Earthquake Research in China  2020, Vol. 34 Issue (2): 187-209     DOI: 10.19743/j.cnki.0891-4176.202002006
Progress and Prospect of Geophysical Research on Underground Gas Storage: A Case Study of Hutubi Gas Storage, Xinjiang, China
ZHANG Bo1, SONG Lili1, JI Zhanbo1, WANG Baoshan1,2, WU Zhide3     
1. Key Laboratory of Seismic Observation and Geophysical Imaging, Institute of Geophysics, China Earthquake Administration, Beijing 100081, China;
2. School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, China;
3. Research Institute of Petroleum Exploration and Development, Langfang 065007, China
Abstract: Hutubi Underground Gas Storage (UGS) is an important part of China's West-East Gas Pipeline Project. Its production safety also plays an essential role in ensuring peaking, emergency and reserve of natural gas in the region. Geophysical observations and research conducted on Hutubi UGS provide significant support for understanding the operation status and ensuring safe operation of the UGS. Since the beginning of the construction of Hutubi UGS and its production, several works including gas field observation experiments, rock experiments and numerical simulations have been conducted. Preliminary observational results show that during the initial operation stage of the UGS, the "breathing phenomenon" in caprock and microseismicity in the vicinity are closely related to the operation of the gas storage. However, rock experiments and numerical simulations show that these activities may gradually weaken with the multi-cycle operation of gas storage. The impact of the operation of Hutubi UGS in the surrounding areas is gradually weakening, and its operation tends to be stable.Implementing long-term and multi-method geophysical observations is able to provide us a better understanding of the relationship between the operation of UGS and regional geological hazards. On this basis, the corresponding geomechanical model can be established to form an effective risk management mechanism for gas storage operation. Thus, it is of great significance to understand its operation status, monitor storage conditions, guide production and operation, and ultimately guarantee the safe production of the gas storage.
Key words: Hutubi Underground Gas Storage     Deformation     Seismicity     Rock experiment    

INTRODUCTION

With the increasing attention to the environment, natural gas has been widely known as a kind of green, economical alternative energy. Its application on industrial production (industrial fuel, chemical production, etc.) and daily life (urban gas, transportation, etc.) has led to a significantly increasing demand of natural gas in most cities. Meanwhile, as one of the alternative energy sources for petroleum, natural gas is also an important component to ensure national energy security. In 2018, the national consumption of natural gas reached up to 276.6 billion m3, of which the external dependence of natural gas will exceed 45%, and maintain a sustained growth (Liu Chaoquan et al., 2019). However, several problems between supply and demand of natural gas appear correspondingly, such as the mismatch between storage area and demand area, the inconsistency between the actual natural gas production and demand in different seasons, etc. Such imbalance between supply and demand has become increasingly prominent with the increase of demand. In order to solve this problem, underground gas storage (UGS), a safe and reliable gas storage system, provides protection for peak shaving, emergency and reserve of natural gas. By the end of 2018, China has built 26 underground gas storage units, and the peak shaving capacity has reached 13 billion m3 (Yang Jing et al., 2019). However, its annual total effective working capacity accounts for only 3% of the total natural gas consumption, which is far lower than the world average level (11%) (Lei Hong, 2018). According to the world average, the domestic gas storage capacity demand by 2020 is estimated to be 385 million m3 (Ding Guosheng et al., 2015). There is still a gap in China's gas storage, which requires further construction and development.

At present, a number of types of gas storage have been explored and developed in the world, including depleted oil and gas storage, aquifer gas storage, salt cavern gas storage, etc. (Ding Guosheng et al., 2006).The operation safety of such large-scale industrial facility not only directly affects itself, but also exerts influence on the safety and property of its surrounding areas. In recent years, industrial activities, especially those related to underground fluids, such as hydraulic fracturing (Holland A. A., 2013; Kim W. Y., 2013; Lei Xinglin et al., 2008, 2013; Schultz R. et al., 2017), deep well water injection (Zoback M. D., 2012), geothermal exploitation (Allis R. G., 1982; Glowacka E. et al., 1996; Martínez-Garzón P. et al., 2013; Segall P. et al., 1998), long-term development of groundwater and oil and gas mining (González P. J. et al., 2012; Maury W. M. R. et al., 1992) and CO2 storage (Nicol A. et al., 2011), may cause changes in stress state and physical properties of media, and further lead to ground subsidence, induced earthquakes and other disasters (Ellsworth W. L., 2013; Yang Hongfeng et al., 2017; Zoback M. D., 2012). Compared with the industrial activities related to underground fluids, the main composition of underground gas storage is gas, which characterized by flammability and explosibility (natural gas itself is flammable and explosive), ability to reach high temperature and pressure (the temperature of natural gas compressed to critical state can reach 100-200℃, other pressures can reach 20-30 MPa), and many dividing points of high and low pressure (between gas injection and production systems, gas phase and liquid of production system) (Ying Qingdang, 2004). Furthermore, its production process is mainly divided into two parts: gas injection and gas extraction, which exert different effects on the region with gas storage and other industrial facilities related to underground fluids, and finally, the similar conclusions cannot be obtained directly. In order to have a better understanding of the impact of UGS operation, it is urgent to conduct corresponding investigations.

Geophysical observation related to gas storage is mainly based on seismic monitoring and deformation observation. Numerous microseismic monitoring investigations have been conducted in the vicinity of UGS. However, only limited seismicity cases are caused by natural gas injection and storage operations. In addition, few microseisms have been observed in the Bergermeer Gas Field in Netherlands accompanied by some natural gas storage operations (Kraaijpoel D. et al., 2012; Kwee J., 2012), and only seismic events which are less than MW0.4 have been reported in Haje Gas Storage in Czech Republic (Benetatos C. et al., 2013). The Castor UGS in Valencia Gulf, Spain, has induced a series of seismic activities in the early stage of its injection operation, resulting in the shutdown of gas storage after completing only a small part of the injection activities. Furthermore, a series of subsequent seismic events with a maximum magnitude of MW4.2 have also been reported to be related to the injection activities of the gas storage (Cesca S. et al., 2014; Gaite B. et al., 2016; Ruiz-Barajas S. et al., 2017). The observation of Collalto UGS in Italy is mainly conducted after a period of gas storage operation, and no microseismic phenomenon related to gas storage has been observed (Evans D.J. et al., 2008; Moratto L. et al., 2019). The monitoring of microseisms has formed a risk control system similar to traffic light (Zoback M.D., 2012; Ellsworth W.L., 2013). Similarly, deformation observation which is mainly based on satellite technology, including GPS and InSAR, has been widely applied to the monitoring of surface deformation related to underground gas injection (Onuma T. et al., 2009; Vasco D.W. et al., 2010; Morris J.P. et al., 2011; Teatini P. et al., 2011; Fokker P.A. et al., 2013; Gee D. et al., 2016). With the combination and analysis of deformation observation, rock mechanical properties, gas storage operation data, geological structure and some other information, the operation risk is assessed through geomechanical modeling, and then the operation strategy is modified to reduce the risk (Rutqvist J. et al., 2010; Teatini P. et al., 2011). However, neither earthquake nor deformation monitoring lacks the complete observation and study of the gas storage from the initial stage of construction to the stable operation stage. Thus, conducting a complete observation may help us better understand the impact of the operation of the gas storage, and provide support for understanding its production and operation dynamics and ensuring operational safety.

Hutubi UGS is the first large-scale gas storage after the introduction of the second pipeline of the "West-East Gas" project (Zhang Huaiwen et al., 2015). It is also an ideal experimental site for us to study the impact of underground gas storage operation in the region. Additionally, Hutubi UGS which is situated in the southern margin of the Junggar Basin, belongs to depleted oil and gas reservoir that is the most commonly used underground storage form at present. It is approximately 7 km from the west of Hutubi county and about 78 km from the southeast to Urumqi, and is adjacent to the Tianshan tectonic belt with active tectonic activities (DeMets C. et al., 1990; Sobel E.R. et al., 1997). There is an active source located about 30 km to the north of the storage, providing convenience for the study of regional underground media and other repeatable sources (Su Jinbo, 2015; Wei Yunyun, 2016; Wei Bing et al., 2016). Since the operation of Hutubi UGS in June 2013, plentiful geophysical observations have been conducted around the gas storage and its surrounding areas, including ground deformations, seismic monitoring processes and rock experiments. Besides, the numerical simulations related to the injection production process have also made progress. This paper aims to review and summarize the geophysical field observations, rock experiments and related numerical simulation works that have been conducted for the Hutubi gas storage and to look forward to the possible research and discussion aspects in the future.

