Characteristics of Fractured Lacustrine Carbonate Reservoirs in the Zhongshi Area, Jianghan Basin, China

Abstract

The fractured lacustrine carbonate oil reservoir in the Lower submember of Member 4 (Qian-4) of the Qianjiang Formation in the Zhongshi area, Jianghan Basin, represents an important target for hydrocarbon exploration and exhibits substantial exploration and development potential. To clarify the mechanisms by which fractures control reservoir effectiveness, this study integrates core description, thin-section petrography, petrophysical measurements, and geophysical interpretation to systematically characterize matrix properties and fracture development. Results show that the reservoir matrix is dominated by micritic carbonate rocks and grain-dominated carbonate rocks, and overall exhibits low-porosity and ultra-low-permeability characteristics, with an average porosity of 5.19% and permeability generally below 5 mD. Fractures are well developed within the matrix, mainly comprising non-tectonic bedding-parallel fractures and tectonic high-angle fractures. Fracture-related porosity averages 8.42%, and permeability can reach 10–100 mD or higher. The fracture attributes and their spatial distribution are the key controls on hydrocarbon enrichment and deliverability; the occurrence of different fracture types across lithologies and sublayers can significantly enhance reservoir flow capacity. Moreover, natural-fracture characteristics provide critical geological constraints for hydraulic fracturing design and implementation. These findings offer a theoretical basis for fine-scale exploration and development of fractured lacustrine carbonate reservoirs.

1. Introduction

Recent exploration of lacustrine carbonate rocks in the Jianghan Basin has achieved notable breakthroughs, mainly concentrated in the Tankou and Zhongshi areas of the Qianjiang Sag. These advances have been driven by updated geological understanding, innovations in hydrocarbon accumulation models, and integrated progress in engineering technologies, collectively providing new directions for reserve growth and production enhancement in mature fields [1]. Lacustrine carbonates refer to carbonate sediments deposited in inland lacustrine basins. Depending on water salinity, they can be classified into three major types: freshwater-lake basin, brackish-to-saline lake basin, and salt-lake basin types [2]. As a distinctive class of sedimentary rocks formed in inland lacustrine settings, lacustrine carbonates constitute an important domain of global hydrocarbon exploration and development. The uniqueness of their depositional environments, together with the complexity of diagenetic evolution, jointly controls the strong heterogeneity of their storage space (pore-fracture systems). These rocks have two key implications. On the one hand, they record the evolutionary trajectory of paleolake salinity—from freshwater to saline and ultimately to salt-lake conditions. On the other hand, they commonly exhibit both source-rock and reservoir attributes, enabling “self-generation and self-storage” of hydrocarbons [3]. Worldwide, numerous basins have demonstrated the exploration value of lacustrine carbonate plays, such as the Congo Rift Basin in Africa, the Campos Basin in Brazil, and the Uinta Basin in the United States [4,5,6,7]. Recent integrated geophysical studies continue to highlight the critical role of fracture and porosity characterization in carbonate reservoir assessment [8]. In China, discoveries in the Huanghua Depression, Jiyang Depression, the Bohai Bay Basin, and the Subei Basin, among others, have elevated lacustrine carbonate reservoirs to a prominent strategic position within domestic hydrocarbon resources [9,10,11,12,13,14]. Exploration practices and research outcomes from these representative basins provide important references for better understanding of the sedimentary architecture and reservoir characteristics of lacustrine carbonates. Industrial-scale production breakthroughs have been achieved, making lacustrine carbonates a frontier direction in international petroleum exploration [15,16,17]. In China, lacustrine carbonate hydrocarbon resources are estimated at 4.5 × 10⁹ t, yet the average proved ratio is only about 11%. This contrast indicates that the exploration potential of lacustrine carbonates remains far from fully realized and urgently requires further development [18,19,20,21,22].

Compared with the representative basins discussed above, the Zhongshi area of the Jianghan Basin investigated in this study is a salt-lake basin with distinctive tectonic and depositional conditions which result in stronger reservoir heterogeneity and more complex fracture development. Previous studies indicate that the spatial distribution of lacustrine carbonates is jointly controlled by three factors: tectonic setting, paleoclimatic conditions, and provenance supply [18]. However, key issues—such as the fracture development patterns under salt-lake settings and the lithofacies-stress coupled mechanisms controlling reservoir quality—remain insufficiently studied, leaving notable research gaps [17]. The discovery of a fractured lacustrine carbonate oil reservoir in the Lower submember of Member 4 (Qian-4) in the Zhongshi area therefore provides an important avenue for clarifying the specific characteristics of such reservoirs in salt-lake environments.

