Preliminary Estimation of Paleoproductivity via TOC and Habitat Types: Which Method Is More Reliable?-A Case Study on the Ordovician-Silurian Transitional Black Shales of the Upper Yangtze Platform, South China

INTRODUCTION

As a critical component of the global carbon cycle and of practical importance for oil and gas exploration and the genesis potential of source rocks, primary productivity (PP) has been defined as "the amount of inorganic carbon converted into organic carbon by photosynthesis ('fixed') per unit of ocean area per unit time" (Ryther, 1969). Hence, because of the limitation of finding the primary producers that have been degraded or decomposed during the depositional process and over geological time, only with some reliable and operable proxies can the estimation of the primary paleoproductivity be attempted.

Thus far, many biogenic components in sediments have been used as proxies for surface-water primary paleoproductivity estimation in seas and oceans, such as, total organic carbon (TOC) (Menzel et al., 2002; Meyers, 1997; Bralower and Thierstein, 1984; Müller and Suess, 1979), biogenetic carbonate (Siesser, 1995; Brummer and van Eijden, 1992), biogenic opal (Lange and Berger, 1993), as well as planktonic and benthic foraminifers (Herguera and Berger, 1991). Meanwhile, many biogeochemical parameters related causally to the organic carbon cycle, including some special trace elements (e.g., Ba, Mo, Fe, P, etc.) and stable isotopes (e.g., C, N, S, O), have been introduced into this kind of research (Lyons et al., 2003; Dymond et al., 1992). Most recent is the ongoing SINOPEC project, "Geobiological Processes in the Formation of Marine Source Rocks", where some authors have preliminarily quantified the primary paleoproductivity of the Late Permian basin in South China based on radiolarian fossils (Gu et al., 2007) and trace element Cu (Zhang et al., 2007). Furthermore, some authors have probed the potential to estimate primary paleoproductivity by using iron abundance (Hu et al., 2007), molybdenum isotopes (Zhou et al., 2007), as well as carbon isotopes (Huang et al., 2007), of the related stratigraphical intervals in South China. Thus far, no one has attempted to quantify the primary paleoproductivity of the Ordovician-Silurian transitional interval in South China.

In South China, the Ordovician-Silurian transitional interval exposed in the Wufeng Formation and the lower part of the Longmaxi Formation is a graptolite-bearing, organic-rich, carbonaceous-siliceous, black shale- or cherty slate-dominated succession, which is widely distributed across the Yangtze platform and adjacent areas (Su et al., 2007; Zhan and Jin, 2007; Chen et al., 2004, 1995; Wang et al., 1996). This area, which includes the Hirnantian GSSP, has been studied for decades as one of the most important stratigraphic intervals in South China, especially to investigate the terminal-Ordovician glacial eustasy, the Hirnantian mass-extinction, as well as the aftermath, and significant progress has been made in many respects (Chen et al., 2006, 2005, 2004, 2000; Rong, 2006, 1984; Rong et al., 2002; Su, 2001; Wang et al., 1993; Johnson et al., 1989). Recently, this interval has been regarded as one of the three main high-quality source rocks within the so-called 'lower assemblage strata' of South China, which also include basal Sinian (Ediacaran) and Lower Cambrian rocks, as it is believed that it partially sourced the historic and famous Puguang Gas Field and is related to other significant discoveries in the Yangtze region (Su et al., 2007; Tenger et al., 2007, 2006). Accordingly, estimates of the paleoproductivity of typical black-shale intervals considered herein will probably be helpful for better understanding the high-quality source rock formation and further evaluation of the marine strata in the Yangtze region, to find new potential source rock horizons in South China.

