Characteristics of In Situ Stress and Reservoir Pressure in Deep Coal Seams and Their Influences on Reservoir Depletion: A Field Case Study

,


Introduction
Coalbed methane (CBM), the main component of CH 4 , can be released from coal seams during subsurface mining and ground well extraction.CBM development promotes the safety of coal mining and the consumption of natural gas resources [1,2].In addition, known as a kind of potent greenhouse gas, CBM extraction and utilization are estimated to reduce carbon emissions into the Earth's atmosphere [3][4][5].
Deep coalbed methane (DCBM) resources have become the focus of CBM exploration and development.According to the previous study, CBM resources with depths over 1000 m of coal seams account for 61.2% of the total reserves in China [6].The Uinta, Greater Green River, and Hanna basins in the United States, and the Alberta Basin in western Canada have undergone DCBM development, and some of these areas have been suitably commercialized [7,8].In recent years, in China, some DCBM wells in Fukang, Wuxiang, Dacheng, and southern Yanchuan areas have also achieved high production, indicating that the DCBM resources in China exhibit the possibility of high production.However, the production effect of DCBM wells in China remains unsatisfactory.The gas production rate of quite a few gas wells is less than 1000 m 3 /d, which greatly restricts the development of the CBM industry.Compared to shallow reservoirs, DCBM reservoirs are generally characterized by higher in situ stress, reservoir pressure, and temperature levels, and lower permeability, making reservoir extraction more difficult [9,10].Moreover, the geological conditions of DCBM reservoirs have not been fully explained to date.A generally accepted drainage control system has not yet been established.
Many previous studies have been carried out on geological characteristics such as structure, sedimentation, rock physical properties, and gas occurrence state [11,12].Certain factors notably affect the production of fluid within DCBM reservoirs.For example, coal permeability is known as one key factor affecting fluid (gas and formation water) movement.According to the experimental studies, coal permeability decreases exponentially with the increase of effective stress, and low permeability in the DCBM reservoirs makes fluid flow in the fracture network very difficult during reservoir depletion [13,14].For DCBM reservoirs, characterized by a large difference in the horizontal principal stress, the stimulated reservoir volume is small [10,15].The maximum and minimum horizontal principal stresses as well as vertical stress have positive correlations with the burial depth of coal seams, and the increase of horizontal principal stress can reduce the reservoir permeability [16].Therefore, the in situ stress distinctly restricts the extension of the pressure relief range during the gas production process.Additionally, gas saturation and gas content increase with the burial depth of the coal seam, and this has a positive effect on the production of CBM in the deep coal seam [17].The fluids production rate varies obviously with the depths of coal seams [18].Chen et al. [19] studied the structure of the Ordos Basin and presented three types of geological structures with a high production of DCBM wells, including structural highs, axial parts of anticline and nose structures, and monocline updip structures.To improve the production efficiency of DCBM wells, geological investigations should be closely combined with development engineering to guide CBM production.Therefore, it is urgent to further analyze the fluid production characteristics of DCBM reservoirs and explore the intrinsic connections between fluid production in deep coal seams and the main controlling geological factors.Additionally, these studies can promote the optimization of CBM extraction and the establishment of a suitable depressurization method during DCBM reservoir depletion.However, the influences of stress and reservoir pressure on the fluid production in the DCBM reservoirs have not been further investigated [10,20].Moreover, the field production data of DCBM wells is probably employed to study the implications of stress and reservoir pressure for DCBM depressurization.It is helpful to provide a reference for the formulation of suitable depressurization design aspects depending on the specific conditions of deep reservoirs.
The objective of this study is to investigate the stress and reservoir pressure properties of DCBM reservoirs and determine their effects on reservoir depressurization.We analyze the fluid production characteristics of DCBM wells and investigate the implications of stress and reservoir pressure for DCBM production, in the eastern margin of the Ordos Basin (EMOB).On this basis, a recommended depressurization method for DCBM reservoirs during reservoir depletion is proposed.This study is aimed at providing guidance for the depressurization control of DCBM reservoirs.

