Journal of Earth Science  2017, Vol. 28 Issue (6): 1006-1019   PDF    
0
Division and Characteristics of Shale Parasequences in the Upper Fourth Member of the Shahejie Formation, Dongying Depression, Bohai Bay Basin, China
Jing Wu1,2,3, Zaixing Jiang4    
1. Petroleum Exploration and Production Research Institute of SINOPEC, Beijing 100083, China;
2. State Key Laboratory of Shale Oil and Gas Enrichment Mechanisms of SINOPEC, Beijing 100083, China;
3. SINOPEC Key Laboratory of Shale Gas/Oil Exploration & Production, Beijing 100083, China;
4. Faculty of Energy Resources, China University of Geosciences, Beijing 100083, China
Abstract: Shale parasequence analysis is an important part of sequence stratigraphy sudies. This paper proposed a systematic research method for analyzing shale parasequences including their delineation, division, characteristics and origins. The division method is established on the basis of lithofacies. Multi-method analysis and mutual verification were implemented by using auxiliary indicators (such as mineral compositions, geochemical indicators and wavelet values). A typical shale parasequence comprises a lower interval of deepening water-depth and an upper interval of shallowing water-depth (e.g., a shale parasequence including a high-total organic carbon (TOC) shale-low-TOC limy shale). Abrupt increases in pyrite content, TOC value, relative hydrocarbon generation potential ((S1+S2)/TOC), and wavelet values are indicative of parasequence boundaries. The proposed research method was applied to study the upper fourth member of the Shahejie Formation in the Dongying depression, Bohai Bay Basin. Results show that there were seven types of parasequences developed. A singular and a dual structured parasequences were identified. Three factors controlling the development of the shale parasequences were identified including relative lake level change, terrestrial input and transgression. The development of high-TOC ( > 2%) shale parasequences was mainly controlled by biological and chemical sedimentation. The low-TOC ( < 2%) shale parasequences were mainly deposited by chemical sedimentation. The diversities of shale parasequences were caused by four major controlling factors including climate, relative lake level change, terrestrial input and emergency (e.g., transgression).
Keywords: shale    parasequence    Dongying depression    Shahejie Formation    
0 INTRODUCTION

Shale oil and gas has recently become a research focus for petroleum exploration (Liang et al., 2017a, b, 2016; Li et al., 2016). Shale parasequence analysis is crucial in predicting organic richness zones, mineralogical affinity and fracture potential of fine-grained sediments, all of which play significant roles in controlling the porosity, permeability and hydrocarbon generation potential (S1+S2) of shales (Singh, 2008). Such research can provide useful insights into the exploration of shale oil and gas. As shales appear to be homogeneous and do not exhibit significant vertical changes in lithofacies or sediment grain sizes, it is quite challenge to delineate and divide shale parasequences. Most of the recent third-order sequence (Angulo and Buatois, 2012; Smith and Bustin, 2000), a few of which are designed for parasequence analysis. To date, there has been no systematic research method for shale parasequence analysis available. Therefore, the study of shale parasequences is of great importance and urgently needed (Slatt and Rodriguez, 2012; Catuneanu, 2006).

The existing division methods only rely on one particular aspect of geological characteristics. For example, the gamma ray profile can be classified into the upward-increasing API (American Petroleum Institute) intervals, upward-decreasing API intervals, and intervals of constant API. These three kinds of intervals are bounded by gamma ray kicks, i.e., well-log based flooding surfaces, which can be termed as gamma ray parasequences (Slatt and Abousleiman, 2011; Singh, 2008). Additionally, the oil-prone shale are commonly formed by a series of superimposed depositional TOC units (high TOC values at the base, decreasing upward) (Creaney and Passey, 1993). One TOC unit was thought to represent a parasequence (Liu et al., 2011). However, no single unique feature in a stratigraphic section can be used to make an interpretation (Loucks and Rupple, 2007). The interpretation should base on a variety of characteristics indicators, especially for shale. The existing methods are limited and lack of contacting with third-and fourth-order sequences which providing the frame for parasequences.

To figure out these problems, this study aims to establish a systematic research method including division, characteristics and origins analysis for shale parasequences. Lithofacies is the most direct, comprehensive, and essential indicator to reflect sequence stratigraphy change. Therefore, lithofacies characteristics and vertical superimposition patterns are chosen as the mainline during sequence stratigraphy research. Firstly, petrologic features are analyzed. Then, multi-method analysis and mutual verification are implemented by incorporating vertical changes in mineral compositions and geochemical indicators to divide third-and fourth-order sequence that lay a framework for parasequences research. After that, a division method of parasequences can be built. Various types of parasequences with different characteristics can be divided in a stratigraphic framework of third-and fourth-order sequence. Origins of different parasequences are analysed finally. Our method is verified by using the upper fourth member of the Shahejie Formation (Es4s) in the Dongying depression, Bohai Bay Basin.

1 GEOLOGICAL SETTING

Dongying depression is a third-order tectonic unit of Bohai Bay Basin and a Mesozoic-Cenozoic fault-depression basin (Figs. 1a, 1b). It is surrounded by four uplifts and characterized by north-steep and south-glacis (Fig. 1c). The depression has an exploration area of approximately 5 760 km2(Wu et al., 2014). It develops Paleozoic, Mesozoic, and Cenozoic strata from the bottom to the top, and the Cenozoic strata include the Paleogene, Neogene, and Quaternary strata. The Paleogene strata are generally thick and consist of Kongdian (Ek), Shahejie (Es), and Dongying (Ed) formations. The Es Formation can be divided into four members (Es1-4). The objective stratum in this study is Es4s, the upper part of Es4.

