Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2026 Jul 17;11(29):43385–43396. doi: 10.1021/acsomega.6c01024

Coalbed Methane Reservoir Characterization and Geological Modeling: A Case Study of Block A in the Shenfu Gas Field, Eastern Margin of the Ordos Basin

Di Yang †,*, Yongjing Tian ‡, Maojun Fang †, Yuhu Bai †, Yu Qi †, Fen Liu †, Ruyong Feng †, Hao Lu †
PMCID: PMC13425314  PMID: 42540083

Abstract

In recent years, with the growth of global demand for oil and gas, coalbed methane, as an important unconventional resource, has become a research hotspot in the field of exploration and development. However, coal reservoirs typically exhibit strong heterogeneity, with significant variations in microscopic components, pore characteristics, and coal body structures, making it challenging to identify the distribution of favorable reservoirs. To address this issue, this study proposes a novel approach to coal reservoir characterization and modeling based on coal macrolithotypes. The method uses core observation and experimental analysis to determine the geological characteristics of favorable macrolithotypes, thereby guiding the construction of the coal reservoir geological models. It not only delineates the spatial heterogeneity of coal reservoirs but also provides an effective framework for predicting favorable coal reservoirs. The method is successfully applied to block A of the Shenfu gas field, located on the eastern margin of the Ordos Basin. Results indicate that bright coal and semibright coal are favorable coal reservoirs in this area, characterized by superior gas generation and storage capacities. They are primarily distributed in the central part of the study area horizontally and concentrated in the upper middle part of the coal seam vertically, forming the key target interval for development. Finally, the reliability of the proposed approach is verified through comparison to production performance from actual wells. Therefore, this method provides a robust basis for predicting favorable coal reservoirs and offers a valuable guide for improving the coalbed methane development efficiency and achieving increased production.

graphic file with name ao6c01024_0020.jpg

graphic file with name ao6c01024_0018.jpg

1. Introduction

With the rapid development of the global economy, energy demand has surged dramatically. Coalbed methane (CBM), as an important unconventional resource, has received extensive attention worldwide in recent years. − However, traditional research often focuses on the hydrocarbon source rock properties of coal seams while ignoring their reservoir rock attributes and inherent heterogeneous characteristics, posing a significant challenge to the prediction of favorable coal reservoirs. − The heterogeneity of coal seams is primarily characterized by two aspects: first, different coal lithotypes exhibit significant variations in resource characteristics such as material composition, adsorption capacity, and reservoir properties. − Second, the susceptibility of different coal intervals to reservoir stimulation varies considerably, strongly controlling fracture propagation and connectivity during hydraulic fracturing and thereby limiting the spatial distribution of induced fractures. , These heterogeneity-driven factors result in significant production variability between wells, making the characterization and modeling of coal reservoirs essential to delineate the distribution of favorable coal reservoirs, thereby providing a solid foundation for optimizing development plans.

Coal macrolithotypes represent a classification system widely used in the CBM reservoir evaluation, coal mining, and coal geology. − It can be categorized into four types according to the total proportion of vitrain and clarain: bright coal (>80%), semibright coal (50%–80%), semidull coal (20%–50%), and dull coal (<20%). These distinct coal macrolithotypes exhibit significant differences in hydrocarbon-generation capacity, adsorption properties, reservoir potential, and logging responses. − In recent years, scholars have conducted studies on coal macrolithotypes from different perspectives: Roslin and Esterle (2016) proposed a method using cluster analysis of logging data for the automatic identification of different coal macrolithotypes; Zhao et al. (2018) comprehensively characterized the fracture and pore features of coal macrolithotypes and established a structural model based on these observations; Liu et al. (2019) investigated the impact of coal macrolithotypes on CBM production and numerically simulated fracturing patterns for different coal macrolithotypes; Guo et al. (2022) introduced an index to quantitatively analyze logging curve response characteristics of various coal macrolithotypes; Li et al. (2023) summarized differences in pore structure among various coal macrolithotypes, providing new insights for CBM reservoir characterization; Zhang et al. (2024) compared pore characteristics and CBM diffusion-seepage behavior across different coal macrolithotypes, while Hou et al. (2024) further explored the relationship between coal porosity parameters and macrolithotypes and assessed their influence on CBM production. Collectively, these studies demonstrate that significant differences exist among coal macrolithotypes, which contribute to the heterogeneity of coal reservoirs. Therefore, accurate characterization of coal macrolithotypes is essential for understanding reservoir heterogeneity and guiding predictions of favorable coal reservoir distribution.