Research Location and
Investigation Procedure The coal rank in the study area largely encompasses middle-rank coal, rising gradually from north to south and becoming high-rank coal in the southeast margin.The logging porosity and gas content in the coal seams vary between 2.7% and 6.3% and between 5 and 20 m 3 /t, respectively [19].The karst aquifer of the Ordovician System in the area is rich in water, and the water level is generally higher than that of the coal seam.Generally, according to the change characteristics of gas and water production of CBM wells, the production process can 2 Geofluids be divided into four stages, including the single-phase water production stage, rapid-growth gas production stage, stable gas production stage, and declining gas production stage [23].Among these stages, the last three stages comprise the gas production stage.Due to the artificial influence of drainage equipment damage, abnormal power failure, and  3 Geofluids incorrect production control, the production curves of many CBM wells do not fully reflect the geological conditions and are thus not representative.Therefore, we select typical wells, free from abnormal interference, to analyze the duration of single-phase water production.As shown in Figure 2, the daily gas production volume of DCBM wells is only approximately 500 m 3 /d, indicating a very low gas production rate.The duration of the single-phase water production stage in this area varies between 7 and 462 d, with a large difference.The average duration of this stage is calculated to reach 211 d, indicating a long initial dewatering period.Regarding the shallow CBM wells near the study area, the duration of the first production stage is generally shorter than 100 d, while for 72% of the CBM wells in the EMOB, the duration exceeds 100 d.

Difficult Stable Gas Production
Stage.The duration and fluid production at different gas production stages for typical DCBM wells were measured, as indicated in Table 1.The results suggest that after reaching a gas production peak, the wells in the study area do not experience the stable gas production stage but directly enter the declining gas production stage.Therefore, it is very difficult to maintain a stable gas production level in DCBM wells, and the declining stage occurs too early.

Consistent Variation Trend of the Gas and Water
Production Curves.Among shallow CBM wells, there exists a complex relationship between the gas and water production rate curves.When the production process of the DCBM wells in the study area reaches the rapid gas production growth stage, the changing trend of gas and water production curves is highly consistent, both showing an initial rapid increase and a subsequent gradual decrease, as shown in Figure 3.The data in this figure were obtained from field test by China United Coalbed Methane Co. Ltd.Even when fluid production is discontinuously induced by artificial operations, the change patterns of gas and water production are uniform.As the CBM production process involves dewatering depressurization, the consistent gas and water production variation trends indicate that water production distinctly restricts the production of gas.Furthermore, at a specific time (please refer to the arrows in Figure 3), the gas production rate declines with an abrupt decrease in the water production rate, which indicates the flow channel in the reservoir is single.This occurs because the water production rate can reflect the reservoir flow conductivity.A highwater production rate indicates unimpeded fluid migration channels and easy gas flow.A low water production rate suggests blocked fluid migration channels, with gas hardly produced continuously.In multichannel reservoirs, CBM can flow along various channels.Therefore, the gas production rate is not restricted by water production.