MATERIAL AND METHODS

Here the authors try to quantify the primary paleoproductivity of the two Ordovician-Silurian transitional black shales in the Upper Yangtze platform, initially following the previously presented method of using TOC contents as a proxy (Menzel et al., 2002; Meyers, 1997; Bralower and Thierstein, 1984; Müller and Suess, 1979). Several essential factors support the use of TOC as a proxy for primary productivity: (1) TOC contains organic carbon and kerogen (Ⅰ, Ⅱ, Ⅲ, Ⅳ) present in the sediments. When the observed sediments are marine strata, TOC can be regarded as a mixture of type-Ⅰ and Ⅱ kerogens that result directly from the marine primary productivity within the photic zone; (2) Both the graptolite-bearing black-shale intervals have been deposited under typical anoxic sea water, in which organic carbon can be better preserved than under oxic conditions; (3) Detailed stratigraphic data, including the biozones and the absolute age-based time-scale, are available elsewhere (Bergström et al., 2008; Chen et al., 2006, 2000; Gradstein et al., 2004). Consequently, the sedimentation rate (or accumulation rate) of the related stratigraphic interval, supported further by high-resolution integrated stratigraphy (Su et al., 2006, 2003), can be calculated more accurately; (4) Increasingly reliable TOC data have been reported from the Yangtze region of South China in recent years (e.g., Tenger et al., 2007, 2006; Liu et al., 2006; Zhou and Liang, 2002).

In this study, the authors analyzed new samples from the two black shale intervals both at the Nanbazi Section in northern Guizhou (Fig. 1, location 2) and the Wangjiawan Section (Hirnantian GSSP; Chen et al., 2006) in Yichang of West Hubei (Fig. 1, location 8). Six samples from the black shales at the base of the Wufeng Formation and four samples from the base of the Longmaxi Formation were collected at the Wangjiawan Section in Hubei Province and the Nanbazi Section in Guizhou Province, respectively (Table 1), whereas, both sections could be correlated quite well based on the high-resolution integrated stratigraphy (Su et al., 2006, 2003). All the TOC samples were collected from the fresh rock after surface weathering had been removed. When all the samples had been crushed to 200 Φ, the TOC of all the samples was analyzed using the electric-analysis method at the National Center for Geological Exploration Technology, Institute of Geophysical and Geochemical Exploration, CAGS, and the analytical error was less than 0.1%.

Figure  1.  Sketch map showing the tectonic units and facies zones in South China and the surroundings (Late Ordovician) (modified after Su et al., 2007; Chen et al., 2004; Wang, 1985). 1. Dongkala, Guizhou; 2. Nanbazi, Guizhou; 3. Xikou, Chongqing; 4. Sanbaoling, Hubei; 5. Wangjiagou, Sichuan; 6. Luojiang, Chongqing; 7. Bajiaokou, Shaanxi; 8. Wangjiawan, Hubei. At the Nanbazi Section (location 2) and the Wangjiawan Section (location 8), the TOC samples were collected for the present study. Moreover, all the sections connected with a broken-line will be discussed a little later in the text, for habitat type analyses (see Fig. 2). AH. Anhui; CQ. Chongqing; FJ. Fujian; GD. Guangdong; GS. Gansu; GX. Guangxi; GZ. Guizhou; HB. Hubei; HN. Hunan; HEN. Henan; JS. Jiangsu; JX. Jiangxi; QH. Qinghai; SC. Sichuan; SD. Shandong; SAX. Shanxi; SX. Shaanxi; TW. Taiwan; YN. Yunnan; XZ. Xizang (Tibet); ZJ. Zhejiang.

DownLoad: Full-Size Img PowerPoint

Figure  2.  Habitat type (HT) changes around the Ordovician-Silurian transition in the Upper Yangtze platform, South China. Chronostratigraphic subdivision and graptolite zones are from Zhan and Jin (2007), Chen et al.(2006, 2005, 2000) and Gradstein et al. (2004). 1. Limestone with shrinkage crack; 2. oolite limestone; 3. nodular argillaceous limestone; 4. graptolite-bearing black shale; 5. marl; 6. mudstone; 7. siltstone and sandstone; 8. turbidite sandstone/siltstone. All the sections have been observed recently by the authors in this study.

