Potential assessment of CO2 geological storage based on injection scenario simulation: A case study in eastern Junggar Basin

Xin Ma, Dong-guang Wen, Guo-dong Yang, Xu-feng Li*, Yu-jie Diao, Hai-hai Dong, Wei Cao, Shuguo Yin, Yan-mei Zhang

1 Center for Hydrogeology and Environmental Geology Survey, China Geological Survey, Baoding 071051, Hebei, China.

2 Key Laboratory of Carbon Dioxide Geological Storage, China Geological Survey, Baoding 071051, Hebei, China.

3 College of Resources and Environmental Engineering, Wuhan University of Science and Technology, Wuhan 430081, China.

4 Hubei Key Laboratory for Efficient Utilization and Agglomeration of Metallurgic Mineral Resources, Wuhan University of Science and Technology, Wuhan 430081, China.

5 Exploration and Development Research Institute, PetroChina Xinjiang Oilfield Company, Karamay 834000, China.

6 Institute of Geology of Pudong Oil Production Plant, Sinopec Zhongyuan Oilfield, Puyang 457001, China.

7 Exploration and Development Research Institute of Zhundong Oil Production Plant, Petro China Xinjiang Oilfield Company, Fukang,831511, Xinjiang, China.

Abstract: Carbon Capture and Storage (CCS) is one of the effective means to deal with global warming,and saline aquifer storage is considered to be the most promising storage method. Junggar Basin, located in the northern part of Xinjiang and with a large distribution area of saline aquifer, is an effective carbon storage site. Based on well logging data and 2D seismic data, a 3D heterogeneous geological model of the Cretaceous Donggou Formation reservoir near D7 well was constructed, and dynamic simulations under two scenarios of single-well injection and multi-well injection were carried out to explore the storage potential and CO2 storage mechanism of deep saline aquifer with real geological conditions in this study.The results show that within 100 km2 of the saline aquifer of Donggou Formation in the vicinity of D7 well,the theoretical static CO2 storage is 71.967 × 106 tons (P50)①, and the maximum dynamic CO2 storage is 145.295 × 106 tons (Case2). The heterogeneity of saline aquifer has a great influence on the spatial distribution of CO2 in the reservoir. The multi-well injection scenario is conducive to the efficient utilization of reservoir space and safer for storage. Based on the results from theoretical static calculation and the dynamic simulation, the effective coefficient of CO2 storage in deep saline aquifer in the eastern part of Xinjiang is recommended to be 4.9%. This study can be applied to the engineering practice of CO2 sequestration in the deep saline aquifer in Xinjiang.

Keywords: CO2 geological storage; Deep saline aquifer; Potential assessment; Injection scenarios;Numerical simulation; Junggar Basin

Introduction

The massive emission of greenhouse gases, mainly CO2, has caused many environmental and ecological problems (Wang et al. 2021). Reducing CO2emission has become a common concern domestically and internationally. The experience from European and American countries and Japan shows that Carbon Capture and Storage (CCS) is one of the effective technologies to deal with global warming (Bachu, 2003; IPCC, 2005), The international community has more than ten years of practical experience in industrial scale CO2geological sequestration projects. The Sleipner and Snovit projects in the North Sea of Norway have injected nearly 20 × 106tons of CO2(GCCSI,2016). There are a large number of sedimentary basins on both the land and continental shelf of China, the wide distribution and extensive thickness of which provide a great potential for CO2storage (Li et al. 2006; Wen et al. 2013; Guo et al.2015; Li et al. 2016).