Download:
Figure 1. The geological background of the upper fourth member of Shahejie Formation in the Dongying depression, Bohai Bay Basin, China. (a) Dongying depression is boarded by the Luxi uplift in the south, the Chenjiazhuang uplift in the north, the Qingtuozi uplift in the east, and the Binxian-Qingcheng uplift in the west. During the period of the Es4s, there are four sags which are Boxing, Niuzhuang, Lijin and Minfeng sag; (b) geological setting of Bohai Bay Basin during the Paleogene. Dongying depression is located in the south of Jiyang depression of Bohai Bay Basin; (c) N-S trending section in the study area.

The Es4s is developed by intense rift with fast basin expansion and large extent of subsidence. Vast semideep-deep water shales are distributed in the center of the depression (Zhao, 2005), which are thicker than 1 000 m. Fan delta, nearshore subaqueous fan, and beach-bar siltstone distribute along the margin of the depression (Yang et al., 2011; Wang, 2005; Yan et al., 2005).

The appearance of glauconite(Wu et al., 2014), marine biohermal limestone (Qian et al., 1980), algae (e.g., Chinese Cladosiphon and dinoflagellates), gastropods, clupeomorpha, and Paleodictyon (Yuan et al., 2006; Xu et al., 1997; Zhang et al., 1985; Zhu, 1979) indicate the occurrence of transgression during Es4s in the study area. As the influence of the transgression is paroxysmal and accidental, the study area overall is lacustrine environment.

2 DATABASE AND METHODOLOGY 2.1 Data

A total of 835.94 m continuous cores and 1 141 samples of shales were obtained from 7 wells (NY1, FY1, Niu38, Wang31, Fan120, Li673, Niu872) in the target interval (Fig. 1a, Table 1). All samples were measured in the formation laboratory of Shengli oilfield, Shandong, China. The relative content of each mineral composition of all samples were measured by using X'Pert-MPD diffraction instrument (Philips Corp.): copper butt, pipe pressure 30 kV, conduit flow 40 mA, scanning speed 2°(2θ)/min. When mineral content is more than 40%, relative standard deviation (RSD) is less than 10%. When it is between 20%-40%, RSD is less than 20% (Wang et al., 1996). Three hundred thin sections were observed to analyze the sedimentary characteristic using an optical microscope (Axio Scope A1). Sixty-five samples were grounded to the grain of 200-mesh and subjected by inductively coupled plasma-atomic emission spectroscopy ICP-AES (JY38S) with a focus on the elements of Ca, Al, Mg, etc. Ambient temperature is between 70-75 ℃. Relative humidity is less than 70%. RSD is less than 1-10 ppb.

Table 1 Average value of major mineral compositions of shale in the upper fourth member of Shahejie Formation in Dongying depression, Bohai Bay Basin, China

A total of 256 samples were submitted to Rock-Eval pyrolysis for determination of TOC values, the free hydrocarbons (S1, mg HC/g rock), and the hydrocarbons cracked from kerogen (S2, mg HC/g rock). The TOC values were determined using Rock-Eval-Ⅵ (Cat. No. 2-06-11) from France. Measurement technique is based on the combustion of the sample in an oxygen atmosphere convert the total organic carbon to CO2. With the aid of combustion calculation, the total organic carbon content of the samples can be determined (Wu et al., 2001; Charles and Simmons, 1986). Temperature is between 60-80 ℃ during the process.

For S1 and S2, the samples were heated in helium flow. S1 were detected by using hydrogen flame ionization detector. S2 were detected by thermal conductivity detector. The absolute standard deviation is less than or equal to 0.01 mg/g.

Well logging data have the best continuity among the geological data. However, they contain random noise component. Cyclicity is not obvious. Wavelet is an appropriate tool transforming the signal and one of the most common methods to study stratigraphic cycles (Liu, 2012). Considering the self-similarity and coefficients of various wavelets (Yan et al., 2011), daubechies wavelet (db) can represent the cyclicity of large and small depositional units in the study area. Among the well logging series, gamma-ray logs (GR) match the most correlation to multiscale information of sedimentary cyclicity of source rock (Passey et al., 1990).

The MATLAB software (7.0.1 version, produced from the MathWorks in American) was used to perform one-dimensional continuous transform 11 times of GR curve with db wavelet. Eleven wavelets represent different level of sedimentary cycle.

2.2 Methods 2.2.1 Lithofacies association

The parasequence is a relatively conformable succession of genetically related beds or bedsets bounded by lake flooding surfaces (LFS) and their correlative surfaces indicating that there exists evidence for sudden increase in water depth through the surfaces. A parasequence generally comprises the lower part indicating deepening water depth and the upper part indicating the shallowing water depth (van Wagoner et al., 1990). A typical shale parasequence, which is different from obvious combinations of mudstone-gravel/sandstone of coarse grained sedimentary rocks in the shallow-water, comprises a lower interval deepening water-depth and an upper interval of shallowing water-depth. The change from lower interval to upper interval reflects the shallowing-upward process of water depth.

2.2.2 Mineral compositions

Although mineral compositions change with specific lithofacies, some minerals can reflect the change of water depth. For example, pyrite is proportional to the depth and reducibility of water (Zhang and Ren, 2003; Wilkin et al., 1996; Deng and Qian, 1990; Berner, 1984). The pyrite content changes from high to low indicating the decrease of water depth, which helps to divide parasequence.

2.2.3 Organic geochemical indicators

The change of organic geochemical indicators including TOC and relative hydrocarbon potential (RHP) [(S1+S2)/TOC] levels are proportional to the water depth (Slatt et al., 2012; Arthur and Sageman, 2004). A decreasing after increasing of these indicators indicating a parasequence. The subtle changing points of them are parasequence boundaries. The Th/U ratio is inversely proportional to water depth (Davies and Elliott, 1996). It increases at first and then decreases indicating a parasequence.

2.2.4 Wavelet value

Discontinuities in shales do not show obvious change and cannot be detected by well logs directly. The wavelet analysis can achieve better effect. Wavelet value has periodic changes from high to low levels reflecting the shallowing upward water bodies (Prokoph and Agterberg, 2000; Miall, 1992; Rioul and Vetterli, 1991). It can be used to divide shale parasequence. After wavelet transform, the change of d3 wavelet value is consistent with the other indicators.