DownLoad: Full-Size Img PowerPoint

Table  1.  TOC contents of the investigated samples from black shales

 | Show Table

DownLoad: CSV

For comparison, the authors will also estimate the primary productivity of these black-shale intervals based on the documentation of the related habitat-type (HT) analyses (Yin et al., 1995), with the old actualism. 'The present is the key to the past' has been presented as the new 'forward method' in this SINOPEC project (Xie et al., 2007). Comparison between the two proxies (TOC vs. HT) will be useful to evaluate and test conventional methods of estimating paleoproductivity in ancient strata.

RESULTS

Primary Productivity of the Ordovician-Silurian Transitional Black-Shales: Estimation via TOC Analysis of the residual TOC of the black shale intervals

The analyzed results of the residual TOC of the two group samples are shown in Table 1.

The third group was published recently by Tenger et al. (2007), which is from northern Chongqing and Sichuan near the northern margin of the Yangtze platform. For convenience, their data are shown in Tables 2 and 3.

Table  2.  Primary productivity: Black shale member of Wufeng Fm., Upper Yangtze platform

 | Show Table

DownLoad: CSV

Table  3.  Primary productivity: Black shale member of Longmaxi Fm., Upper Yangtze platform

 | Show Table

DownLoad: CSV

Restitution of the original TOC of the two black-shale intervals

Because of degradation during the hydrocarbon producing process, all the TOC recovered from the sediments in the laboratory will be residual organic carbon, particularly for those highly-mature or post-mature source rocks with a vitrinite-reflectivity of R_o > 1.3 (Qin et al., 2007; Hunt, 1996). Fortunately, the original TOC content can be reconstructed using a special parameter, the restitution coefficient, according to the maturity of the rocks and the type of kerogen present. As suggested by Qin et al. (2007), the restitution coefficient of highly- to post-mature high-quality source rocks with type-Ⅰ kerogen is 1.68, with type-Ⅱ₁ kerogen is 1.48, and so on.

According to published references (Tenger et al., 2007, 2006; Liu et al., 2006; Zhou and Liang, 2002), all the Ordovician and Silurian successions of the Upper Yangtze platform, including the central-northern Guizhou, West Hubei, and North Sichuan-Chongqing of the present study, are highly-mature or post-mature rocks, in which most of R_o values are greater than 2.0, reaching as high as 2.54. Similarly, both the graptolite-bearing black-shale intervals of the Ordovician-Silurian transition in South China contain exclusively type-Ⅰ kerogen (Tenger et al., 2007; Liu et al., 2006; Zhou and Liang, 2002). Hence, the restitution coefficient for the original TOC of the present black-shale samples should be the highest, 1.68 (Qin et al., 2007). With this restitution coefficient, the original TOC values can be estimated and are shown in Tables 2 and 3.

Preservation factor of the original TOC of the two black-shale intervals

As pointed out by many authors (Menzel et al., 2002; Meyers, 1997; Bralower and Thierstein, 1984; Müller and Suess, 1979), the organic carbon typically preserved in sediments represents only a small fraction of the total organic carbon produced by primary producers in the surface-water (euphotic zone). Bralower and Thierstein (1984) have further presented the 'preservation factor' of sedimentary organic carbon as a linear equation between the 'primary production rate' and the 'organic-carbon accumulation rate'

In addition, they found that this factor is influenced by the degree of lamination, which indicates the varying extent of oxygenation and bioturbation and has been practiced broadly (Watanabe et al., 2007). The burrowed intervals (Class 1) are representative of the well-oxygenated end-member, and the finely laminated intervals are an indicator of anoxic environments (Class 4), where the preservation factor can range from roughly 0.001% (Class 1) to more than 10% (Class 4) based on global Holocene data. In general, the preservation factor of an anoxic environment with finely laminated sediments will be greater than 2% (Bralower and Thierstein, 1984).

Regarding the two Ordovician-Silurian graptolite-bearing black-shale intervals in South China, they were typical well-laminated anoxic successions without any significant bioturbation, generated by benthic fauna. In the Holocene data set studied by Bralower and Thierstein (1984), the organic carbon preservation factors in the environments producing laminated sediments were all higher than 2%, and all of the environments with preservation factors below 2% had burrowed sediments. Therefore, as a conservative estimate that should more than make up for any diagenetic loss of organic carbon (e.g., up to 30%, Berner, 1982), the authors have chosen a preservation factor of 2% for their calculations based on these finely-laminated sediments.

