Massive dolomitization driven by MgSO 4 -rich seawater and its effects on thermochemical sulfate reduction,Upper Permian Changxing Formation,northeastern Sichuan,China

Abstract

Massive dolostones replacing build-ups and containing H2S of thermochemical sulfate reduction (TSR) origin occur in the Upper Permian reef and bank deposits, northeastern Sichuan, China, despite the absence of gypsum or anhydrite deposits, which are usually present in TSR cases. Fluid chemistry from fluid incluions and δ13C, δ18O and 87Sr/86Sr values from different diagenetic phases were measured to determine the dolomitization regime, and to assess the relationships between the occurrence of dolomitization and H2S accumulation. Dolomitization was initiated by seawater with slightly increased salinities (penesaline) at shallow depths prior to chemical compaction. The micro- to fine-crystalline cloudy dolomite formed with relatively high Na and Sr contents, and with δ13C, δ18O and 87Sr/86Sr values (2.2‰∼4.8‰, −3.9‰∼-5.0‰ and 0.70724∼0.70746, respectively) inherited from seawater. The evaporation of Permian seawater in back-reef and inter-reef lagoons during sea-level fall and the subsequent seepage reflux into reef-beach bodies led to greater Mg²⁺ and SO₄²⁻ concentrations in higher-salinity pore waters. Further massive dolomitization was promoted by compactional flow of hotter residual seawater at shallow to intermediate depths and resulted in the formation of fine- to medium-crystalline clean dolomite with lower Na and Sr contents, more depleted δ¹⁸O values (-5.1‰∼-6.0 ‰), andδ¹³C (2.5‰∼4.8 ‰) and ⁸⁷Sr/⁸⁶Sr values (0.70726∼0.70741) similar to those of the coeval seawater. The relatively closed hydrodynamic system during burial facilitated SO₄²⁻ preservation. The whole dolomitization process enriched porewater SO₄²⁻, which have been almost exhausted by subsequent TSR, accounting for the high present-day concentrations of H₂S. The output of this study shows that similar scenarios involving dolomitization driven by condensed MgSO₄ seawater and H₂S accumulations can occur in evaporite-free settings across a broad range and deserves special attention during deep oil/gas exploitation.

Keywords

TSR dolomitization penesaline seawater C-O-Sr isotopes fluid inclusion Permian Sichuan Basin

Introduction

Massive dolostone formations have been found to correlate mostly with various fluids in different environments, e.g., normal or modified sea water and deep basinal fluids (e.g. Hardie, 1987; Kaufman, 1994a; Machel, 2004; Warren, 2000). Increasing documents have reported the effects of various models of dolomitization on the quality of dolomitic oil and gas reservoirs (Enayati-Bidgoli and Navidtalab, 2020; Hu et al., 2020; Jones and Xiao, 2005; Purser et al., 1994; Saller and Henderson, 1998; Vandeginste et al., 2009; Wilson et al., 2007). However, previous studies have ignored the fact that massive dolomitization events also result in significant changes in pore water chemistry, which would be succeeded by late diagenetic fluids. The deep extension of diverse dolomitizing fluids linked with different models may generate impressive diagenetic effects. For example, hydrothermal dolomitizing fluids are capable of transporting metal ions and heat, and, therefore, has the potential to provoke mineralization and possible acidification of fluids (Biehl et al., 2016; Davies and Smith, 2006; Shelton et al., 2019).

The presence of evaporitic tidal flats could provide hydrological and hypersaline fluids conducive to the deposition of massive dolostone (Adams and Rhodes, 1960; Jones et al., 2002; Kaufman, 1994a; Shields and Brady, 1995). High sulfate levels are commonly associated with elevated Mg/Ca ratios in hypersaline waters (Hardie, 1987). Warren (2000) emphasized that unlike nonevaporitic dolomitizing fluids, hypersaline fluids flushed from evaporite sources could form anhydrite cements, which might subsequently dissolve (Vandeginste et al., 2009). These sulfate-rich dolomitizing fluids would probably be succeeded by late diagenetic fluids during the following progressive burial stage. Li et al. (2014) claimed that the early refluxing of hypersaline dolomitizing fluids from evaporite-bearing strata had brought abundant reactive sulfates for TSR, leading to gas souring in the Upper Permian reservoirs in the Puguang gasfield (see the location in Figure 1B).

Figure 1.

Paleogeography of (A) the earth (after Weidlich, 2002) and (B) Sichuan Basin (after He and Luo, 2010), including distributional patterns of reefs during late permian-early Triassic. The black box represents the study area. YB = Yuanba, LG = Longgang, SBZ = Shibaozhai, JN = Jiannan, PG = Puguang.

More than 10¹⁰ m³ of gas accumulations have been discovered recently in the Upper Permian dolomitized reef complexes in the Yuanba and Longgang gas fields in the northeastern (NE) Sichuan Basin (Figure 1). Elevated concentrations of H₂S (mostly between 3.3% and 7.2%), of thermochemical sulfate reduction (TSR) origin (Li et al., 2014, 2016; Liu et al., 2014), is an unexpected non-hydrocarbon gas in the reservoirs. That's because rare evaporites, i.e., source of sulfate for TSR, have been found within the strata and the overlying oolitic limestone units in the Lower Triassic Feixianguan Formation (T₁f) (Guo, 2011; Tian et al., 2014). The model proposed by Li et al. (2014) cannot be applied directly to the Yuanba gasfield because the absence of evaporites casts doubt on the hypersaline brine flushing model of pervasive dolomitization. It is noteworthy that the sour gas is present exclusively in dolostones rather than limestones, as indicated in previous studies (e.g. Li et al., 2016). This information yields instructive clues that there is a close relationship between the occurrence of TSR and dolomitization.

Unfortunately, the origin of the Upper Permian dolostones in the study area remains puzzled. Several dolomitization models, e.g., freshwater-seawater mixing models (e.g. Tian et al., 2014) and burial models (e.g. Cheng et al., 2013) have been proposed. Jiang et al. (2014) detected moderate-salinity fluid inclusions in dolomite crystals and claimed that these dolomites had formed from mesosaline brines. New data are therefore required to present and re-assess the hydrologic system, origin and evolution of the dolomitizing fluids, which might provide useful new insights into the source of reactive sulfate for TSR and the reason of H₂S accumulation. Essentially, besides of dissolution of solid calcium sulfate, alternative sulfate sources, e.g., seawater or modified seawater, have been suggested by Machel (2001).

The main objectives of the current investigation are as follows:

To identify the origin and nature of the dolomitizing fluids in the Upper Permian bank and reef complex without extensive associated gypsum/anhydrite deposits;

To account for sulfate evolution during and after dolomitization and explain why aqueous sulfate accumulated exclusively in dolostone horizons;

To assess the intrinsic relation between massive dolomitization of Permian carbonates and TSR in the study area and other fields worldwide.

Geological setting

The Sichuan Basin, which is located in Southwest China, is a large intracratonic basin and hosts a 12 km-thick sedimentary succession deposited on top of the metamorphic basement of the Yangtze platform (Figure 1). Marine sedimentation dominated in the Sichuan Basin from the Upper Sinian to the Middle Triassic (Figure 2). During the late Permian to Early Triassic, the Yangtze area was an epeiric sea on the southeastern side of the Paleo-Tethys Ocean (Algeo et al., 2013). A low-latitude (Figure 1A) climate was prevalent and became increasingly arid into the Early Triassic. Continued petroleum exploration in the northeastern (NE) Sichuan Basin has resulted in the discovery of large gas fields, e.g., the Puguang gasfield on the eastern side of the Kaijiang-Liangping (K-L) Trough, and the Yuanba and Longgang gas fields on the western side (Figure 1B). High-production gas is commonly associated with deeply buried reef and shoal dolostone reservoirs in the Feixianguan (T₁f) and Changxing (P₃ch) Formations (Cai et al., 2004; Li et al., 2020). Geochemical signatures show that the hydrocarbons in the P₃ch and T₁f reservoirs were mainly sourced from the Upper Permian Longtan source rocks (e.g. Cai et al., 2017; Hao et al., 2008; Figure 2).

Figure 2.

Generalized stratigraphic column of the northern sichuan basin.

The barrier reef on the western side, developed in P₃ch, spreads narrowly along and parallel to the platform margin (Figure 1B), consistent with the distribution style of paleo-structural highs. Some patch reefs and bioclastic bank bodies are also scattered in the paleohighs of the backreef or inter-reef belts (Figure 3). The barrier and patch reefs are composed of gray limestones and dolostones, including bafflestone and framestone. The platform-margin bank facies in P₃ch consists of thick-bedded to massive dolarenite, bioclastic dolostone and limestone (Figure 4). The platform-margin bank facies in the Lower Triassic is made up of oolitic limestone and calcarenite (Guo, 2011).

Figure 3.

Map and seismic sections showing the sedimentary facies distribution of the late permian changxing formation (P₃ch) in the Yuanba and longgang gasfields (modified from Huang et al., 2014 and Tian et al., 2014), and locations of gas wells.

Figure 4.

Correlation panel of wells YB12-YB123-YB9-YB3 showing lithology and sequence stratigraphy of the P₃ch formation, and comparison of sea level variation during late Permian between Sichuan Basin and globe (simplified from Yin and Tong, 1996).

The Late Permian sequence stratigraphy of the reef trend in the NE Sichuan Basin has been extensively studied (e.g. Guo, 2011; Duan and Peng, 2014). Two third-order sequences have been recognized from the Changxing Formation and each sequence is composed of a transgressive systems tract (TST) and a highstand systems tract (HST; Figure 4). The TST of the first sequence is composed of interbank deposits within an open platform, and its HST comprises a set of reef-bank deposits belonging to platform-margin facies. The TST of the second sequence is characterized by a set of reef-bank deposits, and its HST is mainly a set of deposits belonging to exposed shallow-bank facies (Guo, 2011). The sea level gradually rose from the Longtanian towards Changxingian, causing reef growth, and regression did not occur until the end of the Permian; this sequence is different from the global sea level, which fell to the lowest point and later rose during the Changxingian (Figure 4). The difference might have been caused by the rapid northward migration of the Cimmerides and the approach of the Eurasian continent, which gradually closed the Paleo-Tethys from east to west (Yin and Tong, 1996).

