al., 2007a) and K/Ar ages of 113–136 Ma (Wang et al.,

Xing'an Mountains in the north and the Zhangguangcai Mountains in 2002) from the upper Yingcheng Formation (K1y) suggest that the age the east (Fig.1a–c). The southern margin of the basin is separated from of the K1y is Aptian and the age of the K1d should be Albian (Fig. 1d). the North China plate by the Late Paleozoic Chifeng–Kaiyuan fault zone The palynological assemblages of Quantonenpollenites crassatus– (Sun et al., 2007). The tectonic evolution of the SLB can be divided into Cranwellia striatella and Trilobosporites–Cyathidites–Tricolpopollenites

(1) a pre-rift phase, (2) a syn-rift phase, (3) a post-rift phase, and (4) a in the upper and lower part of the Quantou Formation (K1q) are of compression phase (Cheng et al., 2006)(Fig. 1d). According to the early Turonian and late Cenomanian, respectively (Li, 2001). This is characteristics of rises and depressions, the SLB can be divided into six consistent with the palynological data from the K1d that have the ages ﬁrst-order tectonic units: central depression zone, north plunge zone, of Early to Middle Albian (Li and Li, 2005). Recently, Wang et al. (2007) west slope zone, northeast uplift zone, southeast uplift zone, and estimated an age of ~94 Ma for the K2qn/K2qboundaryby southwest uplift zone (Gao et al., 1994)(Figs. 1b and 2). extrapolating the stratal thickness of the basin. This is conﬁrmed by

The basement of the SLB consists of Precambrian to Paleozoic their latest, yet to be published radiometric age from the K2qn (Wang, metamorphic and igneous rocks and Paleozoic to Mesozoic granites P.J., personal communications, 2008). (Wang et al., 2006; Pei et al., 2007). Unconformably overlying the basement, up to 7000 m thick Mesozoic and Cenozoic terrestrial strata 2.3. Lacustrine anoxic events (LAEs) in the SLB are unevenly distributed across the basin (Gao et al.,1994). A simplified lithostratigraphy of the Songliao Basin is presented in Fig. 1d. Previous studies suggested that two great lacustrine anoxic events 1 1+2 The Upper Jurassic and Cretaceous strata in SLB is commonly (LAEs) happened in SLB during the deposition of the K2qn and K2n , subdivided into two sequences separated by a regional unconformity and it was proposed that they were related to two great marine at the top of the Yingcheng Formation (K1y). The lower sequence transgressions that may have created warm and wet climate conditions includes the Upper Jurassic Huoshiling Formation (J3h) and the Lower (Fig. 1d; Gao et al., 1994; Hou et al., 2000; Wang et al., 2001; Li and Pang, Cretaceous Shahezi (K1s) and Yingcheng (K1y) Formations (Fig. 1d). 2004). These lacustrine anoxic events was considered crucial for the These formations are composed of volcanic–volcaniclastic rocks and formation of prolific petroleum source rocks in SLB (Hou et al., 2000; alluvial–lacustrine sedimentary rocks. Wang et al., 2001; Li and Pang, 2004; Huang et al., 2007). Evidence 1 The upper sequence consists of seven formations, in ascending supporting the lacustrine anoxic event 1 (LAE1) in K2qn include (1) order including Denglouku (K1d), Quantou (K2q), Qingshankou deposition of finely laminated, organic and pyrite—rich black shale and 13 (K2qn), Yaojia (K2y), Nenjiang (K2n), Sifangtai (K2s) and Mingshui oil shale with a positive δ Corg spike (Wang et al., 2001; Huang et al., (K2m) Formations (Fig. 1d). The Denglouku Formation (K1d) is 2007; Huang, 2007), (2) extinction of deep water-fauna during LAE1 and composed mainly of alluvial and fluvial deposits, while the Quantou flourish of diversified fossil assemblage after LAE1 (Huang et al., 1998;

Formation (K2q) consists of coarse clastic rocks of ﬂuvialorigininthe Hou et al., 2000), and (3) appearance of 28, 30-bisnorhopane which lower part and lacustrine mudstones in the upper part. The originates from a typical anoxic bacteria and preservation of complete C35 lacustrine sandstone, mudstone, and shale of the Qingshankou hopanes series, lower diasteranes content in mudstone (Hou et al., 2003).

(K2qn), Yaojia (K2y) and Nenjiang (K2n) Formations contain the The sedimentary sequence of the red terrigeneous sediments of 1 1+2 majority of the oil source rocks, reservoir rocks and seals of the the K2q, K2y and K2s and the black sediments of the K2qn and K2n 310 H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323 H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323 311

Fig. 2. Cross section of the Songliao Basin (modiﬁed from Wang et al., 2007). Position of sections matches those in Fig. 1b. Sections L2 and D501 are located at the north plunge zone and northeast uplift zone, respectively. Abbreviation N2t represents the Taikang Formation and the other abbreviations are the same as in Fig. 1d.

may record the alternation of anoxic and oxic lacustrine depositional ATS so that they can be correlated with the marine biozones in the environments, similar to the redox cycles recorded in Cretaceous future. marine strata (Fig. 1d; Huang, 2007). The two great lake anoxic events 1 1+2 (LAE1 and LAE2 in Fig. 1d) in K2qn and K2n have been suspected to 3. Data processing be correlatable with the ocean anoxic events (Hou et al., 2000; Huang, 2007; Huang et al., 2007; Wang et al., 2007), but reliable time Natural gamma-ray logging records the intensity of the gamma ray constraints are lacking. emitted during the decay of atomic nuclei of radioactive elements contained in sedimentary rocks. The intensity of gamma ray relates to 40 232 238 2.4. Biostratigraphy of the K2qn the amount of K, Th and U in rocks. Clay and organic particles have strong capacity of absorbing radioactive elements. The gamma-

Fossils are abundant in the K2qn and intensive paleontological ray logging curves can therefore reﬂect changes in the amount of clay studies have been conducted since 1959 (e.g., Gao et al., 1999; Ye et al., and organic materials in sediments, both of which are sensitive to