Time-calibration versus organic-carbon accumulation rate of the two intervals

According to the above-mentioned formula, when the preservation factor and the organic-carbon accumulation rate are 'known', the primary production rate can be determined. However, when any accumulation rate is calculated, the related chronological duration or 'time-span' needs to be considered carefully, as this parameter is often the largest source of uncertainty in the calculation of such rates.

With the updated chronological subdivision around the Ordovician-Silurian transition (Bergström et al., 2008; Gradstein et al., 2004), and the high-precision biostratigraphic framework around the Hirnantian Stage (Chen et al., 2006, 2005, 2000), the time-span of the two black-shale intervals of the Wufeng and Longmaxi formations in South China can be estimated at about 3-4 Ma and 5-6 Ma, respectively (Fig. 2).

Generally, the base of the graptolite-bearing member of the Wufeng Formation formed apparently during the middle to late parts of the Dicellograptus complanatus Zone (Wang et al., 1996; Chen et al., 1995), which is a bit above the base of the former Ashgillian Stage, and now correlates to the middle part of the Katian Stage (Fig. 3; Zhan and Jin, 2007), and is approximately 449 Ma in age (Gradstein et al., 2004). At the top of this interval, the base of the present Hirnantian Stage, correlates with the FAD of Normalograptus extraordinarius (Fig. 3; Chen et al., 2006), which is calibrated with an age of (445.6±1.5) Ma (Gradstein et al., 2004). Thus the duration of this black shale interval is less than (4±1.5) Ma.

Figure  3.  Diagram of habitat type and relationship between habitat type and level bottom benthic assemblage (HT-BA after Yin et al., 1995; GA-BA after Chen, 1990).

DownLoad: Full-Size Img PowerPoint

The base of the graptolite-bearing black-shale interval of the lower Longmaxi Formation in South China correlates with the base of the N. persculptus Zone, which is equivalent to the top part of the current Hirnantian Stage. Limited further by the sensu stricto short-lived Hirnantian Glaciation in Gondwanaland, with a duration of less than 1 Ma, perhaps as short as 0.5 Ma (e.g., Brenchley et al., 1994), the base of the N. persculptus Zone can be assumed to be about 444.5-445 Ma in age (Fig. 3; Chen et al., 2005; Gradstein et al., 2004). In addition, the top of this black-shale interval correlates with the base or the lower part of the Demirastrites triangularis Zone, or the top of the Coronograptus cyphus Zone, which is near the base of the Aeronian Stage, and approximately 439-438 Ma in age (Fig. 3; Gradstein et al., 2004). Hence, the maximum time-span of the black-shale interval of the lower Longmaxi Formation in South China can be constrained to 5.5-7 Ma.

In the study published earlier, some of the authors (Su et al., 2003) presented their estimates for the duration of this Ordovician-Silurian transitional interval. With a high-resolution integrated stratigraphic framework based on precise biostratigraphy (Su et al., 2003; Chen et al., 2000; Wang et al., 1987) and sequence stratigraphy (Su et al., 2003), the time-span of the graptolite-bearing member of the Wufeng Formation in the Yangtze platform is estimated to be as short as 2 Ma, due to the fact that this member has been interpreted as the TST (transgressive systems tract) and HST (highstand systems tract) of the terminal-Ordovician third-order type-Ⅰ sequence (Su, 2007, 2001, 1999) that is about 3 Ma in its entirety. Similarly, the Guanyinqiao Member, as the upper part of the Wufeng Formation, is interpreted as the SMST (shelf-margin systems tract) of the base-Silurian third-order type-Ⅱ sequence (Su, 2007, 2001, 1999), which has been treated as coeval to the Hirnantian glacial lowstand stage (Su, 2007, 2001, 1999; Zhan and Jin, 2007; Chen et al., 2004; Su et al., 2003; Rong, 1984), and the overlying black-shales of the Longmaxi Formation have been interpreted as the TST of the base-Silurian third-order sequence (Su, 2007, 2001, 1999). Therefore, the maximum time-span of the later black-shale interval, that is, the TST of the base-Silurian third-order sequence, is probably less than 3 Ma (Su et al., 2003; Su, 2001).