This study presents a novel approach to coal reservoir characterization and modeling based on coal macrolithotypes. The method systematically summarizes the geological characteristics of different coal macrolithotypes and guides the construction of a three-dimensional geological model, which effectively characterizes the reservoir heterogeneity and reveals the spatial distribution of favorable coal reservoirs. The results are validated by using production data from actual wells in Block A of the Shenfu Gas Field in the Ordos Basin, demonstrating that the proposed method has significant practical value for guiding CBM development and production.

2. Geological Background

The Ordos Basin, located in northwestern China (Figure a), is a multicycle cratonic basin that hosts abundant oil and natural gas resources. , In the eastern basin, the main coal-bearing strata are the Carboniferous-Permian Benxi, Taiyuan, and Shanxi formations. The coal seams are laterally continuous, with highly variable burial depths forming multiple overlapping coal-bearing intervals. , The exploration and development of CBM in the basin began in the 1990s and has since progressed through several stages, including breakthroughs in typical target zones, large-scale development of shallow to middle coal seams, and, more recently, large-scale development of deep coal seams, with continuous and significant advances achieved at each stage. ,

1.

1

Open in a new tab

Geological structure distribution map of the study area. (a) Location map of the Ordos Basin. (b) Location map of Block A in the Shenfu gas field.

The Shenfu gas field is situated on the northeastern margin of the Ordos Basin, in the middle-northern section of the Jinxi Flexural Fold Belt (Figure b), providing highly favorable geological conditions for the occurrence and accumulation of CBM. The study area, Block A, is located in the southern part of the Shenfu gas field (Figure b). It comprises the Ordovician Majiagou Formation, Carboniferous Benxi Formation, Permian Taiyuan Formation, and Shanxi Formation from bottom to top. Based on sedimentary cyclicity and lithologic associations, the Benxi Formation can be subdivided into the upper Benxi (Ben1) Member and the lower Benxi (Ben2) Member. The target layer is No. 8 + 9 coal seam in Ben1Member (Figure ), which is formed by the merging of the No. 8 and No. 9 coal seams within the study area. This coal reservoir is mainly composed of medium-rank coal with a primary structure. The vitrinite reflectance of the coal ranges from 0.80% to 1.46%, showing a spatial pattern of decreasing values from south to north and from west to east. , Before the formation of this coal seam, the area was dominated by extensive barrier-coast deposits. Following the abandonment of this system, widespread peat swamps developed, providing favorable conditions for coal accumulation. ,

2.

2

Open in a new tab

Stratigraphic column of the Shenfu gas field, Ordos Basin.

3. Research Data and Methods

The research is supported by a comprehensive data foundation. All the wells have encountered the target formationthe No. 8 + 9 coal seam of Ben1 Formationwith complete well-logging data sets and experimental analyses. The seismic data are of high quality and provide full coverage of the study area. These multisource data sets provide a solid support and reliable basis for coal reservoir characterization and geological modeling.

This research mainly consists of three parts: (1) the geological characteristics of different coal macrolithotypes are summarized, and well-logging identification criteria are established, clearly identifying bright coal and semibright coal as the favorable coal reservoir in the study area. (2) Quantitative analysis of coal macrolithotypes logging: on the basis of depth correction, the coal macrolithotypes observed in the core are calibrated with a variety of logging curves to analyze the logging response characteristics of different coal macrolithotypes, to establish a quantitative criterion for identifying various types of coal macrolithotypes, and then classify the coal macrolithotypes of the single wells in the study area accordingly. (3) Geological modeling of coal reservoirs based on coal macrolithotypes: under the guidance of geological analysis, combined with the classification of coal macrolithotypes in single wells, a three-dimensional coal macrolithotypes model and its constrained gas content model are established to provide support for favorable coal reservoir prediction.

4. Results

4.1. Geological Characterization of Coal Macrolithotypes

Core observations reveal that four types of coal macrolithotypes are identified in the study area, namely, bright coal, semibright coal, semidull coal, and dull coal (Figure ). The bright coal and semibright coal have strong luster and light texture, with well-developed face cleats that typically form continuous linear or network patterns, indicating good lateral connectivity. In contrast, semidull coal and dull coal have weaker luster and less developed cleats. Furthermore, in the NMR T 2 spectra (Figure ), bright coal and semibright coal display a triple-peak pore structure, dominated by medium pores, large pores, and microfractures, with good pore connectivity and a well-developed pore-fracture system. On the other hand, semidull coal and dull coal exhibit a bimodal pore pattern dominated by smaller pores, with relatively underdeveloped large pores and fractures. The connectivity between small pores and medium pores is also poor, resulting in limited overall flow capacity. Scanning electron microscopy (SEM) observations (Figure ) show that bright coal and semibright coal are predominantly composed of vitrinite. In contrast, semidull coal and dull coal are mainly composed of inertinite and mineral matter, exhibiting smaller pore diameters, with some pores partially infilled by clay minerals. In summary, bright coal and semibright coal possess a more developed pore-fracture network and significantly higher effective porosity compared to semidull and dull coals, giving them greater gas storage potential. Moreover, the well-connected pore system provides favorable conditions for gas adsorption and migration within the coal seam.

