Multi-Phase Dolomitization in the Jurassic Paleo-Oil Reservoir Zone, Qiangtang Basin (SW China): Implications for Reservoir Development

Abstract

The age and dolomitization processes in the Paleo-oil reservoir zone, which is composed of massive dolostones found in the Qiangtang Basin (SW China), are still debated. In this research, the Long’eni-Geluguanna Area was selected. Macroscopic information, thin sections, and geochemical methods were used to investigate the dolomitization characteristics and the processes that controlled dolomitization. Five types of replacive dolomites and two types of dolomite cement were observed. Some of the dolomites displayed ghosts of primary sedimentary structures. Saddle dolomites were prevalent, occurring in the interparticle and moldic pores of the limestone which should have been filled at an early diagenetic stage. Ten microfacies types were identified. The foraminifera assemblage provides evidence that the studied interval is of Early Jurassic age. The δ¹³C values are similar to the contemporaneous seawater signature. The REE+Y patterns of limestones and dolostones exhibit similarities to that of seawater. The mean Na and Sr values are comparable to those of other near-normal seawater dolomites. The δ¹⁸O values of all lithologies are markedly depleted. The dolomitization started penecontemporaneously, with deposition. A general sand shoal setting with patch reefs developed. The dolomitizing fluids, near-normal seawater, was probably formed by slight evaporation on top of the shoal. Saddle dolomites in the interparticle and moldic pores might indicate hydrothermal activity, which also caused the recrystallization of some pre-existing dolomites. The recrystallization might have slightly increased the crystal size, demolished the ghost structures, formed saddle dolomites, and altered the REE+Y patterns. The recrystallization extent diminished with increasing distance from the fluids-providing fracture. Furthermore, the existence of protected areas within the sand shoal settings could enhance the vertical and horizontal heterogeneity of dolostone reservoirs.

1. Introduction

Although dolostone is a widespread lithology in the geological record [1] and is also one of the most significant reservoir types in numerous productive oil fields globally [2,3,4], the scarcity of it in the modern record and the inability to synthesize dolomite in the laboratory under ambient conditions [5,6] have impeded our understanding of dolomitization processes and the uneven distribution of this mineral through time. For decades, researchers have endeavored to resolve this so-called “dolomite problem”. Overtime, numerous dolomitization models have been proposed and subsequently refined [7,8]. These include the sabkha model [9,10,11], the reflux model [12,13,14], the mixing-zone model [15,16,17], the near-normal seawater model [18,19], the microbial model [20,21], and the late burial and hydrothermal model [22,23,24]. Recently, many factors, such as silica content [25,26], pH cycling [26,27], ocean acidification [28], and ocean anoxia [29], have been suggested to play an important role in the dolomitization process. In a recent publication, Kim et al. put forth a model suggesting that episodic changes in saturation with respect to dolomite can offer a solution to the “dolomite problem”. This proposition was based on findings from simulations and laboratory experiments [30]. Nevertheless, none of the existing explanations for dolomitization are entirely satisfactory, including the most recent ones [31,32], and the process remains poorly understood. This is primarily due to the fact that the transformation of the original limestone into dolostone frequently results in the complete destruction of primary facies characteristics.

Unlike other marine sedimentary basins worldwide, the Mesozoic Qiangtang Basin has been the subject of relatively limited scientific investigation due to its elevated altitude. This changed approximately twenty years ago, when a Paleo-oil reservoir zone (exposed former reservoirs proven by remnants of oil and bitumen) was identified in the Southern Qiangtang Basin. This zone is composed of dolostones, and the dolostones were classified as belonging to the Bathonian (Middle Jurassic) Buqu Formation [33] based on the biostratigraphy of the index fossils recovered from limestone beds in the Saigongyao area, which also falls within the reservoir zone [34]. However, due to abundant thrust tectonics in the reservoir zone, the dolostone intervals were deemed to have formed in an area situated 7–15 km to the north and subsequently transported to their current location during the Late Cretaceous to Early Paleogene, rather than forming directly in situ [35,36]. The age of the dolostones has also been called into question by the results of the foraminifera assemblage and U-Pb dating. In light of these findings, the depositional age of the interval is estimated to fall within the Late Triassic to the Early Jurassic period [34].

Besides this age problem, the formation history of this dolostone reservoir remains uncertain, as is the case with many other such reservoirs. In the absence of evaporites, the mixing-zone model was initially employed to interpret the dolostone origin [37,38]. Many researchers, however, have put the mixing-zone model in doubt, citing the rarity of dolomite precipitation in modern mixing-zone settings [39,40]. Subsequently, homogenization temperatures and stable carbon and oxygen isotopes from fluid inclusions indicated that the dolostones formed in a high-temperature burial/hydrothermal setting [41,42,43]. Based on C, O, and Sr isotopes and XRD data, other authors have argued that the dolostones formed penecontemporaneously to deposition due to evaporation and were subsequently altered in the shallow burial realm by buried seawater and also in the deep burial environment by hydrothermal fluids [44,45,46]. The dolomitizing fluids associated with the penecontemporaneous dolomitization processes are considered to have originated in restricted areas between shoals [47], on the platform [48], and/or as a consequence of exposure in the intertidal to supratidal areas [49]. In summary, the origin of the dolostones in this area is still debated.

The present study has focused on investigating samples from the Long’eni-Geluguanna Area, with the following objectives: to discuss the age of the strata based on foraminifera, to determine the characteristics of the depositional environment through microfacies analysis, and to reconstruct the timing and processes of dolomitization. A dolomitization hypothesis that may be applicable to other sequences is proposed, and the implication of this hypothesis for sand shoal dolostone reservoirs is discussed.

2. Geological Background

The Qiangtang Basin is a Mesozoic marine sedimentary basin situated within the Tibetan Plateau (Figure 1a), between the Hoh-Xil Terrane and the Lhasa Terrane [50,51]. It is regarded as a constituent of the eastern Tethys tectonic domain in the context of geological history [52]. The northern part of the basin is bordered by the Hoh Xil—Jinshajiang Suture Zone and the southern part by the Bangong–Nujiang Suture Zone. The basin is further subdivided into two sub-basins, namely the North Qiangtang Basin and the South Qiangtang Basin, by the Central Uplift (Figure 1a) [53]. Our studied area is situated in the westernmost part of the large E-W trending, exposed Paleo-oil reservoir zone in the South Qiangtang Basin [47,54]. This zone has a length of approximately 100 km from east to west, with a width of only about 20 km from south to north [33] (Figure 1b). The sedimentary environment of our studied area was interpreted differently due to the dolomitization of the rocks and the sparse microfacies characteristics. For example, it was interpreted as a reef environment [33], a restricted marine [55], a platform-marginal [56,57] or a subtidal sand shoal [49], or a lagoon surrounded by marginal shoals [58].