Many research organizations or scholars have proposed methods for evaluating the potential of CO2geological sequestration in deep saline aquifer. Both the Carbon Sequestration Leadership Forum (CSLF) and the US Department of Energy(USDOE) have proposed standards and basic methodologies for assessing CO2sequestration capabilities (CSLF, 2005, 2007, 2010; DOENETL, 2006, 2008, 2010). Bachu (2003) proposed the mechanism of CO2sequestration in saline aquifer and obtained the CO2sequestration by calculating the maximum CO2dissolution of formation water (IPCC, 2005)，but due to a lack of data, there would be deviation in the assessment of the storage potential of saline aquifer (Bachu et al. 2007); Goodman et al. (2011) developed a set of more specific estimation method for estimating CO2sequestration resources on regional scale based on USODE method, while Li et al. (2015),Diao et al. (2017) used GIS method in a basinscale assessment. Some other researchers used numerical simulation, combined with field data,indoor experiment and field tests to estimate the relevant parameters based on different scenarios (Xu et al. 2010; Thomas et al. 2012; De Silva et al. 2012; Li et al. 2014; Li et al. 2015; Lee et al.2016; Jin et al. 2017; Wen et al. 2019). Current studies on CO2saline aquifer storage are mainly based on homogeneous saline aquifers (Flett et al.2007, Oh et al. 2013, Liu et al. 2016). Ma et al.(2018) constructed a two-dimensional heterogeneous model and randomly generated heterogeneous model to study the impact of the evaluation results on the heterogeneity potential, and there was a certain deviation between the research results and the CO2storage potential under the influence of actual heterogeneity. Sedimentary basins in China are dominated by continental sediments with complex structures, low porosity and low permeability,and their spatial distribution is generally stratified,showing strong heterogeneity. Existing studies lack the support of fine static reservoir models, so it is difficult to do a reliable potential assessment at the site scale.

The Junggar Basin is located in the northern part of Xinjiang and is the second largest inland basin in China. It is composed of 6 primary structural units, of which the Central Depression is divided into 10 secondary structural units (Fig.1a). According to the CO2emission reduction requirements in eastern Junggar Basin and the geological conditions, China Geological Survey carried out a large number of comprehensive geological surveys of CO2storage near D7 well in the Fukang Depression in 2016 (Fig.1b), including two-dimensional seismic survey. The survey results show that the area has a stable sedimentary environment, with good reservoir and caprock conditions, which is suitable for CO2geological storage. Based on geological survey results, this study focuses on the Junggar Basin, Xinjiang, which has great potential for CCS development in China, and takes an area of 10 km × 10 km in the eastern Junggar Basin as an example site according to the scope of seismic exploration. With well logging and two-dimensional seismic profile data, a three-dimensional heterogeneous static geological model was established. The potential of CO2geological storage was evaluated under two injection scenarios to summarize more scientific potential evaluation methods.The target reservoir is the Cretaceous Donggou Formation in the eastern Junggar Basin, with an elevation from -1 418.40 m to -1 821.88 m and a sedimentary thickness of 340.48 m (Ma et al. 2018,Wen et al. 2019). No regional structure was developed in the study site, and the sedimentary types mainly include fine sandstone, siltstone and mudstone, and the sedimentary environment is delta-lake. The thick mudstone developed in the upper part of the Cretaceous Donggou Formation has low porosity and permeability, which can be used as a natural barrier for CO2storage by saline aquifer in the Donggou Formation.

Fig.1 Location map of the study area and the distribution of the 2D seismic profile (Mi et al. 2018)

1 Potential assessment approach

USDOE (DOE-NETL, 2010) method was adopted to assess the CO2storage potential:

Table1 Recommended values for Esaline (DOENETL, 2010; Bachu, 2015)