1 GEOLOGICAL SETTINGS

The Tianshan mountain is an inland continental mountain system far away from the plate boundary. Influenced by the collision of India plate and Eurasia plate in the early Cenozoic and the strong compression and uplift caused by the subsequent northward pushing, as well as the compression and wedging of Tarim Basin and Junggar Basin, the inner regenerated structural belt of Eurasia continent in the Tianshan area is generated due to the significant subduction collision belt in the Paleozoic (Feng Xianyue, 1985; Avouac J. P. et al., 1993; Hendrix M. S. et al., 1994; Deng Qidong et al., 1999). At present, it is in the period of resurrection and the tectonic activity is strong (Demets C. et al., 1990; Sobel E. R. et al., 1997). There are many depression basins in the South and North Piedmont of Tianshan, such as Urumqi Depression, Kuqa Depression and Kashi Depression in the North Piedmont of Tianshan, among which there are many groups of thin skin structures of reverse fault fold belts in the Mesozoic and Cenozoic strata. On the contrary, in the piedmont area outside the depression basin, Tianshan mountain is directly connected with the Gobi plain due to the high angle reverse faults, including the South Sigu Fault in front of the North Tianshan Mountain and the Korla Fault and Xingdi Fault in front of the South Tianshan Mountain (Yang Xiaoping et al., 2008). Due to the extrusion and thrusting in the northern margin of the Northern Tianshan Mountains, the tectonic deformation in the southern margin of the Junggar Basin gradually expands from the Piedmont basin (Burchfiel B. C. et al., 1999; Deng Qidong, 2000; Guo Zhaojie et al., 2007; Cheng Guangsuo, 2008), and the thrust detachment surface gradually migrates from south to north, and the strength of the fold deformation gradually weakens (Yu Fusheng et al., 2009; Wu Zongtao et al., 2017).

Hutubi anticline is located at the northeast corner of the Piedmont depression (also known as Urumqi Piedmont depression) at the north foot of Tianshan Mountain, the region that is also the east end of the fourth row of reverse fault fold belt in the south margin of Junggar Basin. The anticline is characterized by a slight uplift and a nearly EW long axis fault, with a long axis of 40 km, a short axis of 8 km, and a wing inclination of 6-15°. Under the influence of the ramp generated from the upward extension of the detachment surface in the 7-8 km deep Jurassic strata, Hutubi anticline gradually uplifts and becomes a developing active anticline (Deng Qidong et al., 1999).

Hutubi gas field is located in Hutubi anticline, which is the first medium-sized gas field discovered in Junggar Basin and the largest condensate gas reservoir discovered so far (Wu Zongtao et al., 2017). The formation history is complicated. On one hand, under the strong tectonic activity of the late Himalayan movement, the effective gas source rocks (mainly Jurassic coal series source rocks (Chen Jianping et al., 2019)), which are widely distributed in the southern margin of Junggar Basin generate faults and secondary faults, and form hydrocarbon source ranges and upward migration channels. On the other hand, a series of abnormal high-pressure cap rocks formed by thrust nappe faults in the southern margin of Junggar Basin has shaped a favorable trap condition in the upper part, leading to the convergence and accumulation of the gas reservoirs (Fig. 2) (Chen Jianping et al., 2019).

Fig. 1 Map of main structural traces in Hutubi area

Fig. 2 Gas accumulation model and structural diagram of Hutubi anticline (a) and (b)structural evolution process diagram of Hutubi anticline since about 10 Ma. (modified from Wu Zongtao et al., 2017)

The main reservoir stratum of Hutubi gas reservoir is the Ziniquanzi formation, through which four southern dipping reverse faults sub-parallel to the anticline strike are developing. Among them, three faults cut through the stratum of Ziniquanzi formation, including Hutubi Fault, Hubei Fault and H001 Well North Fault. (Cao Xiqiu, 2013). Hutubi Fault is the largest fault in this area and belongs to overthrust fault. It has an extension of approximately 20 km and a fault distance of 60-200 m. The fault is steep on the top and relatively flat at the bottom, and branches from the west of H2 Well. The vertical fault distance and parallel extension of the Hubei Fault and H001 Well North Fault are relatively short and the fault momentum is also small.

2 DEVELOPMENT OF HUTUBI UNDERGROUND GAS STORAGE

As a clean energy source, the application range of natural gas has increasingly expanded. However, the imbalance between the exploitation and utilization of natural gas has also become more and more prominent with the increase of usage amount. In order to solve the problem of imbalance between the supply and demand of natural gas and meet the requirements of peak shaving, emergency response and strategic reserve of natural gas, an appropriate gas storage system must be established (Zhang Huaiwen et al., 2015). As a result, underground gas storage has become the main natural gas storage system because of its safety, stability, large storage capacity, economic and practical characteristics.

At the end of December 2009, the Exploration and Production Branch of China National Petroleum Corporation (CNPC) organized oil field companies and Langfang Branch to conduct thorough investigation and preliminary demonstration on the target gas storage sites. Finally, it was decided to convert the Hutubi gas field into an underground gas storage. The gas field had been put into production since April 1998, and has reached the recovery rate of 42% by 2013. Zhang Huaiwen et al. (2015) indicates that the cumulative gas production is approximately 53×108 m3 and is in the late stage of stable production.

After three and a half years' project design and construction period, Hutubi UGS began to inject gas on June 9th, 2013. For the first cycle, 3 gas marshalling stations and 16 gas injection wells are put into operation, with a total daily gas injection volume of 1 123×104 m3. As of October 30th, 2013, after 142 days of stable gas injection, the cumulative gas injection volume has reached 12.09 × 108 m3. From November 8th, 2013 to March 16th, 2014, Hutubi UGS has experienced a 68-day peak shaving gas extraction in the first cycle, with a total gas extraction volume of 2.67 × 108 m3. The second cycle of gas injection and extraction has been completed from April 2014 to March 2015, with a total gas injection volume of 17.68 × 108 m3, during which the daily gas extraction reached 753 × 104 m3. After two cycles of operation, 24 injection and extraction wells have been put into operation. The production cycle of the gas storage is divided into three parts: gas injection period (from April 2013 to October 2013), gas extraction period (from November 2013 to March 2014) and balance period (the injection and extraction process and maintain a relatively stable internal dynamic balance state). It marks the completion and safe operation of Hutubi UGS (Zhang Huaiwen et al., 2015).

3 GEOPHYSICAL RESEARCH PROGRESS

As the first large-scale underground gas storage of the second pipeline of the "West-East Gas Pipeline Transmission" project, Hutubi UGS has achieved its design goal, including natural gas peak shaving, accident emergency and strategic reserve. The goal is of great significance to the natural gas application in Xinjiang and even the whole country. In order to deepen the understanding of the impact of underground gas on its surrounding underground medium, and to ensure its long-term and stable operation as well as production safety of the region, relevant geophysical methods have been applied with respect to the operation of Hutubi UGS, such as deformation monitoring, microseismic monitoring, structural imaging, rock physics and numerical simulation.

3.1 Deformation Monitoring Observation (Leveling Surveying, GPS and InSAR)

The operation of underground gas storage consists of natural gas injection and extraction. However, the huge pressure difference generated in the operation may result in great changes in the internal pressure state and the pressure transmission to the ground, thus causing the horizontal and vertical changes of the ground. In addition, the change of load pressure caused by repeated injection and extraction processes may also cause the expansion of cracks, leading to fault activation and seismic activities which will threaten the operation safety (Qiao Xuejun et al., 2018). As one of the methods to directly monitor the expansion, long-term monitoring of surface deformation and related geomechanical simulation can effectively benefit the construction safety of gas storage. In order to study the influence of the operation of Hutubi UGS on the surface deformation, traditional leveling measurement and modern space technology monitoring (GPS and InSAR) are conducted respectively.