However, key issues—such as the quantitative fracture development patterns under salt-lake settings and the lithofacies-stress coupled mechanisms controlling reservoir quality—remain insufficiently studied, leaving notable research gaps [16]. Beyond existing descriptive works, this study provides fundamental advancements in three key aspects: (1) It introduces a quantitative framework for fracture characterization, linking core-scale fracture attributes (density, aperture, orientation) directly to reservoir permeability enhancement and oil-bearing potential. (2) It establishes and validates novel petrophysical relationships (e.g., between fracture density and AC/PEF logs) specifically for saline-lacustrine carbonates, offering a more reliable well-log-based fracture identification method for similar settings. (3) For the first time in this basin, it develops and tests a seismic-attribute-driven predictive model for fracture distribution at the sublayer scale, calibrated with core and image-log data, which significantly enhances the ability to delineate high-productivity belts beyond well control. The discovery of a fractured lacustrine carbonate oil reservoir in the Lower submember of Member 4 (Qian-4) in the Zhongshi area therefore provides an important case for applying this integrated quantitative approach to clarify the specific characteristics and controls of such reservoirs in salt-lake environments.

2. Geological Setting of the Study Area

The Zhongshi area is located in the central-western part of the Qianbei fault belt in the Qianjiang Sag. It is bordered by the Xijiakou monocline to the west, the Tankou uplift to the east, the Banghu syncline to the south, and is bound to the north by the Qianbei Fault. The exploration area covers approximately 80 km². Previous studies have extensively investigated the structural attributes and evolution of the Qianbei fault system and the Zhongshi fault nose (Figure 1), suggesting that the Qianbei fault system is a fault-step zone composed of a series of synthetic faults, and that the Zhongshi fault nose developed by the end of deposition of Member 2 of the Qianjiang Formation and has undergone continuous inherited uplift from Member 1 to the present [18].

Drilling data reveal that, from bottom to top, the Zhongshi area is mainly underlain by the Lower Paleogene Qianjiang Formation and Jinghezhen Formation, the Neogene Guanghuasi Formation, and the Quaternary Pingyuan Formation. The Qianjiang Formation is subdivided (from top to bottom) into Member 1, Member 2, Member 3, Member 4, and the Lower Submember of Member 4, and is lithologically complex, comprising carbonates, gypsiferous mudstones, clastic rocks, and mudstones (Figure 2). The principal oil-bearing carbonate interval within the Qianjiang Formation occurs in the Lower Submember of Member 4. The I oil unit of the Lower Qian-4 submember is further divided into six sublayers, whereas the II oil unit is divided into three sublayers (

Table 1. Simplified stratigraphic column of the study area.

System	Series	Formation	Member	Oil Layer Group
Quaternary	Pleistocene	Pingyuan Formation
Neogene	Middle Miocene	Guanghuasi Formation
Paleogene	Eogene	Qianhe Town Formation
		Qianjiang Formation	First Member	Qian11, Qian12, Qian13
	Upper Eocene		Second Member	Qian21, Qian22, Qian23
			Third Member	Qian31, Qian32, Qian33, Qian34
			Upper Fourth Member	Qian40, Qian41, Qian42, Qian43
			Lower Fourth Member (Qian-4)	I unit: I-1, I-2, I-3, I-4, I-5, I-6, II unit: II-1, II-2, II-3, II-4
	Middle Eocene	Jinsha Formation

). In this study, the target interval is Lower Qian-4 submember, specifically I-2, I-3, I-4, I-5, I-6, II-1, and II-2 sublayers.

The Qianjiang Formation in the Qianjiang Sag as a whole developed a typical delta-salt-lake depositional system. Under a north-derived, unidirectional provenance supply, a series of depositional facies belts were formed from the basin margin toward the sag (depression) belt, transitioning successively from sandstone–mudstone, to carbonate rocks, and then to evaporites (salt rocks). The carbonate facies are distributed in a circumferential, ring-like belt along the margin of the sag and, together with the central evaporite facies and the outer clastic facies, constitutes a complete salt-lake depositional succession. The carbonate facies cover an aerial extent of approximately 500 km² and provide the principal material basis for lacustrine carbonate oil reservoirs in this area.