3 RESULTS 3.1 Lithofacies Classification, Characteristics and Depositional Models 3.1.1 Lithofacies classification

The X-ray diffraction data show that, the major mineral compositions of the Es4s shale in the Dongying depression are calcite, clay minerals and quartz. Other mineral compositions involve dolomite, pyrite and feldspar. The average content of calcite is 33.47%. Its maximum can reach up to 80%. The average content of clay minerals and quartz is 24.13% and 23.51%, respectively. The organic matter value is between 0.11%-11.4% with an average value of 2.49% (Table 2). According to the contents of various mineral compositions and sedimentary characteristics, a total of 7 kinds of lithofacies are identified (Table 1).

Table 2 Mineral compositions of different shale in the upper fourth member of Shahejie Formation in Dongying depression, Bohai Bay Basin, China
3.1.2 Lithofacies characteristics 3.1.2.1 Oil shale

Carbonate content is greater than 50%. TOC values are greater than 4% (ave. 4.32%). Pyrites are abundant with an average content of 3.28% and maximum of 10% (Table 2). Smooth laminae are developed with obvious light and dark laminae. Light laminae are dominated by columnar calcite crystals formed by recrystallization. Dark laminae are organic-rich clay minerals layers. Laminae are less affected by water turbulence (Figs. 2a-2c). For the abundant of TOC and pyrite, combined with the characteristics of sedimentary structure, this type of shale is inferred to form in a strongly reducing environment of deep water.

Download:
Figure 2. Shale characteristics of the upper fourth member of Shahejie Formation in Dongying depression, Bohai Bay Basin, China. (a) Well FY1, 3 324.79 m, grey black oil shale. Arrow points to the calcite laminae with recrystallization. Strong bubble occurred after dropping hydrochloric acid; (b) Well FY1, 3 324.79 m, oil shale with columnar calcite crystals under cross-polarized light (+) (modified after Wu et al., 2016, arrow), boundaries between light and dark laminae are clear; (c) Well LY1, 3 662 m, field emission scanning electron microscope image of oil shale with abundant framboidal pyrite (arrow); (d) Well LY1, 3 662.1 m, dark grey high-TOC limy shale with a certain degree of recrystallization (arrow); (e) Well LY1, 3 662.1 m, (+), high-TOC limy shale with grainy crystals (arrow). Grainy calcite size changes large, boundaries between light and dark laminae are relatively clear; (f) Well NY1, (+), 3 410.86 m, high-TOC limy shale, pyrite content is high (arrow); (g) Well LY1, 3 631.5 m, light grey low-TOC limy shale, boundaries between light and dark laminae are not smooth; (h) Well LY1, 3 631.5 m, (+), low-TOC limy shale, light laminae are micritic calcite and thick (arrow); (i) Well NY1, 3 430.38 m, light grey low-TOC shale; (j) Well NY1, 3 430.38 m, (+), low-TOC shale, dark laminae are thick comprising clay minerals and little organic matter; (k) Well NY1, 3 498.3 m, grey white low-TOC gypsiferous shale; (l) Well NY1, 3 498.3 m, (+), low-TOC gypsiferous shale, light laminae comprise gypse (arrow), influenced by evaporation and water turbulence, boundaries between light and dark laminae are not clear; (m) Well NY1, 3 373.79 m, grey carbonate-bearing silty shale; (n) Well NY1, 3 373.79 m, (+), carbonate-bearing silty shale comprising micritic and grainy calcites, bioclast, clay, pyrite, etc. (arrows); (o) Well NY1, 3 463.59 m, grey dolomitic-bearing silty shale; (p) Well NY1, 3 463.59 m, (+), dolomitic-bearing silty shale, the lower laminae comprise terrigenous material and the upper laminae comprise micritic dolomite (arrows).
3.1.2.2 High-TOC limy shale

Carbonate content is greater than 50%. TOC values are commonly between 2%-4% (ave. 2.7%). Pyrites are relatively abundant with an average content of 2.22%. TOC value and pyrites content are less than oil shale (Table 3). Laminae are developed with relatively pure light and dark laminae. Light laminae comprise grainy calcite by a certain degree of recrystallization. Dark laminae mainly comprise organic matter and clay minerals. Laminae are influenced by weak water turbulence (Figs. 2d-2f). Based on the minerals compositions and sedimentary structure, this type of lithofacies was formed in a reducing environment of semi-deep water.

Table 3 Origins, types and sedimentations of shale parasequences in the upper fourth member of Shahejie Formation in Dongying depression
3.1.2.3 Low-TOC limy shale

Carbonate content is greater than 50%. TOC values are less than 2%. The content of pyrite is low with an average content of only 1.83% (Fig. 2). The quartz particles with angular edge and bioclast can be found. Laminae are developed with wave distribution of laminae. The thickness of single laminae is about 60-120 μm. Light laminae are dominated by micritic calcite and dark laminae are clay minerals layers with little organic matter (Figs. 2g, 2h). This type of lithofacies was formed in a weakly reducing environment of shallow water.

3.1.2.4 Low-TOC shale

Low-TOC shale is similar with low-TOC limy shale. But clay minerals content of low-TOC shale is greater than 50% with average TOC values less than 2% (Fig. 2). Laminae are developed. Light laminae comprise micritic calcite and dark laminae comprise clay minerals. This type of lithofacies was formed in a weakly reducing environment of shallow water (Figs. 3i, 3j).

Download:
Figure 3. Shale depositional model of the upper fourth member of Shahejie Formation in the Dongying depression (modified after Wu et al., 2016).
3.1.2.5 Low-TOC gypsiferous shale

Gypsums, typical products of drought condition and evaporation, with an average content of 27.75%, distribute in light laminae. Dark laminae comprise clay minerals. Average value of TOC is only 0.67% (Fig. 2). This type of lithofacies was formed in shallow water (Figs. 3k, 3l).

