Earthquake Research in China  2019, Vol. 33 Issue (4): 544-556     DOI: 10.19743/j.cnki.0891-4176.201904010
Attenuation Characteristics of Earthquake Ground Motion for Large Volume Airgun
CAI Huiteng1,2, CHEN Yong1, JIN Xing2, XU Yihe1,3, LI Wen4
1. School of Earth Sciences and Engineering, Nanjing University, Nanjing 210046, China;
2. Fujian Earthquake Agency, Fuzhou 350003, China;
3. Institute of Geophysics, CEA, Beijing 100081, China;
4. Geophysical Exploration Center, CEA, Zhengzhou 450003, China
Abstract: In order to further deepen the understanding of seismic wave propagation characteristics induced by the large volume airgun source,experimental data from multiple fixed excitation points in Fujian Province were used to obtain the equivalent single excitation high signal-to-noise ratio velocity and displacement records through linear stacking and simulation techniques. Then the peak displacements of different epicentral distances were used to calculate the equivalent magnitude of the airgun source excitation at different fixed excitation points so as to establish the attenuation relationship between equivalent magnitude,epicenter distance and velocity peak. Our results show that:① Within 270km of epicentral distance,for the large-volume airgun's single shot,the peak velocity range is about 700-4nm/s,and the peak displacement range is about 200.0-0.2nm; ② The equivalent magnitude of the P-wave from the airgun source with a total capacity of 8,000in3 is 0.181-0.760,and the equivalent magnitude of the S-wave is 0.294-0.832. By contrast,the equivalent magnitude of the P-wave from the airgun source with a total capacity of 12,000in3 is 0.533-0.896,and the equivalent magnitude of the S-wave is 0.611-0.946. The S-wave energy is greater than the P-wave energy,and the excitation efficiency varies greatly with different excitation environment; ③ The peak velocity increases with the equivalent magnitude,and decreases with the epicentral distance. The vertical component of the P-wave peak velocity is the largest among those three components,while the S-wave has the smallest vertical component and similar horizontal components. Hence,our research can provide an important basis for the quantitative judgment of the seismic wave propagation distance using the airgun and the design of the observation system in deep exploration or monitoring with airgun.
Key words: Large-volume airgun source     Velocity peak value     Displacement peak value     Equivalent magnitude     Attenuation characteristics

INTRODUCTION

The active transmission and effective reception of seismic waves to the subsurface using artificial seismic source is an important way for detailed regional-scale deep detection and monitoring of crustal medium changes (Chen Yong et al., 2017). As a high performance artificial source, large volume airgun source (hereinafter referred as airgun source) has become the main seismic source in deep ocean exploration and the combination exploration from both onshore and offshore due to its high repeatability, accurate positioning, high energy conversion efficiency, and abundant low frequency components. Besides, it can meet the detection requirements of far propagation and deep penetration after data processing. (Nazareth J. J. et al., 2003; Van Avendonk H. J. A. et al., 2004; McIntosh K. et al., 2005; Melhuish A. et al., 2005; Qiu Xuelin et al., 2007; Zhao Minghui et al., 2008; Lin Jianmin et al., 2010). In recent years, Chinese scholars have conducted experiments and analysis on a variety of artificial seismic sources, including explosive blasting and airguns(Chen Qifu et al., 2004; Ge Hongkui et al., 2006; Li Hui et al., 2007; Chen Yong et al., 2007; Wang Baoshan et al., 2008, 2010), and finally selected the airgun source as the seismic source to study the structure of the continental crust and its changes. Three fixed airgun activation platforms were established in Binchuan, Yunnan, Hutubi, Xinjiang and Zhangye, Gansu, respectively in 2012, 2013 and 2015, and these fixed activation platforms have been used in monitoring and studying the changes of seismic wave propagation velocity in the medium of earths interior(Wang Baoshan et al., 2012; Wei Bin et al., 2016; Zhang Yuansheng et al., 2016; Chen Yong et al., 2017). In order to overcome the deficiency that the activation platforms are immovable, after years of efforts, Fujian Earthquake Agency, independently, developed mobile and ship borne airgun source technical equipment, and performed successfully airgun activation experiments in several reservoirs within Fujian province, Taiwan strait and the Yangtze River Anhui section (Xia Ji et al., 2016; Xu Jiajun et al., 2016; Chen Huifang et al., 2016, 2017; Lin Binhua et al., 2017a, 2017b).