Development of the fractured lacustrine carbonate reservoirs in the Zhongshi area of the Jianghan Basin underwent a prolonged stage of exploration and evaluation in the early period. In 2022, after stimulation, Well Huang20X-4 produced an oil flow of 26.22 m³/d, representing a major breakthrough in carbonate-reservoir exploration in the salt-lake setting of the Qianjiang Sag and marking the start of a new phase of carbonate development in this area. Subsequently, exploration activities in the Zhongshi area were advanced following the strategy of “expanding Tankou, exploring Zhongshi, and concurrently targeting multiple stratigraphic intervals and play types.” In 2024, Well Z99X, deployed in this area, successfully conducted a production test on a 17.4 m-thick carbonate oil interval in the Qian-4 submember. Under controlled flow through a 4 mm choke, the well achieved a daily oil rate of 137.6 m³/d, with cumulative oil production of 427.56 m³ and a flowback ratio of 6%. This result indicates that carbonate exploration in the Jianghan Basin has expanded beyond the Member 3 (Qian-3) play in the Tankou area and has successfully opened up the Lower Qian-4 interval in the Zhongshi area as a new front for reserve growth. To date, 19 wells in the Jianghan oilfield area have been re-evaluated and/or tested (including both old wells and dedicated exploratory wells) for carbonate targets; 17 wells have obtained commercial oil flow, with cumulative production exceeding 17,000 t, fully demonstrating the substantial development potential of fractured lacustrine carbonates in the Zhongshi area.

3. Materials and Methods

The lacustrine carbonate reservoir of Qian-4 submember in Zhongshi Area is a new breakthrough. This study is based on 266 m of cores from the Well Z99X and the Z9901X, together with petrographic descriptions of 288 epoxy-impregnated thin sections, supplemented by fluorescent thin-section analysis, also with the conventional Well logs, FMI image logs, and seismic interpretation data. The dataset is representative, covering key wells from different structural positions (crest, flank) within the 80 km² study area. Core and image-log data from the central Well Z99X provide a continuous, high-resolution calibration benchmark. We also have incorporated quantitative relationship analysis methods from previous works [24,25,26].

To ensure consistency between core-derived and image-log-derived fracture statistics, a rigorous calibration workflow was applied: (1) Depth-matching core fractures to image-log features. (2) Defining consistent fracture classification criteria (open, filled, bedding) for both datasets. (3) Comparing fracture dip/azimuth from oriented core and image logs. This calibration confirmed a strong correlation (R² > 0.75) for fracture density in intervals with good borehole conditions, validating the use of image logs to extend fracture analysis beyond cored sections.

4. Results

4.1. Reservoir Matrix Composition and Characteristics of Petrophysical Properties and Oil-Bearing Potential

4.1.1. Lithology and Lithofacies

The reservoir lithologies in the Zhongshi area are dominated by carbonate rocks, mainly micritic carbonate rocks and grain-dominated carbonate rocks, with subordinate sandstone and other lithologies. Core observations and thin-section petrography confirm that fractures occur in both carbonate types, but the degree of fracture development differs markedly between them.

Micritic carbonate rocks are widely distributed across the study area. Their rock types include micritic carbonate rock, micritic limestone, micritic dolostone, and transitional lithologies such as micritic dolomitic limestone and micritic argillaceous carbonate rock. Mineralogically, these rocks are dominated by calcite and dolomite, with pronounced variations in their relative abundances. For example, calcite content in micritic limestone samples can reach 50–90%, whereas dolomite content in micritic dolostone typically exceeds 50%. These rocks are mainly composed of micrite with grain sizes < 0.03 mm and are characterized by a dense and relatively homogeneous fabric. Reservoir space is chiefly provided by pores and fractures, while the contents of bioclasts and detrital grains are generally below 10% (Figure 3).

Grain-dominated carbonate rocks also occur in the study area, and the principal rock types include grain-bearing micritic carbonate rock, grain-bearing micritic limestone, and grain-bearing micritic dolostone. Their allochem assemblages are diverse, comprising (i) intraclasts with irregular to elliptical or elongate morphologies, some of which are carbonaceous; (ii) bioclasts derived from multiple fossil groups, including bivalves, ostracods, and algae; and (iii) ooids as well as sand-sized terrigenous grains (0.07–0.54 mm) with moderate sorting and roundness. Intergranular cements are dominated by micrite and sparry calcite, and a variety of cementation styles are observed.