In the eastern Tethys tectonic domain, the Cimmerian blocks, which include Sibumasu, Lhasa, South Qiangtang, Afghanistan, and Iran, initiated rifting from the northern margin of Gondwana in the Early Permian. This marks the onset of the Meso–Tethys expansion and the Paleo–Tethys closure [59]. In the Late Triassic, the North Qiangtang Subterrane collided with Eurasia along the Hoh Xil-Jinshajiang Suture, which is indicative of the final closure of the Paleo–Tethys in this region [59,60]. Additionally, during the Late Triassic period, the South Qiangtang Subterrane collided with the North Qiangtang Subterrane, resulting in the closure of a limited ocean basin between them and the formation of the Central Uplift [61]. Subsequently, the Meso–Tethys gradually expanded to its maximum extent during the Middle to Late Jurassic period, after which it began to close [62]. In the Late Jurassic to the Early Cretaceous, the Lhasa Block collided with the Qiangtang Block, resulting in the closure of the Bangong–Nujiang Ocean [59,63].

The Mesozoic stratigraphy exhibits notable differences between the North and the South Qiangtang Basin, as well as between individual sections (Figure 2). The deposition of marine sediments commenced in the Early Jurassic and continued until the Callovian, when a fast sea-level drop caused a transition from a marine to a terrestrial environment [62]. The Hettangian–Bajocian strata in the North Qiangtang Basin are called the Quemocuo Formation, which was deposited in alluvial to marine-continental environments, whereas in the South Qiangtang Basin, the Quse (Hettangian–Aalenian) and the Sewa Formation (Aalenian–Bajocian) are subdivided. The Quse Formation is considered to have been deposited environments ranging from a shallow marine carbonate platform to deep marine environments, whereas the Sewa Formation is considered to have been deposited in delta or deep marine environments (Figure 2). The Bathonian Buqu Formation is considered to have been formed in a stable carbonate platform environment (Figure 2) [58,62,64]. During this period, the North and South Qiangtang Basins were connected [62].

3. Samples and Methods

3.1. Samples

This research encompasses the Geluguanna outcrop (GL, 32°43′19″ N, 88°46′37″ E), the Long’eni outcrop (LEN, 32°46′53″ N, 88°49′11″ E), and five drill cores (WA1 and WA3 from the Geluguanna Area; WB1, WB4, and WB5 from the Long’eni Area).

A total of 118 samples were taken from the GL outcrop (Figure 3) using a hand drill. The diameter of the drill cores obtained by the hand drill was approximately 5 cm, with an average length of approximately 30 cm. The entire section was divided into 79 layers. In dolostone intervals and immediately adjacent layers, samples were taken at higher resolution. The samples obtained from this outcrop were named “GL (outcrop name)-layer number-sample number in this layer”.

A total of 105, 112, 66, 83, and 26 samples were taken from the WA1, WA3, WB1, WB4, and WB5 drill cores, respectively. The distance between the sampled intervals varies from 0.1 to 3 m, with the exception of intervals that are currently composed of dolomite sand and therefore unsuitable for sampling. Samples taken from the drill cores were named “drill core name-the sample number in this drill core”.

3.2. Methods

3.2.1. Classification Schemes

In accordance with the classification scheme proposed by Leighton and Pendexter [66], four rock types were subdivided as follows: limestone (dolomite content < 10%), dolomitic limestone (10% < dolomite < 50%), calcareous dolostone (50% < dolomite < 90%), and dolostone (dolomite > 90%). The Dunham classification [67] (revised by Embry and Klovan [68]) was employed for the study of limestones. In order to classify the dolomite textures, the classification proposed by Sibley and Gregg [69] was utilized. In regard to dolomite size, the classification proposed by Zhao and Zhu was employed [70].

3.2.2. Petrography and Microfacies Analysis

Thin sections were stained with Alizarin-red-S in order to distinguish the calcite from the dolomite [71]. All samples were impregnated with blue resin to show porosity. Some images were cropped and adjusted slightly using the CorelDraw 2022 software (Alludo, Ottawa, ON, Canada).

The criteria of microfacies analysis proposed by Flügel [72] was employed.

3.2.3. C-O Isotopes

C-O isotope analysis was conducted using a Delta V Advantage isotope ratio mass spectrometer, and GBW4405 and GBW4406 standard samples were used for comparison. Approximately 10 mg of powder was utilized from each sample. Before analysis, samples were treated with anhydrous phosphoric acid at 50 °C for 4 h to release CO₂ completely. The results are presented in relation to the V-PDB standard, and the long-term measurement precision for δ¹³C and δ¹⁸O is ±0.06‰ and ±0.08‰, respectively.

3.2.4. Trace Elements and Rare Earth Elements

Trace elements and rare earth elements (REEs) were measured using an iCAP TQ Inductively Coupled Plasma Mass Spectrometry. Before measurement, 40–60 mg of the sample powder was treated with 1 mL of 3 mol/L acetic acid for 12 h and then centrifuged. Thereafter, the removal of the supernatant and the addition of 0.5 mL of 0.1mol/L HNO₃ were repeated until all the acetic acid was removed. The analytical precision was better than 10% (1σ).

The results for the REEs were normalized to post-Archean Australian shale (PAAS) [73,74] for further discussion.

4. Results

4.1. Outcrop Description

The GL outcrop (Figure 3a) has an approximate thickness of 340 m (Figure 3b,c). The strata in this outcrop exhibit a degree of tilting. In the northernmost part, the dip angle is nearly 90 degrees (Figure 3a). The occurrence of dolostone and dolomitic limestone layers is limited to the upper part (northern part, above layer 52) of the section (Figure 3a). These layers intercalate with limestone layers, including bioclastic limestone, peloidal limestone and/or micritic limestone (Figure 3b). Below layer 52, limestone is the only lithology. This undolomitized interval exhibits bioclastic limestone, peloidal limestone, ooidal limestone, and microbial limestone and is dominated by bioclastic limestone, which contains abundant large, well-rounded lithoclasts and fragments of corals, mollusks (Figure 3d), or abundant (but in low diversity) large, well-preserved mollusk shells (predominantly bivalves and/or gastropods, Figure 3e).