The burial and geothermal history has been elucidated in previous studies (Cai et al., 2014). Rapid sedimentation took place during the Early Triassic and the Middle and Late Jurassic, and the strata experienced maximum burial (more than 10,000 m) and heating to about 220°C during the Early-Middle Cretaceous. Significant uplifts occurred during the Cenozoic because of the Hercynian orogeny, leading to a decrease in temperatures in the P₃ch-T₁f reservoirs to 120∼130°C at present.

The diagenetic history of the Upper Permian Changxing Formation limestones of the NE Sichuan Basin has been summarized in our previous work (Li et al., 2012). Submarine diagenesis led to precipitation of acicular or fibrous circumgranular calcite cement-1, which is followed by bladed and granular calcite cement-2 of meteoric origin. Subsequently, calcite cement-3 precipitated and occurred as mosaic to poikilitic crystals in the center of pores. This calcite cement shows dull red to non-luminescent under CL and posses a high range of homogenization temperatures (HTs) from about 90 to 130°C, with a distinct population at around 100°C. Post-bitumen void-filling calcite is present in low amounts as coarse mosaic to poikilotopic crystals, growing outward from bitumen-coated pore walls.

Materials and analytical methods

Late Permian cores of variable length (over 2,500 m in total, with individual cores ranging from 150 to 300 m) were examined from 14 wells, among which 11 were from the Yuanba Field and 3 were from the Longgang Field. The selected wells reach different sedimentary facies, including reef facies (barrier and patch reefs), bank facies, inter-reef or inter-bank sea facies and forereef slope facies (Figure 3). Approximately 120 samples of Upper Permian carbonates were taken from these 14 wells and examined by polished thin section petrography. Samples were collected at high resolution (sampling interval of 3∼5 m) from partially to completely dolomitized intervals. Of these, 30 bulk carbonates and 6 void-filling calcite samples were studied using cathodoluminescence (cold-CL) petrography, electron microprobe (EMP) analyzes and isotope determination. CL analyzes were performed using a Technosyn cold cathodoluminescence stage with a 15 kV beam and a current intensity of 500 μA. EMP analyzes were carried out to determine the major and trace element contents using a Cameca Camebax BX 50™ instrument equipped with three spectrometers. The precision of analyzes was better than 0.5 mol%.

Bulk dolostone samples were drilled using a dentist drill and analyzed for C, O and Sr isotopes. Approximately 50 mg of dolomite powder was reacted with anhydrous H₃PO₄ acid at 25°C for 24 h and for 72 h. The extracted CO₂-gases were analyzed with a Finnigan MAT-252 mass spectrometer. The δ¹³C and δ¹⁸O values are reported as ‰ relative to the Vienna Peedee Belemnite (VPDB) standard. The precision was ± 0.1‰. For ⁸⁷Sr/⁸⁶Sr measurement, carbonate sample powders (50–100 mg) were dissolved in 2.5 N HCl. Then, the sample solution was loaded onto the pre-conditioned resin column with 2 mL of AG50W × 12 (200–400 mesh) for the separation of Sr from the sample matrix. After rinsing four times with 0.5 mL of 2.5 M HCl, the column was washed with 7 mL of 5 M HCl. Afterwards, the Sr fraction was stripped with 3.5 mL of 5 M HCl. The ⁸⁷Sr/⁸⁶Sr ratios were measured on a Finnigan MAT-261 mass spectrometer, and the precision for ⁸⁷Sr/⁸⁶Sr measurement was ± 0.00003–0.00007. Four samples of solid bitumen samples underwent δ³⁴S analysis, with results reported relative to the Canyon Diablo troilite (CDT) standard.

Five post-bitumen calcites showing coarse vug/fracture-filling spar from sour reservoirs were selected for homogenization temperatures and major ion composition analysis of fluid inclusions. For crush-leach analyzes, 1 g pure giant coarse milky white calcite crystals were washed several times, and then crushed in 5 ml of ultraclean water (Vandeginste et al., 2009). After thorough crushing, the water solution was decanted/filtered to yield a clean leachate and then weighed. The solution was then analyzed by a Dionex DX500 ion chromatograph. To test the analytical procedures, triplicate and blank tests were performed for each sample. Absolute ion concentrations were calculated, taking the weight of fluid inclusion water into consideration. SO₄²⁻/Cl⁻ are calculated using the values of SO₄²⁻ and Cl⁻ concentrations. The formation water samples from the Yuanba and Longgang gas fields were analyzed by using ion chromatography. The details of this measurement procedure followed the Oil and Gas Industry Standard of China.

Results

Dolomite distribution and petrography

Pervasive dolomitization of carbonate build-ups in the Changxing Formation occurred mainly in platform–margin reef and bank deposits, and the extent of dolomitization decreases basinward (Figures 3, 4). Reefal build-ups, reef caps and flank deposits show extensive dolomitization whereas scarce to minor dolomitization occurs in the reef cores (Lei et al., 1994; Tian et al., 2014). Bioclastic limestones in the back-reef and intraplatform flats are partially to completely dolomitized. Vertically, the distribution and abundance of the most massive dolostones are stratigraphically controlled. The regressive facies of third-order depositional sequences tend to be dominated by cycles of extensive dolomitization, whereas abundant limestones are not dolomitized in transgressive facies (Figure 4).

Five types of dolomites are distinguished based on crystal size, distribution, and geometry (Sibley and Gregg, 1987; Tucker and Wright, 1990): (1) very finely to finely crystalline non-plana replacement dolomite (RD1), (2) finely to medium crystalline cloudy non-planar replacement dolomite (RD2), (3) medium to coarsely crystalline planar-e clean replacement dolomite (RD3), (4) finely to medium crystalline planar-s dolomite cement (CD1), and (5) medium to coarsely crystalline planar-e dolomite cement (CD2).

RD1 is irregularly distributed, forming a patchy pattern in reef-bank facies where it partially to completely replaces the algal mud matrix and selectively replaces grains in packstones/grainstones. Strata-bound dolostones are also found locally in the reef caps, back-reef or inter-reef belts and restricted to supratidal facies. There is a progressive increase in abundance of RD1 with increasing ration of tidal-flat to subtidal facies in cycles. RD1 is volumetrically minor and usually accounts for less than 10% of the total rock volume. RD1 consists of 10–30 μm crystals, with nonplanar textures and a uniform orange to dull reddish appearance in CL images (Figure 5A, B). The crystals are generally rich in inclusions and retain the fabric. Various types of euryhaline organisms, such as foraminifers, tubiphytes and crinoids, can usually be identified within RD1 (Figure 5C).

RD2 occurs in almost all facies and is volumetrically the most important (accounting for ∼45% by volume of all dolomite phases). It commonly exhibits a decrease in abundance downward from the supratidal facies and preferentially formed in the top reef and shoal build-ups and in the flanks rather than the lower and inner parts. RD2 crystals range from 30 to 200 μm in size and occur as nonplanar textures with a cloudy appearance (Figure 5D, E). RD2 partially preserves fabrics and commonly shows moderately bright orange–red CL colors when it is very finely crystalline, while it forms non-mimetic patches with dull red CL colors when it is coarser (Figure 5D, F).

RD3 occurs as single rhombs or scattered patches of euhedral porphyrotopic dolomite crystals in slightly dolomitized reef and bioclastic limestone facies (Figure 6A, B) or as pervasive replacive dolostone in thin to massive (several m thick) dolostone intervals (Figure 6C, D). RD3 preferentially appears in close association with stylolites (Figure 6 A, B), accounting for 30∼40% of the total volume. RD3 is characterized by fine to medium crystal sizes (average sizes, 80–400 μm) and commonly shows inclusion-rich cores surrounded by more euhedral limpid dolomite rims. It occurs as completely fabric-destroyed and planar-s to planar-e mosaics, showing purple or dull red CL colors (Figure 6C, D). The massive dolostones usually exhibit abundant vuggy and intercrystalline porosities (Figure 6C).

CD1 appears as bladed and granular crystals surrounding isopachous circumgranular calcite cement-1 in partially dolomitized layers in the upper part of the reef-beach bodies (Figure 7A, B). CD1 resembles calcite cement-2 in the limestone with similar crystal sizes and shapes (Figure 7A), indicating the CD1 crystals are more likely replacive rather than cements filling in void space. CD1 crystals range from 50 to 200 μm in size and occur as planar-e and planar-s textures, which resemble the precursor granular calcite cement-2 (Figure 7A, B). CD1 shows moderately bright orange–red luminescence (Figure 7C), which is similar to that of RD2 dolomite (Figure 5D). The formation of CD1 is followed by a mosaic to poikilitic calcite cement-3 with dull brown to nonluminescent staining under CL (Figure 7A, B, C).

CD2 partially to completely fills mouldic, vuggy and fracture porosity in almost all facies, and is characterized by increasing crystal sizes towards the centers of cavities (Figure 7D, E). CD2 occurs as clear and fine to medium-sized crystals and shows dull red colors similar to those of RD3 (Figure 7F).

Figure 5.

Paired polarized and cathodoluminescent photomicrographs showing RD1 and RD2 in the P₃ch formation. (A) RD1 are generally microcrystalline and inclusion-rich. Local solution vugs are filled with fine-crystalline CD2 crystals, Well LG2, 5,635 m. (B) The same view as (A), but taken under cathodoluminescence (CL). RD1 show a uniform orange–dull red CL colors while CD2 dolomites show purple or dull-red CL colors. (C) Various types of euryhaline organisms, e.g., foraminiter, colaniella, tubiphytes and crinoid, can be identified, Well Yb2, 6,592 m; (D) RD2 shows a relatively wide spread in crystal sizes and a cloudy appearance, with euryhaline organisms partially preserved. (E) RD2 shows a relatively wide spread in crystal sizes and a cloudy appearance, Well Yb204, 6,548.6 m. (F) RD2 shows moderately bright orange–red CL colors.