2 Site characteristics and modelling parameters

2.1 Site characteristics

2.1.1 Lithology

The study area is located in the Fukang Depression, the secondary structural unit of the Junggar Basin, with a stable sedimentary environment and no recent tectonic development. The target stratum mainly consists of fluvial deposit, with gentle sedimentation and a dip angle of 3°-5° slightly to the southeast, and dome structures are developed in some areas (He et al. 2002). The total thickness of the Cretaceous Donggou Formation in the eastern Junggar Basin is 356 m, which is mainly composed of mudstone, sandy mudstone with siltstone and argillaceous siltstone. According to the well logging data (D7), the elevations of the top and bottom of the Donggou Formation in the study area are -1 481.40 m and -1 821.88 m, respectively(Fig.2). The overall permeability showed a trend of high value (maximum 441 mD) at the upper and very low at the bottom (minimum 0.001 mD) (Ma et al. 2018). The porosity had the same trend, and varied from 0.01 to 0.30 (Wen et al. 2019). At the top of the Donggou Formation, there is a layer of thick mudstone and silty mudstone (with high shale content) with a cumulative thickness of about 60 meters, which can be used as regional caprock for CO2storage in saline aquifer. Sandstone, siltstone and argillaceous siltstone (with low shale content) are interbedded with unequal thickness in the middle and lower part of the reservoir (Fig.2). The accumulated sedimentary thickness is more than 200 meters, which can be regarded as good reservoir.

Fig.2 Stratigraphic histogram of Cretaceous Donggou Formation (Modified from Yang (2019))

2.1.2 Hydrogeology

The chemistry of the groundwater in the target layer in the study area can be described as Cl-Na type,with pH of 7.5, salinity between 35 g/L and 44 g/L(Ma et al. 2018), and chloride ion concentration between 20 g/L and 26 g/L. The water temperature is about 66℃ (Wen et al. 2019).

2.2 Modelling parameters

The top and bottom elevations of the Donggou Formation around the D7 well area are -1 481.40 m and -1 821.88 m respectively. According to the logging data, the temperature and pressure at the elevation of -1 540 m, in the middle of the model, are set to be 66℃ and 20.6 MPa. The pressure gradient of the formation is about 1.1 MPa /100m, and the fracture pressure of the target reservoir is about 50 MPa. The porosity is 0.010-0.296, and the permeability is 0.001 441 mD (Ma et al. 2018). For the relative permeability model of water and gas, the irreducible water saturation was set to 0.2, while the residual gas saturation set to 0.0 (Fig.3) and the maximum residual gas saturation was set to 0.2. Krw is the relative permeability of the water phase; Krg is the relative permeability of the gas phase; Sw is the water saturation. Henry’s law is used for modeling CO2phase behavior and solubility in the aqueous phase.

Fig.3 Relative permeability model for numerical simulation

Where: fiis the fugacity of component i in the gaseous phase, xiis its mole fraction in the aqueous phase and Hiis the Henry’s law constant of the component.

2.3 Static model

The data inversion was carried out using the well logging data of D7 well and three 2D seismic profiles measured by China Geological Survey in 2016 (Fig.4a). The seismic data shows that the top of the formation has a small amplitude, which indicates that the shale content is high and can be used as a regional caprock; the lower part has a large amplitude, which indicates that the shale content is low and can be used as a reservoir. The well logging data shows that mudstone is relatively well developed at the top of Donggou Formation (Fig.4b,4c) according to the yellow line area at the top of the seismic profiles, and the arenaceous bodies in the middle and upper part of the formation are banded, which is consistent with the measured 2D seismic data. Through the difference fitting analysis, the fitting data were imported into the well grid, and the logging data were basically consistent with the permeability and porosity data in the grid fitting data (Fig.4b, 4c).

Fig.4 Spatial distribution of 2D seismic data and mesh data fitting of borehole

According to the scope of the study area, a 10 km × 10 km static model grid was built around the D7 well, with the grid dimension of 150 m × 150 m ×5 m, and a total of 79 layers and 338 910 grid cells.Based on well logging and 2D seismic data, the permeability and porosity models of the study area were developed by using collaborative simulation method (Fig.5). The permeabilities in X and Y directions are the same, while the permeability in Z direction is 0.3 times that in X direction (Ma et al. 2018,Wen et al. 2019).