The traditional leveling measurement uses baseline measurement method and leveling method to monitor the change of horizontal and vertical components respectively, and the deformation of fault is monitored by manual observation through short leveling and short baseline. The vertical deformation characteristics of the surface in the gas storage area are obtained by using the height difference data obtained from the second-class leveling survey conducted in Hutubi UGS during the period of 2013-2015. The vertical deformation of the regional surface is mainly affected by three aspects: the overall subsidence of the basin under the tectonism of the North Tianshan Piedmont depression (the amplitude is ~-2.86 mm/a), the exploitation of groundwater, and the operation of gas storage. Among them, groundwater is mainly exploited from a large area of farmland in the north of the gas storage. In summer seasons when agricultural cultivation is concentrated, irrigation will cause funnel-shaped settlement in the pumping center of farm well, and then show significant seasonal characteristics (Li Jie et al., 2016). After eliminating the above two aspects, uplift and subsidence are observed in the vicinity of the geometric center of the reservoir with the pressure change in the wellhead, and a negative correlation is shown with the increase of the distance to the center. Compared with the wellhead pressure data of the Hutubi UGS, the surface deformation is 0.625-1.125 mm/MPa. However, due to the fact that the traditional leveling measurement is time-consuming and susceptible to external conditions, the results are limited to the vertical component and the observation of the horizontal component is lacking.

With the development of space monitoring technology, the application of several modern techniques, such as Global Positioning System (GPS) and Interferometric Synthetic Aperture Radar (InSAR) on ground deformation monitoring have become popular methods for geological disaster research. At present, it has been widely used in the fields of landslide, land subsidence, mine depression deformation, etc. Compared with the traditional measurements, GPS has the advantage to overcome the influence of external conditions to a certain extent, and it is more convenient to achieve the vertical and horizontal changes of fixed measuring points. The GPS observation shows similar vertical results with traditional leveling measurements.(Wang Dijin et al., 2016; Qi Wei et al., 2018). In addition, GPS observation also shows a "breathing phenomenon" on the horizontal deformation of the surface cover, that is, the gas storage surface collapses inward with the decrease of the wellhead pressure during the gas extraction and diverges outward during the gas injection. The deformation caused by the change of wellhead pressure during gas injection and extraction is estimated respectively (1.02 mm/MPa in horizontal direction and -1.11 mm/MPa in vertical direction during gas injection; 1.24 mm/MPa in horizontal direction and 0.86 mm/MPa in vertical direction during gas extraction cycle) (Wang Dijin et al., 2016). However, due to the effect of the measuring point distribution, the results are still obtained from the change of discrete measuring points, resulting in limited spatial resolution. By comparison, InSAR which is based on interferometry technology is able to overcome the problem of measuring point deficiency, leading to a higher resolution and continuous space observation. The results obtained by small baseline subset (SBAS) InSAR technology show that the deformation characteristics of the whole area are discontinuous, and the deformation amplitude gradually decreases outwards from the intensive area of gas well on the east side (Fig. 3), which may be related to the complex structure of underground gas storage (Chen Wei et al., 2016).

Fig. 3 Deformation time series of underground gas storage in Hutubi. Time series diagram of HUK9 and HUK19 well location deformation and wellhead pressure (modified from Chen Wei et al., 2016)

The research on the deformation of gas storage cover mainly focuses on the early stage of operation. Under the influence of periodic load and pressure changes during injection and production, the pore pressure will change accordingly. This will lead to the elastic deformation and fatigue failure of the reservoir rock, which manifests as the "breathing" deformation of the reservoir surface cover in the horizontal direction. In this process, the rapid crustal uplift reflects the rapid elastic or plastic rebound of the UGS caprock under the continuous compression (Qiao Xuejun et al., 2018). In the vertical direction, the caprock may be more affected by the ground subsidence associated with the overexploitation of groundwater and manifest as a negative correlation with the wellhead pressure of gas storage. However, due to the lack of hydrological data, the deformation correction caused by groundwater is not accurate enough and further collection and discussion are still needed. In general, the deformation of the caprock is heterogeneous, reflecting complicated underground structure. However, the current observation is mainly focused on the initial stage of gas storage operation, which lacks of long-term and multiperiod continuous observation. In the "breathing phenomenon", the lithology of caprock (including porosity, connectivity, compressive strength, permeability, tensile strength, etc.) changes with the continuous multi-period operation of gas storage. However, whether this phenomenon will change is worth further study and discussion. Based on a deeper understanding of the deformation effect exerted by gas storage operation on the caprock, the corresponding geomechanical model can be gradually established and improved. Besides, the relationship between periodic deformation of caprock and production data of gas storage(injection volume, storage capacity, etc.) is established to further simulate and infer the state of underground reservoir. Further comparison of the actual production data with the simulation results can effectively evaluate the production and operation status of the gas storage. Such comparison plays an important role in ensuring the production and operation safety.

3.2 Structural Imaging

The main storage area is deep buried several kilometers underground, resulting in troubles to directly monitor the state of the reservoir area medium. The inversion of the regional spatial distribution of wave velocity using seismic signals has several advantages. Firstly, the drilling cost is saved and the spatial distribution of the regional medium state is obtained efficiently. Besides, the medium state information in deeper layers can also be collected, providing support to monitor the operating state of gas storage. The 1-D velocity model is an approximation of the regional structure, and is also the basis for developing seismic location and structural imaging. Different from the determination of traditional velocity models in other regions, the large-capacity "Xinjiang" airgun launcher constructed about 25 km in the northern part of the Hutubi UGS can provide a repeatable and known-source earthquake hypocenter for the determination and correction of the one-dimensional velocity model in the Hutubi region. The energy generated each time by the airgun is equivalent to that from a ML0.9 earthquake (Yang Wei et al., 2013). Combined with the high repeatability of its source, the stacked signal can obtain a high signal-to-noise ratio, which can be clearly recorded by stations within 380 km (Su Jinbo, 2015; Wei Bing et al., 2016). Ji Zhan bo (2019) use the dispersion curve of the multimode surface wave signal from the active source to inverse the S-wave velocity model of the shallow surface (within 2 km) (Fig. 4), and discuss the characteristics and causes of the high-order surface wave by identifying and analyzing the five order surface waves (three order Rayleigh waves and two order Love waves). The comparison among results of shallow wave velocity (< 100 km) of three well logs in the Hutubi area shows that the inversion of the 1-D S-wave velocity model ia accurate and efficient. However, limited by the frequency band of surface wave, it is hard to obtain robust constraints in deep structures.

Fig. 4 Map of station distribution of gas gun experiment in Hutubi UGS and shallow S-wave velocity model obtained from inversion of each survey line and comparison with logging data. (modified from Ji Zhanbo, 2019)

Based on the S-wave velocity model proposed by Ji Zhanbo (2019), Zhou Pengcheng et al. (2019) establish the velocity model database using grid search method in which the depth stratification is unchanged and step length is set as 0.1 km/s. The velocity model is mainly screened by forward travel time. Based on the records of portable stations constructed around the gas storage (red triangle in Fig. 5), the velocity model is selected by two-step method which restricts the arrival residual (observation arrival time and theoretical arrival time difference) of stations with different distance. According to the results of surface wave imaging, the model with the lowest velocity in the shallow layer (within 500 m) is selected as the final velocity model. On the other hand, the same method is used to build the velocity model database. Zhang Bo et al. (2016) use the inversion method to determine the one-dimensional velocity model of the region. It is similar to the perforation inversion method to determine the inter well velocity model in industry. Based on the known characteristics of the seismogenic information of the active source, the velocity model is selected by comparing the location recovered by the absolute location method with its actual location. Finally, the velocity model with the smallest distance between them is determined as one-dimensional velocity model suitable for regional study. At the same time, the inversion parameter setting in the recovery process is also used as the benchmark of regional seismic location. The two methods respectively screen the velocity model obtained by grid point search from the perspective of forward modeling and inversion the two eventual velocity models show great similarity. In details, the results of forward modeling have higher velocity at 1 km and 5.5 km. The velocity model determined by Zhou Pengcheng et al.(2019) has better results in forward travel time, while the velocity model determined by Zhang Bo et al. (2016) has better results in inversion of source location.