Figure 6.

(A) RD3 occurs as single rhombs or scattered patches of euhedral porphyrotopic dolomite, closely associated with stylolites, well Yb11, 6,912.2. (B) RD3 is dull red in color under CL. (C) RD3 occurs as pervasive dolomites, Well LG2, 6,129 m. (D) Under CL the RD3 crystals are purple or dull red in color.

Figure 7.

Photomicrographs of two types of dolomite cements (CD1 and CD2) in the P₃ch formation. (A) Stained section showing CD1 dolomite as circumgranular bladed and granular crystals and interparticle stage II calcite cements (red spars), Well Yb2. (B, C) Paired polarized and cathodoluminescent photomicrographs of CD1 dolomite, showing moderately bright orange–red luminescence, Well Yb204, 6,550.6 m. (D) CD2 dolomite occurs as clear fabric-obliterative vug-filling crystals, Well Yb123. (E, F) Paired polarized and cathodoluminescent photomicrographs of CD2 dolomite, showing purple or dull red CL colors, Well LG2, 6,118 m.

Geochemistry

Carbon, oxygen and strontium isotopes

C, O and Sr isotopes of dolomites from the Upper Permian Changxing Formation are listed in Table 1 and plotted in Figure 8. Oxygen isotope values show a gradual decrease from RD1 dolomite to RD2 and RD3 dolomites. RD1 dolomite samples show the least depleted δ¹⁸O values of −3.9 and −4.5‰ (n = 5, mean −4.1‰) relative to other generations of dolomite. However, these values are lower than those of hypothetical Upper Permian marine dolomites of −2.2 to −4.3‰ (e.g. Korte et al., 2004; Tian and Zeng, 1995). RD2 dolomite samples have δ¹⁸O values ranging from −4.1 to −5.0‰ with an average of −4.6‰ (n = 6). RD3 dolomite has the lowest or most depleted δ¹⁸O values (-5.1 to −6.0‰, n = 14) with a mean of −5.5‰. RD1 and RD2 dolomite samples have similar narrow δ¹³C values from 2.2 to 3.5‰ (n = 11). RD3 dolomite shows a slightly wider spread at 2.5∼4.8‰ (n = 8, Figure 8A). Despite minor differences, the δ¹³C values of the massive dolomite are in the range of those of Late Permian seawater (Huang, 1994; Korte et al., 2004).

Figure 8.

Carbon, oxygen and strontium isotopic composition of RD1, RD2 and RD3 dolomites and TSR-calcites. (A) Cross plots showing the relationships of δ¹³C values to δ¹⁸O values for RD1, RD2 and RD3 dolomites; (B) ⁸⁷Sr/⁸⁶Sr ratios of massive dolomite and post-bitumen calcite in comparison with Late Permian and Lower Triassic seawater. Data of Late Permian seawater are from Burke et al. (1982) and Korte et al. (2004) and data of Lower Triassic seawater are from Korte et al. (2003).

Table 1.

Sr, C and O isotopic compositions of various types of dolomite and post-bitumen calcite from the P₃ch reservoirs.

Sample Number	Depth (m)	Sample Occurrence	Crystal size (μm)	δ13C (‰)	δ18O (‰)	87Sr/86Sr
LG2	5,635	RD1	10–30	2.2	−3.9	0.707384
Yb2-1	6,552.4	RD1	10–40	3.1	-3.9	0.707365
Yb101-1	6,903	RD1	20–30	2.9	-4.3	0.707405
Yb101-2	6,913.2	RD1	20–60	2.5	-4.1	0.707342
Yb102-1	6,720	RD1	20–60	3.5	-4.5	0.707373
LG82-1	4,224.43	RD2	50–120	3.1	-4.3	0.707469
LG82-2	4,235	RD2	80–150	3.2	-5.0	0.707395
Yb11-1	6,910.3	RD2	80–170	2.7	-4.5	0.707271
Yb11-2	6,916.7	RD2	60–150	2.6	-4.7	0.707240
Yb204-2	6,548.6	RD2	60–150	2.9	-4.8	0.707280
Yb224	6,653	RD2	50–100	3.0	-4.1	0.707370
LG001-1	6,233.4	RD3	100–300	2.5	-5.7	0.707385
Yb9-1	7,067.51	RD3	100–150	4.3	-5.8	0.707085
Yb22-1	6,647.46	RD3	120–200	3.6	-5.9	0.707261
Yb123-1	--^b	RD3	120–300	3.9	-5.8	0.707096
Yb204-3	6,552	RD3	100–200	4.8	-5.6	0.707377
Yb 27-1 ^a	6,291.6	RD3	120–300	--^b	-5.2	0.707273
Yb 27-2 ^a	6,291.9	RD3	80–250	--	-5.1	0.707352
Yb 27-3 ^a	6,294.8	RD3	100–350	--	-5.8	0.707373
Yb 27-4 ^a	6,294.8	RD3	100–350	--	-5.7	0.707300
Yb 27-5 ^a	6,297.2	RD3	80–400	--	-6.0	0.707269
Yb 27-6 ^a	6,299.2	RD3	120–300	--	-5.1	0.707407
Yb 27-7 ^a	6,299.4	RD3	120–350	--	-5.4	0.707354
Yb 27-8 ^a	6,300.7	RD3	120–280	--	-5.1	0.707362
Yb 27-9 ^a	6,301.3	RD3	150–280	--	-5.2	0.707300
Yb11-3	6,917	Post-bitumen calcite	--	-9.5	-6.0	--
Yb102-2	6,724	Post-bitumen calcite	--	-3.7	-13.6	0.707290
YB204-4	6,422	Post-bitumen calcite	--	2.6	-7.6	0.707423
Yb204-5	6,551	Post-bitumen calcite	--	-1.0	-6.7	0.707284
Yb9-2	6,909	Post-bitumen calcite	--	3.1	-6.1	0.707375
LG82	4,224	Post-bitumen calcite	--	-0.2	-7.1	0.707469

Data from Tian et al. (2014).

No measurement and data.

RD1 represents microcrystalline matrix dolomite; RD2 represents very fine- to medium-crystalline cloudy matrix dolomite; RD3 represents fine- to medium-crystalline floating matrix dolomite; RD4 represents fine- to medium-crystalline clear matrix dolomite.

The replacive dolomite and the post-bitumen calcite samples have Sr isotope ratios ranging from 0.70709 to 0.70747 (n = 25, Table 1, Figure 8B), most of which are within the range of Late Permian seawater (0.7069∼0.7074, Burke et al., 1982; Korte et al., 2004). These values are much lower than those of Lower Triassic seawater (0.7074–0.7082, Korte et al., 2003).

Major and trace element contents

All dolomites, regardless of replacement or cement type, are near-stoichiometric to stoichiometric (49.6 to 52.3 mol% CaCO₃) (n = 32, Table 2). Typically, stoichiometry increases with increasing crystal size (R² = 0.65, Figure 9A). The mole % CaCO₃ generally decreases from RD1 dolomites (51.7∼52.3 mol%, n = 4) to RD2 (51.3∼51.8 mol%, n = 5) and RD3 (49.6∼51.2 mol%, n = 17).

Figure 9.

Cross plots showing the relationships of dolomite crystal size and CaCO₃ content (A), Sr and CaCO₃ content of dolomite (B).

Table 2.

Elemental geochemistry and CL colors of different types of dolomite.

Sample Number	Depth (m)	Sample Occurrence	Na (ppm)	Sr (ppm)	Fe (ppm)	Mn (ppm)	CaCO3 (%)	MgCO3 (%)	CL colors
LG2	5,635	RD1	403	513	148	93	52.1	47.9	Orange-dull red
Yb2-1	6,552.4	RD1	516	446	105	132	51.7	48.3
Yb101-1	6,903	RD1	682	410	120	108	51.8	48.2
Yb101-2	6,913.2	RD1	678	371	134	--	52.3	47.7
LG82-1	4,224.43	RD2	640	146	97	116	51.3	48.7	Moderately bright orange-red
LG82-2	4,235	RD2	463	487	148	120	51.3	48.7
Yb11-1	6,910.3	RD2	614	219	161	86	51.8	48.2
Yb11-2	6,916.7	RD2	357	208	--	108	51.7	48.3
Yb204-2	6,548.6	RD2	592	391	100	152	51.5	48.5
Yb9-1	7,067.51	RD3	245	153	104	219	51.5	48.5	Purple or dull red core with a thin slightly bright red rim
Yb9-3	--	RD3	315	110	238	145	50.8	49.2
Yb27-10	6,309.9	RD3	348	161	210	87	51.1	48.9
Yb22-1	6,647.46	RD3	332	--	206	147	50.6	49.4
LG001-1	6,233.4	RD3	171	167	287	129	51.4	48.6
Yb9-4	7,067.51	RD3	219	100	--	--	51.1	48.9
Yb123-1	--^a	RD3	390	--	298	85	50.2	49.8
Yb204-3	6,552	RD3	671	90	187	92	51.2	48.8
Yb27-1^b	6,291.6	RD3	--	129	300	101	50.7	49.3
Yb27-2^b	6,291.9	RD3	--	107	174	111	50.7	49.3
Yb27-3^b	6,294.8	RD3	--	124	372	102	50.7	49.3
Yb27-4^b	6,294.8	RD3	--	137	877	140	50.6	49.4
Yb27-5^b	6,297.2	RD3	--	396	176	108	50.6	49.4
Yb27-6^b	6,299.2	RD3	--	98	218	78	49.6	50.4
Yb27-7^b	6,299.4	RD3	--	163	288	72	51.1	48.9
Yb27-8^b	6,300.7	RD3	--	149	240	73	50.7	49.3
Yb27-9^b	6,301.3	RD3	--	117	826	95	50.2	49.8
Yb9-5	6,937	CD1	401	217	109	86	51.0	49.0	Moderately bright orange-red
Yb204-6	6,550.6	CD1	310	--	136	127	50.9	49.1	Moderately bright orange-red
LG82-3	4,235	CD2	376	60	521	76	50.9	49.1	Purple or dull red
Yb102-2	--	CD2	470	145	460	220	50.8	49.2
Yb204-7	--	CD2	213	87	311	298	51.2	48.8
Yb27-11	6,304.7	CD2	289	43	236	81	50.4	49.6

No measurement and data;.