During the process of detecting the crustal structure or monitoring the change of medium using airgun source, the propagation ability of seismic waves activated by airgun source directly affects the scope and depth of detection or monitoring. In essence, the propagation ability of seismic waves activated by airgun source depends on whether the ground motion can be distinguished by stations at different epicentral distances. In addition, the magnitude of ground motion is determined by activation power of airgun source (airgun capacity and pressure, airgun array combination, size and depth of sinking), activation environment (size, shape and depth of water body, and properties of medium on the upper and lower layer of the liquid-solid coupling interface), and attenuation of transmission process of underground medium. Given this, based on the experimental data of repeated activation of multiple airgun sources obtained at fixed activation points in Fujian onshore and offshore, linear stacking method is firstly adopted. After the stacking results are divided by the number of stacking times, the results of a single shot can be obtained and are further simulated into displacement records. Then the airgun source activation capability and activation environment are integrated into one influence factor as the activation efficiency of airgun source. In addition, the equivalent magnitude is calculated with the peak displacement based on different epicentral distances to evaluate the activation efficiency at different fixed points, and then the attenuation relation of equivalent magnitude, epicentral distance and peak velocity are established in order to provide an important basis for quantitative judgment of propagation ability of seismic waves activated by airgun source and for designing an observation system for deep detection or monitoring using airgun source.

1 DATA AND PROCESSING METHOD 1.1 Data Sources

During 2014-2017, Fujian Earthquake Agency has selected 16 sites and conducted 20 times of fixed-point large volume airgun activation experiments (shown in Table 1). The onshore and offshore activation platforms are mobile reservoir airgun source system and the "Yanping 2" offshore airgun source system, respectively. Both systems are originally designed to consist of 4 1500LL longlife airguns with a total capacity of 8, 000in3. In 2016, the "Yanping 2" offshore airgun source system was expanded and reconstructed, and was finally upgraded to a combination of 6 1500LL longlife airguns with a total capacity of 12, 000in3. It is worth mentioning that among these 20 fixed-point activation experiments, some were conducted at the same site but in different time periods due to different experimental purposes. Due to the difference of reception effect caused by the variation of background noise and time periods, we consider this type of experiments as independent experimental samples.

Table 1 Overview of fixed-point activation experiment of large-volume airgun

After more than 20 years construction, up to now, Fujian seismic network has 88 stations in operation, 73 of which are equipped with broadband seismometers, including 46 stations with 120s-50Hz bandwidth and 27 stations with 60s-50Hz bandwidth. The sampling rate is 100Hz and most of them have low environmental interference. Continuous digital waveform records of three components (E-W, N-S, U-D) recorded by 73 broadband seismometers during the experiment, as shown in Fig. 1(a), were collected in this study. Each station intercepted 20s before and 300s after each activation as a record and forms an SAC format data file. Eventually, approximately 1.1 million records were intercepted.

 Fig. 1 Distribution map of fixed activation points and stations (a), waveform stacking for the vertical components activated at L01 fixed point according to YXBM station (b) and a scatter plot of peak velocity of the vertical components activated at L01 fixed point according to YXBM station (c) (a)The red five-pointed stars are the fixed activation points, and the blue triangles are stations. (b)The black lines are the superimposed waveforms of the vertical components, with a total of 725, and the red line is the stacking result. (c)The black points are the scatter of peak velocity of the vertical components, the blue line is the mean value of peak velocity, and the red line is the peak value of the stacking result
1.2 Data Processing Methods

We remove instrument response, mean value, and linear trend of each component first, then we rotated the two horizontal components were into radial (R-R) and tangential (T-T) direction. Considering that the dominant frequency band of airgun source is 3-8Hz (Lin Jianmin et al., 2010), the 3-8Hz band-pass filtering was thus performed for each record, and specific processing steps are described as follows:

(1) In order to obtain high-validity amplitude information, we adopt the linear technique to stacking. For each fixed activation point, the RMS screening linear stacking method proposed by Jiang Shengmiao et al.(2017) was applied to stack station by station, and the stacking results were divided by the number of stacking times as the result of a single activation;

(2) All post-stacking velocity records of each fixed activation point were integrated into seismic profiles, and the signal-to-noise ratio of each seismic record in the five-component (U-D, E-W, N-S, R-R and T-T) profiles was calculated. Seismic records with signal-to-noise ratio greater than 8 were selected and peak velocities of P-wave band and S-wave band were extracted respectively;

(3) To simplify the subsequent calculation of equivalent magnitude for fixed activation points, it is necessary to simulate the velocity records into displacement records. Using the recursive method of mutual simulation of different types of ground motion parameters proposed by Jin Xing et al. (2004), the velocity records with SNR greater than 8 in the second step were simulated into displacement records, and peak velocities of P-wave band and S-wave band were extracted respectively.