2002). Fossils of the K2qn include abundant ostracods, conchostraca, controlling factors such as precipitation–evaporation ratios and lake- fish, gastropods, bivalves, sporopollen, algae, and vertebrates (Hou et al., level fluctuations induced by climate changes (e.g., Serra, 1984; 2000; Wan et al., 2007). The identification of these fossils has played an Hinnov, 2004). important role in the stratigraphic correlation and division and the In this study, natural gamma-ray logs from 10 wells penetrating the reconstruction of paleoecology, paleoenvironment and paleoclimate in Qingshankou Formation (K2qn) across the basin were selected for the SLB. However, the biostratigraphic resolution is low and many fossils cyclostratigraphic analysis (Figs. 1b, c and 2). The shortest and the 1 2+3 are restricted to either K2qn or K2qn . Fossil abundance and diversity longest gamma-ray logs are J32 (101 m) and G692 (544 m), respectively also vary among sections across the basin and thus it has been difficult to (Figs. 3–6). In well F64 and J32 of the west slope zone (Figs. 1b and 2), establish a reliable biostratigraphic framework correlatable with that of K2qn unconformably overlies the Upper Jurassic strata, while in the the marine successions (Wan et al., 2005). Recently, a detailed ostracod other 8 wells K2qn conformably overlies the Quantou Formation (K2q). biostratigraphy of the K2qn with 19 ostracod biozones was established, The sampling spacing is from 0.05 m to 0.1524 m for different logs. based on collections from five recovered wells (Zhang et al., 2007b). Spectral analysis and continuous wavelet analysis were conducted These ostracod biozones are going to be calibrated by the new floating in order to investigate whether the cyclicity records signal resulting

Fig. 1. (a) Location of the Songliao Basin (SLB) in northeastern China. (b) Simplified paleogeographic reconstruction of the SLB during deposition of Member 1 of the Qingshankou 1 Formation (K2qn ). Tectonic units in the SLB and locations of the study wells (black dots) are indicated. Lines A–B mark the position of the cross section in Fig. 2. (c) Paleogeographic 2+3 reconstruction of the SLB during deposition of the K2qn . (d) Simplified stratigraphy and geological events of the Upper Jurassic to Cretaceous in the SLB. The sea-level change curve is adopted from Gradstein et al. (2004) and the lake-level curve is from Gao et al. (1994). 312 H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323 from astronomical climate forcing. A band-pass filter was designed to silty sandstone and mudstone towards the top (Fig. 4a). The gamma-ray remove the low (N100 m) and high (b1 m) frequency variability. values are in the range of ~60–180 API. Low values correspond to Spectral analysis was performed with the REDFIT software package calcareous mudstone, whereas high gamma-ray values are produced by

(Schulz and Mudelsee, 2002). Wavelet analysis is used to transform black shale and mudstones. Signiﬁcantly low values at the top of K2qn depth-related sedimentary signature into wavelengths at distinct correspond to the red silty sandstones (Fig. 4). The density log has values depth intervals, useful for detecting and distinguishing abrupt discon- mostly from 1.7 to 2.3 g/cm3, but lower values are found at the top, tinuities, cyclicity and changes in sedimentation rate (Prokoph and possibly related to increased porosity in sandstones (Fig. 4). Agterberg, 2000). The wavelet software was provided by Torrence and Wavelet analysis of both gamma-ray and density logs reveal Compo (1998) and is available at http://paos.colorado.edu/research/ relatively continuous cycles with periods of 68 m, 39 m, 13.5–9m,5– wavelets/. 3.8 m and 2.5–1.7 m (Fig. 3f and g). The lack of a strong shift at periods of 39 m and 13.5–9 m in both gamma-ray and density logging suggests 4. Cyclostratigraphy that relatively stable sedimentation rate prevailed during the deposi-

tion of K2qn, while the variance at shorter periods of 5–3.8 m and 2.5– 4.1. Orbital parameters in Late Cretaceous 1.7 m probably records short-term ﬂuctuations in sediment supply (e.g., Prokoph and Agterberg, 2000). According to geological and paleomagnetic evidence (Zhao et al., To determine whether the observed cycles in sedimentary strata 1990; Chi et al., 2000), the Songliao Basin has been adjacent to the were formed by astronomical forcing, the most commonly used present position in early Late Cretaceous. We used the June 21 insolation method is to compare the relative ratio of the observed cycles with curve at 45°N from 94 to 90 Ma as a reference (Fig. 3a-1), which is that of the Milankovitch cycles (Hinnov, 2000; Weedon, 2003). The calculated using Analyseries 2.0 (Paillard et al., 1996) and the improved ratios of the major periods of ~39 m, 13.5–9m,5–3.8 m and 2.5–1.7 m La2004 solution (Laskar et al., 2004). Four distinct frequency bands can from natural gamma-ray and density logging of section M206 is be observed from the continuous wavelet analysis scalogram (Fig. 3a-2), 20:5:2:1, which matches well with the Milankovitch cyclicities of which are consistent with results from spectral analysis. Spectral 405 ka:100 ka:40 ka:20 ka (long eccentricity, short eccentricity, analysis shows the presence of the principal eccentricity periods of obliquity, precession). Spectral analysis on the full time series of the

405 ka (E3), 125 ka (E2)and100ka(E1), principal obliquity periods of gamma-ray logging reveals four distinct cycles of 14.2–8.9 m, 4.9– 37.5 ka (O1)withlesseronesof48ka(O2), 56 ka and 28 ka, and principal 3.8 m, 2.6–2.3 m and 1.9–1.7 m (Fig. 3f and g). The ratio for the ﬁve precession periods of 22.5 ka (P2),18.4 ka (P1) and 15.9 ka (Fig. 3a-3). We main peaks of these cycles is 13.3 m:10.3 m:4.05 m:2.43 m:1.9 m will use these periods as the Late Cretaceous “canonical” orbital varia- (123:95:37.5:22.5:17.6). These ratios are similar to the results of tions in our study. spectral analysis on the June 21 insolation curve at 45°N from 94 to 90 Ma (Fig. 3a). We thus consider that the sedimentary cyclicity in 4.2. Cycle analysis and cyclostratigraphic interpretation section M206 was formed by orbital forcing (Weedon, 2003). The major periods of ~39 m, 13.5–9m,5–3.8 m and 2.5–1.7 m were Parallel bands in the wavelet scalograms show that the spectral formed most likely by long (405 ka) and short (100 ka) eccentricity, power is conﬁned to distinct and relatively continuous period bands obliquity (40 ka) and precession (20 ka), respectively. for all sections (Fig. 3b-l). Spectral analyses reveal distinctive The small cycle length of the original natural gamma-ray and spectral power peaks and the ratio of corresponding periods is density logging curves range from 1.7 m to 2.5 m, which are interpreted ~20:5:2:1, which is similar to the ratio of Milankovitch cyclicities of as precession-related sedimentary cycles (Fig. 4). Counting from the