4.1.2. Petrophysical Characteristics

The storage space of the fractured lacustrine carbonate reservoirs in the Qian-4 submember in the Zhongshi area, Jianghan Basin, can be broadly divided into pores and fractures. Among them, fractures constitute the primary storage space and the main pathways for fluid flow, whereas pores play a secondary, supportive role.

Pore space is dominated by matrix porosity, including primary intergranular pores and dissolution pores, which mainly occur in micritic and grain-dominated carbonates. However, their overall proportion is low and their petrophysical quality is poor. Statistics show that the average matrix porosity is only 5.19%; 46.33% of the samples have porosities in the range of 1.0–5.0%, and permeability is generally <5 mD (Figure 4), indicating a typical low-porosity, ultra-low-permeability reservoir with limited storage capacity and weak flow capability.

The matrix pores exhibit no obvious oil staining, and only minor oil indications—appearing as yellowish, dull fluorescence—occur along fractures, suggesting poor oil-bearing potential of the matrix pore system (Figure 5).

4.2. Characteristics of Fracture Development in the Reservoir

Core-based fracture observations and related laboratory tests were carried out for key wells in the study area. Core observations were conducted for two wells, with a cumulative core length of 153.78 m examined and 2837 fractures documented. On this basis, the fracture attributes—including type, attitude (orientation and dip), aperture, length, linear density, and filling characteristics—were systematically investigated.

4.2.1. Fracture Types

Based on core observations and thin-section petrography, combined with drilling geology and well-log interpretation, fractures in the target layers in the study area are diverse. In terms of genesis, fractures can be classified into non-tectonic fractures and tectonic fractures [18]. The study area is dominated by non-tectonic fractures (bedding fractures), with tectonic fractures being subordinate. From the perspective of attitude, tectonic fractures can be further divided into high-angle fractures and low-angle fractures (

Table 2. Fracture description table.

Type		Description	Layer	Core Photo	Fracture Density (m−1)
Non-tectonic fractures (bedding-parallel fractures)	Unfilled	Bedding-parallel, distributed concordantly with stratification on core surfaces, showing morphologies such as bending, discontinuity, branching, etc.	I-4		10.9–14.7
	Filled	Calcite, quartz, argillaceous, or gypsum fillings are locally observed in some bedding fractures. Clear boundaries between filled and unfilled segments can be identified.	I-3		5.8–8.9
	With dissolution vugs	Some bedding fractures are affected by dissolution, forming vugs and pores that are commonly elliptical in shape.	I-3		3.4–5.6
Tectonic fractures	Unfilled	The angle between the fracture plane and the plane perpendicular to the wellbore axis is 45–90°.	I-3		10.9–14.7
	Filled	Calcite, quartz, argillaceous, or gypsum fillings are locally observed in some tectonic fractures.	I-4		5.8–8.9
	With dissolution vugs	Dissolution vugs occur within tectonic fractures in a variety of forms, commonly including circular, elliptical, and irregular honeycomb-like shapes.	I-4		3.4–5.6

).

Fracture development varies markedly among different lithologic assemblages, depositional microfacies, and structural positions. Non-tectonic (bedding) fractures are mainly developed in the I-6 and I-3 sublayers, are moderately developed in the I-4 sublayer, and are relatively weak in the I-3 sublayer. Tectonic fractures are mainly developed in the I-3 sublayer, are secondary in the I-5 and I-6 sublayers, and are relatively weak in the I-3 sublayer. The pronounced differences in fracture intensity indicate that fracture development is strongly controlled by depositional microfacies type, lithologic tightness/compactness, and the magnitude of tectonic stress.

4.2.2. Fracture Dips

Fractures can be classified into four categories based on dip angle: horizontal fractures with dip angles < 5°, low-angle fractures with dip angles > 5° and <45°, and high-angle fractures with dip angles > 45° and <90° [21]. Horizontal fractures are the most prevalent in the study area, with a total of 1874 fractures, accounting for 66.06% of all fractures. High-angle fractures are the second most common, with 759 fractures (26.75%). Low-angle fractures total 204 (7.19%) (

Table 3. Statistical analysis of fracture inclination in core drilling wells.