Fig. 5 3-D S-wave velocity structure inversion results and S-wave velocity results of longitudinal section (Wang Juanjuan et al., 2018)

In recent years, the fast-developing noise imaging method has increasingly become a reliable imaging method which is based on velocity structures, and has been applied to multiple areas. Using noise can avoid the restriction brought by the heterogeneity of seismic events. By calculating the noise cross-correlation function between stations, the empirical Green's function between stations can be approximately obtained, and then the dispersion curve corresponding to the path of stations can be extracted. Finally, the distribution of two-dimensional group velocity in the region can be inverted. The shallow S-wave velocity structure of the Hutubi is obtained by inversion of the noise records of 15 portable stations in the gas storage area (Fig. 6). In the longitudinal direction, the shallow velocity is relatively small, which may be related to the relatively thick and younger regional loose sedimentary caprock (Wang Fang, 2017). Affected by the fissures produced by pumping and gas injection, the gas storage area shows low shear wave velocity (Wang Juanjuan et al., 2018). However, an obvious low-speed layer appears at 420-500 m, which may be associated with the loose sand gravel layer according to the regional drilling geological data (Wang Fang, 2017). On the other hand, there are also differences in the S-wave group velocity in the horizontal region, especially in the Hutubi Fault and Hubei Fault, where there are significant transition zones of high and low group velocity. Besides, the vertical distribution difference is small, which is consistent with the distribution of steep faults in the region (Wang Fang, 2017). Because of the complex near-surface structure, it is still necessary to further conduct more accurate imaging work with dense stations.

Fig. 6 The variation of wave velocity measured between July 2016 and April 2017 for different stations in the gas storage (a) one side station pair of the gas storage; (b) cross station pair of the gas storage; (c) the change of air pressure with time for drilling gas from July 2013 to November 2015. (Wang Fang, 2017)

The underground medium may also change due to the continues operation of UGS, whereby the injection of large volume injection of critical state natural gas can change the average density, stress state, porosity, connectivity, etc. However, the heterogeneity of time and space of natural earthquakes restricts study of underground medium change. In order to better understand the variation characteristics of underground media over time, Ji Zhanbo (2019) use the gas gun source signal, which can be continuously and repeatedly excited to replace natural earthquakes, to analyze the P-wave and surface wave travel time, together with Rayleigh surface wave polarization changes. The results suggest that there are two forms in this change: seasonal and daily changes. The daily variation is more correlated with temperature and solid tide, while the seasonal variation is mainly affected by groundwater level. However, there is a strong coupling effect between the daily temperature change and the solid tide, thus leading to further quantitative analysis to distinguish the influence of the two, especially on the shallow velocity. On the other hand, based on the continuity of noise, Wang Fang (2017) discusses the change of shallow medium in Hutubi gas storage using background noise. After analyzing the variation of wave velocity recorded on the stations both at the boundary and across of gas storage, it is pointed out that there is obvious seasonal variations of wave velocity in Hutubi area (Fig. 6). Compared with the air pressure state of gas storage and the change of regional groundwater level, it is considered that the seasonal change of wave velocity may be affected by both. Long term overdraft in Hutubi caused groundwater level to decline year after year. Consequently, this long-term downward trend may has an impact on the underground media, and needs further discussion.

3.3 Microseismic Monitoring and Seismicity

With increasingly reported earthquake events induced by human activities, the relationship between human activities and destructive earthquakes has gradually arose social concern. In particular, the earthquake induced by industrial activities related to the underground fluid has become a hot topic in recent years, but there are few reports related to UGS. During the operation of underground gas storage, there are two main impacts on the underground medium. On one hand, the controlled injection of large volume and high-pressure natural gas which is in the critical state may increase a large number of tension fractures in the underground gas storage (Goertz-Allmann B.P. et al., 2011). It is similar to hydraulic fracturing. On the other hand, the wellhead pressure difference between injection and production processes can reach to more than 10 MPa, and the pressure in the reservoir may fluctuate circularly, leading to great fluctuation of pore pressure in underground medium and unstable effective stress and reservoir porosity of pore throat structure (Witherspoon P.A. et al., 1967). Its operation characteristics increase the possiblity of induced-earthquakes in the gas storage which may further change the regional seismicity, threatening the security of life environment and production in the vicinity.

In June 2013, gas injection officially started in Hutubi UGS, and a series of small earthquakes occurred in the vicinity of the gas field, which has aroused public concerns. In order to study the relationship between seismicity change and the operation of Hutubi UGS, Tang Lanlan et al. (2018) conducted microseismic detection and positioning within 10 km of Hutubi gas storage during 2013-2015 (including three gas injection stages and two gas production stages) based on the waveform records of Xinjiang Seismic Network (Fig. 7). A total of 273 seismic events with ML>1.0 are detected. These events dominantly occur when the injection rate and wellhead pressure suddenly increase in field operations. The correlation between time and space indicates that the operation of gas storage is related to seismic events. According to the preliminary simulation, the physical mechanism is considered to be associated with the pressure-bearing fault near Hutubi gas reservoir. In detail, the fault may have induced seismic activity under the effect of stress mutation caused by natural gas injection into the reservoir. However, due to the limitation of Match & Locate's method in determining the depth of seismic events, the results only include information of the source location in the horizontal direction, making it impossible to determine whether the seismogenic structure is a natural fracture or two main faults. Zhou Pengcheng et al. (2019) conduct ETAS simulation and high-resolution relocation for the events recorded in the network catalogue of Hutubi area from 2013 to 2018, and the results further prove this kind of induction relationship. Although some events may be missed in the network catalogue, the high-resolution relocation results show that the correlations between induced seismicity and geological structure, focal fault geometry and water injection well data are weak. The results also show that the fault earthquake is not related to reservoir formation in hydrology. Therefore, the increase of seismicity in the early stage of operation is caused by the disturbance of pore elastic stress which is resulted from gas injection into gas storage. It is helpful to understand the physical mechanism of the occurrence of large magnitude events (ML>1.0) in the region. However, there is a lack of discussion on small magnitude events (ML < 1.0). The event with smaller magnitude has smaller scale of seismogenic and can reflect the structural stress characteristics of the natural fractures with smaller scale in the region.It is also of great significance for understanding the influence of the gas storage operation on the underground reservoir medium.

Fig. 7 The gas cumulative input volume, average wellhead pressure and injection rate for east (a)-(c) and west wells (d)-(f), respectively. the M-t plot (g) and seismic frequency per day (h) for Hutubi region. (Tang Lanlan et al., 2018)

Hutubi area is located in the North Tianshan Piedmont structure with active tectonic movement. The regional seismicity is the result of a serious factors. Both the results based on the network catalogue or those obtained from microseismic detection show that the change of seismicity in Hutubi area is related to the operation of underground gas storage in the first two periods of operation. It is worth noting that although the gas storage capacity, injection and production rate have increased after two cycles of operation, the earthquake events with ML>1.0 near the gas storage have significantly decreased (Tang Lanlan et al., 2018). This is consistent with the time when the actual reserve of the gas storage reaches the designed maximum storage capacity. Whether the change of seismicity in the process of increasing production is affected by the change of storage capacity and injection rate of the gas storage, and whether it will weaken after reaching the designed storage capacity still need multi period observation and research. Meanwhile, further and more accurate positioning of microseismic events is also helpful for us to decide whether the seismogenic structure is caused by natural fracture or main fault dislocation. Such process is of great significance for deciding whether the fault in Hutubi area slips or whether there is leakage in the gas storage. To conduct long-term seismic monitoring and obtain the seismicity near the UGS, will help us to monitor the operation state.It will provide supports for adjusting operation parameters of gas storage in time to reduce induced disaster.

3.4 Other Observation and Research

In addition to the study of topography and seismology, observation studies around the operation of Hutubi UGS also include the in-situ stress and gravity monitoring.

In-situ stress (also known as initial stress, absolute stress or original stress) is a natural stress existing in the stratum without engineering disturbance. It is the fundamental force that causes deformation, fracture, fold and even earthquake in the crustal rock. The measurement of in-situ stress can provide a basis for the discussion of the dynamic genesis and internal mechanism of tectonic activities. According to the fitting of the displacement vector of the faults in Hutubi area, it is pointed out that the stress structure in this area is strike slip fault (Zhang Jie et al., 2016). The estimation of stress state in Hutubi area by focal mechanism solution shows that the events in North Tianshan Mountain area are mainly strike slip fault, while the stress state in North Tianshan Piedmont depression where Hutubi UGS is located is reverse fault, and there is about 9° deflection of the principal stress in the east and west of the gas storage which has a boundary at 87° (Li Yanyong et al., 2018). Different regions correspond to different types of stress structures, which provide different bases for discussing the dynamic and internal causes of earthquake events in the region.