405 ka:100 ka:40 ka:20 ka (long eccentricity, short eccentricity, base of the K2qn, each cycle is numbered and there are ~253 precession obliquity, precession) (Fig. 3b-l). cycles in both gamma-ray and density logging data (Fig. 4). Two We followed the method of Westerhold et al. (2007) to construct a Gaussian band-pass filters were used to filter the signals of the long cyclostratigraphy for the K2qn for each section (Figs. 4–6). Direct cycle and short eccentricity cycles of ~39 m and 13.5–9 m, respectively. The counting and Gaussian band pass filtering of the data are used. The filtering outputs of the gamma-ray and density logging are shown in first peak of gamma-ray logging (precession), extracted short and long Fig. 4. We counted 12.8 long eccentricity cycles and 51.6 short eccentricity cycles were used as the starting point (Figs. 4–6). Strong eccentricity cycles from the gamma-ray logging, but 12.3 long shifts in the wavelet spectral bands for sections G692 (at 2100 m), L2 eccentricity cycles and 50.1 short eccentricity cycles from the density (at 1770 m), F64 (at 557 m and 500 m) suggest changes in cycle logging (Fig. 4). There are about 0.5 long eccentricity and 1.5 short periodicities due to change in sedimentation rates, so we use different eccentricity offsets. The underlying reason for this difference is Gaussian band pass filters to extract orbital forcing cycles (Figs. 5 and uncertain, but because the cycle bands in the wavelet scalograms 6). The longest section (M206) in the central depression zone and the and the filter outputs of gamma-ray logging demonstrate more stable shortest section (J32) in the west slope zone were selected as end and constant signals than those of the density logging (Figs. 3f, g and 4), members for interpreting the results from wavelet and spectral and because previous studies demonstrated that gamma-ray readings analyses and constructing cyclostratigraphy. are more sensitive to variations in lithology and clay–mineral content

In section M206, the total thickness of K2qn is 496.8 m (from 1286 m than other logging tools (Serra, 1984; Prokoph and Thurow, 2000), we 1 to 1782.8 m in the well log; Fig. 4). The lower part of K2qn (K2qn )is chose the gamma-ray logging data for cyclostratigraphic study for all composed mainly of gray and black mudstone, with thin calcareous sections in this study. mudstone layers and three black oil shale layers. The upper part of K2qn The total thickness of K2qn in section J32 in the west slope zone is 2+3 (K2qn ) consists mainly of thick, gray to dark-gray mudstone with 101 m (from 398 m to 499 m in the well log; Figs. 3k and 6f). It calcareous mudstone interbeds, but changes to thinly laminated, red unconformably overlies the Jurassic volcanic rocks and is

Fig. 3. (a) June 21 insolation curve at 45°N from 94 to 90 Ma (a-1), wavelet scalogram (a-2) and spectral analysis (a-3) of the theoretical insolation curve, which are calculated using Analyseries 2.0 (Paillard et al., 1996) and La2004 solution (Laskar et al., 2004). The shaded contours in wavelet scalograms are normalized linear variances, with blue representing low spectral power and red representing high spectral power. Regions below curves on both ends indicate the cone of inﬂuence where edge effects become signiﬁcant. (b)–(f) and (h)–

(l) Wavelet scalogram and spectral analyses of the gamma-ray logging of the K2qn in different tectonic zones of the Songliao Basin. (g) Wavelet scalogram (g-1) and spectral analysis (g-2) of the density logging of M206. Letters “d” mark the possible discontinuities. See Figs. 1b and 2 for location of sections, and Figs. 4–6 for original logging data. H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323 313 314 H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323

Fig. 3 (continued). H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323 315 unconformably overlain by sandstones of the second and third 5. Discussion 2+3 1 member of Yaojia Formation (K2y ). In this section, K2qn is com- 2+3 posed of purple mudstone and gray muddy siltstone and K2qn 5.1. Astrochronology of the Qingshankou Formation (K2qn) consists mainly of emerald, celadon and purple mudstone with sandstone and muddy siltstone. The records from the K2qn in the SLB display excellent precession, The parallel bands in the wavelet scalogram of section J32 show obliquity, short and long eccentricity signals (Figs. 3–6), and theoreti- three cycle periods of 16–9 m, 5.8 m and 3.2 m. Spectral analysis on cally, every band of signals can be used for tuning to the orbital solutions. the full time series reveals two distinct cycle bands of 6.2–4.3 m and However, because the precision of the orbital solution for Mesozoic is 3.6–2.8 m (Fig. 3k). These periods has a ratio of 5:2:1, which is limited (Laskar et al., 2004), the high frequency cycles may not be correlative to the short eccentricity, obliquity and precession ratio of suitable for constructing the ATS. Changes in resonance between the 100 ka: 40 ka:20 ka (Fig. 3k). Although spectral analysis for this orbits of Earth and Mars may have altered the 95 to 128 ka eccentricity. section does not show the peaks of eccentricity cycles, short and long Tidal friction in the Earth–Moon system could slow the Earth's rotation eccentricity cycles were extracted by the Gaussian band pass ﬁlters rate over time, increasing the periodicities of precession and obliquity according to the wavelet analysis and the variation of gamma-ray (Laskar et al., 2004). Thus a full-spectrum tuning of the Mesozoic records logging (Fig. 3k). We counted 2.8 long eccentricity, 10.9 short to the La2004 solution is impossible at the moment. Because the long eccentricity and 50 precession cycles in section J32 (Fig. 6f). eccentricity cycle (405 ka) remains very stable at least back to ~250 Ma,

Fig. 4. (a) Simpliﬁed lithology and stratigraphy of the K2qn from well M206. (b) Continuous gamma-ray (red line) and density (blue line) logging of the K2qn from M206 with 0.125 m sampling resolution. The red and blue numbers indicate the number of precession-related cycles. (c) Cycle counting of long eccentricity ﬁlter outputs of gamma-ray (red) and density

(blue) logging data of the K2qn from M206. (d) Cycle counting of short eccentricity filter outputs of gamma-ray (red) and density (blue) logging data of the K2qn from M206. The long and short eccentricity-related cycles are extracted by Gaussian band pass filters with filter frequency of 0.0225±0.003 cycles/m and 0.104±0.013 cycles/m, respectively. 316 H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323