Data from Tian et al. (2014).

CD1 represents fine-crystalline circumgranular dolomite cement; CD2 represents fine- to medium-crystalline void-filling dolomite cement.

All dolomite types have low Sr contents (90∼513 ppm, Table 2). There is a weak correlation between the Sr contents and CaCO₃ contents of the matrix dolomites and the dolomite cements, with R² of 0.37 (Figure 9B). Strontium concentrations of RD1 and RD2 dolomites (371∼513 ppm and 146∼487 ppm, respectively) are higher than those of dolomite precipitating from normal seawater (∼125 ppm) (Machel and Anderson, 1989; Malone et al., 1996; Swart et al., 2005). RD3 and CD2 dolomites have relatively lower contents (90∼396 ppm and 43∼145 ppm, respectively). Iron concentrations of RD1 and RD2 dolomites range from 97 to 161 ppm (n = 8), and those of RD3 and CD2 are slightly enriched at 174∼877 ppm (n = 20). The replacive dolomites contain similar manganese concentrations from 72 to 219 ppm (n = 26).

The Na contents of all the dolomites show a wide range from 171 ppm to 682 ppm (n = 17). RD1 and RD2 dolomites have slightly higher Na contents (403∼682 ppm and 357∼640 ppm) than RD3 (mostly 171∼390 ppm, Table 2).

Fluid inclusion and formation water chemistry

The cation and anion compositions of fluid inclusions in post-bitumen calcite crystals and formation water samples from the Yuanba and Longgang gas fields are presented in Table 3. A progressive decrease in SO₄²⁻ concentrations is accompanied by an increase in homogenization temperatures (Th) of the fluid inclusions (Figure 10A). The fluid inclusion waters have SO₄²⁻/Cl⁻ ratios (5.4∼10.4, expressed in weight units) higher than and Na⁺/Cl⁻ ratios similar to those of Late Permian seawater (0.11 and 0.54, respectively, Figure 10B). The P₃ch formation waters within dolostone layers show significantly lower SO₄²⁻/Cl⁻ ratios from 0.001 to 0.019 and lower SO₄²⁻ concentrations from 0.2 g/L to 0.6 g/L. In contrast, the formation waters within limestone layers show SO₄²⁻/Cl⁻ ratios from 0.04 to 0.12 and SO₄²⁻ concentrations from 1.3 g/L to 1.9 g/L, which are similar to those of the contemporary seawater (Horita et al., 2002; Lowenstein et al., 2005). The formation waters within the limestone and dolostone layers show a wide range of Na⁺/Cl⁻ ratios from 0.01 to 0.66.

Figure 10.

Variation of major ions concentrations of the paleo-fluids during the whole diagenesis of the Permian carbonates. (A) SO₄²⁻ concentrations of fluid inclusion waters vs. Th. (B) SO₄²⁻/Cl⁻ ratios vs Na⁺/Cl⁻ in Permian seawater, fluid inclusion waters and formation waters.

Table 3.

Chemical compositions of fluid inclusion waters from post-bitumen calcite crystals and formation water from the P₃ch Formation. The concentrations of H₂S are for the associated gas reservoirs.

Sample Number	Depth (m)	Age	Sample occurrence	F⁻	Cl⁻	SO₄²⁻	Na⁺	K⁺	Mg²⁺	Ca²⁺	Th (°C)	H₂S (%)
Yb204-8	6,422	P₃ch	Post-bitumen calcite	0.3	17.3	97.3	7.8	4.4	1.7	13.2	142.0	−
Yb204-9	6,551	P₃ch	Post-bitumen calcite	−^c	21.5	117.8	11.3	0.8	1.6	10.6	161.5	2.4
Yb9-6	6,909	P₃ch	Post-bitumen calcite	0.1	9.1	86.2	5.1	0.4	0.9	12.8	187.8	−
Yb27-10	6,303	P₃ch	Post-bitumen calcite	−	20.6	215.2	10.4	0.9	0.9	9.7	137	4.1
LG82	4,224	P₃ch	Post-bitumen calcite	0.1	19.7	168.9	11.7	1.6	0.7	18.1	127.1	−
Yb4	6,683- 6,693	T₁f	Limestone	−	11.1	1.3	6.8	0.4	0.08	0.5	−	<1
Yb5	−	T₁f	Limestone	−	38.1	1.4	11.5	0.5	0.8	10	−	<1
Yb9-7 ^a	7,000– 7,020	P₃ch	Limestone	−	36.2	1.4	23.9	0.1	1.2	−	−
LG7^b	6,632.71	P₃ch	Limestone	−	32.6	1.9	19.2	1.9	0.08	1.3	−	−
LG81	3,680–3700	P₃ch	Limestone	−	40.6	1.6	18.3	1.0	0.5	6.7	−	−
Yb1	7081	P₃ch	Dolomite	−	116.3	0.6	2.2	0.6	2.8	54.9	−	7.2
Yb2	6,545	P₃ch	Dolomite	−	149	0.2	8.2	0.3	18.6	40.3	−	4.5
Yb2	6,677	P₃ch	Dolomite	−	106	0.6	12.9	0.2	1.7	42.9	−	4.4
Yb9	6,836– 6,857	P₃ch	Dolomite	−	37.6	−	11.7	0.4	2.1	6.3	−	12.11
Yb16 ^a	6,950– 6,974	P₃ch	Dolomite	−	31.3	0.6	19.6	0.2	0.6	12.16	−
Yb123	6,978– 6,986	P₃ch	Dolomite	−	28.2	−	18.1	0.6	0.05	0.6	−	25.7
LG6^b	5,111– 5,198	P₃ch	Dolomite	−	73.7	0.6	6.3	1.3	5.3	25.3	−	3.3
LG21	5,686– 5,697.5	P₃ch	Dolomite	−	60.4	0.2	16.1	0.4	0.3	13.1	−	−

Data from Li et al. (2016); ^b Data from Qin et al. (2016); ^c Beyond the detection limit or no data available.

Data of H₂S concentrations are for the gas reservoirs and from Liu et al. (2014); Li et al. (2016) and Qin et al. (2016).

The ion concentrations of the fluid inclusion waters are expressed in μg/g and the values of the formation waters are expressed in g/L.

Δ³⁴s values of solid bitumens

Solid bitumen in reservoir rock samples associated with gas containing H₂S > 4.1% shows δ³⁴S values ranging from 17.7‰ to 24.5‰ (n = 4, Table 4), which are close to the range of Late Permian seawater that has δ³⁴S_CAS values of 11.5–26.9‰ (Newton et al., 2004).

Table 4.

S isotopic compositions of solid bitumen from the P₃ch reservoirs. The concentrations of H₂S are for the associated gas reservoirs.

Sample Number	Depth (m)	Age	Sample occurrence	δ³⁴S (‰)	H₂S (%)
Yb22-2	6,639	P₃ch	Solid bitumen	20.4	−^a
Yb27-11	6,295	P₃ch	Solid bitumen	24.5	4.1
Yb27-12	6,297	P₃ch	Solid bitumen	20.9	4.1
Yb205	−	P₃ch	Solid bitumen	17.72	5.3

Discussion

Time and condition of dolomitization

RD1 and RD2 in the extensive dolomitized zones and CD1 in the slightly dolomitized limestone lithofacies may have formed during the same period of dolomitization (Figure 11), as indicated by their identical CL colors, crystal sizes and shapes (Figure 5; Figure 7 C). Dolomite with a near-micritic to finely crystalline texture and relative fabric retention (Figure 6A, C, D) is typically formed under near-surface, low-temperature and saline conditions (Gregg and Sibley, 1984). This interpretation is further supported by the fact that the formation of CD1 dolomite in the slightly dolomitized limestone lithofacies was followed by precipitation of mosaic to poikilitic calcite cement-3 with dull brown to nonluminescent appearance under CL (Figure 7A, B). In our previous study (Li et al., 2012), this calcite cement was demonstrated to have precipitated from hot (90–130 °C) and reduced pore water in shallow to intermediate-depth burial environments. That is, early dolomitization occurred at shallow depths prior to chemical compaction.

Figure 11.

Synthetic paragenetic sequence showing main stages in the diagenetic evolution of P₃ch carbonates and variation of SO₄²⁻ concentrations of paleo-waters through time. Celestite was recognized by Li et al. (2012).

Most of the RD1 and RD2 samples have carbon and oxygen isotope values and strontium isotope ratios within those estimated for Late Permian seawater (Figure 8A, B; Korte et al., 2004), indicating that the dolomite was precipitated from solutions with an isotopic composition comparable to that of contemporary seawater or slightly modified seawater. With the δ¹⁸O_SMOW (Vienna standard mean ocean water, SMOW) value of Late Permian seawater at −1‰ (Melim and Scholle, 2002), the precipitation temperatures of RD1 and RD2 dolomites are calculated to be 47.1–49.3°C and 49.3–53.2°C, respectively (Figure 12), using their δ¹⁸O values (Table 1) and the dolomite–water fractionation equation of Sheppard and Schwarz (1970). The corresponding dolomitization depths of RD1 and RD2 are estimated to be 479–546 m and 512–667 m, assuming a palaeotemperature of 30°C–33°C for Late Permian seawater (Joachimski et al., 2012) and a geothermal gradient of 3.17°C/100 m (Wang and Xiao, 2010). These confirm that replacement took place at shallow burial, which is consistent with the textural features of dolomites.

Figure 12.