Combining the two different estimates of the duration, the averaged time duration can be further refined to 2-4 Ma and 3-5 Ma, for the black shale intervals of the Wufeng and Longmaxi formations, respectively. With broad reasonable values for duration, the bulk deposition/accumulation rate (S) of the two intervals can be calculated, and the organic-carbon accumulation rate can be estimated via the formula suggested by many earlier workers (e.g., Bralower and Thierstein, 1984; Müller and Suess, 1979)

where TOC_ori (%) is the original TOC in the sediments before hydrocarbon generation, which can be inferred via the restitution coefficient mentioned earlier; S (cm/ka) is the bulk accumulation rate that is equivalent to the thickness of black-shale accumulated during a given amount of time; ρ(g/cm³) is the dry density of the sediments/rocks of the related strata interval, for black shale, it is about 2.6 g/cm³; Φ(%) is porosity of the dry sediments/rocks of the related strata interval, for black shale, it is 2%-5% in general; 10 is calculation to convert gC/(cm²·ka) to gC/(m²·a). Note that all percentages (TOC_ori and Φ) should be entered as a decimal into equations (2) and (3).

Primary production rate of the two black-shale intervals

With the above-mentioned formula, the primary production rate can be quantified as follows

Using an organic carbon preservation factor of 2%, as suggested by Bralower and Thierstein (1984), for finely laminated sediments consistent with the two Ordovician-Silurian transition black shale intervals, the estimates for primary paleoproductivity of the two intervals are 32 (43-21) gC/(m²·a) and 21 (27-15) gC/(m²·a), for the Wufeng and Longmaxi formations, respectively. They are significantly lower than the average modern global average values (~140 gC/(m²·a) estimated by Bralower and Thierstein (1984)) or those of the Black Sea (~210 gC/(m²·a) estimated by Karl and Knauer, 1991), or even the central South China Sea (< 100 gC/(m²·a) estimated by Tan and Shi (2006)). However, they are actually closer to the values of the current Mediterranean Sea (~26 gC/(m²·a) estimated by Menzel et al. (2002)) or the Pacific-like open ocean (25-50 gC/(m²·a) cited by Meyers (2006)). In addition, it is noteworthy that the values are also similar to those calculated for the mid-Cretaceous, using the same method (~14 gC/(m²·a) estimated by Bralower and Thierstein (1984)), where oceanic anoxic events are recorded by laminated sediments that form under apparently low productivity oceans.

A series of questions arise: Is this paleoproductivity estimate reliable? Why are the values of the primary paleoproductivity of the two Ordovician-Silurian black shales intervals so low? Are there more similarities with the mid-Cretaceous ocean anoxic event?

At present, it is impossible to answer all of these questions in the current investigation. The authors can take steps towards addressing the first question of reliability, by finding an alternative means of estimating paleoproductivity. If the authors compare their TOC-based values with those estimated via different methods they can begin to assess the reliability of such estimates. Hence they will estimate the paleoproductivity of these two intervals using the 'forward method' recently presented by Xie et al. (2007). Using present paleoecological relationships as a key to understanding ancient environments, habitat-type (HT) analysis (see Fig. 3; Yin et al., 1995), based mainly on the depth of marine environment, has been developed on the basis of conventional benthic assemblages (BA), and can provide a rough comparison of environmental conditions between modern and ancient strata.

Primary Productivity of the Ordovician-Silurian Transitional Black Shale: Estimation via Habitat Type (HT)

According to Xie et al. (2007), the present 'forward method' for the evaluation and estimation of the source rock potential and related processes of hydrocarbon generation utilizes the distribution of habitat types (HT) as well as their evolution in space and time. Through comparison with the modern primary productivity data of various habitat types from different depositional environments around the world, the paleoproductivity of a given succession can be reasonably inferred, based on the analysis of habitat types (HT) in the ancient strata. By analyzing the habitat types of the Ordovician-Silurian transition black-shale intervals the authors can determine an alternative value for paleoproductivity that can be compared to the TOC proxy.

