Gas Content and Geological Control of Deep Jurassic Coalbed Methane in Baijiahai Uplift, Junggar Basin

Abstract

Deep coalbed methane (CBM) resources are abundant in China, and in the last few years, the country’s search for and extraction of CBM have intensified, progressively moving from shallow to deep strata and from high-rank coal to medium- and low-rank coal. On the other hand, little is known about the gas content features of deep coal reservoirs in the eastern Junggar Basin, especially with regard to the gas content and the factors that affect it. Based on data from CBM drilling, logging, and seismic surveys, this study focuses on the gas content of Baijiahai Uplift’s primary Jurassic coal seams through experiments on the microscopic components of coal, industrial analysis, isothermal adsorption, low-temperature CO₂, low-temperature N₂, and high-pressure mercury injection. A systematic investigation of the controlling factors, including the depth, thickness, and quality of the coal seam and pore structure; tectonics; and lithology and thickness of the roof, was conducted. The results indicate that the Xishanyao Formation in the Baijiahai Uplift usually has a larger gas content than that in the Badaowan Formation, with the Xishanyao Formation showing that free gas and adsorbed gas coexist, while the Badaowan Formation primarily consists of adsorbed gas. The coal seams in the Baijiahai Uplift are generally deep and thick, and the coal samples from the Xishanyao and Badawan formations have a high vitrinite content, which contributes to their strong gas generation capacity. Additionally, low moisture and ash contents enhance the adsorption capacity of the coal seams, facilitating the storage of CBM. The pore-specific surface area of the coal samples is primarily provided by micropores, which is beneficial for CBM adsorption. Furthermore, a fault connecting the Carboniferous and Permian systems (C-P) developed in the northeastern part of the Baijiahai Uplift allows gas to migrate into the Xishanyao and Badaowan formations, resulting in a higher gas content in the coal seams. The roof lithology is predominantly mudstone with significant thickness, effectively reducing the dissipation of coalbed methane and promoting its accumulation.

1. Introduction

With the growth of global energy demand and the increasing emphasis on environmental protection, countries have set energy conservation and emission reduction targets [1]. As a low-carbon and renewable clean energy, coalbed methane has a positive effect on mitigating global climate change. CBM is usually located in deep underground coal seams. The gas reservoirs in these areas are often rich and have great development potential. Effective development can significantly aid in the transformation of the energy structure, lessen reliance on conventional fossil fuels, cut greenhouse gas emissions, and advance sustainable energy development [2].

Prospecting and extracting deep CBM have advanced significantly both domestically and internationally in the past several years. Especially in the US, Australia, and Canada, the mining technology of deep CBM has been widely used and optimized [3,4,5]. As China’s CBM resources are gradually explored and developed, the exploration and extraction of deep CBM has emerged as a crucial area of study. China started late in deep CBM exploration, but through the introduction of advanced technology and equipment, combined with the geological characteristics of domestic coal mines, it has also made some breakthroughs. The eploration and development of CBM gradually demonstrate the development trend from single coal seam to coal measures, from high-rank coal to low-rank coal, and from shallow to deep coal seams [6]. Increasing, the research and development of deep CBM are conducive to increasing China’s natural gas production, improving the level of energy supply security, reducing the dependence on imported natural gas, and maintaining national energy security [7].

China’s deep CBM resources are vast; preliminary assessments indicate that the CBM resources at depths below 2000 m are 40.71 trillion cubic meters, with 18.47 trillion cubic meters in the depth range of 2000 to 3000 m [8]. Since the 13th Five-Year Plan, the coalbed methane industry in Xinjiang has accelerated its development, achieving breakthrough exploration in the Junggar Basin, Tianshan series basins, and Santanghu Basin. Among these, the Junggar Basin is the richest in CBM resources, with a gas-bearing area of 26,000 square kilometers at depths shallower than 2000 m and a predicted geological resource amount of 3.11 trillion cubic meters, accounting for 41.3% of Xinjiang’s coalbed methane resources. The Baijiahai Uplift is an important target area in the Junggar Basin with significant exploration and development potential [9].

Previous investigations have revealed that the exploration potential of CBM depends on several key factors, including sedimentary environment, coal rank, structural and geological backgrounds, gas content, permeability, and fluid dynamics [10]. For example, rock structure parameters are one of the most significant aspects affecting the effectiveness of the transport process of useful components in rock mass. Carbonate rocks, especially shale, have great uncertainty and are susceptible to cracks and water effects [11]. Among these, one of the most difficult parameters to evaluate is gas content, which is crucial for the economic viability of CBM development [12,13,14]. The distribution of gas content exhibits strong lateral variation between individual coal seams and vertical variation between individual wells, heavily influenced by various geological factors. Understanding these factors affecting CBM content is essential for formulating effective exploration strategies [15].

The research on deep CBM in the Baijiahai Uplift is still in the preliminary stage. Although this area has rich potential for CBM resources, there are still some shortcomings in the current research. First of all, one of the biggest challenges in the exploration and development of deep CBM is the lack of understanding of the occurrence characteristics of CBM in the region. The geological structure of the Baijiahai Uplift area is complex; especially the burial depth of the deep coal seam is large, and the enrichment and migration law of coalbed methane is not completely clear [16]. Secondly, the existing geological models and reservoir evaluation methods have not fully considered the influence of fracture structure and other factors on the CBM reservoir and migration in the region, resulting in insufficient research on the reservoir characteristics in the Baijiahai area. Meanwhile, the theoretical model of deep CBM gas content needs to be further improved and tested [17].

In light of the most recent data on CBM exploration and development and based on the analysis of the gas-bearing distribution law of the main coal seam in the block, this paper studies the reservoir’s physical characteristics of the Baijiahai Uplift by looking at the conditions of burial depth, thickness, adsorption characteristics, pore characteristics, geological structure and lithology, roof thickness and lithology, and floor thickness and lithology. To explore the geological control factors of gas-bearing characteristics in the study area, the distribution law and exploration and development potential of CBM resources are clarified, which provides a scientific basis for the efficient development and utilization of CBM resources in the region.

2. Geological Setting

The Baijiahai Uplift is situated in the eastern section of the central depression of the Junggar Basin, occupying an area of around 4000 km² [18]. The current structural formation dates back to the late Permian period and has undergone further superposition and modification due to the Indosinian, Yanshan, and Himalayan movements. The area mainly develops two synclines and one anticline, with the Baijiahai Anticline in the center, flanked by the Fukang–Shazhang Syncline to the north and the Wucaiwan Syncline to the south. A series of small faults, arranged in a “stepped” pattern, develop along the NE direction in the axial part of the Baijiahai Anticline. The dip angle of the fault in this area is steep, generally about 60°, and the fault distance is between 20 and 100 m, mostly between 40 and 70 m [19]. Overall, it presents a NE-oriented, three-level positive structural unit, surrounded on three sides by hydrocarbon-generating depressions, making its structural position very favorable (Figure 1). The Jurassic Water Xigou Formation in the Baijiahai Uplift of the Dzungarian Basin is divided from bottom to top into the Badawan Group, the Sangonghe Group, and the Xishanyao Group. The two primary coal-bearing groups among them are the Xishanyao Group and the Badawan Group, primarily consisting of river delta sedimentary systems. The Badawan Group contains 1 to 15 coal seams, with most ranging from 2 to 5 seams; the mineable coal seams range from 1 to 11 layers. The Xishanyao Group contains 1 to 20 coal seams, with individual seams being thick and significantly influenced by sedimentary environments and coal accumulation processes [20].