Fig.5 Geological models of the study area (the left is porosity model, and the right is permeability model)(Modified after Wen et al. 2019)

3 Potential assessment results

3.1 Modelling scenarios

In previous studies, single-well and multi-well injections were commonly used for the dynamic potential evaluation of CO2storage in deep saline aquifers (Wen et al. 2019, Ma et al. 2018, Yang et al.2019). In the future, million tons of CO2will be pumped into and stored in the geological formations in eastern Junggar Basin of Xinjiang to reduce CO2emission. In the actual engineering practice stage, single-well injection or multi-well group injection mode is generally adopted. In order to effectively support the future CO2geological storage engineering practice in eastern Junggar Basin,the scenarios of both single-well and multi-well injections are applied in this study to obtain the CO2sequestration potential and related parameters of the saline aquifer in Donggou Formation by numerical simulation (Fig.6).

Fig.6 3D geological model and well location distribution (Case 1 on the left, Case 2 on the right)

Scenario 1 (Case 1): In this case, one injection well and one production well are simulated for the duration of 300 years, to monitor whether there is leakage of CO2during the injection process. The injection well is the D7 well with the perfprated zone mainly located in the middle to lower part of the well. The injection system is unified, and the maximum allowable injection pressure at the bottom of the well is less than 50 MPa (When the injection pressure is greater than 50 MPa, the injection will be stopped). In this model, the designed CO2injection rate is 2 × 106tons per year for 50 years, followed by 250 years of continuous simulation. The production well is set at the top of the Donggou Formation, and its maximum allowable pressure at the bottom of the well is 50 MPa.

Scenarios 2 (Case 2): In this case, five injection wells and one production well, are simulated for the duration of 300 years. The main injection well is located in the center, while the other four wells equidistantly distribute on the circumference of a circle (R = 3 km) around the D7 well. The perfprated zone is at the lower part and the injection system is unified. The same maximum allowable injection pressure is applied as in Case 1. The designed CO2injection rate is also 2 × 106tons per year but evenly shared by 5 wells (400 000 tons/a per injection well) for 50 years, followed by 250 years of continuous simulation. The production well is set at the top of the Donggou Formation, and the maximum allowable pressure at the bottom of the well is 50 MPa.

3.2 Theoretical storage

The results of static modelling show that the total volume of rocks in the study area is 3.255 1×1010m3. According to Bachu (2015)，the storage efficiency is 2.4%（P50）, and the sequestration of P50is about 71.97 million tons（Table2）(Wen et al. 2019).

Table2 CO2 sequestration capacity based on static modelling result

3.3 Simulated reserve assessment

3.3.1 Total amount of CO2injection

In Case 1 and Case 2 models, CO2is continuously injected via injection wells at a design rate of 2 million tons per year for 50 years, with a cumulative injection amount of 1.919×1012moles (i.e.84.41×106t in total). As can be seen from Fig.7,after CO2is injected into the reservoir, it mainly exists in two forms, namely gaseous and dissolved(aquatic) phases. The CO2contents in both gaseous and dissolved phases increase with the injection amount, of which the increase of CO2in gaseous phase is faster than that in the dissolved phase.However, after the injection is stopped, the dissolved CO2in the reservoir continues to increase while the gaseous CO2gradually decreases (Fig.7). In the whole process of simulation, the amount of CO2captured is continuously increasing, but at the injection stage, the capture rate of CO2is always lower than the dissolution rate. After injection stops, the capture rate of CO2increases rapidly. At the end of the simulation, the total amount of CO2captured in Case 1 and Case 2 reaches 7.059 × 1011moles (31.063 × 106t in total) and 8.653 × 1011moles (38.077 × 106t in total), respectively.