Relationship between δ¹⁸O of dolomite, δ¹⁸O of water, and temperature, according to 10³ lnα = 3.2 × 10⁶/T²-3.3 (Sheppard and Schwarz, 1970). See text for explanations.

The Na contents of RD1 and RD2 dolomites (357∼682 ppm, Table 2) are significantly higher than those of dolomites precipitated from normal seawater (∼185 ppm) (Qing and Mountjoy, 1988), suggesting a saline dolomitizing fluid (e.g. Land and Hoops, 1973; Sass and Bein, 1988). This result is consistent with the high Sr contents of RD1 and RD2 dolomites (Figure 9B, Table 2) because the precipitation rates of dolomite and the Sr distribution coefficient (K_D^Sr) are usually high in evaporative brines (Land, 1980; Lorens, 1981; Montanez and Read, 1992). Generally, the high contents of Na and Sr in these two types of dolomites, the small crystal sizes (Table 1), and the planar-s and nonplanar fabrics (Figure 5A, C, D), are typical of brine reflux dolomites (e.g. Machel, 2004; Melim and Scholle, 2002; Rameil, 2008). However, the absence of gypsum/anhydrite (Figures 2, 4) suggests that evaporated seawater did not reach the salinities required for abundant gypsum precipitation. Therefore, the RD1 and RD2 dolomites are considered to possibly have precipitated in restricted lagoonal seawater with slightly elevated salinity, that is, under penesaline conditions. This situation has been invoked to explain extensive dolomitization in the absence of gypsum/anhydrite deposits in both experimental and field studies (e.g. Eren et al., 2007; Kaufman, 1994b; Qing et al., 2001; Rameil, 2008; Sun, 1994). Such penesaline conditions are in good accordance with the common observation of euryhaline organisms, consisting, for example, of mainly foraminifers, tubiphytes and crinoids in RD1 (Figure 5C).

Considering the paleoenvironment of the study area, seawater with elevated salinity could have developed in back-reef and inter-reef lagoons (Figure 3; Figure 13) during sea-level fall, as suggested in previous studies from other fields (e.g. Hurley and Lohmann, 1989; Mutti and Simo, 1994). Downward flux likely occurred when brines from overlying restricted lagoon facies became dense enough to pass directly through the gently sloping platform margin, resulting in early massive dolomitization (Figure 13). The frequent presence of RD1 and RD2 in the back-reef and inter-reef belts and at the top reef and shoal build-ups indicates the close relationship between dolomitization and sea-level change and therefore supports the validity of the model. The Upper Knox dolomites in the Appalachians (Montanez and Read, 1992), the middle Turonian dolomites in central Tunisia (Touir et al., 2009) and the Capitan Formation forereef dolomites (Melim and Scholle, 2002) are also good examples documenting such a relationship.

Figure 13.

Conceptual model showing the dolomitization and TSR process and the concomitant chemical evolution of the paleo-fluids. (A) During high sea level in Late Permian the back reef lagoon was in relatively free water exchange with the open marine. The mineral deposits were mainly aragonites and the primary pores were filled with MgSO₄ connate water. (B) During low sea level, limit water exchange between the lagoon and the open marine and moderate evaporation caused seawater condensation without evaporite precipitation. Reflux of the penesaline seawater initiated dolomitization of aragonite and/or calcite along the platform rim, which resulted in that dolomite layers held condensed seawater, whereas the limestone layers kept MgSO₄ connate water (C) During the subsequent burial, dewatering of carbonates would liberate seawater and led to further dolomitization and SO₄²⁻ accumulation. (D) The elevated concentrations of SO₄²⁻ in the dolomite layers were succeeded by the deep-buried pore waters and reacted with the substantial oil/gas to produce H₂S, whereas the SO₄²⁻ in the limestone layers kept unchanged or slightly decreased due to minor hydrocarbons accumulated.

Further dolomitization probably occurred at shallow to intermediate depths during the subsequent onset of chemical compaction (Figure 11), as indicated by the close association of RD3 dolomites with bedding-parallel stylolites (Figure 6 A), which commonly form at burials between 500 and 800 m in limestones, some earlier, and at higher burials in dolostones due to their greater resistance to dissolution (Ebner et al., 2009; Mountjoy and Amthor, 1994; Nicolaides and Wallace, 1997). This inference is also supported by the calculated dolomitization temperatures of 53.2–58.9°C (Figure 12), using the δ¹⁸O values of RD3 dolomites, the dolomite–water fractionation equation of Sheppard and Schwarz (1970), and the corresponding depths of 668–848 m. The increasing dolomitization temperatures from RD1 and RD2 to RD3 are consistent with the covariations in stoichiometry, Sr content and crystal size of all the dolomites (Figure 9A, B). CD2 overgrows RD1, RD2 and RD3 dolomites facing intercrystalline pores, and the remaining dolomites are filled with bitumen and late diagenetic blocky calcite cement (Figures 5A, D and 6C, D). These features reveal that the precipitation of CD2 postdates the massive dolomite replacement and predates the occlusion of the intercrystalline porosity by bitumen and post-bitumen calcite (Figure 11).

The dolomitization at shallow to intermediate depths was driven by hotter and more reducing residual seawater derived from compaction. Dewatering of carbonates would liberate seawater, either pristine or evolved/concentrated, trapped within the pore system (Figure 13). This process might have provided Mg²⁺ for local dolomitization. The continuous stabilization of aragonite and high-Mg calcite components, as well as the clay mineral transformations during the formation of stylolites (Figure 6A), could have been an extra source of Mg²⁺ (Wanless, 1979). The flow of Mg²⁺-bearing fluids would account for progressive replacement of limestone (RD3) and precipitation of dolomite cements (CD2) within the reef and shoal build-ups (Figure 13).

Variation in SO₄²⁻ concentrations during and after dolomitization

Late Permian seawaters are classified as aragonite seas (MgSO₄ seas, Figure 14) and share chemical characteristics with modern seawater, including high SO₄²⁻ (2.21 g/L) and Mg²⁺ (1.25 g/L) concentrations (Hardie, 1996; Lowenstein et al., 2005). Seawater can achieve much higher values of SO₄²⁻ and Mg²⁺ concentrations at elevated salinities (Hardie, 1987). Similar salinities likely occurred in the condensed seawater from the restricted lagoons in the study area, where the brine reflux dolomites that formed possess elevated salinities (8∼23 wt%, Jiang et al., 2014) with respect to the original seawater (3.5% salinity). The SO₄²⁻ concentrations of the pore waters in the reef and shoal rocks are supposed to be correspondingly up to 2.3∼6.6 times (change in salinity) the original values after seepage refluxing of condensed seawater (Figure 13). This result is supported by the covariation between SO₄²⁻ concentration and salinity or degree of evaporation (Hardie, 1987; McCaffrey et al., 1987). The subsequent dolomitization for RD3 formation in intermediate-depth burial environments, as mentioned above, was favorable for the preservation of sulfate in the pore waters due to the relatively closed hydrodynamic system. Additionally, during the whole dolomitization process, abundant carbonate-associated sulfate (CAS) was very likely to be released and added to the pore waters due to the significant differences in CAS concentrations between dolostone and limestone (Li et al., 2014). The precipitation of pre-bitumen celestite (Figure 11; Li et al., 2012) provides compelling evidence to support the enrichment of sulfate in the pore waters during dolomitization.

Figure 14.

Secular variation in SO₄²⁻ concentrations and Mg²⁺/ Ca²⁺ of seawater and preliminary mineralogy of marine carbonates and evaporites during Phanerozoic time, estimated from fluid inclusions in marine halites (Horita et al., 2002). The SO₄²⁻ concentration and Mg²⁺/ Ca²⁺ of seawater reached the peak during Late Permian.

The geochemical signals of the pore waters seem to be succeeded by late diagenetic fluids, as indicated by ⁸⁷Sr/⁸⁶Sr ratios of post-bitumen calcites similar to those of dolomites (Figure 8B) and high SO₄²⁻/Cl⁻ concentrations (Figure 10B) of the fluid inclusion water in post-bitumen calcites. These calcites, showing coarse vug/fracture-filling spar, negative shifts in carbon isotope compositions (Table 1) and high precipitation temperatures (142 to 187.8°C), are genetically correlated with TSR (Cai et al., 2003; Li et al., 2012; Machel, 1987; Worden and Smalley, 1996; Worden et al., 1995). The TSR reaction might have occurred in situ, as indicated by the exclusive presence of sour gases in the P₃ch reservoirs rather than in the T₁f reservoirs, and by δ³⁴S values (17.7‰–24.5‰, Table 4) of the solid bitumen samples that are close to the range of P₃ch seawater with δ³⁴S_CAS values of 11.5–26.9‰ (Newton et al., 2004).

The negative correlation (Figure 10A) of SO₄²⁻ concentration with homogenization temperature (T_h) hints at a continuous consumption of sulfate by the TSR reaction. The gradual decrease in SO₄²⁻ concentration caused by TSR is responsible for the SO₄²⁻ deficit in the present-day formation water (Figure 10B, Figure 13). A comparison of the SO₄²⁻/Cl⁻ ratios in the fluid inclusion water (5.4∼10.4) and the formation water (0.001 to 0.019) indicates that almost all of the accumulated sulfate has been removed during TSR. The evolution of SO₄²⁻ concentrations in paleowaters during diagenesis is presented in Figure 11.

Relationship between dolomitization and TSR

One of the critical factors for the occurrence of TSR could be limited amounts of dissolved sulfate in closed systems, i.e., low SO₄²⁻ concentrations in pore waters and volumetrically minor residual pore spaces during deep burial (Bildstein et al., 2001; Li et al., 2019; Worden et al., 2000). Therefore, in hydrocarbon reservoirs, any improvement in these two parameters would enhance the TSR reaction and H₂S accumulation.