The Long’eni outcrop (Figure 4a) has an approximate thickness of 140 m (Figure 4b). The dip angle of the strata in this outcrop ranges from 23 to 59 degrees. The only lithology observed in this outcrop is dolostone (Figure 4b). Some sedimentary structures were preserved to some extent after dolomitization, including meter-scale cross-beddings (Figure 4c), irregular “beddings” that may indicate a reef or reef debris facies precursor (Figure 4d), and laminations (Figure 4e,f).

It was not possible to observe the top and bottom formation boundaries, as they are either weathered or not exposed.

4.2. Dolomite Texture

4.2.1. Replacive Dolomites

The dolomites present in the studied samples can be classified as either replacive or cement. Replacive dolomites (RD) are more prevalent. The crystals exhibit a range of sizes, from very finely crystalline to very coarsely crystalline, and shapes, from planar-e to non-planar. Replacive dolomites are subdivided into five groups based on their crystal size and morphology.

RD1: The group is distinguished by the presence of very finely crystalline dolomites (<0.1 mm). The dolomites frequently exhibit a planar-e to -s and a murky appearance (Figure 5a).

RD2: The group is distinguished by the presence of finely crystalline dolomites (0.1~0.25 mm). In the densely packed portions of the dolomites, they frequently exhibit a planar-s morphology. However, in the porous regions, they often show a planar-e to -s morphology (Figure 5b).

RD3: The group is distinguished by the presence of medium crystalline dolomites (0.25~0.5 mm). The dolomites are frequently planar-s or non-planar, but also exhibit a planar-e to -s texture in the porous parts (Figure 5c).

RD4: The group is distinguished by the presence of coarsely crystalline dolomites (0.5~1 mm). The dolomites frequently display a non-planar shape and exhibit a mosaic fabric (Figure 5d). In the porous potions, the dolomites may display a more planar boundary. Some of the dolomites display undulose extinction, which is a defining characteristic of saddle dolomites (Figure 5d).

RD5: The group is distinguished by the presence of coarsely to very coarsely crystalline dolomites exhibiting undulose extinction, which are identified as saddle dolomites. Some planar-e to -s crystals without undulose extinction are still observed (Figure 5e,f).

4.2.2. Dolomitic Cement

Two different types of dolomitic cements can be distinguished:

CD1: These are brighter cement with planar boundaries. They mainly grow on the crystals adjacent to the pores, forming a lining within the empty space (Figure 5c).

CD2: This cement type is composed of saddle dolomites, and they are more abundant compared to CD1. CD2 is present in both dolostone (Figure 6) and limestone (Figure 7 and Figure 8). In dolostone, it is observed to occur in and in close proximity to veins (Figure 6a–e), as well as within moldic pores (Figure 6f). In limestone, they are frequently observed within veins (Figure 7a), interparticle pores (Figure 7b–d and Figure 8a–d), as well as moldic pores (Figure 7e–h). In some cases, the saddle dolomites are dedolomitized (Figure 8e,f). The curved lattice allows for determination. In some cases, they still exhibit undulose extinction and remnant dolomite (Figure 8e,f).

4.2.3. Ghost Structures

Dolomitization normally destroys primary sedimentary information. In our studied samples, however, the primary information in some samples was found to have been preserved to some extent (Figure 5c,d, Figure 6e,f and Figure 9). The grains are the most readily identifiable components, as their type, size, roundness, and sorting are normally well-preserved. In some instances, even the internal structures of grains were well-preserved (Figure 9e,f). In the case of other components and/or structures, such as micrites, microbial structures are only discernible to a limited extent. Only in rare cases, the ghosts of lamination are recognized (Figure 9g,h). Consequently, only in some samples, these residual structures were sufficient to enable the reconstruction of the original microfacies type. In saddle dolomites, the ghost structures are frequently absent or only minimally preserved (e.g., in RD5, Figure 5e,f, within and in close proximity to veins, Figure 6a–d), but in some instances, they are recognizable and more well-preserved (e.g., in RD4, Figure 5d and Figure 6d,e).

4.3. Microfacies Analysis

MF1

Homogeneous Mudstone

MF1 is mainly characterized by non-laminated lime mudstone (Figure 10a,b). The fossils present in each sample are scarce and exhibit low diversity. In very rare cases, foraminifera, fragments of gastropod and/or bivalve, and echinoderms could be seen. Evaporite pseudomorphs were identified. They concentrated along veins, rather than being distribute throughout the matrix (Figure 10b).

MF2

Peloidal Packstone

The grains in MF2 are rounded to well-rounded and moderately to well-sorted peloids (Figure 10c,d). Fossils are rare. Foraminifera (Figure 10d) and gastropods could be seen in some cases. The grains are supported by each other, with the space between them mainly occupied by micrite or dolomitized micrite.

MF3

Ooidal Grainstone

The grains in this MFT are predominantly very-well sorted and well-rounded ooids (Figure 10e,f). Due to the strong micritization, the internal structures of most of the ooids were demolished, but can still be seen in some ooids (Figure 10e). Marine phreatic cement is observed (Figure 10e). The ooids are often found to have highly fragmented bivalve and/or gastropod shells as nuclei.

MF4

Well-Sorted Bioclastic/Intraclastic Rudstone

MF4 is distinguished by the presence of well- to very well-sorted, well-rounded, relatively large (>2 mm) bioclastic and intraclastic grains, as well as the absence of carbonate mud. In the space between the larger grains, a number of smaller peloid grains, micritized ooids and coated bioclasts were observed (Figure 11a). Fossils show moderate abundance and diversity, though microproblematica is the most prevalent fossil type. Marine phreatic cement (Figure 11b) was observed.

MF5

Poorly Sorted Bioclastic and Intraclastic Rudstone (without Mud) to Grainstone

MF5 contains subangular to well-rounded grains that are poorly sorted. In comparison to MF3, this MFT also exhibits the presence of larger (>2 mm) grains. These larger grains are reworked bioclasts and, in some cases, lithoclasts (Figure 11c,d,f). The occurrence of larger grains is less prevalent in a few samples, so these samples should be classified as grainstones. Fossils show relatively higher abundance and diversity. Mollusks (mainly gastropods, Figure 11f), sponges (Figure 11d), corals (Figure 11d), microproblematica (Figure 11c), echinoderms (Figure 11c), and foraminifera (Figure 11f) were found.