Fig. 1(b) shows the superposition of the vertical component records of 725 times of activation at L01 fixed point received by YXBM station at a distance of 13.63km(black lines), and the red line shows the result of RMS screening linear stacking. It can be seen that the velocity of the noise segment(0-2s) is about ±200nm/s before stacking and about ±10nm/s after stacking with a relatively long signal segment (2-10s). In addition to P-wave and S-wave, surface wave components can also be seen. Combined with Fig. 1(c), it can be seen that the peak velocity in the signal segment appears in P-wave, ranging from 500 to 800nm/s with the average peak value of 698.6nm/s and the post-stacking park value of 689.5nm/s. Furthermore, if we define the amplitude stacking loss rate is the ratio of the difference of the average peak value and the post-stacking peak value to the average peak value, then the amplitude stacking loss rate of the L01 fixed point-YXBM station is 1.3%, indicating the high validity of the post-stacking amplitude value and reflecting the ideal effect of the RMS screening linear stacking method.

Fig. 2 shows the seismic velocity profiles and displacement profiles of post-stacking vertical components activated at L01 fixed point recorded by 73 stations. It also reveals the corresponding peak values in the P-wave and S-wave groups. It can be seen from Fig. 2 that signals of P-wave and S-wave groups after stacking can be continuously distinguished within the epicentral distance of 0-270km. The extracted travel times corresponding to the peak values of P-wave and S-wave groups are in accordance with the corresponding apparent velocity relation. The peak velocity ranges from 700nm/s to 4nm/s, and the peak displacement ranges from 200nm to 0.2nm. The attenuation mode of peak velocity is consistent with that of peak displacement, and the peak velocity or peak displacement of P-wave and S-wave groups are similar. However, some peak velocities collected from observation points which has similar epicentral distances vary by an order of magnitude.We consider that it is caused by the directional differences in the transmission process of the energy excited by airgun source. To sum up, in the subsequent analysis, according to the data obtained by the above processing method, the peak displacement value is taken to calculate the equivalent magnitude activated different fixed points, and the peak velocity is taken to establish the airgun source velocity attenuation relation.

 Fig. 2 Distribution of post-stack velocity and displacement at fixed points (a)Distribution of post-stack peak velocity of vertical component activated at L01 fixed point. (b)Normalized display of post-stack seismic velocity profile at L01 fixed point. (c)Distribution of post-stack peak displacement of vertical component activated at L01 fixed point. (d)Normalized display of post-stack seismic displacement profile at L01 fixed point. The blue circles are the peak velocity value of P-wave group, the red circles are the peak velocity of S-wave group, the blue squares are the P-wave windows, and the red squares are the S-wave windows
2 CALCULATION OF EQUIVALENT MAGNITUDE

In this study, the seismic source activation ability and activation environment at different fixed points are considered as a whole. Meanwhile, the method similar to the conventional local magnitude computing method is used to calculate the equivalent magnitudes of P-wave and S-wave which are further used as the parameters to evaluate the activation efficiency at different activation sites. The equivalent magnitude calculation formulas are as follows:

 $M_{\mathrm{LP}}= \lg \left(A_{\mathrm{P}}\right)+R(\mathit{\Delta})$ (1)
 $M_{\mathrm{LS}}= \lg \left(A_{\mathrm{S}}\right)+R(\mathit{\Delta})$ (2)
 $A_{\mathrm{P}}=A_{\mathrm{Z}}$ (3)
 $A_{\mathrm{S}}=\left(A_{\mathrm{N}}+A_{\mathrm{E}}\right) / 2$ (4)

Where, MLP and MLS are equivalent magnitudes of P-wave and S-wave, respectively. AP and AS are the peak amplitudes of P-wave and S-wave respectively, while AN and AE are the peak displacement of S-wave of the N-S component and E-W component respectively. The unit of AZ, AN and AE is μm. Δ is the epicentral distance in km. R(Δ) is the calibration function of local magnitude, and we take R12 in appendix A of the "Provisions of Local Magnitude(GB 17740—2017)".

For the post-stacking peak displacement value of P-wave and S-wave activated at each fixed site and recorded by each station, formula(1)-(4) are used to calcalute the corresponding magnitude, and the mean value of magnitudes of all stations is taken as the equivalent magnitude at the fixed activation site. The results are shown in Table 1, and the calculation process for the L01 fixed site in land reservoir and the S08 fixed site in the Taiwan Strait are shown in Fig. 3. As shown in Fig. 3, the discreteness of the calculated results with epicentral distance from 100km to 200km is generally better than that of other epicentral distances except for few stations.