Model mass/volume balance calculations were conducted to obtain the concentration of the generated H₂S using the equation of Li et al. (2016). The calculated H₂S concentration is only 2.6% based on a reservoir with 10% initial water saturation (typical for gas-filled carbonate reservoirs as suggested by Worden et al., 1996) and ∼1.0 g/L SO₄²⁻ concentrations of initial pore waters (CaCl₂ sea, Figure 14). In contrast, the calculated value could range up to 5.7% if the initial water was Late Permian seawater with a SO₄²⁻ concentration of 2.2 g/L (Lowenstein et al., 2005), indicating that MgSO₄ seawater is a superior sulfate supplier. However, the value is significantly lower than the concentrations of H₂S within some P₃ch reservoirs, i.e., 12.1% in well Yb9 and 25.7% in well Yb123 (Table 3). The subsequent evaporation of MgSO₄ seawater and CAS released during dolomitization are more important to have multiplied (by 2.3∼6.6 times) the SO₄²⁻ concentrations of the pore waters. The calculated H₂S concentrations range from 13.1∼28.5%, which seems reasonable when considering the reported removal of H₂S in the P₃ch reservoirs, e.g., the formation of pyrite (e.g. Jiang et al., 2014) and sulfur deposits (e.g. Li et al., 2014) and the incorporation of sulfur into solid bitumen (e.g. Cai et al., 2010).

Notably, the formation waters within the Permian-Triassic limestone layers possess SO₄²⁻ concentrations (1.3 g/L to 1.9 g/L, Table 3) close to or slightly lower than those of the original seawater (1.9–2.2 g/L, Horita et al., 2002; Lowenstein et al., 2005). That is, most of the SO₄²⁻ was not involved in TSR throughout deep burial. This interpretation implies that there should have been another critical factor, i.e., transport of aqueous hydrocarbons (Worden et al., 2000), which inhibited the TSR reaction and H₂S production (H₂S% <1, Table 3). The lower porosity (2.3%) and permeability (1.2 × 10⁻³ μm²) of the P₃ch limestone reservoirs than those of the dolostone reservoirs (5.1%∼16.7% and 5.3 × 10⁻³∼306 × 10⁻³ μm², respectively, Guo, 2011) might have significantly reduced the chances of contact between hydrocarbons and dissolved sulfates. Zhu et al. (2007) also highlighted the importance of the workspace for the TSR reaction and found that almost all of the gas pools with high H₂S concentrations in China possess porosities over 3.5%. Therefore, the occurrence of massive dolomitization in the P₃ch Formation seems to be a necessary process to preserve porosities, providing more space for accommodation of both residual water and hydrocarbons. The preference for RD3 dolomites that have abundant vuggy and intercrystalline porosities further indicates that the late dolomitization is preferable to enhance the workspace for TSR.

Regional comparisons

TSR-H₂S enrichment was also reported in the evaporite-free dolostone reservoirs of P₃ch on the eastern side of the K-L Trough (e.g. Cai et al., 2004, 2013; Hao et al., 2015; Li et al., 2014; Liu et al., 2013). Similarly, most of the necessary sulfate was demonstrated to have been inherited from early Mg- and sulfate-rich dolomitizing fluids (Li et al., 2014). However, unlike the study area, the dolomitizing fluids on the eastern side of the K-L Trough were derived from the Early Triassic hypersaline evaporite lagoons, where laminated/nodular anhydrite and gypsum formed. The massive dolomite and the post-bitumen calcite yield radiogenic Sr isotopes (mainly from 0.70753 to 0.70785), significantly higher than those of the Upper Permian carbonates in situ (Jiang et al., 2013; Li et al., 2014) and those of RD1, RD2, RD3 and the post-bitumen calcite in this study (Figure 8B; Table 1).

The differences in dolomitizing fluids between the two sides are also responsible for the presence of different H₂S concentrations, i.e., generally >10% for the eastern side (Cai et al., 2013; Liu et al., 2013; Li et al., 2014) and 3.3∼7.2% for the western side (Table 3). The brines from the Early Triassic hypersaline evaporite lagoons are obviously superior in carrying dissolved sulfate, which is favorable for the consumption of charged hydrocarbons and H₂S production. This superiority is also reflected in remarkably higher H₂S concentrations in the evaporite-carbonate units around the world, i.e., the Crossfield member of the Okotoks gas field in the West Canadian Basin (Dobbin, 1968), the Jurassic Smackover Formation of Black Creek Field, Mississippi (Heydari, 1997), and the Lower Carboniferous rocks in West Siberia (Belenitskaya, 2000). For the occurrence of TSR related to penesaline dolomitizing fluids, relatively limited sulfate places strong constraints on H₂S production. However, the mild to moderate degrees of H₂S enrichment are also worth our special concern for hydrocarbon discovery in deeply buried dolostone reservoirs.

A comparison of the sea-level trend between the Yangtze Platform in South China (Wang et al., 1999; Yin and Tong, 1996) and global eustatic changes (Figure 4; Haq and Al-Qahtani, 2005, Haq and Schutter, 2008) reveals obvious differences during the Late Permian. Regarding long-term changes, there was a distinct transgression in the Yangtze Platform during most of the period, causing reef growth instead of the global sea-level fall to the lowest point. The subsequent regression began at the end of the Permian, and the drop in sea level led to the formation of back-reefs or inter-reef lagoons, where condensed MgSO₄ seawater formed in a (semi) arid climate. Such penesaline conditions are favorable for the development of dolomitizing fluids and the subsequent occurrence of TSR under deep burial. Therefore, similar scenarios involving non-evaporitic dolomitization and TSR, as well as their close relationship, have probably occurred along the platform-margin reef belts elsewhere on Yangtze platform (Figure 1B). Two case studies in point are available to support the proposal.

The Shibaozhai gas field, lying at the southern tip of the K-L Trough (Figure 1B), has high concentrations of TSR-H₂S (3.7%, Cai et al., 2004) in the P₃ch dolostone reservoirs compared with minor to negligible H₂S in the T₁f limestone reservoirs.

The Jiannan gas field, located in the eastern Sichuan Basin, is the only gas field in marine carbonate strata in the middle of the Yangtze plate (Figure 1A). H₂S build-ups of TSR origin (mostly from 1.7% to 3.8%) appear more frequently in the P₃ch dolostone reservoirs than in the T₁f reservoirs (Li et al., 2015; Zhang et al., 2010).

Interestingly, the pattern of sea-level fluctuation during the Late Permian on Yangtze Platform is representative of the Eurasian Tethys and existed in the Alps, Transcaucasia, NE Vietnam and Japan (Figure 1A, Sano, 2006; Yin and Tong, 1996). We tentatively propose that the model mentioned above might be applicable across a broader range, if hydrodynamic system stays closed during diagenesis for SO₄²⁻ preservation and the temperatures of hydrocarbon reservoirs are high enough for the occurrence of TSR (Machel, 2001; Worden et al., 1995). Unfortunately, data are still scarce to support this proposal. More studies of this aspect are in progress.

Conclusions

Penesaline seawater, developed in back-reef or inter-reef lagoons during sea-level fall, dolomitized the reef-flat facies of the P₃ch Formation in the western reef complex, NE Sichuan. Pervasive dolomitization was further promoted and completed by hotter residual seawater during the subsequent onset of chemical compaction. Additional Mg²⁺ are sourced from the stabilization of aragonite and high-Mg calcite components, as well as clay mineral transformations during the formation of stylolite.

During early reflux dolomitization, sulfate continuously accumulated in the pore waters under the combined effects of MgSO₄ seawater evaporation and refluxing. The progressive dolomitization in shallow- to intermediate-depth burial environments facilitated CAS release and sulfate preservation due to the relatively closed hydrodynamic setting.

Model mass/volume balance calculations show that the sulfate accumulated in the dolomitization process was greatly depleted by the reduction of liquid and gas hydrocarbons (TSR), which was responsible for producing of the present-day high concentrations of H₂S in the NE Sichuan Basin.

The occurrence of massive dolomitization driven by MgSO₄ (aragonite) seawater, under the conditions of favorable sea-level fluctuation and restricted carbonate lagoon, is capable of initiating TSR in evaporite-free deep reservoirs around the world, if there is a closed hydrodynamic system during diagenesis that is favorable for sulfate preservation. This correlation has important implications for the deep exploitation of sour oil/gas in dolostone horizons, which, however, have been largely ignored to date.

Footnotes

Acknowledgements

We would like to thank two anonymous reviewers for giving us constructive suggestions, which were helpful in improving the quality of the manuscript. This work is financially supported by National Natural Science Foundation of China (Grant No. 41972154) and the Fundamental Research Funds for the Central Universities (Grant No. 2652019104).

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China, the Fundamental Research Funds for the Central Universities, (grant number 41972154, 2652019104).

ORCID iD

Kaikai Li

References

Adams

Rhodes

(1960) Dolomitization by seepage refluxion. AAPG Bulletin 44: 1912–1920.

Algeo

Henderson

Tong

, et al. (2013) Plankton and productivity during the permian–triassic boundary crisis: An analysis of organic carbon fluxes. Global and Planetary Change 105(jun.): 52–67.

Belenitskaya

(2000) Distribution pattern of hydrogen sulfide-bearing gas in the former Soviet union. Petroleum Geoscience 6(2): 175–187.

Biehl

Reuning

Schoenherr

, et al. (2016) Impacts of hydrothermal dolomitization and thermochemical sulfate reduction on secondary porosity creation in deeply buried carbonates: A case study from the lower Saxony Basin, northwest Germany. AAPG Bulletin 100: 597–621.

Bildstein

Worden

Brosse

(2001) Assessment of anhydrite dissolution as the rate-limiting step during thermochemical sulfate reduction. Chemical Geology 176: 173–189.

Burke

Denison

Hetherington

, et al. (1982) Variation of seawater ⁸⁷Sr/⁸⁶Sr throughout Phanerozoic time. Geology 10: 516–519.

Cai

Jiang

, et al. (2014) Petrological and geochemical constraints on porosity difference between lower Triassic sour- and sweet-gas carbonate reservoirs in the Sichuan Basin. Marine and Petroleum Geology 56: 34–50.

Cai

Zhu

, et al. (2010) TSR Origin of sulfur in the Permian and Triassic reservoir bitumen in East Sichuan Basin, China. Organic Geochemistry 41: 871–878.