Fig. 5. Cycle counting of precession (black), long (red) and short (blue) eccentricity of gamma-ray logging data of the K2qn from sections (a) G692, (b) C503 and (c) Z805 of the central depression zone. The long and short eccentricity cycles are extracted by Gaussian band pass ﬁlters. Filter frequency for long eccentricity cycles: G692–0.025±0.003 (from 1800 to 2100 m) and 0.019±0.002 (from 2100 to 2344 m) cycles/m; C503–0.0187±0.003 cycles/m; Z805–0.0208±0.003 cycles/m. Filter frequency for short eccentricity cycles: G692–0.083±0.026 cycles/m; C503–0.075±0.01 cycles/m; Z805–0.081±0.02 cycles/m. it was recommended that a Mesozoic astronomically calibrated time extracted long eccentricity cycles from most sections of the K2qn are less scale can be established as a ﬂoating time scale using long eccentricity stable than the short eccentricity cycles (Figs. 4–6), which also show cycles (Laskar et al., 2004; Hinnov and Ogg, 2007). However, the more stable and clear signals than other bands of signals in most wavelet

Fig. 6. Cycle counting of precession (black), long (red) and short (blue) eccentricity of gamma-ray logging data of the K2qn from sections (a) L2, (b) W209, (c) D501, (d) F64, (e) L271 and (f) J32. The long and short eccentricity cycles are extracted by Gaussian band pass ﬁlters. Filter frequency for long eccentricity cycles: L2–0.028±0.005 cycles/m; W209–0.0262±0.003 cycles/m; D501–0.03125±0.004 cycles/m; F64–0.02381±0.003 cycles/m; L271–0.01923±0.003 cycles/m; J32–0.02439±0.003 cycles/m. Filter frequency for short eccentricity cycles: L2– 0.108±0.021 (from 1449 to 1770 m) and 0.091±0.01 (from 1770 to1874 m) cycles/m; W209–0.1042±0.015 cycles/m; D501–0.1266±0.025 cycles/m; F64–0.09375±0.03125 cycles/m; L271– 0.0746±0.08 cycles/m; J32–0.1111±0.033 cycles/m. H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323 317 318 H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323

Fig. 7. (a) Astronomical calibration of short eccentricity cycles from ﬁlter output of gamma-ray logging (red) compared with short eccentricity (blue) and full eccentricity (gray) of

La2004 solution (Laskar et al., 2004) for the Qingshankou formation (K2qn) in M206. Also shown are the comparisons between extracted long eccentricity, obliquity and precession cycles from gamma-ray logging of well M206 (red) and those of the La2004 solution (black). The Gaussian band pass ﬁlters for short (long) eccentricity of La2004 solution are 0.0093±

0.0013 (0.002469±0.0003) cycle/ka. (b) Ostracods biozones of the K2qn (modiﬁed from Zhang et al., 2007b) calibrated by the established astronomical times of M206. scalograms and spectral analyses (Fig. 3). Thus we chose to construct the longest accumulation time (Figs. 3f, g and 4), and no identiﬁable

ATS using the short eccentricity cycles. stratigraphic discontinuities. The estimated age of ~94 Ma for the K2qn/ After cycle analyses of all sections across the SLB, we chose section K2qboundary(Wang et al., 2007) was used as a reference for curve M206 at the central depression zone as the reference section to establish matching between the short eccentricitycycles from gamma-ray logging the ATS (Fig. 7) because this section records stable Milankovitch cycles, and the short eccentricity curve of La2004 solution (Fig. 7). The .W ta./ErhadPaeaySineLtes28(09 308 (2009) 278 Letters Science Planetary and Earth / al. et Wu H. – 323