Fig.7 Variation of CO2 in the reservoir over time (Case 1 on the left, Case 2 on the right)

The injection of CO2in the saline aquifer inevitably leads to the increase of reservoir pressure.The overall average pressures of the reservoir increased from an initial pressure of 21.878 MPa to 39.776 MPa in Case 1 and 39.615 MPa in Case 2,respectively (Fig.8), with an increase rate of 81.81% and 81.07%. In Case 1, the pressure at the bottom of the injection well is constantly slightly greater than the average pressure of the reservoir during the injection stage. In Case 2, pressures at the bottom of INJ1 and INJ3 wells are slightly greater than the average pressure of the reservoir, which is mainly related to the depth of the injection wells.On the other hand, pressures of INJ2 and INJ4 wells at the bottom of the hole are slightly lower than the average pressure of the reservoir. Generally, the pressure of the injection well continues to increase during the injection stage, and after the completion of injection, it drops rapidly and then gradually decreases year by year. At the end of injection, the maximum increase of reservoir pressure is less than the allowed pressure of rocks,which indicates that the reservoir has good storage safety.

Fig.8 The reservoir pressure and well bottom-hole pressures over time (Case1 on the left, Case2 on the right)

3.3.2 Spatial distribution of CO2

After CO2was injected into the saline aquifer, it migrated and diffused around under the combined pressure of injection and formation. When CO2was injected into the saline aquifer, it mainly diffused upward and around the formation. Under a single well injection condition, less CO2can be sequestered at the bottom of the formation than the top, mainly because the density of CO2is much lower than that of saline water; thus under the effect of injection pressure and buoyancy, the CO2will rapidly migrate upwards, and spread along particular horizons of the formation. In the case of upward diffusion in a formation with low permeability, the vertical diffusion velocity will be very limited and the horizontal diffusion will be more prominent.

As can be seen from Fig.9, in the process of CO2storage, the gaseous CO2migrated to a maximum distance of 3 500 m in the saline aquifer during the 300 years in Case 1, and mainly migrated to the high terrain along the bedding direction of rock, while the maximum distance of vertical diffusion was about 300 m. In Case 2, the diffusion process was greatly affected by the change of the physical parameters in the formation, and the gaseous CO2preferentially entered the zones with high permeability. The CO2mainly diffused northeastwards, also in southeastern and western directions(Fig.10). The gaseous CO2diffused around the injection well INJ to a certain distance, then migrated to the north and northeast. Similar distribution mode occurs in the INJ1 injection well in Case 2. These indicate that the difference in permeability of rock formation has a great influence on the migration path of CO2. In addition, around the injection well area of INJ2, CO2did not spread around the well after entering the saline aquifer due to the influence of dome structure. It quickly entered the top of the dome structure, which indicates that the trap structure has a significant impact on CO2storage.Compared with single well injection, multi-well injection can take full advantage of reservoir space and increase the reliability of saline aquifer for CO2storage. At the same time, it was observed that CO2did not break through the argillaceous caprocks overlying the Cretaceous Donggou Formation in both Cases This suggests that the Donggou formation has a great self-storage and self-sealing ability to store CO2and is a good reservoir with a high safety factor for CO2sequestration.

Fig.9 The 3D distribution of gaseous CO2 at 300 years (Case1 on the left and Case2 on the right)

Fig.10 The 3D distribution of gaseous CO2 at 300 years (Case1 on the left and Case2 on the right)

Another form of CO2occurrence in the saline aquifer is the dissolved CO2. Distribution of CO2is affected by temperature, pressure, salinity, permeability and other factors in the reservoir. Migration of the dissolved CO2is mainly driven by concentration difference in groundwater. In Case 1, the CO2firstly enters the saline aquifer through the upper perforated zone and then diffuses along the surface,while the dissolved CO2is mainly distributed below the middle-upper part. However, in Case 2, the dissolved CO2mainly accumulates in the middlelower part of the reservoir due to the relatively small injection volume of each well. As soon as CO2dissolves in saline water, the density of the saline water increases and it moves downward by gravity. For example, the dissolved CO2around the INJ well almost reaches the bottom of the reservoir(Fig.11). The distribution of the dissolved CO2near the INJ2 well is obviously affected by the trap structure. The distribution pattern of the dissolved CO2in saline aquifer is roughly the same as that of the gaseous CO2, but the former has a wider extent.The dissolved CO2between the INJ and INJ2 wells is apparently connected (Fig.11-right). In fact,dissolved CO2in the vicinity of all injection wells is well connected (Fig.12), with the maximum northsouth range of about 8.6 km, the maximum eastwest range of 9.3 km and a maximum thickness of nearly 300 m.