Habitat type of the two black-shale intervals and the related succession

Conventionally, it has been thought that most of the planktonic graptolite faunas lived in offshore, deep, stagnant water environments. However, increasing evidence shows that graptolite faunas can live in various environments, in different water layers, and even in epicontinental seas. Accordingly, many authors have indicated that the graptolite depth-zonations (GA, Zhan and Jin, 2007; Chen, 1990) can be identified with certain groups and correlated fairly well with the related BA. Except for the absence of graptolites in BA1, the rest of the depth-zonations of graptolite faunas correspond to the benthic assemblages (GA2-BA2, GA3-BA3, GA4-5-BA4-5). This is useful to connect the two kinds of ecological terminologies with each other. For example, the GA2 is suggested to correlate with BA2, which means that the GA2-bearing succession must be in HT Ⅱ, and so on (Fig. 3).

Following the research of Chen (1990), the present two Ordovician-Silurian transitional graptolite-bearing black-shale intervals in the Yangtze platform (Fig. 2) should be placed in the GA3-GA5 zones because of the graptolite biotopes contained therein. In other words, both the black-shale intervals in the Yangtze platform should belong to the inner-shelf to outer-shelf environments (i.e., HT Ⅲ-Ⅳ) (Figs. 2 and 3). Limited by anoxic water, however, there was almost no benthic fauna in the main part of the intervals anywhere, throughout the region.

The habitat type (HT) of the surrounding strata can also be determined, to help constrain the habitat type (HT) of the two black-shale intervals. First, the two black-shale intervals at the Ordovician-Silurian transition in the Yangtze platform are separated by the shelly-facies limestones or argillaceous beds, which are named the 'Guanyinqiao Bed/Member' bearing the famous BA2-3 Hirnantia fauna (Rong et al., 2002; Rong, 1984). Further, in North Guizhou, which is near the Kang-Dian-Qian-Gui oldland (Fig. 1), the black-shale interval of the Wufeng Formation lies unconformably under the Hirnantian ooide beds (HT Ⅰ: see Dongkala Section in Fig. 2, location 1 in Fig. 1). Moreover, at some places, the two intervals are punctuated by a significant erosional surface or stratigraphic gap (see Dongkala, Taiyanghe, Wangjiagou sections in Fig. 2, locations 1, 4, 5 in Fig. 1). These features, combined with the fact that these black shales occur on the Yangtze platform itself, clearly indicate that, unlike the mid-Cretaceous ocean anoxic event beds found in the deep-ocean cores, the two black-shale intervals considered herein did not develop in abyssal depositional settings, and thus they could be eroded easily by a moderate magnitude sea-level fall or tectonic uplift.

The overlying and underlying successions of the two black-shale intervals indicate the settings of HT Ⅱ-HT Ⅲ (Fig. 3), which are represented by light colored, shelly facies limestones, marls, siltstones, and sandstones. Most of them bear the BA2-BA3 shelly faunas, even containing occasional BA1 faunas, which appear with typical peritidal-zone depositional structures such as herringbone cross bedding, ripple-marks, and mud-cracks. An exception appeared in the Linxiang Formation, which bears the BA4-5 Foliomena fauna in some places (Zhan and Jin, 2007; Rong et al., 1999 and references therein). Similar analyses previously reported by Johnson et al. (1989) satisfactorily support the present study, whereas, they developed their observation along a west-east orientation corridor section from the Upper Yangtze platform (Guizhou) to the Lower Yangtze (Zhejiang). All these characteristics further demonstrate that the two graptolite-bearing black-shale intervals have been deposited in a long-standing continental shelf environment, whereas, episodic anoxia has appeared, which is caused mainly by two eustatic transgressions around the Ordovician-Silurian transition (Su et al., 2007). Coeval strata are deposited on the upper part of the continental slope, only at the northern margin of the Yangtze platform (Fig. 1), represented by the Bajiaokou Section in southeastern Shaanxi Province (location 7 in Fig. 1; also see Fig. 2), which is indicated by the flysch-dominated succession at the top and well-burrowed siltstone succession at the base (Fu and Song, 1986). The authors infer therefore that this section must correlate with the HT V₁ setting.