2+3 1 Fig. 8. Astronomical time scales for K2qn from 10 sections across the Songliao Basin. The gray area is the position of lacustrine anoxic event 1 (LAE1). The dash lines indicate the lithological boundaries of K2y/K2qn, K2qn /K2qn and K2qn/ K2q; all of them show considerable diachroneity across the basin. On the basis of abrupt jumps in cycle thickness (cf. Prokoph and Agterberg, 1999, 2000), potential stratigraphic discontinuities can be identified in sections such as F64, D501, L2, Z805, C503, and G692, but the missing time at these discontinuities is hard to evaluate due to the lack of sufficient absolute age data. Based on the relatively fine-grained lithologies (mudstones) and stable Milankovitch cycles in sections of the central depression zone, the missing time within the K2qn may be limited in these sections. Instead, the unconformity at the top of the K2qn may record significant stratigraphic truncation in response to a major uplifting event of the Songliao basin. 319 320 H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323 correlation was performed using the software of Analyseries 2.0 abrupt jumps in cycle thickness (cf. Prokoph and Agterberg,1999, 2000), (Paillard et al., 1996). consistent with the appearance of relatively coarse-grained lithology In order to get a start point, filtered long and short eccentricity, (silty sandstone) at the discontinuities. Thus the incomplete preserva- obliquity and precession curves of M206 were slidden back and forth tion of Milankovitch cycles in these sections may relate to non- to get the best visual match of peak locations and amplitudes with deposition and/or erosion at the margin of the basin. Stratigraphic La2004 solution. The tuning was done by assigning ages of short discontinuities may also exist in sections D501, Z805, G692 and L2 eccentricity maxima of the La2004 solution to the corresponding (Fig. 8), but the internal time-missing at these discontinuities is hard to peaks of extracted short eccentricity of gamma-ray logging of K2qn as be evaluated due to the lack of absolute age constraints across the basin. 2+3 1 identified and labeled in the cyclostratigraphy (Figs. 4 and 7). The long The traditionally defined K2qn /K2qn boundary show significant eccentricity, however, show a small offset of ~100 ka with that of the diachroneity (Fig. 8), but the fine-grained lithologies (dominated by La2004 solution. This offset is possibly due to the unstable long mudstones) across the boundary in the central parts of the basin suggest eccentricity signal itself or related to uncertainties with the eccen- that the missing time at this level may be limited. The diachroneity at tricity solutions (Laskar et al., 2004). The precession and obliquity this boundary may record facies changes across the basin. A major bands derived from the gamma-ray logging show similar variations stratigraphic unconformity occurs at the top of the K2qn, which may with those of La2004 solution (Fig. 7a). have resulted in significant stratigraphic truncation across the basin. The established ATS from section M206 suggests that the duration The identification of all bands of Milankovitch cycles in the central of the K2qn is about 5.20 Ma, from 94.27 Ma to 89.07 Ma covering Late parts of SLB suggests strong astronomical climate forcing in the Cenomanian to Early Coniacian (Fig. 7a). The ATS provides time development of the sedimentary cyclicity in the K2qn. The sedimen- constraints for the ostracod biozones (Fig. 7b), which will be useful for tary cycles recorded by the gamma-ray logging of relatively homo- future stratigraphic correlations of the ostracod-bearing terrestrial geneous, fine-grained lithologies (mudstones) in the K2qn may reflect strata and their correlations with marine successions. For example, changes in clay mineral input in response to wet and dry periods. Wet the duration of the fossil zone of Ziziphocypris rugosa is about 90.09– periods may have enhanced chemical weathering and clay mineral 91.69 Ma (Fig. 7b). input, resulting in positive gamma-ray peaks; while decreased Using the same method, we tentatively constructed the ATS for all chemical weathering during dry periods may have formed the the 10 sections of K2qn (Fig. 8). The duration of K2qn varies from negative gamma-ray peaks. The occurrence of four periods of high 5.20 Ma in M206 of the central depression zone to 1.09 Ma in section sedimentary rate (Fig. 9) in the K2qn may record times of sustained J32 of the west slope zone (Fig. 8 and Table 1). Significant variations in wet climate and high sedimentary input in the basin. the duration of K2qn indicate internal stratigraphic discontinuities and particularly the stratigraphic truncation at the top of the K2qn. 5.3. The duration of lacustrine anoxic event 1 (LAE1) in the SLB and its correlation to OAE2 5.2. Preservation and origin of Milankovitch cycles 13 Because available δ Corg values are very sparse and cannot define a 13 Wavelet and spectral analyses on the natural gamma-ray logging of positive δ Corg ‘excursion’, the traditional LAE1 in SLB was defined by 1 the K2qn shows excellent preservation of astronomically driven the presence of oil shales in the lower K2qn . The thickness of the oil sedimentary cycles, including long and short eccentricity, obliquity, shales ranges from 17.0 m (D501) to 27.0 m (G692 and L2), which and precession cycles in sections from the central parts of the basin records ~2–3 short eccentricity cycles and ~10–16 precession cycles where thicker and more complete strata are preserved (Fig. 3b–i). (Figs. 4–6 and Table 1). Our new ATS indicate the shortest and the However, in thinner sections of the west slope zone (F64, L271 and J32), longest duration of LAE1 are 210 ka (C503) and 310 ka (M206), long eccentricity cycles were not detected in the wavelet or spectral respectively (Fig. 8 and Table 1). scalograms (Fig. 3j–l). These sections record only 2–3 long eccentricity The LAE1 has been suspected to be time equivalent to the oceanic cycles (Fig. 6d–f), which are difficult to be detected by wavelet and anoxic event 2 (OAE2) at the Cenomanian–Turonian boundary (Huang spectral analyses. In section F64 (Fig. 3l), only precession and short et al., 1998, 2007). The established ATS for the K2qn confirms this eccentricity cycles can be identified. Potential stratigraphic disconti- correlation. The onset of the OAE2 has recently been estimated as nuities can be identified from these sections (Fig. 8) on the basis of 94.09 Ma (Sageman et al., 2006) and 94.21/93.72 Ma (Mitchell et al.,

Table 1

Estimated age of the geological events in the Qingshankou formation (K2qn) according to ﬂoating astronomical time scales and cyclostratigraphy

Wells Boundary (thickness) (m) Estimated age (duration) (Ma) Sedimentary 1 2+3 rates(cm/ka) K2qn LAE1 K2qn K2qn /K2qn boundary LAE1 Southeast uplift zone W209 285.0–704.0 (418.5) 672.0–697.0 (25.0) 89.85–94.25 (4.40) 93.30 Ma/611.0 m 93.96–94.21 (0.25) 9.51

Northeast uplift zone D501 576.0–762.5 (186.5) 739.0–756.0 (17.0) 91.80–94.27(2.47) 93.38 Ma/699.0 m 93.95–94.19 (0.24) 7.55

Central depression zone C503 687.0–1118.0 (431.0) 1090.0–1114.0 (24.0) 90.91–94.23 (3.32) 92.63 Ma/1042.0 m 93.99–94.20 (0.21) 12.98 Z805 1338.0–1685.0 (347.0) 1649.0–1674.0 (25.0) 91.23–94.31 (3.08) 93.83 Ma/1630.0 m 93.98–94.21 (0.23) 11.3 M206 1286.0–1782.8 (496.8) 1750.0–1775.0 (25.0) 89.07–94.27 (5.20) 93.32 Ma/1700.8 m 93.87–94.18 (0.31) 9.55 G692 1800.0–2344.0 (544.0) 2305.0–2332.0 (27.0) 89.67–94.27 (4.60) 93.65 Ma/2270.0 m 93.95–94.20 (0.25) 11.83

North plunge zone L2 1449.0–1874.0 (425.0) 1840.0–1867.0 (27.0) 89.77–94.25 (4.48) 93.57 Ma/1797.0 m 93.92–94.18 (0.26) 9.49

West slope zone F64 446.0–585.5 (139.5) – (1.36) ––10.26 L271 647.5–759.5 (112.0) – (1.25) ––8.96 J32 398.0–499.0 (101.0) – (1.09) ––9.27 H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323 321

Fig. 9. Comparison of sedimentation rates of the K2qn across the tectonic zones of the Songliao basin. Four periods of high sedimentary rates can be identiﬁed (gray markers) and they may record periods of sustained wet climate during which enhanced chemical weathering increased sediment inputs in the basin.