3.

3

Open in a new tab

Core samples of different coal macrolithotypes in the study area: (a) bright coal, (b) semibright coal, (c) semidull coal, and (d) dull coal.

4.

4

Open in a new tab

NMR T2 spectra of coal macrolithotypes in the study area: (a) bright coal, (b) semibright coal, (c) semidull coal, and (d) dull coal. P1 represents micropores and small pores (<10 ms), P2 represents medium pores (10–100 ms), and P3 represents large pores and microfractures (>100 ms).

5.

5

Open in a new tab

SEM observations of different coal macrolithotypes in the study area: (a) bright coal; (b) semibright coal; (c) semidull coal; (d) dull coal.

Significant differences in gas content and adsorption capacity are observed among different coal macrolithotypes. The isothermal adsorption curves of the study area (Figure ) indicate that under the same pressure conditions, bright coal and semibright coal exhibit considerably higher methane adsorption capacities than semidull and dull coal. Moreover, their larger Langmuir volumes and lower Langmuir pressures further confirm their stronger adsorption capacity. In addition, the gas content test (Figure a) showed that the gas contents of bright coal and semibright coal are mainly concentrated in the range of 16–20 m3/t, while the semidull and dull coals fall within the range of 11–17 m3/t, demonstrating that the former has a significantly higher gas-bearing capacity. Industrial component analysis (Figure b–d) reveals that bright coal and semibright coal generally have lower ash content but relatively higher fixed carbon and volatile matter contents. This indicates a higher degree of coalification and a greater abundance of vitrinite and other high-quality organic matter, suggesting a stronger gas-generation potential. On the contrary, semidull coal and dull coal exhibit higher ash content and lower fixed carbon and volatile matter, resulting in insufficient effectiveness of organic matter and an underdeveloped pore system, which in turn significantly reduces gas content and reservoir quality.

6.

6

Open in a new tab

Isothermal adsorption curves and associated data for different coal macrolithotypes in the study area.

7.

7

Open in a new tab

Comparison of core experimental analysis results of different coal macrolithotypes in the study area: (a) gas content; (b) ash content; (c) fixed carbon content; and (d) volatile matter content.

Based on core observations and experimental analyses of different coal macrolithotypes in the study area (Table ), bright coal and semibright coal feature more developed fracture-pore systems and higher contents of high-quality organic matter. They exhibit stronger gas adsorption capacity and higher gas content, indicating greater gas-generation and storage potential, and are more likely to form stable and efficient flow channels during fracturing operations. Therefore, this study identifies bright coal and semibright coal as the favorable coal types and serves as the primary focus of the investigation. A detailed characterization of their spatial distribution not only facilitates the prediction of favorable coal reservoirs but also provides a robust geological basis for well placement and drilling plan optimization.

1. Summary of Coal Macrolithotype Characteristics in the Study Area.

4.1.

Open in a new tab

4.2. Logging Response Analysis of Coal Macrolithotypes

The available logging curves in the study area are relatively diverse, but their responses to coal reservoirs vary considerably. Relying on a single logging curve makes it difficult to fully characterize the coal reservoir, while incorporating all logs may introduce redundant variables and potential inconsistencies. Previous studies indicate that natural gamma-ray (GR), acoustic transit time (AC), and density (DEN) have significant responses to coal reservoirs. ,,− Accordingly, this study focuses on these three parameters, integrating well log data with core observations to systematically analyze the logging response characteristics of different coal macrolithotypes.

Based on observations of macrolithotypes and their logging responses in cored wells (Figure ), bright coal and semibright coal exhibit a typical “two-low and one-high” pattern, characterized by low GR values, low DEN values, and high AC values; in contrast, semidull coal and dull coal display a “two-high and one-low” pattern, namely, high GR values, high DEN values, and low AC values. This distinct well-log response provides a reliable foundation for accurately identifying the coal macrolithotypes using logging data.

8.

8

Open in a new tab

Comprehensive column diagram of a typical well.

Calibrating core observations with depth-corrected well logs, this study further constructed cross-plots of coal macrolithotypes versus different logging parameters (Figure ). The plot is based on DEN, GR, and AC parameters to classify the coal macrolithotypes and is presented in two subplots. The left Figure a shows the cross-plot of DEN–GR, which reflects the relationship of coal macrolithotypes with DEN and GR: bright coal and semibright coal usually exhibit high DEN values and high GR values, whereas semidull and dull coals are mainly distributed in regions of low to medium densities and low GR values. Figure b on the right is a DEN–AC cross-plot showing the distribution characteristics of different coal macrolithotypes in the DEN versus AC space. In this plot, the bright coal and semibright coal correspond to lower DEN values and higher AC values, while the semidull coal and dull coal are concentrated in the region of high DEN values and low AC values.