In Situ Stress and Reservoir Pressure of Deep Coal Seams.
As mentioned above, there are great differences in geological conditions between deep and shallow reservoirs.Consequently, the gas and water production patterns of DCBM wells exhibit different characteristics.In this work, in situ stress, initial reservoir pressure, and critical desorption pressure are employed to explain the potential impact on the fluid production characteristics of DCBM wells.
In situ stress.The in situ stress significantly influences the stability of the wellbore surface and fracture propagation, but its implication for DCBM depressurization should not be ignored [20,24,25].For example, the analysis results indicate that the minimum horizontal principal stress in the No. 8 and No. 9 coal seams in the Linxing block in the EMOB ranges from 18.44 to 38.99 MPa, and the maximum horizontal principal stress varies between 25.30 and 52.88 MPa [26].A high in situ stress leads to greater formation compaction during reservoir depressurization, as expressed in Equation (1), and the extension of the pressure disturbance and effective desorption range during gas production is limited [27,28].During the DCBM production process of the southern Yanchuan block in the southeastern margin of the Ordos Basin, the drainage radius of a single CBM well is calculated to reach only 85 m, which is much smaller than the drainage radius of 167 m in low-stress areas.The high in situ stress conditions of the deep coal seams in the study area increase the effective stress during the dewatering process, and reservoir diversion channels are more easily closed and blocked, which restrict gas migration.This is thought as a very key reason for the low production rate of DCBM wells in the study area.Therefore, it is considerably difficult to extract gas resources from reservoirs under high in situ stress with the present techniques.Therefore, the effect of a high in situ stress should be fully considered before DCBM production.
where σ e is the effective stress, MPa; σ is the total stress, MPa; α is the effective stress coefficient; P is the dynamic reservoir pressure.Initial reservoir pressure.The initial reservoir pressure and critical desorption pressure are key parameters in the production process of CBM wells, indicating the gas output time and production duration, respectively, of CBM wells.The reservoir pressure reflects the energy contained in the reservoir and is an important driving force of CH 4 desorption and flow in the fracture system.The reservoir pressure varies in both spatial and temporal dimensions during gas production [29][30][31].The initial reservoir pressure generally increases with increasing coal seam buried depth, indicating high pressure in deep formations.In this study, the initial reservoir pressure is estimated based on the hydrostatic column pressure before production, as listed in Table 2.This table indicates that the DCBM reservoirs in the study area exhibit a high pressure, with a pressure gradient ranging from 8.9 to 9.8 kPa/m, and most reservoirs are normally pressurized or slightly under pressurized.Based on previous work, the pressure in the shal-low CBM reservoirs in the Qinshui Basin near the EMOB is relatively low, generally ranging from 2.71 to 6.25 MPa, with the reservoir pressure gradient varying between 3.8 and 8.8 kPa/m.Therefore, the initial reservoir pressure and pressure gradient of deep coal seams are much higher than those of shallow CBM reservoirs.This demonstrates that DCBM reservoirs potentially experience a longer duration of the production process.
Critical desorption pressure.The critical desorption pressure indicates the cycle of the gas production stage.The critical desorption pressure of CBM reservoirs probably varies greatly among different regions.The critical desorption pressure is a key parameter controlling the decline rate of the dynamic liquid level in each wellbore at the single-phase water production stage.The critical desorption pressure of the deep coal seams determined through laboratory tests ranges from 0.86 to 13.27 MPa, with a large difference, as summarized in Table 3.This indicates that there also occurs a large gap in the depressurization degree between different gas wells before gas outflow.If the difference between the critical desorption pressure and the initial reservoir pressure (DCI) is large, this indicates that the reservoir energy is greatly lost when CH 4 is desorbed from the coal matrix.This counteracts the high and stable gas production of CBM wells.Moreover, the large difference between the critical desorption pressure and the initial reservoir pressure suggests that the reservoir pressure should be notably reduced before entering the gas production stage.A large pressure difference not only greatly increases the duration of the first  6 Geofluids production stage but also increases the risk of fracture closure due to the notable increase in effective stress with continued dewatering [32].This analysis objectively explains the phenomenon whereby the initial dewatering stage of DCBM wells persists for a long time and the difficulty achieving stable production, as shown in Figure 2 and Table 1 in Section 3. Similar to the DCI, the gas saturation of CBM reservoirs indicates the difficulty of reducing the initial reservoir pressure to reach the critical desorption pressure.According to statistical data shown in Figure 4(a), it is found that with decreasing gas saturation, the DCI value increases.It suggests that a long time is required to induce CH 4 desorption [34].Therefore, gas saturation can be adopted as an indicator to evaluate the duration of the initial dewatering stage.The test results indicate that the gas saturation in the different wells in the study area ranges from 31.5% to 114.7%, with a large gap.For the coal seam with low gas saturation, the reservoir pressure drop range is larger in the single-phase water production stage.However, for the coal seam with high gas saturation or supersaturated coal seam, the water production stage may be absent with the gas production stage directly starting.Therefore, the dewatering duration of the first stage should be dramatically different; the same depressurization design should not be adopted in all wells.
Calculation of the gas saturation requires the determination of the parameter of the initial reservoir pressure, expressed as follows: The initial reservoir pressure acquires hardly and usually misses in practice, but the other parameters in Equation ( 2) can be tested in the laboratory.
The critical desorption pressure can be expressed as As mentioned above, in the study area, both the critical desorption pressure data and gas saturation data show a large-gap pattern.Furthermore, we find that there exists an Symbols: P i is the initial reservoir pressure, MPa; V L is the Langmuir volume, m 3 /t; P L is the Langmuir pressure, MPa; V i is the initial gas content, m 3 /t; P cd is the critical desorption pressure, MPa; S g is the gas saturation, %. 7 Geofluids obvious positive correlation between the gas saturation and critical desorption pressure (Figure 4(b)).It demonstrates that in deep coal seams, the critical desorption pressure can be directly used to evaluate the duration of the singlephase water production stage instead of the gas saturation.In other words, the duration of the first production stage can be evaluated in the absence of initial reservoir pressure data.Additionally, the critical desorption pressure is easier to obtain than gas saturation, as the initial reservoir pressure is tested hardly in practice.