2008) in central Colorado and Italy, respectively. These estimations, Widespread oceanic anoxia at OAE2 has been proposed to have particularly the onset age of 94.21 Ma, are consistent with the onset resulted from the large-scale igneous activity that may have released age of ~94.21–94.18 Ma for the LAE1 in SLB (Figs. 7 and 8). The large quantities of CO2 to the atmosphere and/or reduced hydro- duration of LAE1 and OAE2, however, needs further investigation. thermal fluids to the ocean water column (Sinton and Duncan, 1997; First, the ~210–310 ka duration for LAE1 is based on the occurrence of Kerr, 1998; Turgeon and Creaser, 2008). The coincidence of an anoxic 13 oil shales rather than a chemical anomaly (such as the positive δ Corg (high organic carbon burial) event in both marine and lacustrine excursion) in K2qn; therefore its duration may have been under- environments suggest the involvement of atmospheric CO2. Elevated estimated. Second, existing estimations for the duration of OAE2 vary CO2 level may have enhanced the primary productivity in both marine significantly from ~320 ka to ~900 ka. On the basis of biostratigraphic and terrestrial systems (e.g., Schlanger and Jenkyns,1976; Schlanger et data and stratigraphic interpolation, Arthur et al. (1988) and Caron al., 1987), leading to water column oxygen deficiency and increased et al. (1999) estimated the duration of OAE2 as 0.5–0.8 Ma and 0.4 Ma, organic carbon burial. Mitchell et al. (2008) proposed that the respectively. Estimations based on orbital cyclicity include 563–601 ka Cretaceous OAEs including OAE2 may be astronomically driven, or 847–885 ka in central Colorado (Sageman et al., 2006), 320 ka in resulting from reduced seasonality during protracted period of low western Canada (Prokoph et al., 2001), and 440 ka in the Tarfaya Basin insolation variation. The same onset age of ~94.21–94.18 Ma and in Morocco (Kuhnt et al., 2005). Nonetheless, the ~210–310 ka existence of similar nodes of procession, obliquity and eccentricity duration of LAE1 is, in first order, comparable with the ~320–900 ka around the LAE1 (Fig. 7a) and OAE2 (Mitchell et al., 2008) support this duration of OAE2, although apparently the duration of both LAE1 and hypothesis. However, it is unclear whether such nodes were resulted