Well	Fracture Dip Angle
	High Angle		Low Angle		Horizontal
	per	%	per	%	per	%
Z9901X	541	19.06	125	4.41	1133	39.94
Z99X	218	7.68	79	2.78	741	26.12
Total	759	26.75	204	7.19	1874	66.06

, Figure 6).

Fracture length and aperture were measured and statistically analyzed based on core observations. A total of 1687 complete fractures were identified in core, with lengths of approximately 0.10–0.18 m, mainly concentrated in the range of 0.10–0.15 m, which accounts for 60.5% of the total fractures.

In terms of fracture width, fractures can be divided into large fractures, medium-small fractures, and microfractures [22], defined here as length > 1000 mm for large fractures, 100–1000 mm for medium-small fractures, and 10–100 mm for microfractures. Medium-small fractures are the most developed, with 1491 fractures (52.56%), followed by microfractures with 1114 fractures (39.27%). Large fractures are the least common, totaling 232 (8.17%). Overall, tectonic fractures in the study area are dominated by horizontal fractures and medium-small fractures (

Table 4. Statistical analysis of fracture width in core drilling wells.

Well	Fracture Width
	Big (>10 mm)		Small and Medium (1~10 mm)		Tiny (<1 mm)
	per	%	per	%	per	%
Z9901X	137	1.14	893	7.46	769	6.42
Z99X	95	0.79	598	4.99	345	2.88
Total	232	8.17	1491	52.56	1114	39.27

).

4.2.3. Fracture Density

Fracture linear density is a key metric for quantifying the degree of fracture development. Based on core-statistical data from Well Z99X and Z9901X, fractures were classified and counted by type (tectonic fractures vs. non-tectonic fractures) to clarify differences in fracture development among sublayers and between wells.

In Well Z99X, the linear density of non-tectonic fractures is overall higher in the II oil unit than in the I oil unit. The II-2 sublayer exhibits the best development, with a linear density of 14.36 fractures/m. Within the I oil unit, the linear densities of the I-3, I-5, and I-6 sublayers are relatively comparable, at 8.64 fractures/m, 7.68 fractures/m, and 7.78 fractures/m, respectively. In contrast, the linear density of tectonic fractures is overall higher in the I oil unit than in the II oil unit. The I-6 sublayer shows the highest tectonic fracture density (14.89 fractures/m), whereas the II-2 sublayer displays the lowest density (4.31 fractures/m). The tectonic fracture linear densities of the I-3, I-5, and II-1 sublayers are 11.41 fractures/m, 13.36 fractures/m, and 6.49 fractures/m, respectively (Figure 7).

In well Z9901X, the linear density of non-tectonic fractures shows a pronounced high-value zone centered on the I-5 sublayer, reaching 12.91 fractures/m. The I-1 sublayer exhibits the poorest development, with a linear density of only 5.28 fractures/m. The non-tectonic fracture linear densities of the I-1, I-3, I-4, and I-6 sublayers are 5.28 fractures/m, 11.41 fractures/m, 11.83 fractures/m, and 8.87 fractures/m, respectively.

For tectonic fractures, the I-3 sublayer displays the highest linear density, reaching 18.94 fractures/m. The tectonic fracture linear densities of the I-1, I-4, I-5, and I-6 sublayers are 3.78 fractures/m, 12.57 fractures/m, 7.45 fractures/m, and 6.68 fractures/m, respectively (Figure 8).

4.2.4. Quantitative Impact of Fractures on Porosity and Permeability

The fracture system significantly enhances reservoir flow capacity. Laboratory core-porosity and permeability measurements indicate that unfilled or partially filled fractures can yield permeabilities of 10–100 mD or higher, far exceeding matrix permeability (<5 mD). For example, in the Well Z9901X, fractured samples show permeability increases of one to two orders of magnitude relative to adjacent matrix samples. In full-diameter core tests, the permeability of open fractures can exceed 100 mD (Figure 9).