3. Samples and Methods

3.1. Coal Rock Quality Experiment

The Baijiahai Uplift’s Xishanyao and Badawan formations are where the coal samples were extracted, specifically from wells BJ8, JT2, and DN141 in the Junggar Basin. According to GB/T 40485-2021 “Automatic Determination of Random Reflectance of Vitrinite in Coal—Image Analysis Method [22]”, GB/T 8899-2013 “Methods for the Microscopic Composition and Mineral Composition of Coal [23]”, and GB/T 30732-2014 “Industrial Analysis Methods for Coal—Instrumental Method [24]”, the maximum vitrinite reflectance (Ro), industrial analysis, and microscopic composition content of each sample were determined (

Table 1. Analysis of the coal quality and the maceral findings.

Sample	Proximate Analysis (%)			Maceral (%)			Ro,max/%
	Mad	Aad	Vdaf	V	I	L
BJ8-1-1 *	5.04	5.49	31.33	67.70	21.90	9.40	0.96
BJ8-1-2 *	5.07	2.99	29.30	66.50	24.40	7.50	0.95
JT2-1	5.70	1.96	23.26	55.00	43.50	0.90	0.47
JT2-2	5.53	2.36	23.87	79.20	18.10	0.50	0.45
JT2-3	4.67	2.20	27.06	57.20	40.90	0.00	0.60
JT2-4	5.89	1.34	26.96	71.50	22.20	0.00	0.62
JT2-5	6.04	2.41	29.97	62.70	32.90	0.80	0.60
JT2-6	5.77	1.46	29.67	36.20	60.50	0.90	0.47
JT2-7	3.12	3.57	36.15	95.20	2.30	0.80	0.56
JT2-8	2.00	6.11	44.92	96.00	0.70	2.30	0.50
JT2-9	1.96	7.24	43.44	96.30	0.50	0.00	0.43
JT2-10	2.08	4.98	43.32	98.20	1.70	0.00	0.65
JT2-11	2.21	3.20	38.58	86.40	9.10	3.00	0.50
JT2-12	2.01	3.75	41.73	82.60	6.20	6.60	0.54
JT2-13	2.20	7.10	40.70	86.10	12.60	0.70	0.53
DN141-1-1	1.85	4.83	25.07	33.90	57.40	4.90	0.94
DN141-1-2	1.13	3.63	23.40	/	/	/	/
DN141-1-3	1.01	2.02	27.85	53.90	42.20	1.40	0.70
DN141-1-4	1.06	2.06	27.71	/	/	/	/
DN141-1-5	0.97	1.89	27.37	40.10	/	/	0.72
DN141-1-6	1.23	1.39	30.36	/	54.50	3.40	/
DN141-2-1	0.81	9.89	46.17	86.70	/	/	0.64
DN141-2-2	0.76	6.02	42.92	/	1.50	7.10	/
DN141-2-3	0.97	5.29	44.59	82.90	/	/	0.64
DN141-2-4	0.72	11.10	43.07	/	11.10	2.90	/
DN141-2-5	0.92	9.91	42.47	92.90	/	/	0.73
DN141-2-6	1.14	4.80	45.04	/	0.60	1.20	/

Note: M_ad, moisture content, air-dried basis; A_ad, ash yield, air-dried basis; V_daf, volatile component yield, dry ash-free basis; V, vitrinite; I, inertinite; L, liptinite; Ro, max—maximum reflectance of vitrinite. * Data reference [26], / for no test data.

). The findings show that the highest vitrinite reflectance of the samples from the Badawan Formation is around 0.43% to 0.73%, whereas the samples from the Xishanyao Formation are between 0.45% and 0.96%. According to the international coal classification standard, Ro < 0.5 is low-rank coal, 0.5 ≤ Ro < 2 is medium-rank coal, and Ro ≥ 2 is high-rank coal [25]. Both groups of coal samples are classified as medium- to low-rank coal. The average vitrinite content in the Xishanyao Formation coal samples is 56.72%, while in the Badawan Formation coal samples, it is 90.33%, demonstrating that the Badawan Formation has a substantially higher vitrinite content than the Xishanyao Formation. The coal samples from the Xishanyao Formation have an average moisture content (M_ad) of 3.64%, an average ash yield (A_ad) of 2.57%, and an average volatile matter yield (V_ad) of 27.37%. In contrast, the average moisture content in the Badawan Formation coal samples is 1.61%, with an average A_ad of 6.38% and an average V_ad of 42.55%. The Xishanyao Formation is classified as having extremely low total moisture, extremely low ash, and medium volatile matter coal, while the Badawan Formation is classified as having extremely low total moisture, low ash, and high volatile matter coal.

3.2. Isothermal Adsorption Experiment

The ISO-300 isothermal adsorption desorption equipment, manufactured by the Terratek Company in the US, was used to carry out the isothermal adsorption modeling studies for deep coal seams. The coal sample particle size was 60 to 80 mesh, and the methane concentration used in the experiments was 99.99%. The experimental methods and procedures followed GB/T 19560-2008 [27]. The samples were treated under balanced water conditions for 24 h, ensuring that the coal samples reached equilibrium under each temperature condition. The data processing model was based on the Langmuir monolayer adsorption model to obtain the isothermal adsorption constants, including the Langmuir volume (V_L) and Langmuir pressure (P_L). The simulation design temperatures were set at 30 °C, 70 °C, and 80 °C, with a maximum simulated pressure of 25 MPa (Figure 2). The study region has a geothermal gradient of 2.4 °C/hm and a pressure gradient of 0.93 MPa/hm. According to the temperature and pressure conditions of the reservoir in the study area, the reservoir temperature range of the sample is 70 °C to 96 °C and the reservoir pressure range is 24 to 34 MPa. The maximum temperature simulated by isothermal adsorption experiment is 80 °C, and the maximum pressure is 25 MPa. For the Xishanyao Formation, the Langmuir volume ranged from 7.83 to 14.30 m³/t, with the mean of 11.70 m³/t, and the Langmuir pressure was 6.12 MPa on average, spanning from 2.90 to 9.87 MPa. For the Badawan Formation, the Langmuir volume ranged from 14.35 to 16.22 m³/t, with an average of 13.37 m³/t and the Langmuir pressure ranged from 3.38 to 6.36 MPa, with an average of 5.05 MPa (

Table 2. Isothermal adsorption experimental results under equilibrium water conditions.