Fig.11 The distribution of CO2 in dissolved phase in 300 years (Case1 on the left and Case2 on the right)

Fig.12 The distribution of CO2 in dissolved phase at 300 years (3D) (Case1 on the left and Case2 on the right)

The proportion of reservoir residual gas in Case 2 is higher than that in Case 1, mainly because the distribution area of gaseous CO2in the multiple well injection is larger than that in a single well injection (Fig.13) to a certain extent. Therefore,increasing the contact area between CO2and the formation rocks can lead to a higher residual gas proportion and security of CO2saline aquifer storage.

Fig.13 The distribution of residual trapping CO2 at 300 years (Case1 on the left and Case2 on the right)

3.3.3 The proportion of CO2trapping forms

From the two different injection conditions discussed above, it is estimated that a total of 84.414 million tons of CO2could be injected in 50 years.It can be seen that CO2can exists in both gaseous and dissolved phases in deep saline aquifer. However, the proportion of CO2in gas/dissolved forms is different in different injection scenarios (Table3).Multi-well injection not only makes full use of the reservoir space, but also strengthens the hydraulic connection between the injection wells and expands the distribution of CO2in the saline aquifer. Multi-well injection is significantly better than single-well injection in terms of safety. However,from the perspective of engineering conditions, it will increase the cost to some extent. However, in Junggar Basin, where oil and gas wells are densely distributed, it is easier to apply the multi-well injection approach with a convenient operating environment.

Table3 The total amount of CO2 injection and its occurrence forms

3.3.4 Influence on groundwater environment

When CO2enters the saline aquifer, the groundwater environment in the aquifer is bound to change.It will firstly increase the aquifer pressure near the injection well when the CO2enters the reservoir,and then the dissolution of CO2in the aquifer will cause reaction such as mineralization in the water and then change the density of the water body. As can be seen from Fig.14, the density of water in the saline aquifer changes greatly after 300 years; and the closer to the injection well, the greater the variation of water density is. In the vicinity of the injection well, the density of water is increased by a maximum of 26.1‰. At the same time, when the CO2is dissolved in water, the density of water increases which results in a downward migration. The CO2is hence much safer present in dissolved phase than in the gas phase in the saline aquifers.

Fig.14 Water density distribution in the reservoir at 300 years (Case1 on the left and Case2 on the right)

3.3.5 Influence on reservoir pressure

When a large amount of CO2is injected into the reservoir, the reservoir pressure changes greatly. In general, the pressure in the injection well decreases radially in the surrounding area during the initial injection period. However, when the injection is stopped, the pressure gradually settles down over a period of time. After 300 years, the reservoir pressure would be leveled off. As can be seen from Fig.15, the reservoir pressure increases from the initial maximum pressure of 24.197 MPa to 42.165 MPa and 42.007 MPa, respectively, increasing by 174.257% and 173.604%. The increase of the reservoir average pressure in Case1 is slightly higher than that in Case 2. The buried depth of the reservoir is 1 945.5 m to 2 286 m, the fracture pressure gradient of the formation is 1.63-2.25 MPa/100 m, and the maximum withstanding pressure of the rock formation is about 52.00 MPa. The maximum pressure of CO2injection into the reservoir is less than the fracture pressure of the rocks, so the reservoir has superior storability and high safety factor for the geological sequestration of CO2.

Fig.15 Pressure distribution in the reservoir at 300 years (Case 1 on the left side and Case 2 on the right)

After CO2is dissolved in the saline water, the density of saline water obviously increases, accompanied by a downward movement. The distribution range of the dissolved CO2in the saline water in Case 1 is obviously smaller than that of Case 2. In Case 1, the maximum migration distance of the dissolved CO2is 4 km.