MF6

Poorly Sorted Bioclastic and Intraclastic Rudstone (with Mud) to Packstone

The main difference between MF6 and MF5 is the presence of a greater quantity of carbonate mud (Figure 11e,f) in the former.

MF7

Bioclastic Floatstone to Rudstone with Better Preserved Fossils

This MFT also exhibits a higher carbonate mud content, and the shape of fossils is usually preserved better (Figure 12a–d) than those in MF4-6 (Figure 11). Gastropods (Figure 12a), sponges (probably, Figure 12a), corals (Figure 12b), and foraminifera (Figure 12c,d) are common. It is notable that the fossils are often overgrown by microbes (Figure 12b).

MF8

Subrounded Rudstone

The most distinctive feature of MF8 is the prevalence of subrounded reworked lithoclasts, which exhibit pronounced microbial structures (Figure 12e). Fossils are relatively uncommon in this type. Most of them are foraminifera, gastropod, and cyanobacteria/microproblematica. Isopachous marine phreatic cement is identified (Figure 12e).

MF9

Microbial Bindstone

MF9 is characterized by microbial layers (Figure 12f,g). Fossils are rare in most cases.

MF10

Stromatolite

MF10 is distinguished by the presence of typical stromatolite structures (Figure 12h). In some samples, the microbial layers were fractured and formed a breccia. This suggests that these layers were hard before reworking. No additional fossils were identified within this MFT.

4.4. Geochemical Evidence

4.4.1. Stable Carbon and Oxygen Isotopes

The C-O isotope data are presented in Figure 13a. For the range and mean values, please refer to

Table 1. C-O isotope data.

Lithology		δ13C (VPDB, ‰)	δ18O (VPDB, ‰)
Limestone	Range	−0.49~7.45	−16.51~−7.73
	Mean	2.05	−11.55
Transition	Range	−1.91~4.10	−13.75~−9.28
	Mean	2.12	−10.98
Dolostone	Range	−3.49~5.28	−17.79~−7.62
	Mean	2.10	−10.12

.

4.4.2. Trace Elements and Rare Earth Elements (REEs+Y)

The Na and Sr concentrations are presented in Figure 13b. For the range and mean values, please refer to

Table 2. Na and Sr data.

Lithology		Na (µg/g)	Sr (µg/g)
Limestone	Range	17.74~276.80	57.09~871.30
	Mean	117	167
Transition	Range	79.66~345.40	47.35~115.20
	Mean	240	81
Dolostone	Range	26.61~891.90	20.34~137.50
	Mean	313	51

.

The REE+Y distribution patterns and the plot of values of Eu/Eu* vs. MREE/LREE are shown in the Figure 13c–f. For the range and mean values, please refer to

Table 3. Parameters of the REE+Y.

Rock Types		Σ[REE+Y]	ΣLREE	ΣMREE	ΣHREE	Eu/Eu*
Limestone	Range	0.76~7.66	0.44~5.15	0.08~0.75	0.03~0.29	0.94~1.19
	Mean	3.00	1.97	0.30	0.12	1.04
RD1	Range	1.20~1.32	0.50~0.57	0.10~0.11	0.08~0.09	0.75~0.83
	Mean	1.26	0.54	0.11	0.08	0.79
RD2	Range	0.76~1.63	0.28~0.83	0.07~0.16	0.05~0.10	0.91~1.10
	Mean	1.13	0.54	0.11	0.08	0.97
RD3 *	Range	-	-	-	-	-
	Mean	1.04	0.62	0.11	0.05	0.95
RD4 *	Range	-	-	-	-	-
	Mean	1.29	0.46	0.13	0.09	0.95
RD5	Range	0.42~0.74	0.24~0.38	0.04~0.08	0.02~0.05	1.03~1.25
	Mean	0.58	0.31	0.06	0.03	1.14

* RD3 and RD4 only have one measured sample.

. Eu/Eu* = Eu/[0.67Sm + 0.33Tb] (based on [75]).

Figure 13. The geochemical results for different lithologies. (a) Stable carbon and oxygen isotope values. The values are compared to the contemporaneous seawater calcite background values (blue background, −1.5~4.5‰, from [76]). (b) Na and Sr concentrations of different dolomites. The values are compared to the background values of near-normal seawater dolomites (Na: 100~600 µg/g; Sr: 40~150 µg/g, both from [19]) (c) REE+Y patterns of limestones. (d) REE+Y patterns of RD1-RD4. (e) REE+Y patterns of RD5. (f) Eu/Eu* vs. MREE/LREE plot. The crustal fluids region is from [75].

Figure 13. The geochemical results for different lithologies. (a) Stable carbon and oxygen isotope values. The values are compared to the contemporaneous seawater calcite background values (blue background, −1.5~4.5‰, from [76]). (b) Na and Sr concentrations of different dolomites. The values are compared to the background values of near-normal seawater dolomites (Na: 100~600 µg/g; Sr: 40~150 µg/g, both from [19]) (c) REE+Y patterns of limestones. (d) REE+Y patterns of RD1-RD4. (e) REE+Y patterns of RD5. (f) Eu/Eu* vs. MREE/LREE plot. The crustal fluids region is from [75].

5. Discussion

5.1. Depositional Environment

The essential characteristics and interpretations (based on [72]) of different microfacies types are presented in

Table 4. Key characteristics and interpretations of MFTs.

MFT	Key Characteristic	Interpretation
MF1	Very high micrite content, rare grains and no lamination (due to bioturbation)	Very low water energy, not restricted, e.g., back-shoal [77]
MF2	High micrite content, abundant grains (peloids/micritized ooids)	Low water energy, e.g., back-shoal or protected areas in shoal/shoal fringe
MF3	Very well-sorted and well-rounded ooids, very low micrite content	Very high water energy, e.g., sand shoal [77,78]
MF4	Well- to very well-sorted and well-rounded bioclasts (>2 mm), very low micrite content	High water energy, e.g., sand shoal [79]
MF5	Poorly sorted bioclasts (>2 mm), very low micrite content	Moderate water energy, e.g., shoal fringe [79,80,81]
MF6	Poorly sorted bioclasts (>2 mm), high micrite content	Moderate water energy (lower than that of MF5), e.g., shoal fringe to back-shoal, or protected areas in shoal fringe [79,80]
MF7	Better preserved fossils, high micrite content	Low water energy, e.g., back-shoal [79]
MF8	Subrounded lithoclasts with microbial structures, low micrite content between lithoclasts	In close proximity to autochthonous microbial structures
MF9	Microbial layers	Protected/restricted areas in shoal/shoal fringe [82]
MF10	Stromatolite	Protected/restricted areas in shoal [82]

.