3.1.2.6 Carbonate-bearing silty shale

Compared with oil shale or limy shale, carbonate content of this type of shale is relatively low with an average value of 42.41%. Mainly calcites are micritic and few are grainy calcites. However, the content of quartz and feldspar is higher than other shale. The average value of TOC is 2.64%. Pyrites are abundant with an average value of 3.17% (Table 2). Such lithofacies was formed in a reducing environment of semi-deep water and influenced by terrigenous material (Figs. 2m, 2n).

3.1.2.7 Dolomitic-bearing silty shale

Dolomite is micritic with an average value of 34.2%. TOC value is between 1.5%-3.64% and lower than high-TOC limy shale and carbonate-bearing silty shale. Pyrites are abundant too with an average value of 3.71% (Table 2). This type of lithofacies was developed only in the early stage of Es4s with a reducing environment of semi-deep water (Figs. 3o, 3p).

3.1.3 Lithofacies depositional model

Based on 7 kinds of lithofacies with different sedimentary characteristics and environment, a depositional model of Es4s shales proposed (Fig. 3). Organic matter and pyrite are abundant for the favorable preservation condition of strongly reducing environment of deep water. Smooth laminae are developed for almost without water turbulence. In this kind of environment, lithofacies is dominated by oil shale. TOC value is generally between 2%-4% in reducing environment of semi-deep water. Laminae are developed with weak water turbulence. The lithofacies are developed from the lower to upper sections according to the water depth, including high-TOC limy shale, carbonate-bearing silty shale and dolomitic-bearing silty shale. TOC values and pyrite content are both low in the weakly reducing environment of shallow water. Laminae are developed with wave distribution as influenced by water turbulence. The lithofacies are dominated by low-TOC shale and low-TOC limy shale. In a drought conditions, water evaporation lead to the formation of low-TOC gypsiferous shale and saline lake.

3.2 Division and Characteristics of Third-and Fourth-Order Sequence

One whole third order sequence including lowstand systems tract (LST), transgressive systems tract (TST) and highstand systems tract (HST) was developed in the Es4s Well NY1 is taken as a typical example (Fig. 4).

Download:
Figure 4. Sequence division of the upper fourth member of Shahejie Formation in Well NY1, Dongying depression, Bohai Bay Basin, China.
3.2.1 LST

The interval of 3 492-3 498.5 m (Es4s not penetrated) mainly comprises low-TOC gypsiferous shale and low-TOC shale. The Th/U ratio increases upward while the TOC value decreases upward. These features indicate that water continues to shallow upward. In this interval, pollen are dominated by Ephedripites, Labitricolpites, Caryapollenites, Quercoidites indicating arid climate that formed an evaporation environment and caused continuous decline in water depth. On the top of this interval, there is a surface of discontinuity (gypsiferous deposition turned to limy sediments)—first flooding surface. Based on the comprehensive analysis, this interval is LST.

3.2.2 TST 3.2.2.1 TST-1 (passive lake-level rising phase)

In the interval of 3 434-3 492 m, the lithofacies are dominated by dolomitic-bearing silty shale interbedded with low-TOC limy shale and upward-increasing high-TOC limy shale. In the middle segment, there is a set of turbidity deposition. TOC value and pyrite content increase upward except the segment of turbidite depositsions. Th/U ratio undergoes opposite change. Above-mentioned characteristics indicate that the lake level experienced continuous rise including two retrograding parasequence sets (Fig. 4, S2-S3). One thing is worth to note that four special sets of sedimentary strata (Fig. 4, ①-④) are present with dolomite shale, with unusually high salinity (Sr/Ba is proportion to the salinity, Jiang et al., 1994), Mg levels, TOC value and abnormally low values of trace elements (Cr and Pb). These sets are affected by transgression responding to the geological setting of this study area. The lake level passively rose due to the influence of four phase's transgression, leading to the development of TST-1.

3.2.2.2 TST-2 (normal lake-level rising phase)

In the interval of 3 374-3 434 m, high-TOC limy shale and upward-increasing oil shale are developed. TOC value peaks in the whole well segment, i.e., 11.4% (3 374.19 m). The values of Sr/Ba and Th/U decrease. This interval recovered to a normal lake deposition after the transgression and composed with three retrograding parasequence sets (Fig. 4, S4-S6). In this interval, pollen are dominated by Taxodiaceae, Ulmoideipites indicating warm-damp climate of the study area. Water depth is the greatest on the top of the interval influenced by the climate, associated with the development of the maximum flooding surface. Thus, this interval belongs to TST and is classified as TST-2 in order to distinguish from the TST-1.

3.2.3 HST

In the interval of 3 326-3 374 m, two sets of lithofacies associations are developed and each set include carbonate-bearing silty shale and upward-increasing low-TOC limy shale. The water depth decreases upward. These two sets which have feature of superposition patterns (Fig. 4, S7-S8). Based on the obviously increasing Th/U ratio, sustainably decreasing TOC value and pyrite content, this interval belongs to HST.

3.3 Division Method of Fifth-Order Sequence (Parasequence)

In this study, parasequences are divided with a combination of shale reflecting the shallowing-upward process of water depth and verified with data of mineral compositions, geochemical indicators and wavelet in Well NY1. The thickness of parasequence within the range from 2 to 4 m.

During the division, we notice the consistency of the lithofacies occasionally, with minor changes in mineral compositions and geochemical indicators. TOC and RHP levels undergo increase followed by decrease indistinctively. These features suggest that the deepening-shallowing change in water depth is not significant, accounting for a lack of change in the type of lithofacies. This kind of section also represents a parasequence with thickness of 2-3 m.

Seven types of parasequence are identified using the proposed method. Furthermore, they are divided into two categories, including the unitary and dual structures, according to the amount of lithofacies type and changes of other indicators.