3.3.6 Maximum dynamic storage potential

In the above two models, CO2is continuously being injected for 50 years, and the bottom hole pressure of the injection well has not yet reached the maximum pressure that the reservoir can withstand. In order to explore the maximum CO2dynamic storage potential in this area, the injection condition of Case 1 and Case 2 models are set as: The CO2injection is stopped as soon as the bottom hole pressure of the injection well reaches 50 MPa. The CO2injection amount is regarded as the maximum dynamic storage potential of the reservoir under this condition. The simulation results show that the CO2dynamic storage potential in Case 2 is greater than that of Case 1, indicating that different injection modes may result in different storage potentials. According to the simulation result, the maximum CO2dynamic storage capacity is 145.295-× 106tons and the dynamic storage coefficient can reach 4.9% (Table4). Multi-well injection can improve storage efficiency and achieve greater storage potential.

Table4 Maximum dynamic storage potential of CO2 in Case 1 and Case 2

4 Conclusions

Based on the logging data of D7 Well and the measured 2D seismic data near the well area, a 3D heterogeneous geological model of the Cretaceous Donggou Formation was constructed on the basis of the sedimentary conditions, which reflected the distribution characteristics of the arenaceous reservoir rocks and mudstone caprock to the maximum extent, and described the distribution of porosity andpermeability of the caprock. This study calculated the theoretical CO2sequestration, carried out different scenarios (Case 1- single well injection,Case 2 - multi-well injection) through dynamic simulation, obtained the CO2sequestration of the Donggou Formation around the D7 well and the optimized dynamic simulation model, and vali-dated the effective storage coefficient which is suitable for CO2sequestration in Junggar Basin.This study also revealed the CO2storage mechanism, storage potential and the CO2spatial distribution in the saline water aquifer of the Donggou Formation in the eastern Junggar Basin.

(1) The results of 3D high-reliability geological model show that the theoretical CO2storage capacity is 71.967 million tons at a guarantee rate of P50in the saline aquifer of Cretaceous Donggou Formation in the vicinity of the D7 well, which covers 100 km2. The dynamic simulation results show that the injectable quantities of CO2in Case 1 and Case 2 are 135.919 and 145.295 million tons,respectively, while the injection duration can reach 81 years and 95 years.

(2) The CO2injected into saline aquifer mainly exists in the gaseous phase and the dissolved phase. The heterogeneity of reservoir has a great influence on the spatial distribution of the gaseous CO2,such as the abrupt change of porosity and permeability, or a sudden change of terrain caused by the heterogeneous deposition, where the CO2plume is irregularly shaped or has unidirectional migration.

(3) Multi-well injection (Case 2) is more conducive to the efficient utilization of reservoir space than single well injection (Case 1), and the CO2in the dissolved phase is slightly higher. A possible explanation could be, the multi-well injection increases the contact area between the CO2and rocks,so that the proportion of residual gas is higher than that of single well injection. Also multi-well injection is safer compared with single-well injection.

(4) Based on the comparative analysis between the theoretical calculation results and the dynamic simulation results, it is suggested to set the effective coefficient P50as 4.9% for CO2storage capacity evaluation in the saline aquifer of Junggar Basin;and Case 2 is recommended for the dynamic calculation of CO2storage capacity.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (NSFC, Grant No.41702284, 41602272), National key R&D program of China (Grant No. 2019YFE0100100), the Natural Science Foundation of Hubei Province, China(Grant No. 2019CFB451), and the Open Fund of Hubei Key Laboratory for Efficient Utilization and Agglomeration of Metallurgic Mineral Resources(Grant No.2020zy003). This work was also partially supported by the China Australia Geological Storage of CO2project (CAGS), and the China Geological Survey project (Grant No. DD20160307).We also would like to thank Professor Andrew Fietz and Dr. Liuqi WANG for their guidance and help while I was in Geoscience Australia.