Gravity measurement can reflect the state of underground medium. A series of factors, including groundwater level, soil moisture and surface elevation, can cause the change of gravity. By comparing the change of the gravity value of the measuring points inside and outside the Hutubi UGS area with the time, it is found that compared with the measuring points outside the reservoir area, the change of the gravity value of the internal measuring points is not only affected by the change of the groundwater level, but also influenced by the density change of the underground medium and the ground rise and fall related to the injection and production of the gas storage (Yushan A. et al., 2018). However, the variation of regional gravity is affected by numerous factors. Further strengthening the observation of regional gravity is helpful to understand the influence of the operation of underground gas storage on the underground medium, and then provide support for the operation safety of the gas storage.

3.5 Rock Physics Experiment

In recent years, plenty of field observations and research have been conducted in the Hutubi UGS, including the high-precision monitoring of seismic wave velocity, the investigations using surface GPS and InSAR, and the studies of gravity field change. However, the surface observation data are affected by multiple factors and how to effectively utilize the data to explain the underground gas storage characteristics is problematic. Therefore, the indoor experimental research is beneficial for not only providing basic information for field observation data, but also serving as a reference for better analysis and understanding of field data.

In order to study the physical property change and damage mechanism of the rocks caused by the change of pore pressure during the operation of Hutubi UGS, the cyclic loading experiment is conducted on the rock samples (Berea and Boise sandstone) close to the reservoir rock under the corresponding stress environment (simulated ground stress of 2 750 m underground: confining pressure of 50 MPa, axial pressure of 65 MPa; simulated ground stress of 3 500 m underground: confining pressure of 63 MPa, axial pressure of 81 MPa). The experimental results show that the permeability and wave velocity of rock samples change with cyclic loading, reflecting the change of internal structure of rock samples. During the loading and unloading process of the cycle, with the increase of the cycling number, an irrecoverable deformation occurs in the core. Besides, the permeability increases, especially those of water-saturated rock samples. Higher stress state (indicate deep depth) also makes the permeability increase more significantly with the increase of cycle times. On the other hand, the vP/vS ratio can reflect the degree of closure and anisotropy of the internal pore fissures. With the increase of the cycling number, the change range of the wave velocity ratio increases, but the change of the water-saturated rock and the gas-saturated rock show different properties. The wave velocity ratio of the gas-saturated rock increases with the increase of the effective stress, while the wave velocity ratio of the water-saturated rock decreases with the increase of the effective stress. However, the effect is relatively small. At the same time, the experiment shows that the wave velocity ratio of the former one is obviously higher than that of the latter one. Such feature can be used to analyze the properties of reservoir fluids.

Although the core collected in the field can only be replaced by rock samples with similar composition and porosity due to reasons such as water swelling and weak cementation, the results still provide more intuitive understandings of the physical response of rocks to repeated loading and unloading under reservoir conditions. Furthermore, it provides important information for interpretation of seismic wave velocity, GPS data and gravity field change in field observation data and numerical simulation related to gas storage operation.

3.6 Numerical Simulation

The operation of gas storage is a complex process of fluid migration, especially the depleted oil and gas storage with more complex underground structure. In the process of repeated injection and extraction of gas, the reservoir, rock cap and fault of underground gas storage will produce a variety of mixed deformation processes including elasticity, creep and plasticity under the influence of the changes of several factors, including injection pressure, injection production rate, internal pressure alternating rate and stabilization time. Therefore, it is difficult to describe its characteristics and predict its development by using simple constitutive relation. With the development of computer technology, the numerical simulation method may provide effective solutions for these issues.

With the support of relevant production data and observation data of topographic change in Hutubi UGS, the multi-point source Mogi model (Chen Wei et al., 2016) and 3-D finite element model (Wei Qi et al., 2018) are proposed to simulate the ground displacement of the gas storage. The simulation results show that the caprock of gas reservoir exhibits a "breathing phenomenon" that lasts for several cycles during the gas injection and extraction process. Affected by the operation of the gas storage, the gas injection process shows the trend of outward expansion on the surface under the pressure of the reservoir cap. On the contrary, the gas extraction process shows the trend of inward collapse on the surface. In order to further studying the influence of cyclic injection and production on physical properties of the reservoir, Wang Jianbo et al. (2018) establish a finite element model that describes the elastoplastic damage on reservoirs under pore pressure based on elastoplastic mechanics and fatigue damage mechanics. Under the hypothetical condition with multi cycle injection and extraction, after analyzing the change of pore pressure and porosity, in combination with the simulation of surface deformation, it is considered that the reservoir can bear plastic deformation for no more than 6 years. After 6 years of service life, the accumulated deformation of reservoir pore reaches the limit, leading to collapse of pore structure, decrease of pore, decrease of storage capacity, and the subsequent surface subsidence. Among them, the "breathing phenomenon" on the surface will gradually weaken with the repeated injection extraction cycle, and eventually disappear after 6 years. Meanwhile, after simulating the safety of the two main faults in the reservoir, it is suggested that the Hutubi Fault and the Hubei Fault will reach the deformation limit in 3.6 years and 13.6 years respectively after gas injection under the assumption of numerical simulation.

Relevant simulations of the operation of gas storage, including topography, pore pressure porosity and fault safety are conducted to provide guidance for discussing the interpretation of relevant observation, change of reservoir medium and production safety of gas storage. As the operation of gas storage and related observations proceed, more refined structure, longer time of actual measurement and production data are expected to provide more accurate model basis and constraints for further simulations, and meanwhile provide fundamental information for guiding and ensuring the safety of gas storage.

4 CONCLUSION AND PROPECT

Hutubi UGS is the first large underground gas storage in the second pipeline of the "West-East Gas Pipeline Transmission" project, which plays an extremely important role in ensuring the use of natural gas and national energy security. Various studies including observation experiment (topography change, structure and medium change, seismicity, etc.), rock experiment and numerical simulation have been conducted around the gas storage. The results indicate that in the initial stage of gas storage operation, the reservoir caprock is not only controlled by the groundwater level, but also exhibits a "breathing phenomenon" related to the operation of the gas storage. Furthermore, it shows that plenty of seismic activities are clearly related to the operation of the gas storage. The related rock experiments and numerical simulations show that under the assumption of numerical simulation, the deformation and seismic activities related to the operation of the gas storage will gradually weaken as the operation of the gas storage proceeds. This is consistent with the observation of multi cycle seismicity. At present, the impact of the operation of Hutubi gas storage in the surrounding areas is gradually weakening, and its operation still remains in a safe state.

However, there is a lack of discussion on this multi period effect change in geophysical observation related to the gas storage. Whether this weakening trend will change with the further production and operation of the gas storage still requires attention. Such attention is also beneficial to deepen our understanding of the migration and storage processes of underground gas and liquid as well as the changes of the underground medium caused by them. On this basis, the geomechanical model of Hutubi area is established and gradually improved through geophysical investigations using production data of gas storage, providing references for judging the state of the underground media and storage conditions in Hutubi area. At the same time, a variety of geophysical studies are combined to intensively study and discuss the relationship between the production of underground gas storage and regional geological disasters. Finally, the risk management for the operation safety of gas storage is gradually established. The management can provide guidance for the reasonable control of production parameters and the reduction of geological disasters, and provide guarantee for the work safety of Hutubi UGS. In addition, it can also provide ideas for the prevention and control of geological disasters induced by large-scale industrial facilities such as underground gas storage, CO2 storage, geothermal exploitation, and reservoir water storage.

ACKNOWLEDGEMENT

We thank the peer reviewers for their insightful suggestions and comments.