3.4 Characterization of Parasequence 3.4.1 Dual structure 3.4.1.1 Oil shale and High-TOC limy shale

This kind of parasequence is dominated by grey black oil shale (Fig. 5a) in the lower interval (Fig. 5 blue arrows on core picture), with dark grey high-TOC limy shale (Figs. 5b, 5c) developed in the upper interval (Fig. 5 green arrows on core picture). From the lower to upper interval, TOC, S1+S2 and d3 level rise rapidly and decline slowly. These features reflect a suddenly increase from semi-deep water to deep water through the LFS followed by a decrease back to semi-deep water (Fig. 5 columnar section).

Download:
Figure 5. Characteristics of parasequence comprising oil shale and high-TOC limy shale of the upper fourth member of Shahejie Formation in Well NY1, Dongying depression, Bohai Bay Basin, China. On the core picture, the blue and green arrows respectively represent the lower and upper part of the parasequence. The numbers from 1 to 3 and the five-pointed star respectively show the location of sample in the table and thin section. LFS is lake flooding surface. (a) 3 432.17 m, (+), Oil shale with columnar calcite (CC, arrow); (b) 3 431.59 m, (+), high-TOC limy shale, abundant pyrite (P) and remains of higher plants (R, arrows); (c) 3 431.59 m, (+), high-TOC limy shale, remains of higher plants with grainy calcite (GC, arrow).

This kind of parasequence mainly distributes in the upper TST-1 and lower TST-2 with upward-increasing thickness (Fig. 4). Warm-damp climate and deep water of TST afforded abundant algae and strong reducibility that helped the preservation of organic matter in the lower interval of parasequence. The limy depositions are formed by chemical precipitation with low terrestrial input and recrystallized by organic acid produced by the evolution of biological organic matters (Fig. 5c). Water depth decreased relatively in the upper interval of parasequence. This change reduced the preservation condition of organic matter, which lead to the reduction of TOC value and transition from oil shale to high-TOC limy shale.

3.4.1.2 High-TOC limy shale and Low-TOC limy shale

In the lower interval of parasequence (Fig. 6 blue arrows), low-TOC limy shale quickly tends to dark grey high-TOC limy shale (Fig. 6a), with increasing-upward light grey low-TOC limy shale in the upper interval (Fig. 6 green arrows, b). Pyrite content, TOC value, S1+S2 and d3 level increase significantly followed by decrease. The water depth deepens quickly from shallow water to semi-deep water and then shallows slowly back to shallow water (Fig. 6 columnar section). This kind of parasequence mainly distributes in TST and occasionally occurred in HST (Fig. 4).

Download:
Figure 6. Characteristics of parasequence comprising high-TOC limy shale and low-TOC limy shale of the upper fourth member of Shahejie Formation in Well NY1, Dongying depression, Bohai Bay Basin, China. (a) 3 425.98 m, high-TOC limy shale with grainy calcite under plane-polarized light (-) (GC, arrow); (b) 3 424.29 m, low-TOC limy shale with micritic calcite (-) (MC, arrow).

Influenced by the warm-damp climate, deep water and a small amount of terrigenous material in TST, algae were plenty and organic matters were preserved well. Limy deposition were formed through the chemical precipitation, combined with high TOC value. High-TOC limy shale was developed in the lower interval of parasequence. When water depth decreased relatively with decreasing reducibility in the upper interval of parasequence, organic matters were poorly preserved and low-TOC limy shale was developed.

3.4.1.3 Low-TOC shale and low-TOC gypsiferous shale

After the development of low-TOC gypsiferous shale in the bottom, the lower interval of parasequence is quickly dominated by light grey low-TOC shale (Fig. 7 blue arrows, A), with increasing-upward grey white low-TOC gypsiferous shale in the upper interval (Fig. 7 green arrows, B). At the same time, pyrite content, TOC value, S1+S2 and d3 level obviously increase followed by decrease. These features reflect a sudden increase through the LFS followed by decrease in water depth (Fig. 7 columnar section).

This kind of parasequence only distributes in the parasequence set 1 of LST (Fig. 4) with arid climate, shallow water and high level of salinity (Fig. 7 increasing-upward Sr/Ba). When the terrigenous material decreased (Fig. 7 table, quartz & feldspar), gypsiferous depositions were developed through chemical precipitation.

Download:
Figure 7. Characteristics of parasequence comprising low-TOC shale and low-TOC gypsiferous shale of the upper fourth member of Shahejie Formation in Well NY1, Dongying depression. (a) 3 498.3 m, (+), low-TOC shale; (b) NY1 well, 3 495.29 m, (+), low-TOC gypsiferous shale, gypsum (G, arrow).
3.4.2 Unitary structure 3.4.2.1 Low-TOC limy shale

Lithofacies type is consistently dominated by low-TOC limy shale (Figs. 8a, 8b) from the lower to upper interval of the parasequence (Fig. 8 blue and green arrows). There is no significant change in mineral compositions too. But there are some changes in the color of core (Fig. 8, core of 3 434.78 m is more black than other). TOC value, S1+S2 and d3 level slight increase followed by decrease. Although all of these values are low with weak changes, these still reflect an increase through the LFS followed by decrease in water depth (Fig. 8 columnar section). This kind of parasequence mainly distributes in the interval of TST-1. Climate began to warm-damp and water depth started to deepen in TST-1 (Fig. 4). Algae were deficient. Chemical precipitation lead to the formation of limy deposition.

Download:
Figure 8. Characteristics of parasequence comprising low-TOC limy shale of the upper fourth member of Shahejie Formation in Well NY1, Dongying depression, Bohai Bay Basin, China. (a) 3 434.56 m, (+), low-TOC shale; (b) 3 434.56 m, (+), low-TOC limy shale, laminae of micritic calcite with abundant fragments of Ostracoda and quartz (F and Q, arrows) indicate shallow water.
3.4.2.2 High-TOC limy shale

The parasequence is consistently dominated by high-TOC limy shale from the lower to upper interval (Fig. 9 blue and green arrows, a). The changes of TOC value and S1+S2 level are same with the last kind of parasequence (low-TOC limy shale) and and still reflect a mildly increase through the LFS followed by decrease in water depth (Fig. 9 columnar section). The value of d3 is steady.