Depressurization Method of DCBM Reservoirs.
As CH 4 in CBM reservoirs mainly exists in the adsorbed state, depressurization is the main way to induce CBM desorption and migration.Water transfer and gas transfer are the basic transfer modes of reservoir depressurization during CBM production.Dewatering with depressurization exhibits good adaptability to shallow CBM reservoirs and is widely applied in CBM extraction.However, the basic geological conditions and fluid production characteristics of DCBM reservoirs distinctly differ from those of shallow reservoirs, as mentioned in Sections 1-3.Based on the implications of stress and reservoir pressure for DCBM depressurization, in this study, it is considered that the depressurization method involving only the dewatering of shallow coal seams hardly exploits DCBM resources efficiently.We propose that the major reasons are as follows: (1) Under the high in situ stress conditions of deep coal seams, the damage exerted by the effective stress on the fracture system during dewatering increases distinctly, which greatly limits CBM migration and outflow (2) A large gap between the initial reservoir pressure and critical desorption pressure results in a longer duration of the dewatering stage for DCBM wells, which reduces the depressurization efficiency and increases energy consumption Therefore, the depressurization method should be modified to improve the efficiency of pressure reduction in DCBM reservoirs.Eliminating the negative effect of high in situ stress may be an effective way to overcome the challenge of DCBM production.We propose that the depressurization of DCBM reservoirs during production should be performed via stress release.We think that in situ stress release can be facilitated via artificial means.In situ stress variation causes formation movement with increasing fracture opening and fracture number [35].On this basis, the reservoir permeability is greatly improved.Moreover, as both stress and reservoir pressure decrease in this process, the effective stress reduction during reservoir depletion could vary slowly (Equation (1)).Consequently, it reduces the damage to the conductivity of the fracture system and benefits desorption extension.
In situ stress can be released by artificial space establishment within the coal seam.It is suggested that multiple slots or cavities should be formed in the coal seam through the high pressure hydraulic jet method, which provides space for stress release.As shown in Figure 5  8 Geofluids deformation of coal seams, expand the number of seam pores and fractures, and ultimately promote the drop of reservoir pressure.The advantage of this method is that it significantly increases the number and opening degree of flow channels, improving reservoir permeability.Moreover, due to the stress change of coal seams, the effective stress acting on the fracture system is effectively reduced which can avoid reservoir damage during CBM production.Segmented hydraulic slotting or cavity building along a directional wellbore is a potentially valuable engineering method to realize stress release (Figure 5(b)).The length, width, height, dip angle, and other morphological parameters of the pressure relief space can be calculated and designed according to the detailed geological conditions, such as the stress state, geological structures, and physical properties of rock.A FLAC 3D simulation is performed to study the stress release differences between the general borehole and the slotted borehole.The main initial conditions in the simula-tion are shown in Table 4 [36].The result shows that the stress-disturbed region around the slotted wellbore is distinctly larger than that of the general wellbore in the subsurface coal mines.In addition, the stress release value of the general borehole is smaller than that of the slotted borehole (Figures 6(a) and 6(b)).Furthermore, based on the field test data from the Pingdingshan mining area near the Ordos Basin, the gas flow rate greatly increases by a hydraulic jet slotting method, as shown in Figures 6(c) and 6(d) [37,38].In this mining area, the depths of the target coal seams are also large, exceeding 1000 m in many areas.These studies indicate that the hydraulic jet method can achieve a remarkable depressurization effect for coal seams and has potential application in stress release for deep coal seams.
Hence, the application of the depressurization method based on the stress release probably greatly enhances the rate and degree of reservoir pressure decline in the production process.This is beneficial for the realization of the efficient 9 Geofluids development of DCBM resources.Furthermore, the stress release method is widely used for coal seam stimulations in underground coal mines, and the hydraulic jet slotting technology has been developed quite well [39,40].Therefore, it provides a potential possibility for the application of the stress release method in the surface development of DCBM resources.However, during the stress release process, the design of slot or cavity parameters is aimed at increasing the stress release range of coal seams as much as possible.This requires effectively controlling the slot or cavity parameters, including the length, width, strike, inclination, and spacing.In addition, during the establishment of slots or cavities in coal seams, a large number of solid particles need to be drained back.Furthermore, avoiding fluid power loss in the DCBM wellbores is also a prerequisite for the implementation of the high-pressure hydraulic method.These are the main challenge of the stress release method for DCBM depletion.