9.

9

Open in a new tab

Cross-plot of coal macrolithotypes with different well logs in the study area (a) DEN vs GR and (b) DEN vs AC.

Building on the above relationships, this study established the logging response identification criteria for coal macrolithotypes (Table ), determining the logging response thresholds for each macrolithotype. In order to evaluate the value, generalizability, and reliability of the criteria, ten wells not involved in the modeling are selected as a verification well set. The results showed that the log-derived coal macrolithotypes achieved an overall matching rate exceeding 85% when compared to core descriptions. This indicates that the log-based identification method is both accurate and reliable, effectively supporting the evaluation of the heterogeneity of coal seams and the prediction of favorable coal reservoirs in the study area, which is of great significance in guiding the placement of CBM wells and the optimization of fracturing design.

2. Logging Identification Criteria for Coal Macrolithotypes.

coal macrolithotypes DEN, g/cm–3 GR, API AC, μs/ft
bright coal <1.34 <50 >118
semibright coal 1.34–1.44 50–70 112–118
semidull coal 1.44–1.56 70–85 106–112
dull coal 1.56–1.85 >85 <106
Open in a new tab

4.3. Geological Modeling Based on Coal Macrolithotypes

In order to better characterize the spatial heterogeneity of coal reservoirs, this study conducted geological modeling of coal reservoirs based on coal macrolithotypes, guided by insights into the geological features. As the foundation of the modeling workflow, structure modeling is first performed, aiming to accurately describe the geometry of subsurface geological structures, the fault system, and their spatial configuration. , The structure model established based on the integration of well and seismic data is shown in Figure , which clearly reflects the structural morphology of the target layer. Overall, the structure of the target layer in the study area is generally gentle, with a small stratigraphic dip and weak structural deformation. The elevation of the coal seam top surface ranges from −720 m to −920 m, with little fluctuation. The development of faults in the area is relatively low, predominantly normal faults, with an overall orientation mainly trending north-northeast (NNE) and northeast (NE).

10.

10

Open in a new tab

Structure model of the study area.

On the basis of the logging identification criteria for coal macrolithotypes established in Table , the macrolithotypes of individual wells in the study area are classified. Statistical analysis of well data indicates that the overall proportions of coal macrolithotypes in the study area are as follows: bright coal 14%, semibright coal 22%, semidull coal 18%, dull coal 39%, and 7% of noncoal fractions such as mudstone partings. Within the framework of the structural model, this study performed coal macrolithotype modeling using a geostatistical method constrained by the macrolithotype probability model obtained from Bayesian seismic inversion. Specifically, the macrolithotype interpretations from well data are used as hard data, while the seismic probability models serve as a trend constraint. A three-dimensional coal macrolithotype model in Figure is then constructed through cosimulation to characterize the spatial distribution of coal macrolithotypes.

11.

11

Open in a new tab

Coal macrolithotypes model of the study area.

According to the cross-section of the coal macrolithotype model (Figure a), the coal seam exhibits strong vertical heterogeneity. The middle and upper parts are dominated by bright coal and semibright coal, gradually transitioning downward into semidull coal and dull coal, with locally developed mudstone partings at the base. The vertical distribution pattern is mainly controlled by the sedimentary hydrodynamic conditions and the evolution of the paleopeat swamp: the upper high-quality coal reservoirs developed in a continuously water-covered, strongly reducing swamp environment, which is favorable to the enrichment of vitrinite. In contrast, the lower coal reservoirs are deposited under intermittently water-covered conditions with alternating oxidation–reduction environments, increased content of inertinite and clastic material, and a corresponding decline in coal quality. In the plan view (Figure b), the favorable coal macrolithotypes exhibit a NE-trending, belt-like distribution, primarily developed in the central part of the study area and gradually thinning westward. In this region, the thickness of favorable coal macrolithotypes is relatively large, ranging from 6 to 8 m, making it a promising and highly prospective development zone.

12.

12

Open in a new tab

(a) Cross-section of the coal macrolithotype model between wells; (b) the thickness map of favorable coal macrolithotypes (bright coal and semibright coal).

Gas content, an important property of coal reservoirs, is modeled within the constraints of the established coal macrolithotype framework (Figure ). First, the correlation between coal reservoir gas content and macrolithotypes is analyzed; next, the spatial trend of gas content can be derived from seismic inversion or attribute analysis. Finally, the gas content data from wells are used as hard constraints, and a facies-controlled geostatistical method is employed to establish a three-dimensional gas-content model of the coal reservoir. The resulting model not only characterizes the spatial heterogeneity of gas content in coal reservoirs but also effectively guides the prediction of high-gas-content zones, thereby providing a robust geological foundation for predicting favorable coalbed methane reservoirs and optimizing the development strategies.