REFERENCES
Allis R.GAllis R.G. Mechanism of induced seismicity at the Geysers geothermal reservoir, California[J]. Geophysical Research Letters, 1982, 9(6): 629-632. DOI:10.1029/GL009i006p00629
Avouac J.P., Peltzer GAvouac J.P., Peltzer G. Active tectonics in southern Xinjiang, China:analysis of terrace riser and normal fault scarp degradation along the Hotan-Qira fault system[J]. Journal of Geophysical Research, 1993, 98(B12): 21773-21807. DOI:10.1029/93JB02172
Benetatos C., Málek J., Verga FBenetatos C., Málek J., Verga F. Moment tensor inversion for two micro-earthquakes occurring inside the Háje gas storage facilities, Czech Republic[J]. Journal of Seismology, 2013, 17(2): 557-577. DOI:10.1007/s10950-012-9337-0
Burchfiel B.C., Brown E.T., Deng Qidong, Feng Xianyue, Li Jun, Molnar P., Shi Jianbang, Wu Zhangming, You HuichuanBurchfiel B.C., Brown E.T., Deng Qidong, Feng Xianyue, Li Jun, Molnar P., Shi Jianbang, Wu Zhangming, You Huichuan. Crustal shortening on the margins of the Tien Shan, Xinjiang, China[J]. International Geology Review, 1999, 41(8): 665-700. DOI:10.1080/00206819909465164
Cao Xiqiu. A reaserch on reservoir geomechanic features of a gas storage in Xinjiang after natural depletion[D]. Beijing: China University of Geosciences (Beijing), 2013 (in Chinese with English abstract).
Cesca S., Grigoli F., Heimann S., González Á., Buforn E., Maghsoudi S., Blanch E., Dahm TCesca S., Grigoli F., Heimann S., González Á., Buforn E., Maghsoudi S., Blanch E., Dahm T. The 2013 September-October seismic sequence offshore Spain:a case of seismicity triggered by gas injection?[J]. Geophysical Journal International, 2014, 198(2): 941-953. DOI:10.1093/gji/ggu172
Chen Jianping, Wang Xulong, Ni Yunyan, Xiang Baoli, Liao Fengrong, Liao JiandeChen Jianping, Wang Xulong, Ni Yunyan, Xiang Baoli, Liao Fengrong, Liao Jiande. The accumulation of natural gas and potential exploration regions in the southern margin of the Junggar basin[J]. Acta Geologica Sinica, 2019, 93(5): 1002-1019 (in Chinese with English abstract).
Chen Wei, Yu Pengfei, Xiong Wei, Li jie, Feng Guangcai, Qiao XuejunChen Wei, Yu Pengfei, Xiong Wei, Li jie, Feng Guangcai, Qiao Xuejun. InSAR deformation monitoring and simulation of underground gas storage in Hutubi, Xinjiang[J]. Journal of Geodesy and Geodynamics, 2016, 36(9): 803-807, 812 (in Chinese with English abstract).
Cheng GuangsuoCheng Guangsuo. A study of the balanced sections of the structure in the central part of southern Junggar basin[J]. Acta Geoscientica Sinica, 2008, 29(5): 563-570 (in Chinese with English abstract).
DeMets C., Gordon R.G., Argus D.F., Stein SDeMets C., Gordon R.G., Argus D.F., Stein S. Current plate motions[J]. Geophysical Journal International, 1990, 101(2): 425-478. DOI:10.1111/j.1365-246X.1990.tb06579.x
Deng Qidong, Feng Xianyue, Zhang Peizhen, Yang Xiaoping, Xu Xiwei, Peng Sizhen, Li junDeng Qidong, Feng Xianyue, Zhang Peizhen, Yang Xiaoping, Xu Xiwei, Peng Sizhen, Li jun. Reverse fault and fold zone in the Urumqi range-front depression of the northern Tianshan and its genetic mechanism[J]. Earth Science Frontiers, 1999, 6(4): 191-201 (in Chinese with English abstract).
Deng Qidong. Active Tectonics of the Chinese Tianshan Mountain[M]. Beijing: Seismological Press, 2000 (in Chinese).
Ding Guosheng, Xie PingDing Guosheng, Xie Ping. Current situation and prospect of Chinese underground natural gas storage[J]. Natural Gas Industry, 2006, 26(6): 111-113 (in Chinese with English abstract).
Ding Guosheng, Li Chun, Wang Jieming, Xu Hongcheng, Zheng Yali, Wanyan Qiqi, Zhao YanjieDing Guosheng, Li Chun, Wang Jieming, Xu Hongcheng, Zheng Yali, Wanyan Qiqi, Zhao Yanjie. The status quo and technical development direction of underground gas storages in China[J]. Natural Gas Industry, 2015, 35(11): 107-112 (in Chinese with English abstract).
Ellsworth W.LEllsworth W.L. Injection-induced earthquakes[J]. Science, 2013, 341(6142): 1225942. DOI:10.1126/science.1225942
Evans D.J., West J.M. An appraisal of underground gas storage technologies and incidents, for the development of risk assessment methodology[R]. British Geological Survey, 2008.
Feng XianyueFeng Xianyue. Seismogeological characteristics of the Xinjiang area[J]. Seismology and Geology, 1985, 7(2): 35-44 (in Chinese with English abstract).
Fokker P.A., Wassing B.B.T., van Leijen F.J., Hanssen R.F., Nieuwland D.AFokker P.A., Wassing B.B.T., van Leijen F.J., Hanssen R.F., Nieuwland D.A. Data assimilation of PS-InSAR movement measurements applied to the Bergermeer gas field. In:Proceedings of International EAGE Workshop on Geomechanics and Energy[J]. Lausanne, Switzerland:European Association of Geoscientists & Engineers, 2013. DOI:10.3997/2214-4609.20131971
Gaite B., Ugalde A., Villaseñor A., Blanch EGaite B., Ugalde A., Villaseñor A., Blanch E. Improving the location of induced earthquakes associated with an underground gas storage in the Gulf of Valencia (Spain)[J]. Physics of the Earth and Planetary Interiors, 2016, 254: 46-59. DOI:10.1016/j.pepi.2016.03.006
Gee D., Sowter A., Novellino A., Marsh S., Gluyas JGee D., Sowter A., Novellino A., Marsh S., Gluyas J. Monitoring land motion due to natural gas extraction:validation of the Intermittent SBAS (ISBAS) DInSAR algorithm over gas fields of North Holland, the Netherlands[J]. Marine and Petroleum Geology, 2016, 77: 1338-1354. DOI:10.1016/j.marpetgeo.2016.08.014
Glowacka E., Nava F.AGlowacka E., Nava F.A. Major earthquakes in Mexicali Valley, Mexico, and fluid extraction at Cerro Prieto geothermal field[J]. Bulletin of the Seismological Society of America, 1996, 86(1A): 93-105.
Goertz-Allmann B.P., Goertz A., Wiemer SGoertz-Allmann B.P., Goertz A., Wiemer S. Stress drop variations of induced earthquakes at the Basel geothermal site[J]. Geophysical Research Letters, 2011, 38(9): L09308.
González P.J., Tiampo K.F., Palano M., Cannavó F., Fernández JGonzález P.J., Tiampo K.F., Palano M., Cannavó F., Fernández J. The 2011 Lorca earthquake slip distribution controlled by groundwater crustal unloading[J]. Nature Geoscience, 2012, 5(11): 821-825. DOI:10.1038/ngeo1610
Guo Zhaojie, Deng Songtao, Wei Guoqi, Li BenliangGuo Zhaojie, Deng Songtao, Wei Guoqi, Li Benliang. Comparative study of the foreland thrust belts of South and North Tianshan and implications for hydrocarbon accumulation[J]. Earth Science Frontiers, 2007, 14(4): 123-131 (in Chinese with English abstract).
Hendrix M.S., Dumitru T.A., Graham S.AHendrix M.S., Dumitru T.A., Graham S.A. Late Oligocene-early Miocene Unroofing in the Chinese Tianshan:an early effect of the India-Asia collision[J]. Geology, 1994, 22(6): 487-490. DOI:10.1130/0091-7613(1994)022<0487:LOEMUI>2.3.CO;2
Holland A.AHolland A.A. Earthquakes triggered by hydraulic fracturing in south-central Oklahoma[J]. Bulletin of the Seismological Society of America, 2013, 103(3): 1784-1792. DOI:10.1785/0120120109