The microfacies data suggest that many intervals display characteristics of a high-water energy environment, which is probably a sand shoal (Figure 14, Figure 15 and Figure 16). As illustrated in the reconstructed paleogeographical map of the Qiangtang Basin [57], the Long’eni-Geluguanna Area was situated in a platform-marginal sand-shoal setting. In this general marginal sand-shoal setting, the potential for the development of patch reefs is indicated by the presence of abundant and diverse fragments of reef-building organisms (Figure 10, Figure 11, Figure 12, Figure 14, Figure 15 and Figure 16). Furthermore, microfacies data indicate that the sediments were deposited in relatively low-energy environments, which were interpreted as representing back-shoal areas or protected areas within shoals (Figure 10, Figure 11, Figure 12, Figure 14, Figure 15 and Figure 16, Table 4).

5.2. Reconsideration of the Strata Age

The Middle Jurassic Bathonian, as corroborated by the paleontological evidence found in the Saigongyao outcrop (Figure 1b), was initially assigned to the entire Paleo-oil reservoir zone. However, the age was recently challenged by Zhang et al., based on the foraminifera found in the XGXN outcrop [34] (Figure 1 and Figure 2). The dolostone intervals in the Saigongyao outcrop were not deemed to be part of the same horizontally distributed depositional sequence with the dolostones observed in the outcrops of the Paleo-oil reservoir zone (Xiaogaxiaona and Zharendong). Additionally, regional marker layers are scarce in most of the outcrops and drill cores [34]. It has also been proposed that the dolostones were formed in a location situated 7–11 km to the north. Subsequently, during the Late Cretaceous to Paleogene, the strata were transported by thrust tectonics to their current position [35,36]. This also calls into questions the previous age estimation. Furthermore, two different hydrocarbon systems have been identified in the reservoir zone [83,84], which could indicate that the carbonate sequence in the reservoir zone was not deposited simultaneously [34].

The age of the studied strata in the Long’eni-Geluguanna Area can be assigned to the Early Jurassic, most probably to the Sinemurian-Pliensbachian, based on the foraminifera assemblage observed in the samples. This is consistent with the findings of Zhang et al. [34], which documented a multitude of Early Jurassic foraminifera species. In our samples, the Cyclorbitopsella tibetica Cherchi, Schroeder and Zhang [85], which did not appear until the latest Sinemurian-Pliensbachian and has a unique shape, was identified (Figure 12c,d). Similarly, this species of large benthic foraminifera was also identified in the Tethys Himalaya block, facilitating the establishment of the regional Early Jurassic biostratigraphy [86,87]. Additionally, the Siphovalvulina sp. (Figure 10d), which was documented by Zhang et al. [34] to have been present in the Sinemurian to the Pliensbachian [88], was also identified in our samples.

Although much evidence points to an Early Jurassic age, confirmation of the strata age requires further detailed biostratigraphic studies.

5.3. Timing of Dolomitization

In some of our samples, dolomites (many of them can be identified as saddle dolomites, discussed later, in Section 5.4.2) obviously formed during the lithification phase, before the final phase of calcite precipitation (Figure 7, Figure 8 and Figure 10f). Most authors agree that most tropical carbonates lithified very early [89], probably in the shallow subsurface [64,65]. The source of the cement is, in most cases, dissolved aragonite, which dissolves due to changes in pore water chemistry driven by the bacterial decay of organic matter (see the review in Munnecke et al., 2023 [89]). A comparable phenomenon is observed in Figure 10f. The ooid (Figure 10f) was cemented by shallow burial cement as the first step, followed by the dissolution of the outer aragonitic cortices, which nevertheless retained the texture to some extent. This aragonitic relics phenomenon was previously documented by Sandberg et al. [90,91] and Titschack et al. [92]. Subsequently, dolomite formed in the empty space, which was subsequently occupied by calcite. Additionally, the absence of strong mechanical compaction in our samples lends further support to this process. Wan et al. reported the occurrence of fabric-preserved (mimic replacement), very finely crystalline dolomites from the eastern part of the Paleo-oil reservoir zone [48]. This type of dolomite is believed to have formed in near-surface conditions [48,69,93]. It can thus be argued that the dolomitization process in the Long’eni-Geluguanna Area commenced at the very early diagenetic stage.

5.4. Dolomitization Processes and Hypothesis

5.4.1. Near-Normal Seawater Dolomitization

A reasonable interpretation of the massive dolostones in the studied area is that they were formed by near-normal seawater dolomitization [18,19,94]. The dolomitizing fluids might have been the seawater, modified by slight evaporation, which occurred periodically at the topmost part of the sand shoal.

Our observations indicate that dolomitization is associated with the topmost part of the shoal, encompassing both high-energy areas (MF3) and protected areas (MF2, 9, 10). In contrast, back-shoal and shoal fringe areas (MF1, 5, 6, 7), which are distinguished by deeper water, were less conducive to dolomitization.

The most evident and corroborating evidence is derived from the GL outcrop and the WA1 well (Figure 14 and Figure 15). The interval below layer 52 is predominantly composed of shoal fringe and back-shoal sediments (MF1, 5, 6, 7), and there is no evidence of dolomitization, with the exception of rare saddle dolomites. Above layer 52, where dolostones are much more abundant, intervals that were not strongly dolomitized are primarily shoal fringe and back-shoal sediments (MF1, 5, 6, 7). In contrast, many of the dolostones contain relics of ooid, peloid, bioclast, and lamination (Figure 14), and the occurrence of ooid and/or peloid ghosts (probably MF3 or MF2 prior to dolomitization) is more prevalent than that of bioclasts (probably MF5 or MF6 prior to dolomitization). A comparable phenomenon is also demonstrated in the WA1 (Figure 15). In intervals that were not strongly dolomitized, the sediments were predominantly shoal fringe and back-shoal (MF1, 5, 6, 7). Additionally, dolostone intervals with more ooid and/or peloid ghosts than bioclast ghosts were observed. In the WB1, intervals of protected settings that exhibit partial dolomitization were observed, whereas the intervals of shoal fringe and back-shoal settings remain undolomitized (Figure 15). Although fully dolomitized (WA3, Figure 15) or not well preserved (highly fragmented), probably due to later strong tectonics and weathering (WB4 and WB5, Figure 16), similar features (but less abundant) were also observed in the samples from these drill cores.