13.

13

Open in a new tab

Gas content model of the study area.

The cross-section (Figure a) shows that gas content in the coal reservoir exhibits significant heterogeneity and layering characteristics: the gas content in the middle and upper sections, which are dominated by bright coal and semibright coal, is generally higher, ranging from 16% to 20%; while the lower section is dominated by semidull coal and dull coal, with lower gas content mostly between 8% and 12%. In the planar view (Figure b), the gas content generally displays an overall higher value in the central area and lower values in the western area. Zones of high gas content largely correspond to regions where favorable coal macrolithotypes (bright coal and semibright coal) are concentrated, whereas the western area, dominated by semidull coal and dull coal, exhibits lower gas content. On the basis of the characterization of favorable coal reservoirs in terms of coal macrolithotypes and gas content, CBM development planning should prioritize the middle to upper sections in the central area to effectively enhance gas production and overall development efficiency.

14.

14

Open in a new tab

(a) Cross-section of the gas content model between wells; (b) planar distribution of gas content.

4.4. Validation Based on Production Performance

In order to verify the reliability of the geological model and to further analyze the impact of reservoir heterogeneity on the development performance, this study selects two highly comparable horizontal wells (wells D and E) drilled in the same well pad for a comparative analysis of their production performance. Importantly, both wells are completed using the same fracturing parameters (Table ), including key parameters such as the number of fracturing stages, injection rate, total fluid volume, and total proppant volume, thereby effectively removing engineering-related influences and ensuring that production differences predominantly reflect the geological conditions.

3. Table of Parameters for Fracturing.

well name number of fracturing stages pumping rate (m3/min) total fluid volume (m3) total sand volume (m3)
well D 8 22 20144.3 2595.3
well E 8 22 19762.9 2596.36
Open in a new tab

The cross-section of the coal macrolithotype model along well D (Figure a) indicates that the well primarily encounters bright coal and semibright coal. Along the well trajectory, the cumulative lengths of bright coal, semibright coal, semidull coal, and dull coal are 567, 91, 5, and 340 m, respectively, resulting in a high proportion of favorable coal macrolithotypes (bright coal and semibright coal). Moreover, the trajectory is located in a high-gas-content zone (Figure c) with an average gas content of 17.5%. In contrast, the cross section of the coal macrolithotype model along well E (Figure b) shows that the coal macrolithotypes encountered in this well are more developed with semidull coal and dull coal, amounting to 347 and 571 m, respectively, while bright coal and semibright coal are less abundant, accounting for only 26 and 57 m, respectively. The main part of the well lies within a low-gas-content zone (Figure d) with an average gas content of 11.7%. The two wells exhibit significant differences in coal macrolithotypes and gas contents, reflecting the strong heterogeneity of the coal reservoir.

15.

15

Open in a new tab

Cross sections of coal macrolithotype and gas content models along the trajectories of wells D and E: (a) coal macrolithotype, well D; (b) coal macrolithotype, well E; (c) gas content, well D; (d) gas content, well E.

The production performance of horizontal wells is evaluated based on the 90 day average peak stable gas production volume, revealing significant differences between well D and well E (Figure ). The gas production of well D reaches approximately 17,000 m3/d (Figure a), whereas that of well E on the same platform is only about 5000 m3/d (Figure b). Under comparable fracturing conditions, well D exhibits more than three times the productivity of well E, owing to its higher proportion of bright coal and semibright coal encountered and its overall occurrence within a high-gas-content zone. This indicates that production performance is mainly controlled by geological factors, with the extent to which wells encounter high-quality coal lithotypes being a key control on CBM well productivity. Based on the spatial distribution of coal macrolithotypes and gas content, the horizontal well trajectory can be further optimized to increase the encounter rate of favorable coal reservoirs, thereby significantly improving the production and overall development efficiency.

16.

16

Open in a new tab

Production curves for typical wells in the study area: (a) well D and (b) well E.

5. Discussion

Two major challenges remain for this study. First, there is uncertainty in quantitatively mapping well-log responses to coal macrolithotypes, as different macrolithotypes often exhibit overlapping responses in γ-ray, density, and resistivity logs. Higher-dimensional characterization methods or machine learning models could be employed to achieve more accurate identification of coal macrolithotypes; second, the coupling mechanism between the coal facies and the model has not been fully characterized, and the existing models provide insufficient quantitative descriptions of sedimentary coal-control processes, which limits the construction of reservoir models. Therefore, enhancing the understanding of coal facies and improving the quantitative identification of coal macrolithotypes from well logs are critical directions for future breakthroughs in optimizing macrolithotype-based reservoir models.