Cai

Worden

Bottrell

, et al. (2003) Thermochemical sulfate reduction and the generation of hydrogen sulfide and thiols (mercaptans) in triassic carbonate reservoirs from the Sichuan Basin, China. Chemical Geology 202: 39–57.

10.

Cai

Xiang

Yuan

, et al. (2017) Sulfur and carbon isotopic compositions of the permian to triassic TSR and non-TSR altered solid bitumen and its parent source rock in NE Sichuan Basin. Organic Geochemistry 105: 1–12.

11.

Cai

Xie

Worden

, et al. (2004) Methane-dominated thermochemical sulfate reduction in the Triassic feixianguan formation east Sichuan Basin, China: Towards prediction of fatal H₂S concentrations. Marine and Petroleum Geology 21: 1265–1279.

12.

Cai

Zhang

, et al. (2013) Carbon isotope fractionation during methane-dominated TSR in East Sichuan Basin gas fields, China: A review. Marine and Petroleum Geology 48: 100–110.

13.

Cheng

Zou

, et al. (2013) Geochemical characteristics and fluid origin of the changxing formation dolomitic rock in the Xinglongchang area of East Sichuan Basin. Acta Geologica Sinica 87(7): 1031–1040. (in Chinese with English abstract).

14.

Davies

Smith

(2006) Structurally controlled hydrothermal dolomite reservoir facies: An overview. AAPG Bulletin 90: 1641–1690.

15.

Dobbin

(1968). Geology of natural gases rich in helium, nitrogen, carbon dioxide and hydrogen sulphide. In: Beebe

Curtis

(eds) Natural Gases of North America. American Association of Petroleum Geologists, Memoir. Vol. 9, pp. 1957–1969.

16.

Duan

Peng

(2014) Lower Triassic Feixianguan stratigraphical sequences and sedimentary facies characteristics in yuanba area in the northeast of Sichuan Basin. Marine Origin Petroleum Geology 2(19): 47–53. (in Chinese with English abstract).

17.

Ebner

Koehn

Toussaint

, et al. (2009) Stress sensitivity of stylolite morphology. Earth & Planetary Science Letters 277(3–4): 394–398.

18.

Enayati-Bidgoli

Navidtalab

(2020) Effects of progressive dolomitization on reservoir evolution: A case from the Permian-Triassic gas reservoirs of the Persian Gulf, offshore Iran. Marine and Petroleum Geology 119: 104480.

19.

Eren

Kaplan

Kadür

(2007) Petrography, geochemistry and origin of lower Liassic dolomites in the Aydnck area, mersin, southern Turkey. Turkish Journal of Earth Sciences 16: 339–362.

20.

Gregg

Sibley

(1984) Epigenetic dolomitization and the origin of xenotopic dolomite texture. Journal of Sedimentary Petrology 54: 908–931.

21.

Guo

(2011) Sequence strata of the platform edge in the Changxing and Feixianguan formations in the Yuanba area, northeastern Sichuan Basin and their control on reservoirs. Acta Petrolei Sinica 32(3): 387–394. (in Chinese with English abstract).

22.

Hao

Guo

Zhu

, et al. (2008) Evidence for multiple stages of oil cracking and thermochemical sulfate reduction in the puguang gas field, sichuan basin, China. AAPG Bulletin 92: 611–637.

23.

Hao

Zhang

Wang

, et al. (2015) The fate of CO₂ derived from thermochemical sulfate reduction (TSR) and effect of TSR on carbonate porosity and permeability, Sichuan Basin, China. Earth-Science Reviews 141: 154–177.

24.

Haq

Al-Qahtani

(2005) Phanerozoic cycles of sea-level change on the Arabian platform. GeoArabia 10(2): 127–160.

25.

Haq

Schutter

(2008) A chronology of Paleozoic Sea-level changes. Science (New York, N.Y.) 322(5898): 64–68.

26.

Hardie

(1987) Dolomitization: A critical view of some current views: perspectives. Journal of Sedimentary Research 57(1): 166–183.

27.

Hardie

(1996) Secular variation in seawater chemistry: An explanation for the coupled secular variation in the mineralogies of marine limestones and potash evaporites over the past 600 m.y. Geology 24(3): 279–283.

28.

Luo

(2010) Lithofacies palaeogeography of the late Permian Changxing Age in middle and upper Yangtze region. Journal of Palaeogeography 12(5): 497–514. (in Chinese with English abstract).

29.

Heydari

(1997) The role of burial diagenesis in hydrocarbon destruction and H₂S accumulation, upper Jurassic Smackover formation, Black Creek field, Mississippi. AAPG Bulletin 81: 26–45.

30.

Horita

Zimmermann

Holland

(2002) Chemical evolution of seawater during the Phanerozoic: Implications from the record of marine evaporites. Geochimica et Cosmochimica Acta 66: 3733–3756.

31.

Cai

Pederson

, et al. (2020) Dolomitization history and porosity evolution of a giant, deeply buried Ediacaran gas field (Sichuan Basin. China). Precambrian Research 338: 105595.

32.

Huang

Yang

Yan

, et al. (2014) Comparison and analysis of Reef-bank Gas reservoirs in Longgang and Yuanba areas in Sichuan Basin. China Petroleum Exploration 19(3): 12–20.

33.

Huang

(1994) Carbon isotopes of Permian and permian-Triassic boundary in upper Yangtze platform and the mass extinction. Geochimica 23: 60–68. (in Chinese with English abstract).

34.

Hurley

Lohmann

(1989) Diagenesis of devonian reefal carbonates in the Oscar Range, Canning Basin, Western Australia. Journal of Sedimentary Petrology 59: 127–145.

35.

Jiang

Cai

Worden

, et al. (2013) Reflux dolomitization of the upper Permian Changxing formation and the lower Triassic Feixianguan formation, NE Sichuan Basin. China. Geofluids 13: 232–245.

36.

Jiang

Worden

Cai

, et al. (2014) Dolomitization of gas reservoirs: The upper Permian Changxing and lower Triassic Feixianguan formations, northeast Sichuan Basin, China. Journal of Sedimentary Research 84(10): 792–815.

37.

Joachimski

Lai

Shen

, et al. (2012) Climate warming in the latest permian and the permian-triassic mass extinction. Geology 40(3): 195–198.

38.

Jones

Whitaker

Smart P

, et al. (2002) Fate of reflux brines in carbonate platforms. Geology 30(4): 371–374.

39.

Jones

Xiao

(2005) Dolomitization, anhydrite cementation, and porosity evolution in a reflux system: Insights from reactive transport models. AAPG Bulletin 89(5): 577–601.

40.

Kaufman

(1994a) The role of fluid flow in dolomitization: Constraints from numerical models Annual Meeting Expanded Abstracts-American Association of Petroleum Geologists: 184.

41.

Kaufman

(1994b) Numerical models of fluid flow in carbonate platforms: Implications for dolomitization. Journal of Sedimentary Research 64(1): 128–139.

42.

Korte

Kozur

Bruckschen

, et al. (2003) Strontium isotope evolution of late Permian and Triassic seawater. Geochimica et Cosmochimica Acta 67: 47–62.

43.

Korte

Kozur

Joachimski

, et al. (2004) Carbon, sulfur, oxygen and strontium isotope records, organic geochemistry and biostratigraphy across the Permian/Triassic boundary in Abadeh, Iran. Int J Earth Sci (Geol Rundsch 93: 565–581.

44.

Land

(1980) The Isotopic and Trace Element Geochemistry of Dolomite: The State of the art. SEPM Special Publication, No. 28. pp. 87–110.

45.

Land

Hoops

(1973) Sodium in carbonate sediments and rocks. Journal of Sedimentary Petrology 43: 614–617.

46.

Lei

Qiang

Wen

(1994) Dolomitization of the upper permian origanic reefs in eastern Sichuan and its adjacent areas. Geological Review 40(6): 534–543. (in Chinese with English abstract).

47.

Zhang

(2015) Geochemical characteristics and genesis of natural gas in jiannan gas field, the western mid-Yangtze area. Acta Petrolei Sinica 36(10): 1199–1209. (in Chinese with English abstract).

48.

Hao

Guo

, et al. (2016) Origin and distribution of hydrogen sulfide in the Yuanba gas field, Sichuan Basin. Southwest China. Maine and Petroleum Geology 75: 220–239.

49.

Cai

Hou

, et al. (2014) Origin of high H₂S concentrations in the upper Permian Changxing reservoirs of the northeast Sichuan Basin, China. Marine and Petroleum Geology 57: 233–243.

50.

Cai

Jiang

, et al. (2012) Sr evolution in the upper Permian and lower triassic carbonates, northeast Sichuan Basin, China: Constraints from chemistry, isotope and fluid inclusions. Appl. Geochem 27: 2409–2424.

51.

George

Cai

, et al. (2019) Fluid inclusion and stable isotopic studies of thermochemical sulfate reduction: upper permian and lower triassic gasfields, northeast Sichuan Basin, China. Geochimica et Cosmochimica Acta 46: 86–108.

52.

George

Cai

, et al. (2020) Comparison of differential diagenesis of two oolites on the lower triassic platform margin, NE sichuan basin: implications for the co-evolution of rock structure and porosity. Marine and Petroleum Geology 119: 1–25.

53.

Liu

Wang

, et al. (2014) Geochemical characteristics and origin of Permian and Triassic natural gas in Yuanba-Tongnanba area, northeastern Sichuan Basin. Acta Petrolei Sinica 35(3): 417–428.

54.

Liu

Worden

Jin

, et al. (2013) TSR Versus non-TSR processes and their impact on gas geochemistry and carbon stable isotopes in carboniferous, Permian and lower Triassic marine carbonate gas reservoirs in the eastern Sichuan Basin, China. Geochimica et Cosmochimica Acta 100: 96–115.

55.

Lorens

(1981) Sr, Cd, Mn and Co distribution coefficients in calcite as a function of calcite precipitation rate. Geochimica et Cosmochimica Acta 45: 553–561.