This feature may facilitate the near-normal seawater dolomitization process. The top most portion of the shoal was deposited in the shallowest water depths. Dolomitizing fluids formed in different environments may also have different characteristics (e.g., Mg/Ca, salinity); therefore, they might have caused different degrees of dolomitization [13]. During the time period of low accommodation rate, restricted waters might have been present in this part, proven by the relics of lamination (Figure 9g,h). Even tidal flat [95] or sabkha settings [96] are possible. In contrast, other parts (e.g., shoal fringe) might have remained well connected to the open marine environment. This restricted setting is also conducive to evaporative dolomitization, as evidenced by several studies [96,97,98,99,100].

The occurrence of evaporation can be substantiated by the (limited) presence of evaporite pseudomorphs, and the reported mimic, very finely crystalline dolomites [48]. These reported dolomites are believed to be formed under near-surface, low temperature, and elevated salinity conditions [48,69,93]. However, these reported dolomites are not associated with evaporites. This suggests that the salinity of the dolomitizing fluids was not as high as that of the fluids which were supersaturated with respect to evaporites. Furthermore, the absence of evaporites is a common feature in numerous studies on the dolomitization of near-normal seawater [19,101,102,103,104].

The occurrence of the Early-to-Middle Jurassic (Quemocuo Formation) evaporites (10~30 m, mainly gypsum and anhydrite) in the nearby Shuanghu Area (Figure 1), reported by Li et al. [105], is indicative of an environment conducive to evaporation within the basin. Although many evaporative dolomites have been observed in association with thick-bedded evaporites [106,107,108], it is possible that the reported evaporites in the nearby Shuanghu Area are not directly related to the studied dolostones. If a direct link existed, it would be expected that pervasive dolomitization would occur in the area where evaporites are present, as well as between that area and the one in our research (see reflux model [12,14]). Nevertheless, this phenomenon has not been documented. In instances where dolostones occur in conjunction with thick-bedded evaporites, it is also possible that the dolomitizing fluids may be derived from near-normal seawater, rather than the highly condensed evaporative seawater associated with the thick-bedded evaporites [19].

The REE+Y distribution patterns can also be explained by our hypothesis. A considerable number of dolostone patterns exhibit a distinct seawater signal (Ce negative anomaly and Y positive anomaly [109]). This could indicate that dolomitizing fluids are seawater-like or that the fluids’ signal is buffered due to a low fluid–rock ratio [110,111], and the signal is mainly inherited from the precursor [112]. Given that dolomitization is a process that requires large fluid–rock ratios [113], the second possibility seems less probable. The REE+Y distribution pattern characteristic is also similar to that observed in numerous other studies on seawater-like mediated dolomitization [103,104,114]. In comparison to other studies on near-normal seawater dolomitization, the Na and Sr concentrations are also similar (Figure 13) [19,94,115,116].

A comparable interpretation of the dolostones in the marginal sand shoal setting was put forth by Lu et al., with regard to the pervasive dolomitization observed in marginal sand shoals in the Permian Qixia Formation, the Sichuan Basin [117]. In the Lower Triassic Feixianguan Formation, the Sichuan Basin, the dolomitization of the marginal sand shoal is similarly interpreted as occurring in a short-lived evaporative setting (sabkha type, without pervasive evaporites), above the sand shoal, during periods of decreased accommodation [96]. In our studied area, however, no evidence for subaerial exposure was found, and thus an intertidal to supratidal setting is not assumed.

5.4.2. Saddle Dolomite in Limestones

Saddle dolomite is believed to be formed in high-temperature environments (60~150 °C, most probably >100 °C [118]; >80 °C [119]). Its distinctive undulose extinction [120] allows for straightforward identification. It can be formed as cement in open spaces, through the replacement of calcite, or by the alternation of previous dolomites [75,113,118].

It is noteworthy that in limestones, a considerable number of saddle dolomites are present within the interparticle pores (Figure 7b–d) and the moldic pores of aragonitic fossils (Figure 7e–h), both of which should be expected to be filled by calcite cement very early [89]. No empty pores were observed. Three potential scenarios could potentially explain this phenomenon:

(1)

Saddle dolomites are directly formed in a subsequent diagenetic stage (after lithification) during deeper burial conditions. This, however, is probably not the case because the majority of the saddle dolomites are not associated with veins, but rather occur preferentially in empty space (Figure 7 and Figure 8). Furthermore, it might also be difficult for saddle dolomite to replace calcite cement (Figure 8a,b). It has been documented that in instances when saddle dolomite grows together with calcite cement, they exhibit competitive growth [118].

(2)

Non-saddle dolomites formed in the pore space during lithification, and in a later diagenetic stage (after lithification), they recrystallize into saddle dolomites. It is difficult to prove this hypothesis, given the limited evidence available. One observation might support this hypothesis. If saddle dolomite crystals present only straight outlines, it might suggest a transformed “normal” dolomite crystal into a saddle dolomite. This, however, was not convincingly documented in the investigated samples.

(3)

Saddle dolomites are formed early, either through the recrystallization of former “normal” dolomites or through direct precipitation, after an initial lithification (at least after partial lithification) and the subsequent dissolution of aragonite, but before calcite precipitation in the moldic pores or in the remaining interparticle pore space (Figure 17). Some saddle dolomites, which are surrounded by calcite cement, exhibit curved outlines (Figure 7c–f). This phenomenon indicates that these saddle dolomites were already “saddle” before the calcite precipitation in the moldic pore and the remaining interparticle pores. Saddle dolomite within interparticle pores was also observed by Martín-Martín et al. [121]. It is unlikely that the temperature required for saddle dolomite formation can be provided by the burial depth, where early lithification occurs. However, hydrothermal dolomitization might occur even in near-surface to shallow burial depths (syn. to early post-deposition [122], <100 m [123]), and it can even breach the sea-floor [124]. This potential scenario appears to be more plausible.