Primary productivity estimation via the habitat type of the black-shale intervals

Because the two graptolite-bearing black-shale intervals in the Yangtze platform have been correlated mainly with HT Ⅲ and HT Ⅳ, which represent the vast area from the inner shelf to the outer shelf with the absolute water-depth ranging from several tens of meters to roughly 200-300 m, the authors may infer the primary productivity in this setting using relevant data from the current continental shelf. Considering the available paleomagnetic data and paleocontinent reconstructions around the Ordovician-Silurian transition, the Yangtze platform is suggested to have been located in the low to mid-latitudes of the southern hemisphere. The modern South China Sea and southern East China Sea represent analogous depositional settings, demonstrating that current primary productivity data from the South China Sea and East China Sea, especially those of the shelf area with similar depth and habitat-types (see Fig. 4, the short, white color dashed lines show the related candidate areas, where the maximum depth is less than 300 m), can provide an estimate of paleoproductivity based on the habitat type (HT). Consequently, based on the recently published data of Tan and Shi (2006), the primary productivity of the two Ordovician-Silurian transitional black-shale intervals ranges from ~100 gC/(m²·a) to ~400 gC/(m²·a).

Figure  4.  Average annual integrated primary productivity of the China surrounding seas from 2003 to 2005 (modified after Tan and Shi, 2006). Dashed lines show the related candidate areas, where the maximum depth is less than 300 m. Unit: gC/(m²·a).

DownLoad: Full-Size Img PowerPoint

DISCUSSION AND CONCLUSIONS

The estimated primary productivity values of 32 (44-21) gC/(m²·a) and 21 (27-16) gC/(m²·a) of the two black-shale intervals via the TOC contents are generally lower than the estimates made via the HT method (100-400 gC/(m²·a)). In addition, the average values via TOC are only roughly similar to those of the present central South China Sea (65-100 gC/ (m²·a), see Fig. 4; Tan and Shi, 2006). Why? Moreover, which method is more reliable, TOC or HT? If the present estimates via TOC can be accepted, what is the reason for the discrepancy with the HT-based estimates? Although the present case study is quite a preliminary attempt at a complicated topic, and all of the questions mentioned earlier cannot be answered thoroughly at once, something enlightening for a further detailed study can be discussed in short here.

The bulk accumulation rate or the linear depositional rate of these intervals could be skewed by a possible misestimation of the time spans based on the current geological time-scale.

As the authors have mentioned earlier, the time-span of the two black shale intervals is estimated as 2-4 Ma, 3-5 Ma, based on quite a precise biostratigraphic subdivision (Chen et al., 2006), the high-resolution sequence stratigraphy (Su et al., 2003), as well as the newly updated geological time-scale (Gradstein et al., 2004), and chronostratigraphic correlation (Bergström et al., 2008). However, the estimated depositional rates of the two intervals at each section are all less than 1 cm/ka (0.128-0.533 cm/ka; see Tables 2 and 3), suggesting the possibility of a depositional setting more similar to that of the current pelagic ocean areas, instead of a continental shelf, as the habitat-type analysis reasonably inferred it to be HT Ⅲ to HT Ⅳ.

As many authors have indicated earlier, the preserved strata are only a very small part of the real and entire depositional process of geological time (Andel, 1981). Controlled mainly by two third-order eustatic transgression events during the Ordovician-Silurian transition, the two black-shale intervals in the Upper Yangtze platform are believed to represent the two most starved stages and anoxic episodes (Su et al., 2007). Therefore it is possible that there may be multiple, short-duration diastems or hiatuses. In other words, perhaps the time-spans of the intervals are acceptable based on the current age-dating, but the record of these two intervals is less than 100% complete, making the time-spans of the intervals measured by the authors, at the outcrop, shorter than anticipated. The length of time recorded by these two black shales represents a major uncertainty at present, and with more study, also represents one of the areas where the greatest improvements in resolution need to be made.