56.

Lowenstein

Timofeeff

Kovalevych

, et al. (2005) The major-ion omposition of Permian seawater. Geochimica et Cosmochimica Acta 69: 1701–1719.

57.

McCaffrey

Lazar

Holland

(1987) The evaporation path of seawater and the coprecipitation of Br (super -) and K (super+) with halite. J. Sediment. Res 57: 928–937.

58.

Machel

(1987) Saddle dolomite as a by-product of chemical compaction and thermochemical sulfate reduction. Geology 15: 936–940.

59.

Machel

(2001) Bacterial and thermochemical sulfate reduction in diagenetic settings-old and new insights. Sedimentary Geology 140: 143–175.

60.

Machel

(2004) Concepts and models of dolomitization: A critical reappraisal. Geological Society, London. Special Publications 235: 7–63.

61.

Machel

Anderson

(1989) Pervasive subsurface dolomitization of the Nisku formation in central Alberta. J. Sediment. Petrol 59: 891–911.

62.

Malone

Baker

Burns

(1996) Hydrothermal dolomitization and recrystallization of dolomite breccias from the miocene monterey formation, Tepuquet area, California. J. Sediment. Res 66: 976–990.

63.

Melim

Scholle

(2002) Dolomitization of the capitan formation forereef facies (Permian, west Texas and New Mexico): Seepage reflux revisited. Sedimentology 49: 1207–1227.

64.

Montanez

Read

(1992) Eustatic control on early dolomitization of cyclic peritidal carbonates: evidence from the early ordovician upper Knox group. Appalachians. Geological Society of America 104: 872–886.

65.

Mountjoy

Amthor

(1994) Has burial dolomitization come of age? Some answers from the Western Canada Sedimentary Basin. In: Purser

Tucker

Zenger

(eds) Dolomites: A Volume in Honour of Dolomieu. International Association of Sedimentologists, Special Publication no. 21, pp. 203–229.

66.

Mutti

Simo

(1994) Distribution, petrography and geochemistry of early dolomite in cyclic shelf facies, yates formation (Guadalupian) capitan Reef complex, USA. In: Purser

Tucker

Zenger

(eds) Dolomites, A Volume in Honour of Dolomieu, Vol. 21. Int. Assoc. Sedimentol. Spec. Publ., Oxford: Blackwell Publishing Ltd, pp. 91–107.

67.

Newton

Pevitt

Wignall

, et al. (2004) Large shifts in the isotopic composition of seawater sulfate across the Permo-Triassic boundary in northern Italy. Earth and Planetary Science Letters 218: 331–345.

68.

Nicolaides

Wallace

(1997) Pressure-dissolution and cementation and in Oligo-Miocene non-tropical limestone (Clifton formation) Otway Basin. Australia. SEPM. Spec. Publ 56: 249–262.

69.

Purser

Tucker

Zenger

(eds). (1994) Dolomites-A Volume in Honour of Dolomieu. International Association of Sedimentologists, Special Publications, 21.

70.

Qin

Yang

, et al. (2016) The gas origin in Changxing-Feixianguan gas pools of Longgang Gasfield in Sichuan Basin. Natural Gas Geoscience 27: 41–49.

71.

Qing

Bosence

DWJ

Rose

EPF

(2001) Dolomitization by penesaline sea water in early Jurassic peritidal platform carbonates, Gibraltar, western Mediterranean. Sedimentology 48(1): 153–163.

72.

Qing

Mountjoy

(1988) Multistage dolomitization in rainbow build-ups, middle Devonian Keg river formation, Alberta, Canada. J. Sediment. Petrol 59: 114–126.

73.

Rameil

(2008) Early diagenetic dolomitization and dedolomitization of late Jurassic and earliest cretaceous platform carbonates: A case study from the Jura mountains (NW Switzerland, E France). Sedimentary Geology 212(1–4): 70–85.

74.

Saller

Henderson

(1998) Distribution of porosity and permeability in platform dolomites — insight from the Permian of west Texas. American Association of Petroleum Geologists Bulletin 828: 1528–1550.

75.

Sano

(2006) Impact of long-term climate change and sea-level fluctuation on Mississippian to Permian mid-oceanic atoll sedimentation (Akiyoshi limestone group. Japan). Palaeogeography, Palaeoclimatology, Palaeoecology 236(3–4): 169–189.

76.

Sass

Bein

(1988) Dolomites and salinity: A comparative geochemical study. In: Shukla

Baker

(eds) Sedimentology and Geochemistry of Dolostones. Society of Economic Paleontologists and Mineralogists, Vol. 43. Tulsa: Special Publications, pp. 223–233.

77.

Shelton

Hendry

Gregg

, et al. (2019) Fluid circulation and fault- and fracture-related diagenesis in Mississippian syn-rift carbonate rocks on the northeast margin of the Metalliferous Dublin Basin, Ireland. Journal of Sedimentary Research 89(6): 508–536.

78.

Sheppard

SMF

Schwarz

(1970) Fractionation of carbon and oxygen isotopes and magnesium between metamorphic calcite and dolomite. Contrib. Mineral. Petrol 26: 161–198.

79.

Shields

Brady

(1995) Mass balance and fluid flow constraints on regional-scale dolomitization, late Devonian. Western Canada Sedimentary Basin. Bull. Can. Petrol. Geol 43(4): 371–392.

80.

Sibley

Gregg

(1987) Classification of dolomite rock texture. J. Sediment. Petrol 57: 967–975.

81.

Sun

(1994) A reappraisal of dolomite abundance and occurrence in the phanerozoic. Journal of Sedimentary Petrology 64(2): 396–404.

82.

Swart

Cantrell

Westphall

, et al. (2005) Origin of dolomite in the Arab-D reservoir from the Ghawar field Saudi Arabaia: Evidence from petrographic and geochemical constraints. J. Sediment. Res 75: 476–491.

83.

Tian

Zeng

(1995) The evolution pattern of the carbon and oxygen isotopes in the Permian marine carbonate rocks from guizhou. Journal of Chengdu Institute of Technology 22(1): 78–82. (in Chinese with English abstract).

84.

Tian

Liu

, et al. (2014) Dolomitization of the upper Permian Changxing formation in Yuanba gas field, NE Sichuan Basin. China. Acta Petrologica Sinica 30(9): 2766–2776. (in Chinese with English abstract).

85.

Touir

Soussi

Troudi

(2009) Polyphased dolomitization of a shoal-rimmed carbonate platform: Example from the middle turonian bireno dolomites of central Tunisia. Cretaceous Research 30: 785–804.

86.

Tucker

Wright

(1990) Carbonate sedimentology. Oxford: Blackwell Scientific. 82, pp. 1–27.

87.

Vandeginste

Swennen

Reed

, et al. (2009) Host rock dolomitization and secondary porosity development in the upper devonian cairn formation of the Fairholme carbonate complex (south-west Alberta. Canadian Rockies): diagenesis and geochemical modeling. Sedimentology 56: 2044–2060.

88.

Wang

Chen

, et al. (1999) Permian sea -level changes and rising -falling events in south China. Acta Sedimentologica Sinica 17(4): 536–541. (in Chinese with English abstract).

89.

Wang

Xiao

(2010) An investigation of paleogeothermal gradients in the northeastern part of Sichuan Basin. Marine Origin Petroleum Geology 15(4): 57–61. (in Chinese with English abstract).

90.

Wanless

(1979) Limestone response to stress: Pressure solution and dolomitization. J. Sed. Petrol 49: 0437–0462.

91.

Warren

(2000) Dolomite: Occurrence, evolution and economically important associations. Earth-Science Reviews 52: 1–81.

92.

Weidlich

(2002) Permian reefs re-examined: Extrinsic control mechanisms of gradual and abrupt changes during 40 Myr of reef evolution. Geobios 35: 287–294.

93.

Worden

Ruffell

Cornford

(2000) Palaeoclimate, sequence stratigraphy and diagenesis. Journal of Geochemical Exploration 69–70: 453–457.

94.

Worden

Smalley

(1996) H₂S-producing reactions in deep carbonate gas reservoirs: khuff formation, Abu Dhabi. Chem. Geol 133: 157–171.

95.

Worden

Smalley

Oxtoby

(1995) Gas souring by thermochemical sulfate reduction at 140°C. Am. Assoc. Pet. Geol. Bull 79: 854–863.

96.

Worden

Smalley

Oxtoby

(1996) The effect of thermochemical sulfate reduction upon formation water salinity and oxygen isotopes in carbon gas reservoirs. Geochimica et Cosmochimica Acta 60: 3925–3931.

97.

Yin

and Tong JN (1996) Late permian-middle triassic Sea level changes of Yangtze platform. Journal of Earth Science (English Edition) 19(1): 101–104.

98.

Zhang

Liu

Tenger , et al. (2010) Characteristics of natural gas in P₂ch-T₁f layers in jiannan gas field and gas source correlation. Natural Gas Geoscience 21(6): 1004–1013. (in Chinese with English abstract).

99.

Zhu

Zhang

Liang

(2007) The controlling factors and distribution prediction of H2S formation in marine carbonate gas reservoir, China. Chinese Science Bulletin 52(Supp. 1): 150–163.

Massive dolomitization driven by MgSO 4 -rich seawater and its effects on thermochemical sulfate reduction,Upper Permian Changxing Formation,northeastern Sichuan,China

Abstract

Keywords

Introduction

Geological setting

Materials and analytical methods

Results

Dolomite distribution and petrography

Geochemistry

Carbon, oxygen and strontium isotopes

Major and trace element contents

Fluid inclusion and formation water chemistry

Δ34s values of solid bitumens

Discussion

Time and condition of dolomitization

Variation in SO42− concentrations during and after dolomitization

Relationship between dolomitization and TSR

Regional comparisons

Conclusions

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

ORCID iD

References

Δ³⁴s values of solid bitumens

Variation in SO₄²⁻ concentrations during and after dolomitization