The fracture system profoundly enhances reservoir flow capacity. Laboratory measurements reveal that fracture porosity contributes an average of 8.42% to the total porosity, compared to the matrix average of 5.19%. More significantly, permeability exhibits a strong power-law correlation with fracture linear density measured from core: (R² = 0.78, n = 45). Unfilled or partially filled fractures elevate permeability by 1–2 orders of magnitude. For instance, in Well Z9901X, samples with fracture density > 10 m⁻¹ show permeabilities ranging from 15 to 105 mD, whereas adjacent matrix samples with fracture density < 2 m⁻¹ have permeabilities consistently below 1 mD (Figure 9). Full-diameter core tests confirm that interconnected open fractures can provide effective permeabilities exceeding 100 mD. The flow behavior is likely anisotropic. The dominant NNE-SSW and NNW-SSE fracture orientations (Figure 13) create preferred permeability pathways along these directions. Fracture aperture, more than length in this system, controls the flow capacity of individual conduits. The coexistence of widely opened (aperture > 100 μm) tectonic fractures and narrower, more numerous bedding fractures results in a dual-permeability system where the former dominates bulk flow directionality and the latter enhance connectivity.

4.2.5. Quantitative Relationship Between Fractures and Hydrocarbon Occurrence

The fracture density shows a pronounced positive correlation with oil-bearing potential. High-fracture-density belts commonly coincide with favorable zones of hydrocarbon accumulation (oil-bearing intervals are marked in red): oil-bearing sections generally contain dense fractures, with linear densities reaching up to 17.9 fractures/m (see Figure 10).

Comparative analysis indicates that oil-bearing intervals are characterized by dominance of non-tectonic fractures and subordinate, locally conductive tectonic fractures. These intervals not only exhibit higher fracture densities, but also the largest fluctuations in rock hardness, with values up to 714 HV. This suggests degradation of rock fabric integrity. The high hardness values (up to 714 HV) indicate a highly brittle mechanical behavior of the micritic and grain-dominated carbonates. Such high brittleness is a key prerequisite for the development of dense fracture networks under tectonic stress. The large fluctuations in hardness within oil-bearing intervals likely represent strong mechanical contrasts between different lithological laminae or between rock matrix and early diagenetic cements. These contrasts promote stress concentration and heterogeneous failure, leading to the complex fracture networks that facilitate fluid migration and storage.

A robust positive correlation exists between fracture density and hydrocarbon enrichment. Quantitative analysis shows that normalized oil fluorescence intensity (from thin sections) and tested oil productivity (m³/d) both increase linearly with fracture density in the pay zones (R² = 0.71 and 0.65, respectively). High-productivity intervals (>50 m³/d)) are exclusively associated with fracture densities exceeding 12 fractures/m (Figure 10). Furthermore, these intervals exhibit the highest contrasts in rock hardness (up to 714 HV), indicating brittle behaviour favorable for fracturing. The quantitative relationship can be expressed as a risk factor for hydrocarbon occurrence: intervals with greater fracture density (>10 fractures/m) and high hardness variability (Delta HV > 200) have a >80% probability of being productive, based on our dataset.

5. Discussion

5.1. Characterization of the Fractures in Reservoir

5.1.1. Relative Timing of Fracture Development

The relative timing of fracture formation can be inferred from cross-cutting relationships and cement fill characteristics observed in cores and thin sections. The predominant non-tectonic (bedding-parallel) fractures are often the earliest formed, frequently associated with syn-sedimentary or early diagenetic processes. These fractures are commonly partially or fully filled with micrite or early-stage calcite cements (Figure 3C,D). Later tectonic fractures, primarily high-angle fractures, cross-cut these early bedding fractures and associated cements. Two major sets of tectonic fractures are identified: an earlier set (NNW-SSE trend) and a later set (NNE-SSW trend), with the latter occasionally offsetting the former. The latest-stage fractures, often with minimal fill or lined with bitumen, show clear associations with dissolution events. Dissolution pores and vugs are frequently concentrated along fracture walls or at fracture intersections (e.g., Figure 3B,F), indicating that fracturing, particularly the late-stage tectonic events, created preferential pathways for corrosive fluids, significantly enhancing secondary porosity. This paragenetic sequence indicates that the fracture system evolved through multiple phases: initial compaction-related bedding fracture formation, followed by tectonic fracturing associated with the Qianbei Fault activity, and subsequent diagenetic modification involving cementation and dissolution.