6. Conclusion

This study proposes a novel approach to delineate the spatial distribution of favorable coal reservoirs based on coal macrolithotypes and applies it to Block A of the Shenfu Gas Field on the eastern margin of the Ordos Basin. The key findings are as follows: (1) bright coal and semibright coal are identified as the favorable coal reservoir types in the study area, and corresponding logging identification criteria are established; (2) a three-dimensional lithological model and a gas content model based on coal macrolithotypes are constructed to effectively characterize the heterogeneity of the coal reservoir; (3) in the study area, the favorable coal reservoirs are mainly concentrated in the central part of the block laterally and are vertically characterized by the development of bright coal and semibright coal in the upper part. The results demonstrate that the spatial distribution of favorable coal reservoirs has a significant impact on the production behavior, which is verified by single-well production data. The method has both theoretical and practical values, providing a reliable geological basis for optimizing well placement and well trajectories in CBM development.

Acknowledgments

The authors are grateful for the support provided by colleagues from China National Offshore Oil Corporation (CNOOC).

Di Yang: Writingoriginal draft, methodology, conceptualization. Yongjing Tian: Review and editing, supervision. Maojun Fang: Review and editing, supervision. Yuhu Bai: Review and editing, supervision. Yu Qi: Writingreview and editing, supervision. Fen Liu: Review. Ruyong Feng: Review. Hao Lu: Review.

The authors declare no competing financial interest.