Download:
Figure 9. Characteristics of parasequence comprising high-TOC limy shale of the upper fourth member of Shahejie Formation in Well NY1, Dongying depression, Bohai Bay Basin, China. (a) 3 375.87 m, (+), high-TOC limy shale, laminae of grainy calcite (GC, arrow); (b) 3 376.85 m, (+), high-TOC limy shale, laminae of pyrite (P, arrow).

This kind of parasequence mainly develops in the mid-late stage of TST-2 and occasionally in the stage of TST-1 and HST (Fig. 4). Rich organic matter supplied by algae and calcite deposited through chemical precipitation with semi-deep reducing water (Fig. 9b) that composed the parasequence. TOC value and S1+S2 level weakly decrease and indicate the upper interval of parasequence.

3.4.2.3 Carbonate-bearing silty shale

Although lithofacies are consistently dominated by carbonate-bearing silty shale (Fig. 10, blue and green arrows, A), the TOC value (generally greater than 2%), S1+S2 and d3 level increase followed by decrease and reflect an increase through the LFS followed by decrease in water depth (Fig. 10 columnar section). This kind of parasequence mainly develops in the HST, gradually thinning upward and transiting to high-TOC limy shale and low-TOC limy shale for the shallowing upward water depth (Fig. 4). The compositions are varied including calcite, remains of higher plants, clay and quartz (Fig. 10 table, a and b). Micritic calcites were mainly formed by chemical precipitation. Quartz particles with angular edge came from terrigenous material and were formed by mechanical action after erosion of parent rocks. Higher plants and algae help the biological sedimentation.

Download:
Figure 10. Characteristics of parasequence comprising carbonate-bearing silty shale of the upper fourth member of the Shahejie Formation in Well NY1, Dongying depression, Bohai Bay Basin, China. (a) 3 372.65 m, (+), carbonate-bearing silty shale, remains of higher plants, quartz and calcite (R, Q and C, arrow); (b) 3 372.65 m, (+), carbonate-bearing silty shale, calcite and remains of higher plants (C & R, arrow).
3.4.2.4 Dolomitic-bearing silty shale

This kind of parasequence mainly comprises dolomitic-bearing silty shale (Fig. 11 blue and green arrows, a). From the lower to upper interval, TOC value and S1+S2 levels increase followed by decrease with abnormal high values of salinity (Sr/Ba) (Fig. 11 columnar section). This kind of parasequence only develops in the TST-1 affected by transgression and comprises micritic dolomite (Fig. 11b) by chemical precipitation. Sea water offered the material basis. The overlying strata is high-TOC limy shale for the high lake level of water after the transgression (Fig. 4).

Download:
Figure 11. Characteristics of parasequence comprising dolomitic-bearing silty shale of the upper fourth member of the Shahejie Formation in Well NY1, Dongying depression, Bohai Bay Basin, China. (a) 3 466.15 m, dolomitic-bearing silty shale; and (b) 3 466.15 m, (+), dolomitic-bearing silty shale with micritic dolomite (MD, arrow).
4 DISCUSSIONS

In the dual structure, there are obvious changes in the lithofacies associations, mineral compositions, geochemical characteristics and wavelets to divide parasequence easily. In the unitary structure, a special or extreme form of dual structure, lithofacies type does not change because of the minor change in water depth during parasequence development. The lithofacies can be developed in relatively deep water (e.g., high-TOC limy shale) or shallow water (e.g., low-TOC limy shale). The division of shale parasequence of unitary structure most depends on slight changes of mineral compositions and geochemical indicators.

The origins and sedimentations of shale parasequence are diverse. Based on the characteristics of each kind of shale parasequence, geological setting and relationship with third-and fourth-order sequence, three kinds of origins and sediments are summarized (Table 3). The development of high-TOC ( > 2%) shale parasequence was mainly controlled by biochemical sedimentation. The development of low-TOC ( < 2%) shale parasequence was mainly dominated by chemical sedimentation.

The geological conditions of the study area were complex. The climate changed from arid-cold in LST to warm-damp in TST and formed a large climate span. The water depth changed from shallow water to deep water under different climate conditions and formed a large water depth span. Due to the changes of climate and water depth, the development of shale was affected by different levels of terrestrial input. The changes of climate and water depth also affected the biochemical condition which influence the TOC value and pyrite content in shale. Arid climate helped the development of gypsiferous deposition by chemical sedimentation. In addition, The TST-1 interval was affected by transgression. Thus, the diversities of shale parasequences were caused by the major controlling factors including climate, relative lake level change, terrestrial input and emergency (transgression).

Because of limited target interval and study area in this study, the kinds of shale parasequences may be more diverse. However, a systematic research method and specific research process has been established with wide applicability. For example, although the lithofacies are different under specific geological conditions of different areas, the combination of a lower interval deepening water-depth and an upper interval of shallowing water-depth is applicative to divide shale parasequences. The auxiliary indicators (mineral compositions, geochemical indicators and wavelet values) can be replaced by the other indicators in other area to divide shale parasequences. Periodical changes of these indicators show the development of parasequences.

5 CONCLUSIONS

Shale parasequence research is generally lacking in sedimentary geology and sequence stratigraphy. The division, characteristics and origins of shale parasequences are still poorly understood due to the absence of any systematic research methods. Our research marks a step forward in shale parasequence research with the following key contribution.

(1) A division method with four aspects of boundary identification for shale parasequence has been proposed in a reasonable stratigraphic framework of third-and fourth-order sequences, which overcomes the limitation of existing research methods relying only on a single indicator.