OAE2 needs to be refined. from reduced seasonality at elevated atmospheric CO2 or directly from 322 H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323 the influences of long-term insolation change. The time range of the Chen, P.J., 2000. Comments on the classification and correlation of non-marine Jurassic – fi and Cretaceous of China. J. Stratigr. 24, 114 119 (in Chinese with English abstracts). data is insuf cient to evaluate the response of other anoxic events in Cheng, J.M., McIntosh, J.C., Xie, X.N., Jiao, J.J., 2006. Hydrochemistry of formation water SLB and therefore the long-term “oceanic anoxic cycles” proposed by with implication to diagenetic reactions in Sanzhao depression and Qijia–gulong Mitchell et al. (2008). depression of Songliao Basin, China. J. Geochem. Explor. 88, 86–90. Cheng, R.H., Wang, G.D., Wang, P.J., 2008. Sedimentary cycles of Quantou–Nenjiang, Cretaceous and Milankovitch cycles from the south hole of CCSD–SK in Songliao 6. Conclusions basin. Acta Geol. Sin. 1, 48–55 (in Chinese with English abstracts). Chi, Y.L., Wang, P.J., Wan, C.B., Shan, X.L., 2000. Study on non-marine deep stratigraphy of – Wavelet and spectral analyses on the natural gamma-ray logging of oil-bearing basin in China. Jilin Science and Technology Press, Changchun, pp. 1 24 (in Chinese with English abstracts). 10 wells of the Late Cretaceous lacustrine Qingshankou Formation (K2qn) Fiet, N., Quidelleur, X., Parize, O., Bulot, L.G., Gillot, P.Y., 2006. Lower Cretaceous stage in the Songliao Basin reveal Milankovitch cycles of precession (20 ka), durations combining radiometric data and orbital chronology: towards a more – obliquity (40 ka) and eccentricity (100 ka and 405 ka). The sedimentary stable relative time scale? Earth Planet. Sci. Lett. 246, 407 417. Gao, R.Q., Zhang, Y., Cui, T.C., 1994. Cretaceous Petroleum Bearing Strata in the Songliao cyclicity may have been formed by cyclic changes from wet to dry climate Basin. Petroleum Industry Press, Beijing, pp. 1–333 (in Chinese with English abstract). in response to astronomical forcing. The number of cycles decreases from Gao, R.Q., Zhao, C.B., Qiao, X.Y., Zheng, Y.L., Yan, F.Y., Wan, C.B.,1999. Cretaceous Oil Strata the central depression zone to the margin of the basin, possibly resulting Palynology From Songliao Basin. Geological Publishing House, Beijing, pp. 1–373 (in Chinese with English abstract). from non-deposition and/or erosion associated with stratigraphic Gradstein, F.M., Ogg, J.G., Smith, A.G., 2004. A Geologic Time Scale 2004. Cambridge discontinuities towards the basin margin and paleogeographic highs. University Press, pp. 55–62. Floating astronomical time scales (ATS) for the sections in the central Hinnov, L.A., 2000. New perspectives on orbitally forced stratigraphy. Annu. Rev. Earth Planet. Sci. 28, 419–475. parts of basin were established by matching the extracted short Hinnov, L.A., 2004. Earth's orbital parameters and cycle stratigraphy. In: Gradstein, F.M., eccentricity signals to the La2004 solution. The accumulation time of Ogg, J.G., Smith, A.G. (Eds.), A Geologic Time Scale 2004. Cambridge University Press, pp. 55–62. K2qn varies from 1.09 Ma to 5.20 Ma across the basin. The longest ATS Hinnov, L.A., Ogg, J.G., 2007. Cyclostratigraphy and the astronomical time scale. Stratigra- constructed from section M206 in the central depression zone constrains phy 4, 239–251. the age of the K2qn as from Late Cenomanian to Early Coniacian (94.27– Hou, D.J., Li, M.W., Huang, Q.H., 2000. Marine transgressional events in the gigantic 89.07 Ma). The ATS also provides time constraints for the ostracod freshwater lake Songliao: paleontological and geochemical evidence. Org. Geo- – biozones, which may be useful for future stratigraphic correlation among chem. 31, 763 768. Hou,D.J.,Feng,Z.H.,Huang,Q.H.,2003.Geologicalandgeochemicalevidencesof ostracod-bearing terrestrial successions and potentially, between terres- anoxic event in the Songliao Basin. Geoscience 17, 311–317 (in Chinese with trial and marine successions. The mean sedimentation rates range from English abstract). 7.55 cm/ka to 12.98 cm/ka. Four periods of high sedimentation rates were Huang, Q.H., 2007. Cretaceous stratigraphy and micropalaeontology of Songliao Basin. Unpublished PhD thesis, Chinese Academy of Geologcal Sciences, Beijing, China, 1– detected across the basin, possibly recording periods of sustained wet 122 (in Chinese with English abstract). climate during which enhanced chemical weathering may have Huang, Q.H., Chen, C.R., Wang, P.Z., Han, M.X., Li, X.J., Wu, D.Q., 1998. The late Cretaceous significantly increased sediment inputs to the Songliao basin. bio-evolution and anoxic events in the ancient lake in the Songliao Basin. Acta Micropalaeontol. Sin. 15, 417–425 (in Chinese with English abstract). Based on the established ATS, the duration of the lacustrine anoxic Huang, Q.H., Liang, W.L., Ye, D.Q., Gao, W.F., 2007. The characteristics of Cretaceous event 1 (LAE1) at the basal Qingshankou Formation is estimated as microbiotas and formation of hydrocarbon-rich source rocks in Songliao Basin. Acta 210–310 ka. The onset age of LAE1 (~94.21–94.18 Ma) is comparable Palaeontol. Sin. 46, 380–386 (in Chinese with English abstract). – Kerr, A.C., 1998. Oceanic plateau formation: a cause of mass extinction and black shale with that of the oceanic anoxic event 2 (OAE2) at the Cenomanian deposition around the Cenomanian–Turonian boundary. J. Geol. Soc. Lond.155, 619–626. Turonian boundary, which has been estimated as 93.72/94.21 Ma or Kuhnt, W., Luderer, F., Nederbragt, S., Thurow, J., Wagner, T., 2005. Orbital-scale record 94.09 Ma. The duration of LAE1 (~210–310 ka) and OAE2 (~320– of the late Cenomanian–Turonian oceanic anoxic event (OAE-2) in the Tarfaya Basin fi (Morocco). Int. J. Earth Sci. (Geol. Rundsch.) 94, 147–159. 900 ka) is in the rst order agreement, but further investigations are Laskar, J., Robutel, P., Joutel, F., Gastineau, M., Correia, A.C.M., Levrard, B., 2004. A long- needed to better constrain the duration of the anoxic events in both term numerical solution for the insolation quantities of the Earth. Astron. marine and terrestrial basins. The occurrence of a time-equivalent Astrophys. 428, 261–285. anoxic (high organic carbon burial) event in both terrestrial and Li, D.S., 1995. Hydrocarbon habitat in the Songliao Rift Basin, China. In: Lambiase, J.J. (Ed.), Hydrocarbon Habitat in Rift Basins. Geological Society, London, Special marine basins suggest a shared trigger mechanism that involves Publications, vol. 80, pp. 317–329. fl increased atmospheric CO2 from igneous activities. Elevated CO2 may Li, W.B., 2001. Palyno ora from the Quantou Formation of Songliao Basin, NE China and – have facilitated primary productivity and anoxic water column its bearing on the Upper Lower Cretaceous boundary. Acta Palaeontol. Sin. 40, 153–176 (in Chinese with English abstract). chemistry in both marine and terrestrial basins. Li, W.B., Li, J.G., 2005. Albian palynological assemblage from the borehole Yu2302 in Yushu—with focus on the age of the Denglouku Formation in Songliao Basin. Acta Acknowledgements Palaeontol. Sin. 44, 209–228 (in Chinese with English abstract). Li, M.W., Pang, X.Q., 2004. Contentious petroleum geochemical issues in China's sedimentary basins. Pet. Sci. 1, 4–22. The authors are grateful for the help from Chengshan Wang, Li, Y.X., Bralower, T.J., Montañez, I.P., Osleger, D.A., Arthur, M.A., Bice, D.M., Herbert, T.D., Xiaoqiao Wan, Pujun Wang, Yanguang Ren and Yimin Dang. We Erba, E., Silva, I.P., 2008. Toward an orbital chronology for the early Aptian Oceanic Anoxic Event (OAE1a, 120 Ma). Earth Planet. Sci. Lett. doi:10.1016/j.epsl.2008.03.055. express our sincere appreciation to the journal editor (Prof. Peter B. Locklair, R.E., Sageman, B.B., 2008. Cyclostratigraphy of the Upper Cretaceous Niobrara deMenocal) and four anonymous reviewers for their careful review Formation, western interior, U.S.A.: a Coniacian–Santonian orbital timescale. Earth and constructive suggestions that significantly improved the paper. Planet. Sci. Lett. doi:10.1016/j.epsl.2008.03.021. Mitchell, R.N., Bice, D.M., Montanari, A., Cleaveland, L.C., Christianson, K.T., Coccioni, R., This study was jointly supported by the National Key Basic Research Hinnov, L.A., 2008. Oceanic anoxic cycles? Orbital prelude to the Bonarelli Level Development Program of China (Grant 2006CB701400), the National (OAE 2). Earth Planet. Sci. Lett. 267, 1–16. Science Foundation of China (Grant 40802012), “111 project” (B07011) Olsen, P.E., Kent, D.V., 1999. Long-period Milankovitch cycles from the Late Triassic and and Program for New Century Excellent Talents (Grant NCET-04-0727) Early Jurassic of Eastern North America and their implications for the Calibration of the Early Mesozoic time-scale and the long-term behaviour of the planets. Philos. of Ministry of Education of China. Trans. Math. Phys. Eng. Sci. 357, 1761–1786. Paillard, D., Labeyrie, L., Yiou, P., 1996. Macintosh program performs time-series analysis. Eos 77, 379. References Pei, F.P., Xu, W.L., Yang, D.B., Zhao, Q.G., Liu, X.M., Hu, Z.C., 2007. Zircon U-Pb geochronology of basement metamorphic rocks in the Songliao Basin. Chin. Sci. Bull. 52, 942–948. Arthur, M.A., Dean, W.E., Pratt, L.M., 1988. Geochemical and climatic effects of increased Prokoph, A., Agterberg, F.P., 1999. Detection of sedimentary cyclicity and stratigraphic marine organic carbon burial at the Cenomanian/Turonian boundary. Nature 335, completeness by wavelet analysis: an application to Late Albian Cyclostratigraphy 714–717. of the Western Canada Sedimentary Basin. J. Sediment. Res. 60, 862–875. Caron, M., Robaszynski, F., Amedro, F., Baudin, F., Deconinck, J.F., Hochuli, P., von Salis- Prokoph, A., Agterberg, F.P., 2000. Wavelet analysis of well-logging data from oil source Perch Nielsen, K., Tribovillard, N., 1999. Estimation de la durée de lévénement rock, Egret Member, offshore eastern Canada. AAPG Bulletin 84, 1617–1632. anoxique global au passage Cénomanien/Turonian: Approche cyclostratigraphique Prokoph, A., Thurow, J., 2000. Diachronous pattern of Milankovitch cyclicity in late dans la formation Bahloul en Tunisie central. Bull. Soc. Géol. Fr. 170, 145–160. Albian pelagic marlstones of the North German Basin. Sediment. geol.134, 287–303. H. Wu et al. / Earth and Planetary Science Letters 278 (2009) 308–323 323