5.1.2. Controlling Factors of Fracture Development

Lithology is the fundamental intrinsic factor governing fracture development. Rocks with higher hardness and brittleness (e.g., grain-dominated carbonates with HV often >600) are more prone to brittle failure under stress, leading to higher fracture density, especially of tectonic fractures. Strata with higher clay contents tend to inhibit fracture formation and exhibit lower fracture densities. Combined with the systematic analysis of fractures and rock mechanical properties presented above, fracture development in the study area is evidently lithology-controlled.

Non-tectonic fractures mainly occur in micritic carbonate rocks, with an average density of approximately 9.09 fractures/m, higher than that of grain-dominated carbonates (8.36 fractures/m). They are characterized by the coexistence of dense fracture networks and dissolution-related microporosity. In contrast, tectonic fractures are concentrated in grain-dominated carbonate rocks, with an average density of about 15.05 fractures/m, significantly higher than that in micritic intervals (8.63 fractures/m). These fractures commonly extend along bedding planes or grain-contact surfaces. The two fracture sets show a complementary vertical distribution and together constitute the primary framework of the fracture system in this interval (Figure 11). Overall, grain-dominated carbonates preferentially develop high-density tectonic fractures. The orientation of these fractures (NNW-SSE and NNE-SSW sets) correlates strongly with the regional stress field evolution of the Jianghan Basin. The NNW-SSE set likely relates to an earlier NE-SW extensional phase during the Paleogene, while the dominant NNE-SSW set aligns with the later, long-lasting N-S to NW-SE extensional regime associated with the Qianbei Fault system, indicating a strong tectonic control.

5.2. Fracture Interpretation from Well Logs and Borehole Image Analysis

In this study, borehole image logs were used to identify and interpret fractures and fracture types, and attitude parameters (e.g., strike and dip) were extracted for comparison and validation against core-based observations. Fractures are generally well developed in the target interval. Image log interpretation indicates that high-resistivity fractures are dominant, with microfault features locally observed. Bedding interfaces are widely recognizable throughout the logged interval, and the intensity of high-resistivity fractures varies markedly along the borehole, indicating pronounced vertical heterogeneity in fracture development (Figure 12).

Statistical analysis of fracture attitudes shows that the strike of high-resistivity fractures is relatively clustered, with two principal sets trending NNW-SSE and NNE-SSW (Figure 13). Dips are mainly moderate to high, suggesting a strongly directional fracture system (Figure 14). Vertically, intervals with microfault development correspond to zones with dense high-resistivity fractures to some extent, implying that both may be governed by a similar tectonic-stress regime. Comparison using the integrated log column indicates that bedding is overall well developed, whereas high-resistivity fractures exhibit a segmented pattern of enhanced development. This suggests that, in addition to regional stress, fracture formation and distribution may be jointly influenced by lithologic assemblages, interbedded interfaces, and variations in local structural deformation. These observations provide borehole-scale constraints for subsequent analyses of fracture–oil–bearing relationships and for optimizing seismic/log-based parameters for fracture prediction. The strong AC/PEF correlations are likely specific to the mineralogical composition and pore structure of these saline-lacustrine carbonates, where fractures significantly alter sonic transit time and photoelectric absorption. While the exact coefficients may vary, the underlying physical principle suggests that similar relationships could be sought in other lacustrine carbonate reservoirs with low primary porosity and contrasting matrix/fracture properties.

5.3. Fracture Distribution Prediction from Seismic Attribute Analysis and Sublayer Characteristics

At the seismic scale, fracture-sensitive attributes (coherence, curvature, ant-tracking) were used to predict fracture density and generate distribution maps for each sublayer (Figure 15). Validation via blind-well tests confirmed model robustness. The predicted patterns exhibit systematic variation controlled by lithology and structure (

Table 5. Summary of predicted fracture characteristics by sublayer.