Sample	Xishanyao Formation		Sample	Badaowan Formation
	VL (m3/t)	PL (MPa)		VL (m3/t)	PL (MPa)
JT2-1	11.80	7.41	DN141-1-2	10.79	7.65
JT2-2	12.31	2.90	DN141-1-3	10.76	7.54
JT2-3	13.95	3.37	DN141-1-4	10.13	6.26
JT2-4	14.30	3.70	DN141-1-5	13.31	7.39
JT2-5	12.59	9.87	DN141-1-6	12.85	7.72
JT2-6	13.77	4.79	DN141-2-1	10.87	6.15
JT2-7	15.62	3.65	DN141-2-2	11.85	5.91
JT2-8	15.93	3.39	DN141-2-3	10.14	5.26
JT2-9	14.35	6.248	DN141-2-4	10.83	6.36
JT2-10	15.64	3.818	DN141-2-5	9.83	5.36
JT2-11	15.81	3.98	DN141-2-6	12.05	6.34
JT2-12	14.64	5.34	BJ8-1-1	9.81	6.69
JT2-13	16.22	3.87	BJ8-1-2	7.83	4.90
DN141-1-1	9.61	5.47

). The isothermal adsorption curves of the coal samples from both the Xishanyao and Badawan formations conformed to the Langmuir equation, exhibiting a logarithmic increase in adsorption with rising pressure. The rate of adsorption increase gradually diminished, and the adsorption amount approached a constant value. Generally, compared to the samples from the Badaowan Formation, the coal samples from the Xishanyao Formation had a greater adsorption capability.

3.3. Pore Characteristics

The pore characteristics of coal include porosity, specific surface area, pore volume, pore size distribution, and pore connectivity. The classification of pore size at the microscopic scale is the basis for studying the pore characteristics of coal reservoirs. This paper used overburden porosity experiments to simulate the porosity of coal samples under reservoir conditions. Low-pressure CO₂ adsorption experiments, low-temperature N₂ adsorption experiments, and high-pressure mercury intrusion experiments are employed to characterize the micropore structure of the coal samples.

The overburden porosity experiments are conducted using an AP-608 automatic permeability–porosity instrument from Coretest Systems, following the petroleum industry standard SY/T 6385-2016 [28]. The low-pressure CO₂ experiments primarily aim to study the micropores (0–2 nm) in the coal samples, utilizing a JW-BK200C surface area and pore size analyzer, with the experiments conducted according to GB/T 21650.3-2011 [29]. The low-pressure N₂ adsorption experiments focused on characterizing the mesopores (2–50 nm) in the coal, using the same instrument as the low-temperature nitrogen adsorption experiments and following the standard for gas adsorption BET methods to test the specific surface area of solid materials (GB/T 19587-2017) [30].The pressure mercury intrusion experiment reflects the pore volume characteristics and pore shapes at different scales in the coal body through the intrusion and extrusion mercury curves. This experiment is carried out using a PoreMaster 33 GT fully automatic mercury intrusion apparatus, adhering to GB/T 21650.1-2008 [31]. Each sample consisted of 1–2 g of flat block samples. First, the oil was extracted and removed, then the sample was dried in a drying oven at 70 °C for 24 h and finally placed in the instrument for testing.

4. Gas Content Prediction

4.1. Prediction of Adsorbed Gas Gontent

4.1.1. Gas Adsorption Theory

Methane adsorption on coal is explained by a number of theories and equations, such as the Langmuir equation based on monolayer adsorption theory, the multi-layer adsorption foundational BET equation, and the D-R equation based on adsorption potential theory. Additionally, there are also potential theory and statistical thermodynamics theory. Among these, methane adsorption on coal is frequently researched by employing the Langmuir equation [32,33]. The mathematical expression of the Langmuir equation is:

$$V={\displaystyle \frac{V_{L}P}{P_L+P}}$$

(1)

In the formula, V is the amount of gas adsorption at the relevant pressure, m³/t; P is the pressure, MPa, ${\mathrm{P}}_{\mathrm{L}}$ is the Langmuir pressure, and MPa; ${\mathrm{V}}_{\mathrm{L}}$ is the Langmuir volume, cm³/g.

4.1.2. Prediction of Adsorbed Gas Content

The main quantitative prediction models for adsorbed gas include gas content gradient prediction models, isothermal adsorption prediction models, and multi-factor regression analysis prediction models.

In light of the research area’s reservoir features, combined with isothermal adsorption experiments, the Langmuir equation was applied to create a prediction model for a quantity of adsorbed gas. The Langmuir constants obtained from the isothermal adsorption experiments were used to calculate the adsorbed gas content ${\mathrm{V}}_{\mathrm{e}}$ at the temperature of the experiments using Equation (1). Given that the experimental temperature and geologic circumstances of deep coal seams are different, we needed to correct the adsorbed gas content. This means separately calculating the adsorbed gas content ${\mathrm{V}}_{\mathrm{r}}$ at the temperature of the associated reservoir gas.

$$V_r=V_e-{\Delta V}_T·\left(T_r-T_e\right)$$

(2)

In the formula, the gas content adsorbed at reservoir temperature in the dry ashless coal sample is denoted by ${\mathrm{V}}_{\mathrm{r}}$, m³/t; ${\mathrm{V}}_{\mathrm{e}}$ is the amount of adsorbed gas present in dry ashless coal samples at the experimental temperature of isothermal adsorption, which is m³/t; ${\mathrm{T}}_{\mathrm{r}}$ is the reservoir temperature, °C; ${\mathrm{T}}_{\mathrm{e}}$ is the experiment’s temperature for isothermal adsorption, °C; and ΔV_T is the methane adsorption decay rate with respect to temperature, m³/(t·°C). The impact of reservoir temperature on methane adsorption at various buried depth circumstances is represented by ΔV_T. Using industrial analysis, the two parameters of A_ad and M_ad can be utilized to convert into the amount of adsorbed gas in samples of air-dried basis coal.

$$V_A={\displaystyle \frac{V_r·(100-M_{\mathit{ad}}-A_{\mathit{ad}})}{100}}$$

(3)

In the formula, V_A is the in situ adsorbed gas content, m³/t. The overall calculation equation for the adsorbed gas content is obtained via a combination of Equations (1)–(3) [34]:

$$V_A={\displaystyle \frac{[V_{L}P-{\Delta V}_T·(P+P_L\left)\right(T_r-T_e\left)\right]·(100-M_{ad}-A_{ad})}{100·(P+P_L)}}$$

(4)

4.1.3. Methane Adsorption Decay Rate

It is commonly accepted that the cumulative impacts of reservoir pressure and temperature govern the methane ability to absorb the seams during formation conditions.

This results from the increased methane adsorption due to rising reservoir pressure and the decreased adsorption caused by elevated reservoir temperature. At lower reservoir temperatures and pressures, methane adsorption in coal is mainly pressure-controlled; at higher reservoir temperatures and pressures, it is predominantly influenced by temperature. The impact of temperature on gas adsorption under various buried depth circumstances is shown in this research by the attenuation rate of methane adsorption. The researched area’s gradient of pressure is 0.93 MPa/hm; the geothermal gradient is 2.4 °C/hm, with a constant temperature zone depth of 50 m and an average surface temperature of 10 °C. Isothermal adsorption experiments conducted by Chen Gang on samples from the Baijiahai Cainan area show that when the temperature rises, the samples’ gas adsorption ability falls [35]. Using isothermal adsorption data from samples at 40 °C and 96 °C, the following equation illustrates how the decrease in gas ability to adsorb at various temperatures changes logarithmically with increasing pressure.

$$\Delta V=\mathit{aln}\left(P\right)+b$$

(5)

In this context, ΔV represents the reduction in methane adsorption at different isothermal experiment temperatures, m³/t. The constants a and b in the relationship are linearly related to the maximum vitrinite reflectance of coal. Based on the fitted formula, the methane adsorption decay rates for coal samples of varying metamorphic degrees in different burial depth ranges can be derived. The methane adsorption decay rate ΔV_T is calculated using Formula (6):

$${\Delta V}_T={\displaystyle \frac{\Delta V}{\Delta T_e}}$$

(6)

where ΔT_e is the difference in temperatures of the different isothermal adsorption experiments, and it is dimensionless. In this study, the temperature difference for the isothermal adsorption experiments is 56.

4.1.4. Prediction Results of Adsorbed Gas Content

Taking wells BJ8, JT2, and DN141 as examples, we utilized the Langmuir isothermal adsorption model. According to Formula (4), the amount of adsorbed gas in the Xishanyao Formation is estimated to be between 5.13 and 10.25 m³/t at 8.17 m³/t on average. For the Badawan Formation, the average adsorbed gas concentration is 9.45 m³/t, with a range of 6.21 to 11.92 m³/t.

4.2. Prediction of Free Gas Content

4.2.1. Quantitative Prediction of Free Gas Content

The majority of the free gas in the seams is found in the coal’s pores, and its content is affected by many factors. The interaction of temperature, pressure, porosity, water saturation, and other factors affects the free gas content in coal seam. This work integrated earlier studies to determine the free gas content in deep coal deposits and the NB/T 1018-2015 “Method for Determining Gas Content in Low-Rank Coal Seams”, based on an equation for free gas content. To increase the accuracy of the free gas computation, the amount of adsorbed gas should be eliminated when forecasting the free gas content, as adsorbed methane occupies a certain amount of pore space. The calculation formula for the porosity of adsorbed methane is [36]:

$${\Phi }_a={\displaystyle \frac{V_a{\rho }_b{\rho }_g}{{\rho }_a}}$$

(7)

The porosity of adsorbed gas is represented by Φ_a in the formula, %; V_a is the adsorbed gas content, m³/t; ρ_a is the adsorbed methane’s density, t/m³; ρ_b is the coal rock‘s bulk density, t/m³; methane density under standard circumstances is ρ_g, 0.66673 × 10⁻³ t/m³.

Free methane’s porosity is:

$${\Phi }_f=\Phi -{\Phi }_a$$

(8)

where Φ is the gas measurement porosity, %. Adsorbed methane density is:

$${\rho }_a={\rho }_d\exp \left[-{\alpha }_e\left(T-T_d\right)\right]$$

(9)

In the formula, methane density at its boiling point at atmospheric pressure is denoted as ρd, 0.423 t/m³; α_e is the thermal expansion coefficient, 2.5 × 10⁻³ K⁻¹; T is the reservoir temperature, K; T_d is the temperature at which methane boils at atmospheric pressure, 111.65 K.

The free gas content forecast method for unconventional natural gas is the source of the formula used to determine the free gas content in deep coal deposits [37]:

$$V_F={\displaystyle \frac{{\Phi }_f{(1-S}_w)}{{\rho }_b{B}_g}}$$

(10)

In the formula, V_F is the free gas in the reservoir, m³/t; Φ_f is the in situ porosity of free methane, %; S_w is the water saturation, %; B_g is the gas volume factor of methane which is dimensionless. The calculation formula for the gas volume factor B_g of methane is:

$$B_g={\displaystyle \frac{V_g}{V_{\mathit{sc}}}}$$

(11)

In the formula, V_g represents the nmol gas volume in a reservoir, cm³; V_sc is an amount of nmol gas at surface normal conditions, cm³. Combined with the real gas state equation it is:

$$PV=nRTZ$$

(12)

In the formula, P stands for pressure, MPa; T for temperature, K; Z for gas compressibility factor; n for gas amount, measured in mol; and R for gas constant. The methane gas system coefficient expression is derived as:

$$B_g={\displaystyle \frac{P_{\mathit{sc}}\mathit{ZT}}{P{T}_{\mathit{sc}}}}$$

(13)

By combining Formulas (10)–(13), the following is the prediction formula for the amount of free gas in CBM:

$$V_F={\displaystyle \frac{{\Phi }_f\left({1-S}_w\right)P{T}_{\mathit{sc}}}{\rho ·P_{\mathit{sc}}·Z·T}}$$

(14)

4.2.2. Prediction Results of Free Gas Content

The Xishanyao Group in the Baijiahai Uplift is classified as saturated to oversaturated coal seams, while the Badawan Formation is classified as undersaturated coal seams. Taking wells BJ8, JT2, and DN141 as examples, as to this formula, the Xishanyao Formation’s free gas levels variate from 3.95 to 5.80 m³/t, with a mean of 5.09 m³/t (14).

4.3. Varification of Gas Content Prediction Results

The adsorbed and free gas can be found in the seams of coal that form the Xishanyao Formation in the Baijiahai Uplift, while the Badawan Formation’s coal seams are predominantly composed of adsorbed gas. According to the prediction results from wells BJ8, JT2, and DN141, the Xishanyao Formation’s overall gas content varies from 10.94 to 14.48 m³/t, having an average of 13.24 m³/t. For the Badawan Formation, the total gas content varies between 6.21 and 11.92 m³/t, having an average of 9.45 m³/t (Figure 3).

Due to the occurrence of mud blockages in the Xishanyao Formation’s samples, the measured gas content results were significantly lower, with the gas desorption phenomenon being most severe in well JT2, making it unsuitable as a validation indicator.

Table 3. Gas content prediction values of Baijiahai Uplift.

Sample	Depth (m)	M (m3/t)	P (m3/t)	Proportion of Gas Content Phases (%)
				Adsorbed Gas Ratio (%)	Free Gas Ratio (%)
BJ8-1-1	3362.99	13.34	11.29	54	46
BJ8-1-2	3360.07	16.11	10.94	47	53
JT2-1	3190.98	3.36	12.46	68	32
JT2-2	3192.38	3.18	14.67	67	33
JT2-3	3193.58	3.25	14.54	63	37
JT2-4	3194.54	3.40	14.84	61	39
JT2-5	3195.40	3.73	11.94	67	33
JT2-6	3195.90	3.93	14.28	72	28
JT2-7	3614.73	9.15	10.74	100	0
JT2-8	3615.70	9.27	11.89	100	0
JT2-9	3616.71	9.16	10.88	100	0
JT2-10	3617.52	9.68	9.19	100	0
JT2-11	3618.84	9.72	11.92	100	0
JT2-12	3619.19	10.14	11.09	100	0
JT2-13	3620.38	10.11	11.32	100	0
DN141-1-1	2582.50	8.86	11.96	55	45
DN141-1-2	2584.09	10.81	13.10	57	43
DN141-1-3	2585.65	12.71	12.94	58	42
DN141-1-4	2586.53	9.07	13.24	57	43
DN141-1-5	2587.23	15.35	14.63	65	35
DN141-1-6	2587.66	12.39	14.55	64	36
DN141-2-1	3007.59	5.81	7.15	100	0
DN141-2-2	3009.82	8.13	8.95	100	0
DN141-2-3	3010.86	7.68	7.14	100	0
DN141-2-4	3012.04	10.33	7.62	100	0
DN141-2-5	3013.44	9.34	6.21	100	0
DN141-2-6	3014.73	8.32	8.72	100	0

Note: M, Measured gas content; P, Predicted gas content.

compares the measured and projected gas levels of the remaining samples. The average ratio of measured to predicted gas content for the Xishanyao Group is 0.97, while for the Badawan Formation, the average ratio is 0.98, indicating that the overall prediction results are satisfactory.

4.4. Calculation of Gas Content in Baijiahai Uplift

4.4.1. Calculation of Adsorbed Gas Content in the Baijiahai Uplift

Due to the limited core data and laboratory test results from deep coal seams, accurately identifying key parameters such as gas content is challenging. Furthermore, the adsorbed gas content is controlled by several geological causes, resulting in a complex nonlinear relationship among them. Traditional methods struggle to express these intrinsic connections. In contrast, neural network methods possess strong nonlinear approximation capabilities, allowing for a realistic representation of the nonlinear relationships between input and output variables. To address this, we studied the correlation between adsorbed gas content in wells JT2 and DN141 and their logging parameters. The Baijiahai Uplift’s adsorbed gas content was forecast using a feedforward neural network prediction model.

A feedforward neural network model is widely used. An input layer, hidden layers, and an output layer are among its several layers. Forward propagation and backpropagation are the two primary phases of a feedforward neural network’s learning process. Input data are sent from the input layer to the hidden layers during the forward propagation step. Each neuron performs a weighted sum and generates an output through an activation function, which is then passed layer by layer until reaching the output layer, where the final prediction is obtained. Subsequently, a function of loss is computed to assess how the actual values differ from the expected values.

During the backpropagation stage, the gradients of each layer’s parameters are computed using the chain rule, and the weights and biases are updated to minimize the loss. This process is repeated until the model converges or reaches a predetermined number of iterations [38], as illustrated in Figure 4.

The logging parameters acoustic (AC), density (DEN), neutron (CNL), true formation resistivity (RT), shallow resistivity (RI), flushed zone formation resistivity (RXO), spontaneous potential (SP), borehole diameter (CALI), and natural gamma ray (GR) were used as inputs to the neural network, with the hidden layer’s neuron count set to nine. The output for modeling was the content of adsorbed gas.

The statistical results of the prediction errors from the neural network are presented in

Table 4. Training sample set and prediction errors of neural network model.

Sample	Formation	The Predicted Value of Adsorbed Gas Content		Relative Error (%)
		Adsorption Model (m3/t)	Neural Network Model (m3/t)
JT2-1	J2x	8.51	8.51	0
JT2-2	J2x	9.85	9.85	0
JT2-3	J2x	9.20	9.20	0
JT2-4	J2x	9.07	9.07	0
JT2-5	J2x	7.97	10.18	27.69 **
JT2-6	J2x	10.25	10.25	0
JT2-7	J1b	10.74	10.69	0.50
JT2-8	J1b	11.89	11.73	1.34
JT2-9	J1b	10.88	11.73	7.87
JT2-10	J1b	9.19	11.64	26.65 **
JT2-11	J1b	11.92	11.67	2.14
JT2-12	J1b	11.09	11.31	1.98
JT2-13	J1b	11.32	11.18	1.27
DN141-1-1	J2x	6.62	6.63	0
DN141-1-2	J2x	7.43	7.43	0
DN141-1-3	J2x	7.55	7.64	1.21
DN141-1-4	J2x	7.59	6.84	9.87
DN141-1-5	J2x	9.48	9.47	0
DN141-1-6	J2x	9.28	9.21	0.69
DN141-2-1	J1b	7.15	7.16	0.25
DN141-2-2	J1b	8.95	8.94	0.16
DN141-2-3	J1b	7.14	7.15	0.15
DN141-2-4	J1b	7.62	4.69	38.35 **
DN141-2-5	J1b	6.21	6.26	0.76
DN141-2-6	J1b	8.72	8.75	0.34

Note: ** is the abnormal value.

.

Figure 5 illustrates the content of adsorbed gas. The adsorbed gas content calculated by the neural network model closely matches the experimental results from core samples. Specifically, the relative error of the predicted adsorbed gas content for the Xishanyao Formation ranges from 0% to 9.87%, while the relative error for the Badaowan Formation ranges from 0.15% to 7.87%, indicating a relatively good correlation.

4.4.2. Calculation of Free Gas Content in Baijiahai Uplift

By analyzing the relationship between the logging parameters of wells JT2 and DN141 and the overburden porosity and water saturation, the following calculation formula was established:

$${\Phi }_{f1}=13.447-0.012\mathit{CNL}-0.211\mathit{DEN}$$

(15)

where Φ_f1 is the original porosity of the Xishanyao Formation, %; CNL is the compensated neutron porosity, %; and DEN is the compensated density, g/cm³.

$${\Phi }_{f2}=0.62-0.005\mathit{CNL}$$

(16)

where Φ_f2 is the original porosity of the Badaowan Formation, %.

$$1-S_W=-9.42359+0.74374\mathit{CALI}+0.07893\mathit{RXO}$$

(17)

where CALI is the well diameter, cm; RXO is the resistivity of the washed zone, Ω·m. Equation (14) is still used to determine the free gas content.

5. Results and Discussions

5.1. Horizontal Distribution of Gas Content

5.1.1. Adsorbed Gas Content

The adsorbed gas of the Xishanyao Formation in the Bajiahai Uplift varies between 7.91 and 10.21 m³/t, having an average of 8.70 m³/t, based on the anticipated results. A trend of greater values in the west is evident in the spatial distribution and reduced figures in the east, with high-value areas mainly located near wells C39 and C34, where the adsorbed gas content in the coal seams exceeds 9 m³/t. The low-value area is found near well C26. For the Badaowan Formation, the adsorbed gas is between 3.70 and 11.29 m³/t, having an average of 7.16 m³/t. High-value areas are primarily near well C39, where values exceed 10 m³/t, while low-value areas are found near wells C25 and C26 (Figure 6). Overall, the adsorbed gas content in the Xishanyao Formation is greater than that in the Badaowan Formation.

5.1.2. Free Gas Content

The Xishanyao Formation in the Baijiahai Uplift is projected to have free gas ranging from 6.19 to 12.21 m³/t, with an average of 10.97 m³/t. High-value areas are located near wells C32 and C36, while low-value areas are near well C19 (Figure 7). The spatial distribution shows a trend of higher values in the southwest and lower values in the northeast. The Badaowan Formation is characterized as undersaturated coal seams and therefore contains no free gas.

5.1.3. Gas Content

The Baijiahai Uplift’s Xishanyao Formation has an average gas content of 18.04 m³/t, with a range of 16.40 to 23.84 m³/t. High-value areas are located near well C32, while low-value areas are near wells C11 and C19. The spatial distribution shows a trend of higher values in the southwest and lower values in the northeast (Figure 8).

5.2. The Impact of Geological Factors on Gas Content

5.2.1. Burial Depth of Coal Seams

The burial depth is a significant factor affecting the gas content since it affects coalification and, in turn, CBM output. In addition, the buried depth of coal seams also plays a very important role in the preservation of coalbed methane [39]. The present study demonstrates that there are two primary ways in which the control influence of the coal seam’s buried depth on gas content is manifested.

(1) The coal reservoir’s temperature and pressure gradually rise as the buried depth increases. The influence of the reservoir’s pressure on the adsorbed gas is larger than that of the reservoir temperature below the critical depth; in other words, the adsorbed gas increases as burial depth does. This is due to the fact that an increase in reservoir pressure causes the free gas to change into adsorbed gas, which helps to preserve adsorbed gas. The adsorption capacity tends to decrease as the burial depth approaches the critical depth, and the negative influence of temperature increases and the positive effect of pressure diminishes.

(2) As the burial depth of the coal seam increases, the pathway for coalbed methane migration upward lengthens. Additionally, due to compaction effects, the coal seam’s permeability declines, and its sealing capacity improves, which is beneficial for the storage of CBM [40].

The burial depth of the Xishanyao Formation is between 1610 and 3363 m, with an average of 2719 m; the burial depth of the Badawaan Formation is between 1830 and 3620 m, with an average of 2991 m. The two groups of coal seams show the plane distribution characteristics of high southwest and low northeast (Figure 9). It is roughly the same as the plane distribution of the gas content. Figure 10 displays the correlation analysis between the research area’s gas content and burial depth. In line with the pattern where gas content increases and then decreases in deeper coal seams, the results demonstrate that the gas content of both the Xishanyao and Badawaan formations generally exhibits a trend of first increasing and then decreasing with increasing burial depth. The turning depth for the gas content is between 2500 and 3000 m; when the burial depth is between 2500 and 3000 m, the gas content is relatively high, likely due to the higher gas retention capacity [12].

5.2.2. Coal Thickness

The coal seam acts as both a reservoir and a source rock for CBM. Variations in coal seam thickness are closely related to geological structures and sedimentary environments.

In addition to influencing CBM generation, coal thickness also affects gas dispersion and migration paths [41]. The amount of organic matter in coal increases in tandem with the coal seam’s thickness. Coal reservoir is a kind of high-density and low-permeability rock stratum. The thicker the coal seam is, the greater the resistance of coalbed methane to spread to the roof and floor, so as to prevent the escape of coalbed methane [42].

In the Baijiahai Uplift, the thickness of the Xishanyao Formation’s coal seams is 8 to 20 m, with an average of 11.97 m, progressively increasing from west to east. The thickness of the Badawan Formation’s coal seams is 5 to 22.80 m, with an average of 14.64 m; the thickness is greater in the east–west direction and smaller in the north–south direction (Figure 11). The distribution patterns of coal thickness and total gas content in the Xishanyao Formation and Badowan Formation of the Baijiahai Uplift show obvious differences. The Xishanyao Group has a higher gas content in the area of the thick coal seam, demonstrating a favorable relationship between gas content and thickness (Figure 12). In contrast, the correlation between coal seam thickness and gas content in the Badawan Formation is weak, indicating that the enrichment mechanism of Jurassic coalbed methane in the Baijiahai Uplift is more complex. This may be related to the relatively small variations in coal seam thickness and sedimentary environments in the study area.

Furthermore, elements like the coal’s chemistry and structure have an impact on the gas content. Compared to the Xishanyao Formation, the Badawan Formation has a higher ash yield and lower porosity, resulting in an overall lower gas content. Therefore, the control of coal seam thickness on gas content in the Badawan Formation is relatively small.

5.2.3. Coal Rock Quality

The vitrinite content in coal not only serves as the material for coalbed methane generation but also has the capacity to adsorb methane. Its content directly determines the ability of coal to adsorb gas. In the Baijiahai Uplift, the microcomponents of the Xishanyao and Badaowan formations are predominantly vitrinite, followed by inertinite, with a sporadic presence of liptinite, which overall favors gas generation. The vitrinite content and the total Langmuir volume exhibit a positive association, as seen in Figure 13 in coal samples from the region of the Baijiahai Uplift [43]. As the vitrinite content increases, the Langmuir volume also tends to increase, indicating that higher vitrinite content leads to greater gas adsorption. The vitrinite content in the Badaowan Formation’s coal samples is higher than that in the Xishanyao Formation’s samples.

Based on the experimental results, an analysis of the relationship between Langmuir volume and moisture and ash contents was carried out for samples in the research region.

The analysis shows that the Langmuir volume decreases with increasing moisture and ash contents, showing a negative correlation [44]. This is primarily because the internal surface area available for adsorption in coal is limited and coal has a greater affinity for water than for methane. Therefore, excessive moisture in coal negatively impacts the gas adsorption capacity. The presence of ash primarily affects adsorption by blocking some of the coal’s pores, reducing its specific surface area and, consequently, its adsorptive capacity [45,46].

5.2.4. Pore Structure

The pore properties of coal are fundamental to studying the occurrence of CBM, the methane’s movement in coal seams, and its adsorption/desorption properties within coal.

The gas content in coal seams is closely related to the distribution characteristics of pores in coal. Coalbed methane primarily exists in two forms: free gas and adsorbed gas. Adsorbed coalbed methane is mainly stored in the organic matter and pore surfaces within the coal, while free coalbed methane primarily accumulates in the pore–fracture spaces of the coal. Therefore, the amount of adsorbed coalbed methane is mainly influenced by the specific surface area of the coal [47,48], whereas the amount of free coalbed methane is primarily affected by the pore volume [49,50,51].

Based on comprehensive low-pressure CO₂ adsorption, low-temperature N₂ adsorption, and high-pressure mercury intrusion experiments, the pore structure of coal samples from the Xishanyao Formation and the Badawan Formation in the Baijiahai Uplift was characterized at all scales (

Table 5. Pore structure parameters.

Sample	Pore Volume (cm3/g)				Pore Specific Surface Area (m2·g)
	Micropores	Mesopores	Mecropores	Total	Micropores	Mesopores	Mecropores	Total
DN141-1-1	0.077	0.047	0.909	1.033	236.375	0.348	1.365	238.088
DN141-1-3	0.065	0.054	0.518	0.637	206.816	0.502	4.605	211.923
DN141-1-5	0.058	0.026	0.456	0.540	194.535	0.346	5.237	200.118
DN141-2-1	0.032	0	0.027	0.059	98.230	0.035	0.333	98.598
DN141-2-3	0.034	0	0.025	0.059	105.861	1.294	2.184	109.339
DN141-2-5	0.035	0	0.024	0.059	110.950	1.273	2.018	114.241
Average	0.050	0.021	0.327	0.398	158.795	0.633	2.624	162.052

). The findings show that the Xishanyao coal samples have a total pore volume ranging from 0.540 to 1.033 cm³/g. On average, it is 0.737 cm³/g. Among these, the average pore volume of micropores is 0.067 cm³/g, accounting for approximately 9.466% of the total pore volume. Mesopores contribute roughly 5.947% of the total pore volume, with an average of 0.042 cm³/g. Macroscopic pores have an average pore volume of 0.628 cm³/g, or roughly 84.586%. For the Badawan coal samples, the total pore volume is 0.059 cm³/g, with an average micropore volume of 0.035 cm³/g, which constitutes approximately 57.062% of the total pore volume, while the average macropore volume is 0.025 cm³/g, accounting for about 42.938%. Thus, the primary contributor to the pore volume in the Xishanyao coal samples is macropores, followed by micropores, with mesopores contributing the least. In contrast, the main contributor to the pore volume in the Badawan coal samples is micropores, followed by macropores.

With an average of 216.710 m²/g, the specific surface area of the coal samples from the Xishanyao Formation varies between 200.118 and 238.088 m²/g. Among these, the average specific surface area of micropores is 212.575 m²/g, accounting for approximately 98.027% of the total specific surface area. The average specific surface area of mesopores is 0.399 m²/g, representing about 0.185%, while the average specific surface area of macropores is 3.736 m²/g, contributing approximately 1.788%. For the Badaowan Formation’s coal samples, the specific surface area averages 107.393 m²/g, with a range of 98.598 to 114.241 m²/g. In this case, the micropores’ average specific surface area is 105.014 m²/g, or around 97.855% of the total specific surface area. The average specific surface area of the mesopores is 0.867 m²/g, contributing around 0.778%, while the average specific surface area of the macropores is 1.512 m²/g, accounting for approximately 1.367%. Thus, the primary contributor to the specific surface area in the Baijiahai Uplift is micropores, followed by macropores, with mesopores contributing the least.

As shown in Figure 14, within the Baijiahai Uplift, there are notable variations in the distribution of pore volume at various strata. The pore volume of the Xishanyao Formation’s coal samples is relatively well developed in both micropores and macropores, while the pore volume of the Badaowan Formation’s coal samples is predominantly developed in micropores. Generally, the micropore segment dominates the pore volume and exhibits a multi-peak characteristic. Micropores account for the majority of the specific surface area of the coal samples in the research area, and each coal sample’s specific surface area clearly decreases as pore width increases (Figure 15).

The complicated pore structure of coal, which has pores of different sizes and shapes, is probably the reason why Figure 16 demonstrates that there is no discernible relationship between the adsorbed gas concentration and specific surface area. While a larger specific surface area typically indicates more surface area available for gas adsorption, the shape and distribution of micropores also influence gas adsorption capacity. Coal’s particular surface area and the interactions between gas molecules and the coal surface determine how much gas is adsorbed by the coal. Additionally, the ability of coal to adsorb methane depends on its physical and chemical properties, such as maturity, chemical composition, and structure. In summary, the lack of a significant correlation between the adsorbed gas content and the micropore-specific surface area may result from the complexity of micropore structures, the interactions between gas molecules and the coal surface, the effects of pore size, and various physical and chemical properties of the coal. Conversely, the macropore volume and the free gas content are clearly positively correlated. This is mainly due to the fact that the pore volume has a significant impact on the amount of free gas in coal; this is primarily because the free gas content in coal is largely influenced by the pore volume and larger pore volumes can accommodate more free gas. Coal has a dual pore structure as a porous and heterogeneous medium, with macropores and mesopores having larger throat diameters, serving as storage spaces and flow channels for free gas. In contrast, transitional pores and micropores have relatively smaller throat diameters and are more influenced by pore morphology and connectivity, making their corresponding pore volumes not directly equivalent to effective free gas storage.

5.2.5. Geological Structure

Different geological structures have distinct characteristics. A trap is a location that can prevent oil and gas from continuing to migrate and allow them to accumulate, consisting of three main components: the reservoir layer, the cap rock, and the barriers that impede the migration of oil and gas. These barriers can be the bending and deformation of the cap rock itself, such as an anticline, or obstructions formed by faults or changes in rock types. Previous studies have categorized traps into four main types: structural traps, stratigraphic traps, lithological traps, and composite traps. As a key site for oil and gas accumulation, traps play a crucial role in hydrocarbon formation. Oil and gas can only be stored in a trap if the formation of the trap coincides with or precedes the time of hydrocarbon accumulation, making structure and trapping the primary controlling factors for accumulation.

A detailed 3D seismic interpretation in the study area shows that the Bajiahai Uplift predominantly features three faulted anticline traps (

Table 6. Structural trap elements.

Trap Number	Closed Height (m)	Type	Area (km2)	High Point of Structure (m)	Low Point of Structure (m)
1	15	Faulted anticline	1.95	−1650	−1665
2	50	Faulted anticline	5.60	−1550	−1600
3	20	Faulted anticline	6.07	−1940	−1960

). These traps are located near wells C39, C43, and C2 (Figure 17), covering a cumulative area of 13.62 km², which contributes to the higher gas content near well C39 in the Badaowan Formation.

One of the most prevalent structural forms in geological tectonic movements is the fault, and their appearance disrupts the continuity of coal seams, playing an important role in controlling the localized accumulation of CBM [52,53,54]. The control exerted by faults on CBM accumulation depends on various factors, including the mechanical properties of the faults, the lithology of the rocks on either side of the fault, the size of the fault, and the tectonic stress field [55].

Faults can both constructively create barriers that seal and obstruct flow and destructively act as conduits for fluid movement [56]. Throughout geological history, faults may develop well-permeable fracture zones, becoming pathways for the accumulation and concentration of coalbed methane. Such fault zones facilitate the lateral migration and enrichment of coalbed methane, resulting in a higher gas content near the faults. Conversely, laterally displaced faults may lead to gas leakage or loss; fractures and pores extending along the faults can cause coalbed methane to accumulate toward the fault zones while potentially escaping from these areas. These pathways may also compromise cap rock integrity, affecting the fault’s ability to trap coalbed methane and reducing gas content within the coal seams. Therefore, faults significantly influence the preservation and migration of coalbed methane.

Research indicates that reverse faults can impede gas flow and prevent gas escape, resulting in a higher coalbed methane content, while normal faults can act as conduits, facilitating gas release and leading to lower coalbed methane content. In the eastern part of the study area, the development of reverse faults is favorable for the preservation of coalbed methane [57]. Seismic interpretation profiles of wells such as DN11, DN 18, C49, C504, C17, BJ3, and C32 show that both J₂x and J₁b have faults that connect to Triassic to Permian (T-P ) (Figure 18). The sourced gas migrates into the coal seams of J₂X and J₁b along these pathways, resulting in a higher free gas content near wells such as C32, C36, and C39 in the Xishanyao and Badaowan formations. Thus, the free gas content in the Xishanyao Formation is primarily controlled by faults.

5.2.6. Roof Lithology and Thickness

Sedimentary conditions govern the Junggar Basin’s Jurassic coal-bearing strata, with the primary coal seam’s roof consisting of both dense mudstone and coarse-grained sandstone. Coalbed gas is mainly stored in the coal seam in the form of adsorbed gas, and the gas-retaining and permeability characteristics of the surrounding rock directly affect the conditions for gas accumulation. Good cover conditions can slow down the loss of coalbed gas and indirectly suppress gas desorption, allowing the coal seam to achieve a higher gas content. The denser and thicker the roof and floor rocks are, the more favorable they are for the preservation of coalbed gas [58]. The gas concentration of primary coal seams with mudstone or sandy mudstone as the roof should be higher than those with sandstone roofs because, generally speaking, these materials provide better sealing conditions than sandstone roofs [59]. Additionally, an effective sealing layer must have a certain degree of toughness to avoid crack formation during tectonic deformation, making muddy rocks the best sealing layers for coalbed gas [45].

Taking the relationship between the roof and floor rock types of the Baijiahai Uplift and the gas content of the coal reservoir as an example, the roof rock types of wells C19, C32, C34, C36, and C39 in the Xishanyao Formation are interbedded mudstone and sandstone. In wells C32, C36, and C39, the proportion of mudstone in the roof rock is significantly higher, resulting in a noticeably higher gas content, with a gas content exceeding 20 m³/t in these three wells. Furthermore, a greater thickness of the roof and floor rocks enhances the suppression of vertical gas escape, facilitating gas accumulation in the coal seam.

In contrast, the roof rock types of the Badaowan Formation in wells C19, C32, C34, C36, and C39 primarily consist of sandstone, mudstone, and conglomerate. Compared to the Xishanyao Formation, the proportion of mudstone in the Badaowan Formation’s roof is significantly lower, and the roof thickness is also smaller, leading to a notably lower gas content in the Badaowan Formation’s coal seams compared to those in the Xishanyao Formation (Figure 19).

Additionally, a statistical analysis of the roof rock types and thicknesses of 11 coalbed methane wells in the study area reveals that the main roof rock type of the Xishanyao Formation is predominantly mudstone, which accounts for approximately 74% of the total content. The thickness of the mudstone is generally substantial, with most exceeding 5 m and an average of 8.88 m, resulting in an average gas content of 18.04 m³/t in the coal seams. Sandstone and sandy mudstone make up a smaller proportion, around 26% (Figure 20). The higher percentage of mudstone provides an effective sealing effect for gas diffusion, which is beneficial for the accumulation and preservation of coalbed methane [60]. In contrast, the roof of the Badaowan Formation primarily consists of mudstone (including mudstone and sandy mudstone), which accounts for approximately 52% of the total, comparable to the proportion of sandstone. The thickness of the roof is mostly greater than 5 m, with an average of 7.33 m, resulting in an average gas content of 7.16 m³/t in the coal seams. Overall, both the Xishanyao and Badaowan formations have thick roof layers; however, the higher mudstone content in the Xishanyao Formation results in better sealing capabilities, leading to a higher gas content compared to the Badaowan Formation.

6. Conclusions

The gas content distribution and geological control factors of the Jurassic coal seams in the Baijiahai Uplift of the Junggar Basin were analyzed based on geological data and gas content simulation, leading to the following conclusions:

• The Jurassic coal seams in the Baijiahai Uplift are located at greater depths and have larger thicknesses. The microscopic components of the coal primarily consist of vitrinites, with inertinites being secondary. The Xishanyao Formation is characterized by very low total moisture, very low ash content, and medium volatile matter, while the Badaowan Formation’s coal samples exhibit very low total moisture, low ash content, and high volatile matter. The Xishanyao Formation shows well-developed micropores and macropores, whereas the Badaowan Formation has better developed micropores.
• Overall, the gas content in the Jurassic coal seams of the Baijiahai Uplift is relatively high. The total gas content of the Xishanyao Formation ranges from 16.40 to 23.84 m³/t, while that of the Badaowan Formation ranges from 3.70 to 11.29 m³/t. The Xishanyao Formation displays a high southwest to low northeast gas content distribution, whereas the Badaowan Formation shows a high west to low east pattern. The gas presence in the Xishanyao Formation consists of both adsorbed gas and free gas, while the Badawan Formation primarily contains adsorbed gas, with the Xishanyao Formation exhibiting a higher gas content than the Badawan Formation.
• The gas content of both the Xishanyao and Badaowan formations exhibits a trend of first increasing and then decreasing with depth. However, their correlation with coal seam thickness differs. The gas content of the Xishanyao Formation shows a positive correlation with coal seam thickness, while the Badaowan Formation does not exhibit a significant correlation, potentially due to the stability of the sedimentary environment and the structure and composition of the coal. The adsorption capacity of the Jurassic coal samples is positively correlated with the vitrinite content and negatively correlated with the moisture content and ash yield.
• The Xishanyao Formation has well-developed micropores and macropores, with macropores providing a larger pore volume conducive to the retention of free gas. The specific surface area of both formations mainly derives from micropores, which is favorable for the presence of adsorbed gas. Structural conditions such as traps and faults also play an important role in gas accumulation in the coal seams. Additionally, the roof rock of both formations is primarily mudstone, with considerable thickness, effectively preventing gas escape. Notably, the mudstone content in the Xishanyao Formation is higher than that in the Badaowan Formation, contributing to its greater gas content.