Prokoph, A., Villeneuve, M., Agterberg, F.P., 2001. Geochronology and calibration of Wang, P.J., Liu, W.Z., Shan, X.L., Bian, W.H., Ren, Y.G., Du, X.D., Wan, C.B., Cheng, R.H., global Milankovitch cyclicity at the Cenomanian–Turonian boundary. Geology 29, 2001. Depositional Events: Introduction, Example, Application. Jilin Science and 523–526. Technology Press, Changchun, pp. 1–182 (in Chinese with English abstract). Sageman, B.B., Meyers, S.R., Arthur, M.A., 2006. Orbital time scale and new C-isotope Wang, P.J., Liu, W.Z., Wang, S.X., Song, W.H., 2002. 40Ar/39Ar and K/Ar dating on the record for Cenomanian–Turonian boundary stratotype. Geology 34 (2), 125–128. volcanic rocks in the Songliao basin, NE China: constraints on stratigraphy and Schlanger, S.O., Jenkyns, H.C., 1976. Cretaceous anoxic events: causes and consequences. basin dynamics. Int. J. Earth Sci. (Geol. Rundsch.) 91, 331–340. Geol. Mijnb. 55, 179–184. Wang, Y., Zhang, F.Q., Zhang, D.W., Miao, L.C., Li, T.S., Jie, W.Q., Meng, Q.R., Liu, D.Y., 2006. Schlanger, S.O., Arthur, M.A., Jenkyns, H.C., Scholle, P.A., 1987. The Cenomanian– Zircon SHRIMP U-Pb dating of meta-diorite from the basement of the Songliao Turonian Oceanic Anoxic Event I, stratigraphy and distribution of organic carbon- Basin and its geological significance. Chin. Sci. Bull. 51, 1877–1883. rich beds and the marine δ13C excursion. Geological Society, London, Special Wang, P.J., Xie, X.A., Frank, M., Ren, Y.G., Zhu, D.F., Sun, X.M., 2007. The Cretaceous Publications, vol. 26, pp. 371–399. Songliao Basin: valcanogenic succession, sedimentary sequence and tectonic Schulz, M., Mudelsee, M., 2002. REDFIT: estimating red-noise spectra directly from evolution, NE China. Acta Geol. Sin. 81, 1002–1011. unevenly spaced paleoclimatic time series. Comput. Geosci. 28, 421–426. Weedon, G., 2003. Time-Series Analysis and Cyclostratigraphy. Cambridge University Serra, O., 1984. Fundamentals of well-log interpretation. 1. The acquisition of logging Press, Cambridge, pp. 1–259. data. Amsterdam, Elsevier. Dev. Pet. Sci. 15A, 1–423. Westerhold, T., Röhl, U., Laskar, J., Raffi, I., Bowles, J., Lourens, L.J., Zachos, J.C., 2007. On the Sha, J.G., 2007. Cretaceous stratigraphy of northeast China: non-marine and marine duration of magnetochrons C24r and C25n and the timing of early Eocene global correlation. Cretac. Res. 28, 146–170. warming events: implications from the Ocean Drilling Program Leg 208 Walvis Sinton, C.W., Duncan, R.A., 1997. Potential links between ocean plateau volcanism and Ridge depth transect. Paleoceanography 22, PA2201. doi:10.1029/2006PA001322. global ocean anoxia at the Cenomanian–Turonian boundary. Econ. Geol. 92, 836–842. Wu, H.C., Zhang, S.H., Sui, S.W., Huang, Q.H., 2007. Recognition of Milankovitch Cycles in Stratigraphy Committee of China, 2002. Introduction to the regional chronostratigraphic the natural gamma-ray logging of Upper Cretaceous terrestrial strata in the (geological time) chart of China. Geological Publishing House, Beijing, pp. 1–72 (in Songliao Basin. Acta Geol. Sin. 81, 996–1001. Chinese). Ye, D.Q., Huang, Q.H., Zhang, Y., Chen, C.R., 2002. Cretaeeous Osrracoda Biostratigraphy Sun, S.W., Shu, L.S., Zeng, Y.W., Cao, J., Feng, Z.Q., 2007. Porosity–permeability and in Songliao Basin. Petroleum Industry Press, Beijing, pp. 11–144 (in Chinese with textural heterogeneity of reservoir sandstones from the Lower Cretaceous Putaohua English abstract). member of Yaojia Formation, Weixing Oilfield, Songliao Basin, Northeast China. Zhang, F.Q., Pang, Y.M., Yang, S.F., Dong, C.W., Chen, H.L., Su, P., 2007a. Geochronology of Mar. Pet. Geol. 24, 109–127. zircon SHRIMP, geochemistry and its implication of the volcanic rocks from Torrence, C., Compo, G.P.,1998. A practical guide to wavelet analysis. Bull. Am. Meteorol. Yingcheng Formation in depression Area, north of Songliao Basin. Acta Geol. Sin. 81, Soc. 79, 61–78. 1248–1258 (in Chinese with English abstract). Turgeon, S.C., Creaser, R.A., 2008. Cretaceous oceanic anoxic event 2 triggered by a Zhang, Z.L., Liu, Z.W., Wang, B.C., Zhang, Y., Ye, D.Q., 2007b. Ostracod biostratigraphy of massive magmatic episode. Nature 454. doi:10.1038/nature07076. the Late Cretaceous Qingshankou formation in the Songliao Basin. Acta Geol. Sin. 81, Wan, X.Q., Li, G., Chen, P.J., Yu, T., Ye, D.Q., 2005. Isotope stratigraphy of the Cretaceous 727–738. Qingshankou Formation in Songliao Basin and its correlation with marine Zhao, X., Coe, R.S., Zhou, Y., Wu, H., Wang, J., 1990. New paleomagnetic results from Cenomanian stage. Acta Geol. Sin. 79, 150–156 (in Chinese with English abstract). Northern China: collision and suturing with Siberia and Kazakhstan. Tectonophysics Wan, X.Q., Chen, P.J., Wei, M.J., 2007. The Cretaceous system in China. Acta Geol. Sin. 81, 181, 43–81. 957–983.