Theoretically and practically, most of the parameters utilized and measured in this investigation have their own systematic errors.

Just as it cannot be expected that all the TOC data from the observed strata are an absolutely accurate and reliable measure of the original organic carbon content, considering the array of postdepositional factors that can alter the TOC content in the sedimentary strata (Hayes et al., 1999). The organic-carbon preservation factor derived from the Holocene scenarios, even that of the Mid-Cretaceous oceanic anoxic deposits, may not be appropriate for any other place and time in the geological record. In fact, if the authors have changed the preservation factor from 2% to a smaller value, they would get higher paleoproductivity values, and vice versa. At present, this study represents a first order approximation highlighting the need to further refine many of these variables through future research.

It seems that the authors have to accept that in different geological periods, similar or even identical HT does not always mean the same depositional environmental conditions, especially the degree of oxygenation, which will greatly influence the organic carbon concentration in the sediments.

Except for some restricted seas and basins, such as the Black Sea, most of the current seas and oceans around the world are well-oxygenated at the sediment-water interface. Regarding the seas surrounding China, especially the northern part of the South China Sea as well as the East China Sea (Fig. 4), they have been open and oxic neritic shelf seas since the Pleistocene. These are quite different from the anoxic Ordovician-Silurian transitional "Upper Yangtze Sea" (Fig. 1) although they all belong to the neritic shelf seas according to the HT method (HT Ⅲ to HT Ⅳ).

Furthermore, the Ordovician-Silurian transition atmospheric p_CO2 was estimated to be as high as 10-16 PALs (present atmospheric level) (Kump et al., 1999; Berner, 1994), and the colonization of the terrestrial biosphere had not yet taken place (Algeo and Scheckler, 1998). Therefore, probably the authors could not directly and simply correlate the Ordovician-Silurian transitional Upper Yangtze platform with the current South China Sea or East China Sea. Similarly, Neogene analogs for organic-carbon-rich deposition, such as, the modern Black Sea or the Mediterranean Sea, during times of sapropel formation, were also different from the Ordovician-Silurian transitional "Upper Yangtze Sea" (Fig. 1), because they represented relic oceans generated by continent-continent collision. Apart from the anoxia, common to all of these cases, the paleogeography, tectonic setting, and eustatic sea levels were combined to make each situation unique. The common link between all of them however was anoxia. Therefore, future research should focus on determining a way to differentiate between the various causes of anoxia possible in Paleozoic oceans.

In summary, this study was an initial estimation of primary paleoproductivity, using two different methods. The conventional inverse method was used via the TOC content of the Ordovician-Silurian transitional black-shale intervals of the Yangtze platform, and the forward method, via the habitat type (HT) analyses, and correlation to those of the current seas surrounding China, by using the present relationships as the key to the past. Although the paleoproductivity estimates made via the TOC method were acceptable within the current biozone-based chronostratigraphic framework, this method was not consistent with the more generalized forward method (HT). Because of different conditions within the depositional environment, particularly the oxygenation characteristics, the quantification of the primary productivity via the HT method should be operated cautiously.

A lot more study still needs to be done to explain the discrepancy between the two methods, as well as to further increase the chronostratigraphic resolution of the Ordovician-Silurian transition, which will consequently improve the reliability of the estimates of paleoproductivity.

ACKNOWLEDGMENTS

The authors are indebted to Profs. Yin Hongfu, Xie Xinong, Tenger, Gerald R Baum, Markes E Johnson, Warren D Huff, and Frank R Ettensohn, for fruitful discussion. Colleagues and students of China University of Geosciences, Gong Shuyun, Li Quanguo, Wang Wei, Ma Chao, Li Lu, Zhang Lei, are thanked greatly for their assistance both in the field and the laboratory. Prof. Xie Shucheng read the manuscript critically. Sincere thanks go to Mrs. Huang Yutao and Mr. Deng Mingliang for their selfless assistances. This is also a contribution to the IGCP 503 project.