Technically, saddle dolomite can be formed in the different scenarios mentioned above, but according to our observations, it seems that at least some of it formed early (near-surface), but under the influence of hydrothermal fluids.

Saddle dolomites occur in every studied drill core and in the GL outcrop. However, in the outcrop, saddle dolomites are more abundant in the upper part (Figure 14). This might be attributed to the distance between the outcrop and the main dolomitizing fluid-providing fracture or fault. As illustrated by fault-related hydrothermal dolomitization models, in which hydrothermal dolomitizing fluids were observed to flow upwards [124,125,126,127], dolostones would be formed both in close proximity to the fault, forming a dolostone pipe, and at varying distances laterally, depending on the permeability of the rocks [124,125,126,127]. Even distances exceeding 7 km from the source fault have been observed [126]. If the outcrop is not situated in close proximity to a main fault, it is possible that the lower section is subjected to a lesser degree of influence from the fluids, resulting in a reduction in the prevalence of saddle dolomites. Similarly, Martín-Martín et al. [121] also documented that the dolomitic cement with undulose extinction is confined to the immediate vicinity of the dolomitization front (up to 1~2 m).

All saddle dolomites from the lower part of the GL outcrop (lower than the layer 40) were dedolomitized (Figure 8e,f and Figure 14). This phenomenon might be attributed to the intrusion of Ca-rich fluids into the system, resulting from the conversion of gypsum-anhydrite (gypsum lost 49% of its volume in the form of intercrystalline water, which is saturated with respect to CaSO₄ [128]), as well as by pressure dissolution within anhydrite in the burial setting [129]. The reported occurrence of Early–Middle Jurassic gypsum and anhydrite [105] may lend support to this hypothesis. Although egogenetic and telogenetic meteoric water can also result in dedolomitization, this process is expected to occur from above.

5.4.3. Hydrothermal Dolomitization and Recrystallization

Hydrothermal activity is indicated by the presence of abundant saddle dolomites (Figure 6, Figure 7, Figure 8, Figure 14, Figure 15 and Figure 16) and by the positive Eu anomalies in some samples (Figure 13) [109,130].

The occurrence of these probably early saddle dolomites raises the question of whether the massive dolostones might have been formed during the same process. Hydrothermal dolomitization can result in the formation of pervasive dolostones that exhibit both stratabound characteristics and a facies selectivity in certain instances [125,126]. The dolomitizing fluids might be seawater-like, resulting in the geochemical signals of dolostones, analogous to those of seawater [127,131]. In consideration of numerous very finely to finely, planar-e to -s dolomites (may indicate a temperature <50 °C [132]) in our samples, and of the very finely crystalline, mimic dolomites documented by Wan et al. [48], it can be postulated that hydrothermal dolomitization might not be the main reason for the formation of massive replacive dolostones in the study area. Lu et al. also reported a dolomitization setting in which hydrothermal dolomitization occurred shortly after (penecontemporaneous to shallow burial. <100 m) the evaporative dolomitization [123], indicating that two dolomitization processes with significant temperature differences can likely occur nearly simultaneously.

The sample WA1-042 is an ideal specimen for showing the interrelationship between saddle dolomites and dolomite recrystallization (Figure 6a–e). The thin section can be subdivided into four parts (Figure 6b–e). This is analogous to the formation model of the zebra structure in hydrothermal dolostones proposed by McCormick et al. [75] and also the “retreating” dolomitization front proposed by Koeshidayatullah et al. [133]. The hydrothermal dolomitizing fluids flow through the fracture in the replacive dolostone body, precipitating saddle dolomite in the empty space and causing recrystallization of the former dolomites.

It can be reasonably inferred that, in WA1-042, part I might be the saddle dolomite cement precipitated in the empty fracture. It is possible that parts II to IV consist of recrystallized replacive dolomites, with the degree of recrystallization decreasing with increasing distance from the fracture. It can thus be postulated that recrystallization caused by hydrothermal fluids might have been responsible for the slight increase in crystal size, the presence of a greater number of saddle dolomites, and the destruction of the ghost structures.

It has been demonstrated that the recrystallization caused by hydrothermal fluids can alter the geochemical signals of dolostones, such as %Ca, trace element substitution [134], values of δ¹³C and δ¹⁸O, as well as REE+Y [75]. In our samples, the recrystallization caused by hydrothermal fluids might have altered the REE+Y signals. In comparison to other replacive dolomites (RD1-RD4), the Eu/Eu* values of RD5 are more positive (>1, Table 3), and the REE+Y distribution patterns are less similar to the seawater pattern [109] (reduced Ce negative anomaly and Y positive anomaly, Figure 13). In RD2, it is possible that some samples might have undergone recrystallization by hydrothermal dolomitizing fluids, as evidenced by the positive Eu/Eu* values (Table 3) and the REE+Y distribution patterns (Figure 13).

In many samples, the presence of evident hydrothermal characteristics (e.g., saddle dolomite, Eu positive anomaly) was not observed. However, the possibility exists that recrystallization may have influenced these samples. Recrystallization has been demonstrated to occur subsequent to the rapid phase of dolomite replacement [135], and can occur extensively in young dolomites, which were never deeply buried, due to warm, slightly evaporated seawater [136]. It has been demonstrated that recrystallization can increase crystal size [137,138], alter geochemical signals [139,140], and enhance the cation ordering [141,142]. Wan et al. documented that the cation ordering of dolomites in the Paleo-oil reservoir zone increases with the crystal size (very finely-0.57, finely-0.67, medium to coarsely-0.87) [46]. This relationship might be attributed to the recrystallization of dolomites. In many medium to coarsely crystalline dolomites, ghost structures are still visible. This might be attributed to the fact that ghost structures are not susceptible to recrystallization. Although recrystallization caused by hydrothermal fluids was observed to have demolish the ghosts, only in part II (which is likely to have suffered the most significant recrystallization in comparison to III and IV) the ghosts are observed to be essentially entirely destroyed (Figure 6a–e). The ghost structures observed in the context of hydrothermal dolomitization [121,126,143] might also suggest that these structures can withstand elevated temperatures and certain degrees of alteration.

5.4.4. Paragenetic Sequence

According to the results, the paragenetic sequence is established (Figure 17).