Sublayer	Dominant Lithology	Fracture Pattern	Dominant Type	Density Range (m−1)	Main Control
I-2	micritic carbonate	network-like	non-tectonic	<8	depositional fabric
I-3	grain-dominated	continuous belt	tectonic (high-angle)	15–18.9	fault-proximal stress
I-4	micritic (minor grain-rich)	sparse network	non-tectonic	5.8–8.9	dissolution associated
I-5	micritic carbonate	scattered clusters	non-tectonic	5–12.9	shallow-water microfacies
I-6	grain-dominated	continuous belt	tectonic (high-angle)	14.9–16	fault-derived stress field
II-1	micritic & salt-gypsum	discontinuous belt	tectonic (segmented)	6.5–9.4	interbedded interfaces

). In summary, the dense belt is aligned with the derived stress field orientation associated with the Qianbei Fault. The strong spatial correlation between high-density tectonic fracture belts and the fault-derived stress concentration zones (e.g., in I-3 and I-6) underscores a robust, predictable association between specific tectonic phases (later extension) and fracture development in competent lithologies., making them the most favorable exploration targets. The micrite-dominated sublayers (I-2, I-4, I-5, II-1) are characterized by lower-density and network- or patch-like distributions of non-tectonic fractures, with local “sweet spots” associated with paleo-topographic highs or grain-rich interbeds.

Uncertainty and Limitations of the Upscaling Method: The primary uncertainty in upscaling from core/log to seismic scale stems from the non-unique relationship between seismic attributes and fracture intensity. We limited this by (1) using multi-attribute analysis (coherence, curvature, ant) to reduce noise, and (2) calibrating the seismic attribute blends directly to the well-based fracture density using a neural network approach at well locations. The main limitations are as follows: (a) The seismic resolution (~20 m) prevents detection of fractures below a certain size and spacing, causing an underestimation of total fracture density in intensely fractured zones. (b) The method assumes a consistent lithological framework; abrupt lateral facies changes not captured by the seismic inversion could lead to prediction errors. (c) The predictive power decreases with distance from calibration wells, as indicated by the increasing uncertainty in the blind-well tests for the most distal areas.

6. Conclusions

(1)

The reservoirs in the study area constitute a typical fracture-pore dual-porosity/dual-permeability system. The reservoir matrix is characterized by low porosity and ultra-low permeability, with an average porosity of 5.19% and permeability generally <5 mD. In contrast, an effective fracture network markedly enhances flow capacity: the equivalent permeability of unfilled or partially filled fractures can increase by 1–2 orders of magnitude, locally reaching > 10–100 mD. Fractures are therefore the primary factor controlling seepage efficiency and production variability.

(2)

The dominant fracture types are non-tectonic (bedding-parallel) fractures and high-angle tectonic fractures, whose development is jointly governed by lithology, tectonic setting, and diagenetic filling. Micrite-dominated intervals are characterized by densely distributed bedding fractures, whereas grain-dominated carbonate intervals preferentially develop high-density tectonic fractures. Fracture filling intensity increases markedly along interlayer interfaces and within structural core zones, directly influencing fracture conductivity and spatial connectivity.

(3)

Fracture development exhibits pronounced inter-sublayer differences and plan-view zonation. The grain-dominated I-3 and I-6 sublayers form NE-trending, continuous belt-like high-density fracture zones, consistent with the orientation of the derived stress field associated with the Qianbei Fault. In contrast, the micrite-dominated I-2, I-4, I-5, and II-1 sublayers are dominated by network- or patch-like non-tectonic fractures with lower overall densities, although localized “sweet spots” with reservoir potential still occur.

(4)

Geophysical responses indicate strong monotonic relationships between fracture density and AC and PEF logs… providing a basis for subsequent integrated well-seismic modeling. These relationships, while particularly robust in this formation, highlight the potential of using integrated log analysis for fracture detection in analogous low-porosity carbonate settings.

(5)

A fracture-prediction model based on seismic attributes is effective at the sublayer scale. High-fracture-density belts are mainly distributed along zones of positive-curvature anomalies and in areas proximal to faults. Blind-well tests and offset-well comparisons validate the reliability of the prediction model, showing a mean absolute error of ~2.1 fractures/m within a 500 m radius of calibration wells. This provides a confidence level sufficient for guiding well placement in high-priority areas, though with increasing uncertainty in undrilled, structurally complex zones.

(6)

This study advances beyond traditional descriptive models by establishing a quantitative, multi-scale predictive framework. It quantitatively links fracture attributes to permeability enhancement and productivity, provides calibrated well-log proxies (AC, PEF) for fracture density specific to saline-lacustrine carbonates, and delivers a seismically derived predictive model validated at the sublayer scale. In future exploration and development, priority should be given to belt zones such as I-3 and I-6, where fractures are more continuous and fracture filling is low. Localized sweet spots within micrite-dominated intervals that show relatively high fracture density and low-to-moderate filling should also be targeted.