(2) On the basis of division, specific characteristics and origins of shale parasequences, a systematic research method was established. The lithofacies, mineral compositions, geochemical characteristics, as well as the wavelet features of a typical shale parasequence have been documented. Two categories including seven types caused by three origins of parasequences have been identified in consideration of the geological setting and features of the third-and fourth-order sequences features.

(3) The diversities of shale parasequences are caused by a combination of controlling factors including climate, relative lake level changes, terrestrial input and transgression.

(4) This systematic research method for shale parasequences analysis enriches the theory of sequence stratigraphy and our understanding of shale sedimentary geology and provides useful insight for guiding unconventional oil and gas exploration.

ACKNOWLEDGMENT


The research presented in this paper was supported by the National Science and Technology Special Grant of China (No. 2017zx05036-004). The final publication is available at Springer via https://doi.org/10.1007/s12583-016-0943-6.


REFERENCES
Angulo S., Buatois L. A., 2012. Integrating Depositional Models, Ichnology, and Sequence Stratigraphy in Reservoir Characterization: The Middle Member of the Devonian-Carboniferous Bakken Formation of Subsurface Southeastern Saskatchewan Revisited. AAPG Bulletin, 96(6): 1017-1043. DOI:10.1306/11021111045
Arthur M. A., Sageman B. B., 2004. Sea-Level Control on Source-Rock Development: Perspectives from the Holocene Black Sea, the Mid-Cretaceous Western Interior Basin on North America, and the Late Devonian Appalachian Basin. In: Harris N. B., ed., The Deposition of Organic-Carbon-Rich Sediments: Models, Mechanisms, and Consequences. SEPM Special Publication, 82: 35-59.
Berner R. A., 1984. Sedimentary Pyrite Formation: An Update. Geochimica et Cosmochimica Acta, 48(4): 605-615. DOI:10.1016/0016-7037(84)90089-9
Catuneanu O., 2006. Principles of Sequence Stratigraphy. Petroleum Industry Press, Beijing..
Charles M. J., Simmons M. S., 1986. Methods for the Determination of Carbon in Soils and Sediments—A Review. The Analyst, 111(4): 385-390. DOI:10.1039/an9861100385
Creaney S., Passey Q. R., 1993. Recurring Patterns of Total Organic Carbon and Source Rock Quality within a Sequence Stratigraphic Framework. AAPG Bulletin, 77(3): 386-401. DOI:10.1306/bdff8c18-1718-11d7-8645000102c1865d
Davies S. J., Elliott T., 1996. Spectral Gamma Ray Characterization of High Resolution Sequence Stratigraphy: Examples from Upper Carboniferous Fluvio-Deltaic Systems, County Clare, Ireland. Geological Society, London, Special Publications, 104(1): 25-35. DOI:10.1144/gsl.sp.1996.104.01.03
Deng H. W., Qian K., 1990. The Genetic Types and Association Evolution of Deep Lacustrinefacies Mudstones. Acta Sedimentologica Sinica, 8(3): 1-20.
Jiang N. Y., Jia R. F., Wang Z. Y., 1994. Permian Paleogeography and Geochemical Environment in Lower Yangtze Region, China. Petroleum Industry Press, Beijing..
Li Y. Z., Wang X. Z., Wu B., et al., 2016. Sedimentary Facies of Marine Shale Gas Formations in Southern China: The Lower Silurian Longmaxi Formation in the Southern Sichuan Basin. Journal of Earth Science, 27(5): 807-822. DOI:10.1007/s12583-015-0592-1
Liang C., Jiang Z. X., Cao Y. C., et al., 2016. Deep-Water Depositional Mechanisms and Significance for Unconventional Hydrocarbon Exploration. AAPG Bulletin, 100: 773-794. DOI:10.1306/02031615002
Liang C., Jiang Z. X., Cao Y. C., et al., 2017a. Sedimentary Characteristics and Origin of Lacustrine Organic-Rich Shales in the Salinized Eocene Dongying Depression. GSA Bulletin. DOI:10.1130/B31584.1
Liang C., Cao Y. C., Jiang Z. X., et al., 2017b. Shale Oil Potential of Lacustrine Black Shale in the Eocene Dongying Depression: Implications for Geochemistry and Reservoir. AAPG Bulletin. DOI:10.1306/01251715249
Liu Z. H., 2012. Orbital Cycles Analysis and Its Genesis Significance for the Sequence Hierarchy: A Case Study of Carboniferous Karashayi Formation, Central Tarim Basin. Journal of Earth Science, 23(4): 516-528. DOI:10.1007/s12583-012-0272-3
Liu Z. J., Sun P. C., Jia J. L., et al., 2011. Distinguishing Features and Their Genetic Interpretation of Stratigraphic Sequence in Continental Deep Water Setting: A Case from Qingshankou Formation in Songliao Basin. Earth Science Frontiers, 18(4): 171-180.
Loucks R. G., Ruppel S. C., 2007. Mississippian Barnett Shale: Lithofacies and Depositional Setting of a Deep-Water Shale-Gas Succession in the Fort Worth Basin, Texas. AAPG Bulletin, 91(4): 579-601. DOI:10.1306/11020606059
Miall A. D., 1992. Exxon Global Cycle Chart: An Event for Every Occasion?. Geology, 20(9): 787-790. DOI:10.1130/0091-7613(1992)020<0787:egccae>2.3.co;2
Passey Q. R., Creaney S., Kulla J. B., et al., 1990. A Practical Model for Organic Richness from Porosity and Resistivity Logs. AAPG Bulletin, 74(12): 1777-1794. DOI:10.1306/0c9b25c9-1710-11d7-8645000102c1865d