Ji Zhanbo. The study on shallow underground velocity structure and its variations using artificial seismic source[D]. Beijing: Peking University, 2019 (in Chinese with English abstract).
Kim W.YKim W.Y. Induced seismicity associated with fluid injection into a deep well in Youngstown, Ohio[J]. Journal of Geophysical Research, 2013, 118(7): 3506-3518.
Kraaijpoel D., Nieuwland D., Wassing B., Dost B. Induced seismicity at an underground gas storage facility in the Netherlands[C]. In: Proceedings of EGU General Assembly Conference Abstracts. 2012: 13298.
Kwee J. Micro-seismicity in the Bergermeer gas storage field[D]. Utrecht: Utrecht University, 2012.
Lei HongLei Hong. Opportunities and challenges to the construction of underground gas storages in China[J]. Oil & Gas Storage and Transportation, 2018, 37(7): 728-733 (in Chinese with English abstract).
Lei Xinglin, Yu Guozheng, Ma Shengli, Wen Xueze, Wang QiangLei Xinglin, Yu Guozheng, Ma Shengli, Wen Xueze, Wang Qiang. Earthquakes induced by water injection at -3 km depth within the Rongchang gas field, Chongqing, China[J]. Journal of Geophysical Research, 2008, 113(B10): B10310. DOI:10.1029/2008JB005604
Lei Xinglin, Ma Shengli, Chen Wenkang, Pang Chunmei, Zeng Jie, Jiang BingLei Xinglin, Ma Shengli, Chen Wenkang, Pang Chunmei, Zeng Jie, Jiang Bing. A detailed view of the injection-induced seismicity in a natural gas reservoir in Zigong, southwestern Sichuan basin, China[J]. Journal of Geophysical Research, 2013, 118(8): 4296-4311.
Li Jie, Li Rui, Wang Xiaoqiang, Shi Xinpu, Zheng Liming, Abudutayier·Yasen, Sun Xiaoxu, Chen Shujiang, Fang Wei, Cheng RuizhongLi Jie, Li Rui, Wang Xiaoqiang, Shi Xinpu, Zheng Liming, Abudutayier·Yasen, Sun Xiaoxu, Chen Shujiang, Fang Wei, Cheng Ruizhong. Research on surface vertical deformation in the Hutubi underground gas storage reservoir[J]. Earthquake Research in China, 2016, 32(2): 407-416 (in Chinese with English abstract).
Li Yanyong, Wang Chenghu, Yang JiajiaLi Yanyong, Wang Chenghu, Yang Jiajia. The focal mechanism solution and stress field inversion of earthquakes along Hutubi[J]. Journal of Geodesy and Geodynamics, 2018, 38(12): 1246-1250 (in Chinese with English abstract).
Liu Chaoquan, Jiang XuefengLiu Chaoquan, Jiang Xuefeng. Overview of the domestic and foreign oil and gas industry development in 2018 and outlook for 2019[J]. International Petroleum Economics, 2019, 27(1): 27-33, 60 (in Chinese with English abstract).
Martínez-Garzón P., Bohnhoff M., Kwiatek G., Dresen GMartínez-Garzón P., Bohnhoff M., Kwiatek G., Dresen G. Stress tensor changes related to fluid injection at the Geysers geothermal field, California[J]. Geophysical Research Letters, 2013, 40(11): 2596-2601. DOI:10.1002/grl.50438
Maury V.M.R., Grassob J.R., Wittlinger GMaury V.M.R., Grassob J.R., Wittlinger G. Monitoring of subsidence and induced seismicity in the Lacq gas field (France):the consequences on gas production and field operation[J]. Engineering Geology, 1992, 32(3): 123-135. DOI:10.1016/0013-7952(92)90041-V
Moratto L., Romano M.A., Laurenzano G., Colombelli S., Priolo E., Zollo A., Saraò A., Picozzi MMoratto L., Romano M.A., Laurenzano G., Colombelli S., Priolo E., Zollo A., Saraò A., Picozzi M. Source parameter analysis of microearthquakes recorded around the underground gas storage in the Montello-Collalto Area (Southeastern Alps, Italy)[J]. Tectonophysics, 2019, 762: 159-168. DOI:10.1016/j.tecto.2019.04.030
Morris J.P., Hao Yue, Foxall W., McNab WMorris J.P., Hao Yue, Foxall W., McNab W. A study of injection-induced mechanical deformation at the in Salah CO2 storage project[J]. International Journal of Greenhouse Gas Control, 2011, 5(2): 270-280. DOI:10.1016/j.ijggc.2010.10.004
Nicol A., Carne R., Gerstenberger M., Christophersen ANicol A., Carne R., Gerstenberger M., Christophersen A. Induced seismicity and its implications for CO2 storage risk[J]. Energy Procedia, 2011, 4: 3699-3706. DOI:10.1016/j.egypro.2011.02.302
Onuma T., Ohkawa SOnuma T., Ohkawa S. Detection of surface deformation related with CO2 injection by DInSAR at in Salah, Algeria[J]. Energy Procedia, 2009, 1(1): 2177-2184. DOI:10.1016/j.egypro.2009.01.283
Qi Wei, Wang Xiaoqiang, Wang Dijin, Li Jie, Shi XinpuQi Wei, Wang Xiaoqiang, Wang Dijin, Li Jie, Shi Xinpu. Finite element numerical simulation analysis of Hutubi underground gas storage[J]. Journal of Geodesy and Geodynamics, 2018, 38(5): 477-481 (in Chinese with English abstract).
Qiao Xuejun, Chen Wei, Wang Dijin, Nie Zhaosheng, Chen Zhengsong, Li Jie, Wang Xiaoqiang, Li Yu, Wang Tan, Feng GuangcaiQiao Xuejun, Chen Wei, Wang Dijin, Nie Zhaosheng, Chen Zhengsong, Li Jie, Wang Xiaoqiang, Li Yu, Wang Tan, Feng Guangcai. Crustal deformation in the hutubi underground gas storage site in China observed by GPS and InSAR measurements[J]. Seismological Research Letters, 2018, 89(4): 1467-1477. DOI:10.1785/0220170221
Ruiz-Barajas S., Sharma N., Convertito V., Zollo A., Benito BRuiz-Barajas S., Sharma N., Convertito V., Zollo A., Benito B. Temporal evolution of a seismic sequence induced by a gas injection in the Eastern coast of Spain[J]. Scientific Reports, 2017, 7: 2901. DOI:10.1038/s41598-017-02773-2
Rutqvist J., Vasco D.W., Myer LRutqvist J., Vasco D.W., Myer L. Coupled reservoir-geomechanical analysis of CO2 injection and ground deformations at In Salah, Algeria[J]. International Journal of Greenhouse Gas Control, 2010, 4(2): 225-230. DOI:10.1016/j.ijggc.2009.10.017
Schultz R., Wang Ruijia, Gu Y.J., Haug K., Atkinson GSchultz R., Wang Ruijia, Gu Y.J., Haug K., Atkinson G. A seismological overview of the induced earthquakes in the Duvernay play near Fox Creek, Alberta[J]. Journal of Geophysical Research, 2017, 122(1): 492-505.
Segall P., Fitzgerald S.DSegall P., Fitzgerald S.D. A note on induced stress changes in hydrocarbon and geothermal reservoirs[J]. Tectonophysics, 1998, 289(1/3): 117-128.
Sobel E.R., Dumitru T.ASobel E.R., Dumitru T.A. Thrusting and exhumation around the margins of the western Tarim basin during the India-Asia collision[J]. Journal of Geophysical Research, 1997, 102(B3): 5043-5063. DOI:10.1029/96JB03267
Su Jinbo. Data processing and preliminary application of Airgun source in Hutubi Xinjiang[D]. Lanzhou: Lanzhou Institute of Seismology, China Earthquake Administration, 2015 (in Chinese with English abstract).
Tang Lanlan, Lu Zhou, Zhang Miao, Sun Li, Wen LianxingTang Lanlan, Lu Zhou, Zhang Miao, Sun Li, Wen Lianxing. Seismicity induced by simultaneous abrupt changes of injection rate and well pressure in Hutubi Gas Field[J]. Journal of Geophysical Research, 2018, 123(7): 5929-5944.