References

  1. Banerjee A., Chatterjee R.. A methodology to estimate proximate and gas content saturation with lithological classification in coalbed methane reservoir, Bokaro field, India. Nat. Resour. Res. 2021;30(3):2413–2429. doi: 10.1007/s11053-021-09828-2. [DOI] [Google Scholar]
  2. Chen S., Tao S., Tang D.. In situ coal permeability and favorable development methods for coalbed methane (CBM) extraction in China: From real data. Int. J. Coal Geol. 2024;284:104472. doi: 10.1016/j.coal.2024.104472. [DOI] [Google Scholar]
  3. Lang Z., Chen W., Li L., Xu S., Liu F., Fan J., Lu Q.. A general framework to optimize fracture parameters in tight gas reservoir of heterogeneous channel sand bodies. Pet. Sci. Technol. 2024;43:2880–2900. doi: 10.1080/10916466.2024.2405006. [DOI] [Google Scholar]
  4. Kędzior S.. The occurrence of coalbed methane in the depocentre of the Upper Silesian Coal Basin in the light of the research from the Orzesze-1 deep exploratory well. Int. J. Coal Geol. 2024;292:104588. doi: 10.1016/j.coal.2024.104588. [DOI] [Google Scholar]
  5. Mohamed T., Mehana M.. Coalbed methane characterization and modeling: review and outlook. Energy Sources, Part A. 2025;47(1):2874–2896. doi: 10.1080/15567036.2020.1845877. [DOI] [Google Scholar]
  6. Petersen H. I., Brekke T.. Source rock analysis and petroleum geochemistry of the Trym discovery, Norwegian North Sea: a Middle Jurassic coal-sourced petroleum system. Mar. Pet. Geol. 2001;18(8):889–908. doi: 10.1016/S0264-8172(01)00033-2. [DOI] [Google Scholar]
  7. Barbato A. K., Armentrout J. M., Magoon L. B., Demchuk T., Barrie C., Warny S.. Organic geochemical characterization of deltaic paleogene rock units in coos bay, oregon: kerogen type and richness in response to depositional environments. AAPG Bull. 2024;108(9):1767–1798. doi: 10.1306/01252423011. [DOI] [Google Scholar]
  8. Li G., Ji C., Zh Q., Zh T., Ga J.. Coal-rock gas accumulation mechanism and the whole petroleum system of coal measures. Petrol. Explor. Dev. 2025;52(1):33–49. doi: 10.1016/s1876-3804(25)60003-6. [DOI] [Google Scholar]
  9. Li G., Yao Y., Wang H., Li M.. Deep coalbed methane resources in the Shenmu-Jiaxian block, Ordos Basin, China: Geological characteristics and potential for exploration and exploitation. Coal Geol. Explor. 2024;52(2):70–80. doi: 10.12363/issn.1001-1986.23.07.0436. [DOI] [Google Scholar]
  10. Ashraf U., Shi W., Zhang H., Anees A., Jiang R., Ali M., Mangi H. N., Zhang X.. Reservoir rock typing assessment in a coal-tight sand based heterogeneous geological formation through advanced AI methods. Sci. Rep. 2024;14(1):5659. doi: 10.1038/s41598-024-55250-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Li R., Hou X., Chen L., Fang H., Zheng C.. Multifractal investigation on multi-scale pore structure heterogeneity of high rank coal reservoirs. Nat. Resour. Res. 2022;31(3):1665–1685. doi: 10.1007/s11053-022-10046-7. [DOI] [Google Scholar]
  12. Wang K., Guo L., Xu C., Wang W., Yang T., Lin S., Cai Y.. Multiscale characteristics of pore-fracture structures in coal reservoirs and their influence on coalbed methane (CBM) transport: A review. Geoenergy Sci. Eng. 2024;242:213181. doi: 10.1016/j.geoen.2024.213181. [DOI] [Google Scholar]
  13. Tao G., Wang Z., Jin Y., Wang H., Xia D., Pan J.. Multiscale pore–fracture structure characteristics of deep coal reservoirs in the eastern margin of the Ordos Basin, China. Nat. Resour. Res. 2025;34(5):1557–1584. doi: 10.1007/s11053-025-10463-4. [DOI] [Google Scholar]
  14. Hou H., Liang G., Shao L., Tang Y., Yan Z.. Dual control of macrolithotype and coal structure on the pore parameters of middle Jurassic coals in the southern Junggar Basin, NW China. Acta Geol. Sin. 2025;99(1):194–211. doi: 10.1111/1755-6724.15202. [DOI] [Google Scholar]
  15. Chang X., Ming D., Zhang J., Vandeginste V., Wang H., Yin T., Liu Y., Lang H.. Adsorption energy variation and prediction of different coal facies in deep coal reservoirs and spatial distribution using logging curves. Nat. Resour. Res. 2025;34(6):3127–3153. doi: 10.1007/s11053-025-10515-9. [DOI] [Google Scholar]
  16. Zhang W., Bian L., Hu M., Wu P., Moore T. A., Shen J.. Quantitative identification and distribution pattern of macrolithotype of coal in deep coal seams of the Daji Block. Energy Sci. Eng. 2025;13(10):4839–4852. doi: 10.1002/ese3.70212. [DOI] [Google Scholar]
  17. Applied Coal Petrology: The Role of Petrology in Coal Utilization; Suárez-Ruiz, I. , Crelling, J. C. , Eds.; Academic Press, 2008. [Google Scholar]
  18. Speight, J. G. Handbook of Coal Analysis; John Wiley & Sons, 2015. [Google Scholar]
  19. Yuan M., Lyu S., Wang S., Xu F., Yan X.. Macrolithotype controls on natural fracture characteristics of ultra-thick lignite in Erlian Basin, China: Implication for favorable coalbed methane reservoirs. J. Pet. Sci. Eng. 2022;208:109598. doi: 10.1016/j.petrol.2021.109598. [DOI] [Google Scholar]
  20. Yong L., Xu F., Tang S., Zhang H., Xu F.. Progress and development direction of coalbed methane (coal-rock gas) exploration and development in the Ordos Basin. Nat. Gas. Ind. 2024;44(10):63–79. [Google Scholar]