Prokoph A., Agterberg F. P., 2000. Wavelet Analysis of Well-Logging Data from Oil Source Rock, Egret Member, Offshore Eastern Canada. AAPG Bulletin, 84(10): 1617-1632. DOI:10.1306/8626bf15-173b-11d7-8645000102c1865d
Qian K., Wang S. M., Liu S. F., et al., 1980. Tertiary Reef Limestone Found in the Northern East China and Its Significance in Petroleum Geology. Science Bulletin, 25(12): 1025-1029.
Rioul O., Vetterli M., 1991. Wavelets and Signal Processing. IEEE Signal Processing Magazine, 8(4): 14-38. DOI:10.1109/79.91217
Schutter, S. R., 1998. Characterization of Shale Deposition in Relation to Stratigraphic Sequence Systems Tracts. In: Schieber, J., Zimmerle, W., Sethi, P., eds., Shales and Mudstones I. E. Schweizerhbart'sche Verlagsbuchhandlung, Stuttgart, Germany
Singh, P., 2008. Lithofacies and Sequences Stratigraphic Framework of the Barnett Shale: [Dissertation]. University of Oklahoma, Texas
Slatt R. M., Abousleiman Y., 2011. Merging Sequence Stratigraphy and Geomechanics for Unconventional Gas Shales. The Leading Edge, 30(3): 274-282. DOI:10.1190/1.3567258
Slatt R. M., Philp P. R., Abousleiman Y., et al., 2012. Pore-to-Regional-Scale Integrated Characterization Workflow for Unconventional Gas Shales. In: Breyer J. A., ed., Shale Reservoirs—Giant Resources for the 21st Century. AAPG Memoir, 97: 127-150.
Slatt R. M., Rodriguez N. D., 2012. Comparative Sequence Stratigraphy and Organic Geochemistry of Gas Shales: Commonality or Coincidence?. Journal of Natural Gas Science and Engineering, 8(8): 68-84. DOI:10.1016/j.jngse.2012.01.008
Smith M. G., Bustin R. M., 2000. Late Devonian and Early Mississippian Bakken and Exshaw Black Shale Source Rocks, Western Canada Sedimentary Basin: A Sequence Stratigraphic Interpretation. AAPG Bulletin, 84(7): 940-960. DOI:10.1306/a9673b76-1738-11d7-8645000102c1865d
van Wagoner, J. C., Mitchum, R. M., Campion, K. M., et al., 1990. Siliciclastic Sequence Stratigraphy in Well Logs, Core, and Outcrops: Concepts for High Resolution Correlation of Time and Facies. AAPG Methods in Exploration Series, No. 7. The American Association of Petroleum Geologists, Tulsa. 43-62
Wang B., Lin X., Bao Y., et al., 1996. SY/T 6210-1996 Quantitative Analysis of Total Contents of Clay Minerals and Common Non-Clay Minerals in Sedimentary Rocks by X-Ray Diffraction. Petroleum Industry Press, Beijing..
Wang J. F., 2005. Sedimentary Facies of the Shahejie Formation of Paleogene in Dongying Sag, Jiyang Depression. Journal of Paleogeography, 7(1): 45-58.
Wilkin R. T., Barnes H. L., Brantley S. L., 1996. The Size Distribution of Framboidal Pyrite in Modern Sediments: An Indicator of Redox Conditions. Geochimica et Cosmochimica Acta, 60(20): 3897-3912. DOI:10.1016/0016-7037(96)00209-8
Wu J., Jiang Z. X., Pan Y. W., et al., 2016. Lacustrine Fine-Grained Depositional Model—A Case Study of the Upper Submember of the Fourth Member of Paleogene Shahejie Formation in Dongying Sag. Acta Petrolei Sinica, 37(9): 1080-1089.
Wu J., Jiang Z. X., Qian K., et al., 2014. Characteristics of Salinization Mechanism on the Upper Part of Fourth Member of Shahejie Formation in the Dongying Sag, Shandong Province. Acta Geoscientica Sinica, 35(6): 733-740.
Wu L., Zhang Z., Li B., et al., 2001. GB/T 18602-2001 Rock Pyrolysis Analysis. Standards Press of China, Beijing..
Xu J. L., Pan Z. R., Yang Y. M., et al., 1997. Dinophyceae and Acritarchs on Early Tertiary in Shengli Oil Area. University of Petroleum Press, Dongying..
Yan J. H., Chen S. Y., Jiang Z. X., 2005. Sedimentary Characteristics of Nearshore Subaqueous Fans in Steep Slope of Dongying Depression. Journal of the University of Petroleum, China, 29(1): 6-9.
Yan, J. P., Sima, L. Q., Li, Y., 2011. Method of Wavelet Multiscale Analysis and Its Application in Division of Sedimentation Cycle of Lake Facies Mudstone. International Conference on Computational and Information Sciences, June, 2011, Shanghai (in Chinese) http://ieeexplore.ieee.org/document/6086348/
Yang Y. Q., Qiu L. W., Jiang Z. X., et al., 2011. A Depositional Pattern of Beach Bar in Continental Rift Lake Basins: A Case Study on the Upper Part of the Fourth Member of the Shahejie Formation in the Dongying Sag. Acta Petrolei Sinica, 32(3): 417-423.
Yuan W. F., Chen S. Y., Zeng C. M., 2006. Study on Marine Transgression of Paleogene Shahejie Formation in Jiyang Depression. Acta Petrolei Sinica, 27(4): 40-49.
Zhang M. M., Zhou J. J., Qin R. D., 1985. Tertiary Fish Fauna from Coastal Region of Bohai Sea. Memoirs of Institute of Vertebrate Paleontology and Paleoanthropology. Academia Sinica, 17: 55-60.
Zhang S. Q., Ren Y. G., 2003. The Study of Base Level Changes of the Songliao Basin in Mesozoic. Journal of Chang'an University (Earth Science Edition), 25(2): 1-7.
Zhao, Y., 2005. The Tectonic, Sedimentary and Oil Formation Kinetics Research of the North Slope in Dongying Depression: [Dissertation]. Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou (in Chinese with English Abstract)
Zhu H. R., 1979. Algae Fossil from Shahejie Formation of Early Tertiary in Binxian of Shandong. Acta Paleontologica Sinica, 18(4): 324-341.