Figure 17. The paragenetic sequence (not to scale). Every phase is separated into high-energy sediments (left part) and low-energy sediments (right part). The boundary of the shallow and intermediate burial is 600~1000 m, and the boundary of the intermediate and deep burial is 2000~3000 m [144]. Saddle dolomites can also form in the shallow (not near-surface), intermediate to deep burial settings [48,75,123,145]. (1) a–e: There was no early dolomitization. Technically, saddle dolomites formed in a later diagenetic stage would occur along the fracture (Section 5.4.2, hypothesis 1), but this phenomenon was not observed in our samples. (2) a–c,f–i: Partial dolomitization. Saddle dolomites occur in moldic and interparticle pores as cement (through both recrystallization and precipitation, Section 5.4.2 hypothesis 3). (2*) a–c,f,j,k: Partial dolomitization. Technically, saddle dolomites can be formed through recrystallization in a later diagenetic stage (Section 5.4.2, hypothesis 2). (3) a–c,l–n: Pervasive dolomitization. The extent of recrystallization increases. Saddle dolomites are formed through precipitation and recrystallization. The fractures have control on the formation of saddle dolomite and recrystallization. It also shows the difference in porosity. (3*) a–c,l,n: Pervasive dolomitization. Saddle dolomites are only formed in a later diagenetic stage.

Figure 17. The paragenetic sequence (not to scale). Every phase is separated into high-energy sediments (left part) and low-energy sediments (right part). The boundary of the shallow and intermediate burial is 600~1000 m, and the boundary of the intermediate and deep burial is 2000~3000 m [144]. Saddle dolomites can also form in the shallow (not near-surface), intermediate to deep burial settings [48,75,123,145]. (1) a–e: There was no early dolomitization. Technically, saddle dolomites formed in a later diagenetic stage would occur along the fracture (Section 5.4.2, hypothesis 1), but this phenomenon was not observed in our samples. (2) a–c,f–i: Partial dolomitization. Saddle dolomites occur in moldic and interparticle pores as cement (through both recrystallization and precipitation, Section 5.4.2 hypothesis 3). (2*) a–c,f,j,k: Partial dolomitization. Technically, saddle dolomites can be formed through recrystallization in a later diagenetic stage (Section 5.4.2, hypothesis 2). (3) a–c,l–n: Pervasive dolomitization. The extent of recrystallization increases. Saddle dolomites are formed through precipitation and recrystallization. The fractures have control on the formation of saddle dolomite and recrystallization. It also shows the difference in porosity. (3*) a–c,l,n: Pervasive dolomitization. Saddle dolomites are only formed in a later diagenetic stage.

5.4.5. Dolomitization Hypothesis

The process of dolomitization remains a complex phenomenon that has yet to be fully understood. Furthermore, the paucity of research in this region gives rise to many uncertainties. Further research is required to establish a more detailed multi-generational dolomitization model and early saddle dolomite forming mechanism in this region. In consideration of the present findings, we could put forth the dolomitization hypothesis for the Long’eni-Geluguanna Area (Figure 18).

5.5. Implication for Dolostone Reservoirs Exploration

Dolomitization (replacement) is considered to be a porosity-forming process [146,147,148]. Additionally, it has been demonstrated that the penecontemporaneous dolomitization process can inherit primary porosity and maintain it during the burial realm [149], which makes this a beneficial process for the formation of high-quality oil and gas reservoirs. In this instance, dolostones originating from highly porous and permeable sediments and/or limestones, such as ooidal sediments and/or ooidal grainstones, are more likely to exhibit superior reservoir properties than those that have their precursors in fine-grained sediments and/or limestone, such as lime mud and/or mudstones. Furthermore, larger grains, such as ooids and bioclasts, may exhibit a greater prevalence of undolomitized parts at the core of the grains, whereas smaller grains may have undergone complete dolomitization. These undolomitized parts would typically be dissolved [28,150,151], thereby creating porosity (Figure 9 and Figure 17).

This phenomenon facilitates the prediction and exploration of high-quality dolostone reservoirs. Indeed, in sand shoal settings, in which high porosity sediments and/or limestones are formed, numerous high-quality dolostone reservoirs have already been identified and exploited. Notable examples include the Permian Qixia-Maokou Formation [152] in the Sichuan Basin; the Ordovician Penglaibai Formation [153] in the Tarim Basin; and the Central Basin Platform of the Permian Basin [154].

Nevertheless, the presence of fine-grained sediments with less primary porosity in protected areas of shoals, evidenced by the microfacies analysis, may enhance the heterogeneity of the high-quality sand shoal dolostone reservoir properties, which should be considered in the prediction of the high-quality sand shoal dolostone reservoirs.

6. Conclusions

The foraminifera assemblage indicates that the age of the studied strata from the Paleo-oil reservoir zone is Early Jurassic, most probable Sinemurian-Pliensbachian. The microfacies analysis revealed the presence of a sand shoal setting with patch reefs in the Long’eni-Geluguanna Area, as well as back-shoal and protected areas within the shoal.

The dolomitization processes in the Long’eni-Geluguanna Area started at a very early diagenetic stage close to the former seafloor. The near-normal seawater formed periodically at the topmost parts within the shoal might have served as the dolomitizing fluids for the massive dolostones, as evidenced by the following: (1) the dolomitization processes were more prevalent at the top of the shoal, where the water depth was relatively shallow, as opposed to deeper settings such as the back-shoal and shoal fringe; (2) a potential evaporation-favorable climate context, along with the documented presence of very finely, mimic dolomites; (3) δ¹³C values aligning with those of contemporaneous seawater, seawater-like REE+Y patterns, and the similarity of Na and Sr values to those observed in other near-normal seawater dolomites.

The presence of saddle dolomites within the interparticle and moldic pores of limestone might be indicative of hydrothermal activity. It is possible that hydrothermal fluids altered the pre-existing dolomites, resulting in a number of observed petrographic and geochemical changes. These include a slight increase in crystal size, the demolition of ghost structures, the formation of saddle dolomites, and alterations to the REE+Y patterns. The extent of recrystallization is likely controlled by the proximity to the fluid-providing fracture. The regions in close proximity to the fracture exhibit a higher degree of recrystallization. In industrial prediction and exploration of high-quality dolostone reservoirs in sand shoal settings, it is important to take into consideration the fact that especially sediments deposited in protected areas might increase the heterogeneity of dolostone reservoir properties.