  21. Zhao H., Wang X., Liu Z.. Experimental investigation of hydraulic sand fracturing on fracture propagation under the influence of coal macrolithotypes in Hancheng block, China. J. Pet. Sci. Eng. 2019;175:60–71. doi: 10.1016/j.petrol.2018.12.031. [DOI] [Google Scholar]
  22. Tao S., Pan Z., Chen S., Tang S.. Coal seam porosity and fracture heterogeneity of marcolithotypes in the Fanzhuang Block, southern Qinshui Basin, China. J. Nat. Gas Sci. Eng. 2019;66:148–158. doi: 10.1016/j.jngse.2019.03.030. [DOI] [Google Scholar]
  23. Roslin A., Esterle J. S.. Electrofacies analysis for coal lithotype profiling based on high-resolution wireline log data. Comput. Geosci. 2016;91:1–10. doi: 10.1016/j.cageo.2016.03.006. [DOI] [Google Scholar]
  24. Zhao J., Tang D., Qin Y., Xu H.. Experimental study on structural models of coal macrolithotypes and its well logging responses in the Hancheng area, Ordos Basin, China. J. Pet. Sci. Eng. 2018;166:658–672. doi: 10.1016/j.petrol.2018.03.073. [DOI] [Google Scholar]
  25. Liu Y., Xu H., Tang D., Mathews J. P., Zhai Y., Hou W., Li S., Tao S., Xiong X., Wang W.. The impact of the coal macrolithotype on reservoir productivity, hydraulic fracture initiation and propagation. Fuel. 2019;239:471–483. doi: 10.1016/j.fuel.2018.10.150. [DOI] [Google Scholar]
  26. Guo G., Li H., Tian Y., Du K.. Prediction of coal macrolithotypes based on logging parameters and three-dimensional geological modeling. China Geol. Surv. 2022;9(1):125–130. [Google Scholar]
  27. Li C., Yang Z., Tian W., Peng L., Liu L., Liu C., Lu B., Liang Y., Gu J.. Accurate prediction of the macrolithotype of a high-rank coal reservoir by logging interpretation: A case study of the Junlian Block, Sichuan Province, China. Nat. Resour. Res. 2023;32(5):2289–2311. doi: 10.1007/s11053-023-10233-0. [DOI] [Google Scholar]
  28. Zhang K., Jin Y., Meng Z., Wang X., Li M.. Experimental investigation on pore characteristics of heterogenous coal structures reservoir and coalbed methane diffusion/seepage behaviors. Phys. Fluids. 2024;36(7):076612. doi: 10.1063/5.0211848. [DOI] [Google Scholar]
  29. Hou H., Zhang H., Shao L., Guo S., Zhao M. E., Wang S.. Coal macrolithotype distribution and its genetic analyses in the deep Jiaozuo coalfield using geophysical logging data. ACS Omega. 2021;6(51):35523–35537. doi: 10.1021/acsomega.1c05012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Wang Z., Liu L., Hu J., Wang F., Li D., Zhang J., Zhu S., Zhang R., Zhao F., Zhang C.. et al. Dispersion of sandy sediments during marine–continental transition: an integrated study from the late paleozoic western Ordos Basin. Mar. Pet. Geol. 2024;160:106620. doi: 10.1016/j.marpetgeo.2023.106620. [DOI] [Google Scholar]
  31. Zhang J., Wu J., Yang S., Bai W., Zhuo Q.. The characteristics of petroleum seepage in coal seams: A case study of Ordos Basin. Pet. Sci. Technol. 2024;42(5):529–547. doi: 10.1080/10916466.2022.2120502. [DOI] [Google Scholar]
  32. Ma H., Meng Y., Zhang X., Cui X., Wang J.. Suitability evaluation of CO2 geological storage in deep coal seam: A case study in the eastern margin of Ordos Basin, China. Geol. J. 2025;60:1346–1369. doi: 10.1002/gj.5127. [DOI] [Google Scholar]
  33. Liu J., Zhu G., Liu Y., Chao W., Du J., Yang Q., Mi H., Zhang S.. Breakthrough, future challenges and countermeasures of deep coalbed methane in the eastern margin of Ordos Basin: A case study of Linxing-Shenfu block. Acta Pet. Sin. 2023;44(11):1827–1839. doi: 10.7623/syxb202311006. [DOI] [Google Scholar]
  34. Changgui X., Ji H., Wang C., Zhu X.. Enrichment patterns and exploration countermeasures of deep coalbed methane in the Linxing-Shenfu block on the eastern margin of the Ordos Basin. Coal Geol. Explor. 2024;52(8):2. doi: 10.12363/issn.1001-1986.24.01.0026. [DOI] [Google Scholar]
  35. Sun L., Liu J., Tian Y., Li N., Wang J., Feng R.. Early practical experience and key research directions for deep coalbed methane development in the Shenfu block. China Offshore Oil Gas. 2025;37(1):116–128. [Google Scholar]
  36. Xu H., Tang D., Mathews J. P., Zhao J., Li B., Tao S., Li S.. Evaluation of coal macrolithotypes distribution by geophysical logging data in the Hancheng Block, Eastern Margin, Ordos Basin, China. Int. J. Coal Geol. 2016;165:265–277. doi: 10.1016/j.coal.2016.09.002. [DOI] [Google Scholar]
  37. Cui C., Chang S., Yao Y., Cao L.. Quantify coal macrolithotypes of a whole coal seam: A method combing multiple geophysical logging and principal component analysis. Energies. 2021;14(1):213. doi: 10.3390/en14010213. [DOI] [Google Scholar]
  38. Shi J., Zhao X., Zeng L., Zhang Y., Dong S.. Identification of coal structures by semi-supervised learning based on limited labeled logging data. Fuel. 2023;337:127191. doi: 10.1016/j.fuel.2022.127191. [DOI] [Google Scholar]
  39. Pyrcz, M. J. ; Deutsch, C. V. . Geostatistical Reservoir Modeling; Oxford University Press, 2014. [Google Scholar]
  40. Ma, Y. Z. ; Zhang, X. . Quantitative Geosciences: Data Analytics, Geostatistics, Reservoir Characterization and Modeling; Springer International Publishing: Cham, 2019. [Google Scholar]
  41. Hadavand M., Deutsch C. V.. Facies proportion uncertainty in presence of a trend. J. Pet. Sci. Eng. 2017;153:59–69. doi: 10.1016/j.petrol.2017.03.036. [DOI] [Google Scholar]

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES