Comprehensive Assessment of Coalbed Methane Content Through Integrated Geophysical and Geological Analysis: Case Study from YJP Block

Abstract

The study block is located on the eastern edge of the Ordos Basin and is one of the typical medium coalbed methane blocks in China that have previously been subjected to exploration and development work. The rich CBM resource base and good exploration and development situation in this block mean there is an urgent need to accelerate development efforts, but compared with the current situation for tight sandstone gas where development is in full swing in the area, the production capacity construction of CBM wells in the area shows a phenomenon of lagging to a certain degree. In this study, taking the 4 + 5 coal seam of the YJP block in the Ordos Basin as the research object, we carried out technical research on an integrated program concerning CBM geology and engineering and put forward a comprehensive seismic geology analysis method for the prediction of the CBM content. The study quantitatively assessed the tectonic conditions, depositional environment, and coal seam thickness as potential controlling factors using gray relationship analysis, trend surface analysis, and seismic geological data integration. The results show that tectonic conditions, especially the burial depth, residual deformation, and fault development, are the main controlling factors affecting the coalbed methane content, showing a strong correlation (gray relational value greater than 0.75). The effects of the depositional environment (sand–shale ratio) and coal bed thickness were negligible. A weighted fusion model incorporating seismic attributes and geological parameters was developed to predict the gas content distribution, achieving relative prediction errors of below 15% in validation wells, significantly outperforming traditional interpolation methods. The integrated approach demonstrated enhanced spatial resolution and accuracy in delineating the lateral CBM distribution, particularly in structurally complex zones. However, limitations persist due to the seismic data resolution and logging data reliability. This method provides a robust framework for CBM exploration in heterogeneous coal reservoirs, emphasizing the critical role of tectonic characterization in gas content prediction.

1. Introduction

As an important basic datum for coalbed methane enrichment evaluation, development area selection, and coalbed methane (CBM) resource reserve calculation, the coalbed methane gas content is an essential parameter in coalbed methane exploration and development which can be obtained directly through measurement, but the measurement results cannot represent the distribution of gas-bearing areas, and the prediction results for the gas content have an important impact on the calculation of the extent of the reserve and the evaluation of the abundance of the resource [1,2,3]. However, due to the strong non-homogeneity of coal reservoirs, the matter of how to accurately predict the gas content has become an urgent problem in CBM exploration and development, so the prediction of the gas content distribution based on geologic and geophysical analyses has received more and more attention [4,5,6,7,8].

Traditional geological methods for predicting the coalbed gas content typically rely on stratigraphic data and CBM parameters obtained from drilling, combined with stress field simulations [9,10]. These methods often employ techniques such as hierarchical analysis, gray fixed-weight clustering, and artificial neural networks to construct evaluation models [11,12]. However, the limited number of drill holes and the subjective nature of weight assignments hinder the precise characterization of the gas content distribution over large areas. Recent advancements in gas content prediction, utilizing logging interpretation, logging parameter preferences, and machine learning methods (e.g., support vector machines), have made considerable progress [13]. Nevertheless, due to the inherent heterogeneity of coal seams, relying solely on logging-based techniques to accurately predict the lateral distribution of the gas content remains challenging [14]. As the variation in a coal seam’s gas content can induce anomalous seismic wave responses, significant research has focused on using seismic data for gas content prediction. Various seismic-based methods have been proposed for this purpose [15]. However, in reflection seismic exploration, the coal seam is typically a thin layer within the coal formation, and the reflection wave anomalies are caused by the interaction between the coal seam and surrounding rock [16]. Consequently, using seismic technology alone for gas content prediction often leads to multidimensional and ambiguous results, making it difficult to meet the practical needs of exploration in most cases.

Based on this, this research was guided by the geological theory of CBM enrichment. The analysis examined the tectonic, sedimentary, coal thickness, and gas content data from the drilling wells in the area and utilized the gray relational analysis and weighted fusion of the main control elements to develop a prediction model for the CBM content based on seismic–geological integration analysis. The distribution characteristics of the gas content in the coal beds of the 4 + 5 coal seam in the YJP block in the eastern part of the Erdos Basin are forecasted, and the effectiveness and feasibility of this method are thoroughly analyzed and discussed.

2. Geological Background

The YJP block is situated at the eastern edge of the Ordos Basin, bordered by the Yellow River to the west and the Lvliang Mountain Range to the east [4]. The block is primarily mountainous, characterized by significant ground undulations. The terrain is inclined from northeast to southwest, and tectonically, it lies within the flexure belt of West Jin, as shown in Figure 1. The geographic coordinates of the study area are 110°29′15″~111°14′15″ E, 38°00′00″~38°40′00″ N.

As shown in Figure 2, no magma intrusion has been observed in the area, and the strata remain stable and relatively undisturbed, which is typical of the stratigraphy in north China, and the thickness of the Carboniferous–Diabase System in the area is generally 500–700 m. The main coal-bearing formations are the Lower Permian Shanxi Formation (P₁s) and the Upper Carboniferous Taiyuan Formation (P₁t). The Shanxi Formation, characterized by deltaic deposition, has a total thickness of 60–90 m and contains the 4 + 5 coal seam. The Taiyuan Formation, which formed during a land–sea interaction phase, has a thickness of 50–90 m and contains the 8 + 9 coal seam. The 4 + 5 coal seam is the dominant seam in the district and was the central focus of this research.

3. Analysis of Main Control Elements of 4 + 5 Coal Gas Content

Generally, sites with pressurized structures are conducive to the enrichment of coalbed methane (CBM). When mudstone is present in the surrounding rock of the coal-bearing section, the permeability is low, the porosity is reduced, and both displacement and driving pressures are elevated. These factors contribute to a more effective seal for CBM within the reservoir [5,17]. In contrast, the higher porosity of sandstone reservoirs results in a lower gas content. Under comparable coalification conditions, the gas content increases with the thickness of the coal seam [18].

This chapter quantitatively characterizes the tectonic conditions, sedimentary environment, and coal seam thickness within the study area. Through cross-plot analysis, we systematically investigated the control mechanisms of these three critical geological factors regarding the gas content characteristics of the No. 4 + 5 coal seam.

In order to improve the accuracy of this research, the process of seismic interpretation was also particularly important; in the preparation of seismic data, the first step is to determine the geological layer of the main reflected waves. This study utilized the acoustic logging data for the study area to produce a synthetic record for the well–seismic data co-calibration, and through the spectral characterization of the seismic data, it could be seen that the main frequency of the seismic data in the study area was about 30 Hz (Figure 3), with a bandwidth of 15~50 Hz. Therefore, a 30 Hz Reich sub-wave and well logging acoustic curve were selected to produce synthetic seismic records for layer calibration.

3.1. Analysis of Role of Tectonic Conditions in Controlling Gas Content

To study the influence of tectonic conditions on the gas content in this area, coal bed depth and seismic attribute data were obtained from the seismic data and interpretation results for the 4 + 5 coal.

(1)

A quantitative characterization of tectonic deformation was conducted by constructing deformation parameters through trend surface analysis.

The trend surface analysis method utilizes the least squares method and other mathematical tools to perform regression fitting. The goal is to decompose an abstract spatial surface into two components: the trend portion, which characterizes the regional overall change and slower variations (trend value), and the residual portion, which represents local anomalous changes and more rapid variations (residual value) [19]. The trend surface reflected the overall depth changes of the coal seam; the larger the trend value, the deeper the coal seam. The residual component indicated the degree of deformation; the larger the absolute value of the residual, the greater the degree of deformation in the tectonic structure. Positive and negative residual values corresponded to the positive and negative deformation of the strata. The trend values and residuals were

$$Z_i=P_0+P_1{f}_1+P_2{f}_2\cdots +P_n{f}_n$$

(1)

$$a_1=Z-Z_i$$

(2)

where Z_i is the trend value (m); P_n represents the undetermined coefficients; f_n is the curve function; n is the maximum fit; a₁ is the residual (m); and Z is the ground elevation.

(2)

We quantitatively characterized the degree of fault development in the region by calculating the fault dimension values using the cartridge dimension method.

The fault fractal dimension is a way of quantitatively evaluating faults based on the theory of dimensioning, which is an emerging nonlinear discipline that utilizes the tools of dimensioning, self-similarity, statistical self-similarity, and power functions to study the similarity of a class of irregular localities to the whole and provides an effective means of quantitatively describing the intricate, complex, and irregular phenomena of the natural world. The fault fractal dimension value represented the degree of fault development within the formation and was calculated using the counting box dimension method [20]. A square grid with varying side lengths, r (where r = L, L/2, L/4, L/8, …, an isometric series with a positive first term and a geometric progression of 0.5), was used to cover the study area. The study block was divided into a square grid with grid side lengths of r = 1000 m, 500 m, 250 m, 125 m, 62.5 m, and 31.25 m to calculate the number of grids covering the faults, N(r), and if the relationship between N(r) and r satisfied the following equation, then the study object would be fractal as

$$N\left(\mathrm{r}\right)=C{\mathrm{r}}^{-D_0},$$

(3)

Taking the logarithm of both sides of the equation yields

$$\ln N\left(\mathrm{r}\right)=-D_0\ln \mathrm{r}+\mathrm{lnC},$$

(4)

where C is a constant; D₀ is a sub-dimensional value.

If ln N(r) and ln r were linear, then the study object would be a fractal, and the fractal dimension value D₀ was the absolute value of the slope of this linear relationship.

By constructing parameters that characterized the tectonic conditions and extracting seismic attribute data, six sets of parameters—namely, the burial depth, trend surface, residual, dip, curvature, and fault dimension—were ultimately obtained, as shown in Figure 4. Of these, the specific well point coordinates are shown in Appendix A.

Figure 4 shows the following maps: (a) the burial depth, (b) trend surface, (c) residual, (d) dip, (e) curvature, and (f) fault dimension. The burial depth (a) and trend surface (b) maps show that the depth of the 4 + 5 coal seam in the study area exhibited a gradual downward trend from east to west, with a shallow burial depth in the east and a deeper burial depth in the west. The maximum depth variation reached over 1000 m. The residual plot (c) shows that the tectonic deformation of the 4 + 5 coal seam in the study area could be divided into two regions: east and west. In the western region, the deformation alternated from north to south, forming a syncline–anticline–syncline pattern, with the residuals being predominantly negative. In contrast, the eastern region was characterized by two broad, gentle anticline patterns, with the residuals being predominantly positive. The residuals exhibited a wide range of both positive and negative variations, indicating significant stratigraphic deformation. The dip, curvature, and fault dimension (d, e, f) show that the tectonically complex zone of the 4 + 5 coal seams in the study area was primarily concentrated in the central, northwestern, and southwestern regions, with a strip-like distribution. Faults developed within this complex zone, where the dip angle of the coal seams could reach up to 20°, accompanied by significant bending. In contrast, the rest of the study area exhibited relatively simple tectonics, with most coal seams having a dip angle of less than 4°, and faults were not developed.

The above parameters of the tectonic conditions were analyzed in relation to the gas content at the well points, as shown in Figure 5.

Figure 5a illustrates that the gas content was influenced by the burial depth, with the general trend indicating that a greater burial depth corresponded to a lower trend surface and higher coalbed methane content. Figure 5b shows that the gas content was affected by the degree of deformation. The gas content tended to be higher in regions with negative residuals (syncline), while areas with positive residuals (positive deformation) exhibited a slight increase in the gas content. Figure 5c–e demonstrate that in regions with a small curvature and low dip, where faults were not developed, changes in the coal seam gas content were minimal. However, in regions with a significant dip and high curvature, and especially at locations of fault development, the gas content was sharply reduced.

To summarize, the tectonic conditions had a significant influence on the gas content, which aligned with the principles of gas control and served as the primary factor governing the gas content of the 4 + 5 coal seam.

3.2. Analysis of Role of Sedimentary Environment in Controlling Gas Content

To study the influence of the sedimentary environment on the gas content in this area, the sand ratio parameter was selected to quantitatively characterize the sedimentary environment before and after the coal aggregation in the 4 + 5 coal seam (Figure 6). The relationship between the gas content in the boreholes and the sand-to-shale ratio was then analyzed using cross-plots, as shown in Figure 7. Figure 7 reveals that the relationship between the sand ratio and gas content, both before and after the coal aggregation, was not significant, suggesting that the sedimentary environment did not exert strong control over the gas content. Therefore, the sedimentary environment was not the primary factor influencing the gas content of the 4 + 5 coal seam.

3.3. Analysis of Role of Coal Seam Thickness in Controlling Gas Content

To study the influence of the coal thickness on CBM enrichment in this area, the relationship between the CBM content and coal thickness was analyzed using cross-plots based on drilling and logging data from the study area, as shown in Figure 8 and Figure 9. Figure 8 demonstrates that the thickness of the 4 + 5 coal seam varied significantly across the region, with a difference of up to 4.6 m. Figure 9 indicates that the coal thickness did not exert strong control over the gas content. Therefore, the coal thickness was not the primary factor influencing the gas content of the 4 + 5 coal seam.

4. Coalbed Methane Content Prediction Based on Integrated Seismic–Geological Analysis

The analysis of the main controlling factors revealed that tectonic conditions were the primary control on the gas content of the 4 + 5 coal seam, while the depositional environment and coal seam thickness were secondary factors. In this chapter, the gas content of the 4 + 5 coal will be modeled and predicted based on tectonic conditions.

4.1. Weighting Analysis

To assess the influence of various tectonic parameters on the gas content of the 4 + 5 coal seam, the gray correlation method was employed to rank the tectonic parameters and identify the primary controlling factors; gray relation analysis is a multifactor statistical analysis method used to determine the degree of influence of the factors in a system on its development. The basic idea is to determine the closeness of the connection by comparing the geometric shapes of the sequence curves, and the more similar the curves are, the greater the correlation is [21], as shown in

Table 1. Gray correlation degree table for different constructive factors.

Factor (x)		Relevance (a)		Arranged in Order	Weights
x5	depth of burial	a5	0.796	1	0.20
x1	residual	a1	0.757	2	0.19
x6	trend surface	a6	0.670	3	0.17
x2	fault dimension (geology)	a2	0.634	4	0.15
x3	dip	a3	0.589	5	0.14
x4	curvature	a4	0.576	6	0.14

. It is generally accepted that when the correlation degree of a particular element is 0.6 or higher, the element can be considered a significant factor that should not be overlooked.

Table 1 shows that the tectonic elements in this area, in a descending order of influence, were the burial depth, residual, fault dimension, dip, and curvature. Among these, the correlation between the burial depth and residual parameters exceeded 0.75, making them the primary factors influencing the overall trend of the coalbed methane content. Additionally, the correlation between fault dimension values exceeded 0.6, indicating that the fault dimension was the main factor driving local variations in the gas content.

4.2. Convergent Modeling

Due to the large number of parameters involved in the main controlling elements, a linear model was constructed to predict the distribution of the gas content in the 4 + 5 coal seam by integrating each controlling element. This was achieved using weighted fusion, with the parameters of the main controlling elements and the gray relational values serving as the data basis. The modeling process was divided into two steps, as follows.

(1)

The Calculation of the Composite Coefficient (K):

The parameters of the main controlling elements were normalized, and the composite coefficients were calculated using the corresponding gray relational values as weights for each element (Figure 10) as

$$K=a1x1-{\displaystyle \sum _2^i{{\displaystyle a}}_i{{\displaystyle x}}_i},$$

(5)

where K is the integration factor; a_i is the weighting coefficient; and x_i is the normalized value of the quantitative parameters of the main control elements.

(2)

The Calculation of the Gas Content Prediction:

The integrated coefficients and the gas content at the well sites were analyzed using the cross-plot method to obtain the fitting function (Figure 11). Gas content predictions were then calculated based on this fitting function as follows:

$$H=3.3085K+4.9422$$

(6)

where H is the predicted value of the gas content.

4.3. Gas Content Prediction and Error Analysis

To evaluate the prediction accuracy of this method, the coalbed methane content prediction model H, along with seismic transverse high-density data, was used to calculate the distribution of the coalbed methane content across the entire region, as shown in Figure 12a. Subsequently, gas content prediction was performed based on drilling data using the traditional interpolation method, as shown in Figure 12b. Finally, an error analysis of the results was conducted using drilling data that were not included in the modeling and kriging interpolation, and the prediction results of the two methods were compared. The comparison results are presented in

Table 2. Error analysis of gas content prediction.

Name of Well	Measured Gas Content/m3·t−1	Seismic–Geological Integration Approach			Traditional Geological Interpolation
		Predicted Gas Content/m3·t−1	Absolute Error/m3·t−1	Relative Error	Predicted Gas Content/m3·t−1	Absolute Error/m3·t−1	Relative Error
L-101	10.11	9.01	−0.5	−4.94%	11.76	−1.65	16.32%
L-102	10.76	6.45	−1.11	−14.68%	6.68	−0.88	−11.64%
L-103	4.33	9.89	−0.87	−8.09%	7.76	−2.13	−19.80%
L-104	7.56	3.85	−0.48	−11.08%	5.53	1.2	27.71%

.

(1)

The CBM content prediction results from both methods showed a consistent trend, with the 4 + 5 coal’s CBM content decreasing from west to east. The two methods exhibited a strong correspondence, indicating that this method provided a good indication of the overall coalbed methane content changes in the coal seam.

(2)

The relative error in gas content prediction for all validated well sites using the integrated seismic–geologic method was less than 15%. In contrast, the relative error for most validation wells when using the traditional geological interpolation method exceeded 15%, reaching a maximum of 27.71%. This demonstrates that the proposed method was effective in predicting the gas content of the 4 + 5 coal in the study area, providing significantly improved accuracy compared to the traditional method. However, due to limitations in the quality of the logging and seismic data, large prediction errors for the local coalbed methane content persisted.

5. Conclusions and Future Directions

5.1. Conclusions

(1)

This research proposes a seismic–geological integrated analysis method based on key control elements to predict the gas content. The method was validated using data from the YJP block, where the error of the traditional difference method was 27.72% and the relative error between the wells not involved in the validation and the prediction results obtained using the method was controlled within 15%. This confirms the validity and feasibility of the method and its improved accuracy for coalbed methane content prediction.

(2)

The primary controlling elements influencing the gas content of the 4 + 5 coal in the study area were the tectonic conditions. Among these, the burial depth was positively correlated with the gas content, while the trend surface, dip, curvature, and fault dimension values were negatively correlated with the gas content. The coalbed methane content was generally higher in areas with negative residual values, such as syncline areas, and increased in areas with positive residual values, such as anticline areas. The correlation between the sedimentary environment, coal seam thickness, and gas content of the 4 + 5 coal was weak, indicating that these factors had a lesser impact on the gas content. Therefore, the tectonic conditions were the main controlling factor regarding the coalbed methane content.

5.2. Discussion

Compared to traditional drilling and logging interpolation methods, the integrated seismic–geological analysis method based on key control elements offers significant spatial and accuracy advantages. This is reflected in the higher lateral resolution of gas content prediction across the study area and a smaller relative error compared to that of the traditional method. This method can provide an effective framework for the exploration of complex tectonic areas. However, this method also has limitations: it relies on seismic, drilling, and logging data, and the prediction results are significantly affected by the resolution of the seismic data and the reliability of the drilling and logging data, and the requirements for the seismic data and drilling and logging data are high. The accuracy of this method for CBM content prediction will be continuously improved in future studies.