Teatini P., Castelletto N., Ferronato M., Gambolati G., Janna C., Cairo E., Marzorati D., Colombo D., Ferretti A., Bagliani A., Bottazzi FTeatini P., Castelletto N., Ferronato M., Gambolati G., Janna C., Cairo E., Marzorati D., Colombo D., Ferretti A., Bagliani A., Bottazzi F. Geomechanical response to seasonal gas storage in depleted reservoirs:a case study in the Po River basin, Italy[J]. Journal of Geophysical Research, 2011, 116(F2): F02002.
Vasco D.W., Rucci A., Ferretti A., Novali F., Bissell R.C., Ringrose P.S., Mathieson A.S., Wright I.W.Vasco D.W., Rucci A., Ferretti A., Novali F., Bissell R.C., Ringrose P.S., Mathieson A.S., Wright I.W. Satellite-based measurements of surface deformation reveal fluid flow associated with the geological storage of carbon dioxide[J]. Geophysical Research Letters, 2010, 37(3): L03303.
Wang Dijin, Li Yu, Nie Zhaosheng, Wang Tan, Qiao Xuejun, Li Jie, Yu Pengfei, Cheng RuizhongWang Dijin, Li Yu, Nie Zhaosheng, Wang Tan, Qiao Xuejun, Li Jie, Yu Pengfei, Cheng Ruizhong. Study on the cap rock deformation of Hutubi underground gas storage by GPS observations[J]. Earthquake Research in China, 2016, 30(3): 339-450.
Wang Fang. Shallow structure and temporal variation monitoring around Hutubi underground gas storage in Xinjiang based on seismic ambient noise[D]. Beijing: Institute of Geophysics, China Earthquake Administration, 2017 (in Chinese with English abstract).
Wang Jianbo, Ge Hongkui, Wang Xiaoqiong, Liu Tongyuan, Liu Dunqing, Zhang Yanjun, Wu ShanWang Jianbo, Ge Hongkui, Wang Xiaoqiong, Liu Tongyuan, Liu Dunqing, Zhang Yanjun, Wu Shan. Study of the reservoir damage during cyclic injection and extraction process in the Hutubi gas storage[J]. Earthquake Research in China, 2018, 32(2): 288-300.
Wang Juanjuan, Yao Huajian, Wang Weitao, Wang Baoshan, Li Cheng, Wei Bin, Feng LeiWang Juanjuan, Yao Huajian, Wang Weitao, Wang Baoshan, Li Cheng, Wei Bin, Feng Lei. Study of the near-surface velocity structure of the Hutubi gas storage area in Xinjiang from ambient noise tomography[J]. Chinese Journal of Geophysics, 2018, 61(11): 4436-4447 (in Chinese with English abstract).
Wei Bing, Su Jinbo, Wang Haitao, Zheng Liming, Wang Qiong, Zhang Wenlai, Yuan Shun, Wei Yunyun, Chen HaoWei Bing, Su Jinbo, Wang Haitao, Zheng Liming, Wang Qiong, Zhang Wenlai, Yuan Shun, Wei Yunyun, Chen Hao. Site selection and construction of Hutubi airgun source signal transmitting seismic station and its characteristic of source[J]. Earthquake Research in China, 2016, 32(2): 222-230 (in Chinese with English abstract).
Wei Qi, Wang Xiaoqiang, Wang Dijin, Li Jie, Shi XinpuWei Qi, Wang Xiaoqiang, Wang Dijin, Li Jie, Shi Xinpu. Finite Element umerical Simulation Analysis of Hutubi Underground Gas Storage[J]. Journal of Geodesy and Geodynamics, 2018, 38(5): 477-481 (in Chinese with English abstract).
Wei Yunyun. Preliminary application of airgun source signal in Hutubi, Xinjiang[M]. Lanzhou: Lanzhou Institute of Seismology, China Seismological Bureau, 2016 (in Chinese with English abstract).
Witherspoon P.A., Neuman S.P.Witherspoon P.A., Neuman S.P. Evaluating a slightly permeable caprock in aquifer gas storage:1. Caprock of infinite thickness[J]. Journal of Petroleum Technology, 1967, 19(7): 949-955. DOI:10.2118/1634-PA
Wu Zongtao, Wang Yadong, Liu Xingwang, Zheng Youwei, Zheng Jianjing, Chao Xianqiu, Li ZhengheWu Zongtao, Wang Yadong, Liu Xingwang, Zheng Youwei, Zheng Jianjing, Chao Xianqiu, Li Zhenghe. Natural gas accumulation process and main controls factors of Hutubi condensate gas field[J]. Contributions to Geology and Mineral Resources Research, 2017, 32(3): 403-408 (in Chinese with English abstract).
Yang Hongfeng, Liu Yajing, Wei Meng, Zhuang Jiancang, Zhou ShiyongYang Hongfeng, Liu Yajing, Wei Meng, Zhuang Jiancang, Zhou Shiyong. Induced earthquakes in the development of unconventional energy resources[J]. Science China Earth Sciences, 2017, 60(9): 1632-1644. DOI:10.1007/s11430-017-9063-0
Yang Jing, Liu XiaoliYang Jing, Liu Xiaoli. Review of natural gas development in 2018 and its outlook in 2019[J]. Energy of China, 2019, 41(2): 13-18 (in Chinese with English abstract).
Yang Wei, Wang Baoshan, Ge Hongkui, Wang Weitao, Chen YuYang Wei, Wang Baoshan, Ge Hongkui, Wang Weitao, Chen Yu. The active monitoring system with large volume airgun source and experiment[J]. Earthquake Research in China, 2013, 29(4): 399-410 (in Chinese with English abstract).
Yang Xiaoping, Deng Qidong, Zhang Peizhen, Xu XiweiYang Xiaoping, Deng Qidong, Zhang Peizhen, Xu Xiwei. Crustal shortening of major NAPPE structures on the front margins of the Tianshan[J]. Seismology and Geology, 2008, 30(1): 111-131 (in Chinese with English abstract).
Ying Qingdang. Underground gas storage and its operation management[C]. In: Proceedings of 2004 China International Conference on Oil and Gas Pipeline Technology, China Petroleum and Natural Gas Pipeline Administration. Xi'an: China Petroleum Institute, 2004: 252-260 (in Chinese).
Yu Fusheng, Li Guozhi, Yang Guangda, Ma Baojun, Chen ShupingYu Fusheng, Li Guozhi, Yang Guangda, Ma Baojun, Chen Shuping. Deformation feature and genesis simulation of fold-and-thrust belts in the southern margin, Junggar basin[J]. Geotectonica et Metallogenia, 2009, 33(3): 386-395 (in Chinese with English abstract).
Yushan A., Wang Xiaoqiang, Liu Daiqin, Yasen AYushan A., Wang Xiaoqiang, Liu Daiqin, Yasen A. Preliminary analysis on characteristics of gravity change and its influencing factors in Hutubi underground gas storage reservoir[J]. Inland Earthquake, 2018, 32(3): 252-260 (in Chinese with English abstract).
Zhang Bo, Wang Baoshan, Ji Zhanbo. Seismic location of Hutubi gas storage and its surrounding areas[C]. In: Proceedings of 2016 Annual Meeting of Chinese Geoscience Union. Beijing: Chinese Geophysical Society, Seismological Society of China, Organizing Committee of National Symposium on Petrology and Geodynamics, Special Committee of Structural Geology and Geodynamics, Chinese Geological Society, Regional Geology and Metallogeny Committee of China Geological Society, 2016: 1558-1560 (in Chinese).
Zhang Huaiwen, Gao Yan, Chen Hao, Zhou JiangZhang Huaiwen, Gao Yan, Chen Hao, Zhou Jiang. Construction and operation of underground gas storage in Hutubi, Xinjiang[J]. Shihezi Science and Technology, 2015(5): 46-48 (in Chinese).
Zhang Jie, Huang Luyuan, Jiang Yinghao, Liu ZhuoyanZhang Jie, Huang Luyuan, Jiang Yinghao, Liu Zhuoyan. The analysis about characteristics of in-situ stress field in Xinjiang Tianshan area[J]. Journal of Engineering Geology, 2016, 24(S1): 1370-1377 (in Chinese with English abstract).
Zhou Pengcheng, Yang Hongfeng, Wang Baoshan, Zhuang JiancangZhou Pengcheng, Yang Hongfeng, Wang Baoshan, Zhuang Jiancang. Seismological investigations of induced earthquakes near the Hutubi underground gas storage facility[J]. Journal of Geophysical Research, 2019, 124(8): 8753-8770.
Zoback M.DZoback M.D. Managing the seismic risk posed by wastewater disposal[J]. Earth, 2012, 57(4): 38-43.