Conclusions
In this study, the fluid production characteristics of DCBM wells and implications of in situ stress for reservoir depletion are analyzed.Our study indicates that the in situ stress, initial reservoir pressure, and critical desorption pressure are important properties that predominantly control the depletion process of DCBM reservoirs.On this basis, a new depressurization method for DCBM reservoirs is examined.The main conclusions are as follows: In terms of a field case investigation in the EMOB, North China, the duration of the single-phase water production stage of DCBM wells varies, but the overall duration is longer.Stable gas production is difficult to achieve, and an early decline occurs.The variation trends of the gas and water production curves are consistent.A high in situ stress is likely to lower the flow conductivity of the fracture system due to the increase in effective stress during gas production, which limits CBM migration and outflow from reservoir channels.The high DIC value and low gas saturation, in principle, explain the universal phenomenon of a long duration of the initial dewatering stage in DCBM reservoirs.
Based on the in situ stress characteristics of DCBM reservoirs and their implications for reservoir depletion, we propose that reservoir depressurization can be promoted via stress release with hydraulic jet slotting or cavity building.It can help greatly release the stress and improve the reservoir permeability, which avoids the conductivity damage attributed to the effective stress increase.In other words, this proposed depressurization method could help to enhance the depressurization efficiency of DCBM resources.It is suggested to perform further research to make this proposed depressurization method apply to the practice of DCBM extraction.

Figure 1 :
Figure 1: Location and major geological structures in the EMOB.(a) Structural units of Ordos Basin.(b) Location of Ordos Basin.(c) Major geological structures.

Figure 3 :
Figure 3: Fluid production curves of one typical DCBM well.

Figure 4 :
Figure 4: Relationships among gas saturation, DCI, and critical desorption pressure.(a) Gas saturation and DCI.(b) Gas saturation and critical desorption pressure.

Figure 5 :
Figure 5: Sketch of depressurization induced by stress release.(a) Stress release by hydraulic slots method.(b) Segmented hydraulic slotting or cavity building along the directional wellbores.

Figure 6 :
Figure 6: The differences in gas production effect for general borehole and slotted borehole.(a) Vertical stress distribution of general borehole.(b) Vertical stress distribution of the slotted wellbore.(c) The difference in gas flow rate with varied time.(d) The difference in gas flow rate with varied boreholes.

Table 2 :
Hydrostatic column pressure and pressure gradient in the study area.

Table 3 :
Initial gas content and adsorption isothermal parameters of DCBM reservoirs.

Table 4 :
The main conditions in the simulation.