 Fig. 3 Calculation of equivalent magnitude (a) MLP at L01 fixed site; (b) MLS at L01 fixed site; (c) MLP at S08 fixed site; (d) MLS at S08 fixed site The black dot is the calculation result of each station, and the red line is the average result
3 ESTABLISHMENT OF ATTENUATION MODEL

Before the establishment of attenuation model, it is necessary to consider the choice of attenuation model form. Because the equivalent magnitudes selected by airgun source are all less than 1 (Table 1), the seismic source body is relatively small and the source depth can be approximate to 0km, the impact of magnitude saturation term and near-earthquake saturation term are not considered. Therefore, we select an attenuation model that only takes into account the influence of magnitude term M and geometric diffusion term lg(Δ)(Jin Chaoyu et al., 2009) as follows:

 $\lg (Y)=C_{1}+C_{2} M+C_{3} \lg (\Delta)+\varepsilon$ (5)

where Y is the peak velocity, the unit of which is uniformly converted to nm/s. Δ is the epicentral distance in km. ε stands for random variable.

According to formula (5), multiple linear regression method is adopted to fit the peak velocities of P-wave and S-wave of the three components respectively, and the attenuation law of the peak velocity in the bedrock field triggered by the airgun source is obtained. The specific results are shown in Table 2, Fig. 4 and Fig. 5. It is worth mentioning that the magnitude terms of P-wave and S-wave use MLP and MLS respectively in the fitting process, and the two components of the two groups in the horizontal direction are not only fitted as independent parameters but also as the same samples, that is, the results of (E-W + N-S) and (R-R + T-T) in Table 2, Fig. 4 and Fig. 5 shows the fitted attenuation results of P-wave and S-wave, respectively. The original samples are also shown for comparative analysis.

Table 2 Peak velocity attenuation coefficient and standard deviation of fitting

 Fig. 4 Distribution of P-wave peak velocities of 5 components and fitting results (a)U-D; (b)E-W; (c)N-S; (d)R-R; (e)T-T. The fitting curves from top to bottom corresponded to MLP =0.1, 0.3, 0.5, 0.7, 0.9 respectively

 Fig. 5 Distribution of S-wave peak velocities of 5 components and fitting results (a)U-D; (b)E-W; (c)N-S; (d)R-R; (e)T-T. The fitting curves from top to bottom corresponded to MLS =0.3, 0.5, 0.7, 0.9 respectively
4 ANALYSIS AND DISCUSSION 4.1 Analysis of Airgun Source Activation Efficiency

For a more convenient analysis, Fig. 6 shows the distribution of equivalent magnitudes of 20 fixed activation sites given in Table 1. It can be seen from Table 1 and Fig. 6 that: ① MLS is greater than MLP, indicating that the energy of S-wave generated by airgun source activation is greater than that of P-wave, which is consistent with the research results of Zhang Wei et al.(2009). In addition, Fig. 2 indicates that the travel time characteristics of S-wave show that it is generated by sources, which further indicates that S-wave is mainly a converted wave generated by P-wave activated by airgun on the bottom surface of water body near the seismic source, i. e., the liquid-solid interface, and due to topographic influence or energy redistribution caused by the media characteristics in the upper and lower layers, S-wave may be stronger than P-wave. ② The activation ability of airgun source is the same, but different activation environment leads to big differences in efficiency. For example, L01-L04 are inland water bodies; meanwhile, the airgun capacity, pressure and size are similar but has different efficiency, L02 and L03 sites are examples for the latter situation. According to the relation between energy and magnitude (Zhou Yun, 2005), the ratio of P-wave energy is about 8.5, and the ratio of S-wave energy is about 6.0. ③ When airgun activation environment is similar, the efficiency improves with the total capacity increase of the airgun source array. For example, the activation sites S03 and S10, both locate in Zhangzhou sea area with the total capacity of 8, 000in3 and 12, 000in3 respectively, has the similar activation environment. However, the energy ratios of P-wave and S-wave are both about 2.0, indicating that when the total capacity of the gun array increases by half, the energy is nearly doubled.

 Fig. 6 Histogram of equivalent magnitudes at fixed activation sites
4.2 Analysis of the Attenuation Characteristics of Peak Velocity of Airgun Source

Fig. 7 is the comparison of the attenuation relations of the 5-component peak velocities of P-wave and S-wave. It can be concluded from Fig. 4 and Fig. 5 that: ① In general, the attenuation trends of peak velocities are basically consistent with the distribution of statistical samples, reflecting the attenuation trends of sample values. According to the attenuation coefficients in Table 2, the peak velocity increases with equivalent magnitude and decreases with epicentral distance. ② According to the fitting results, the most applicable range for the aforementioned attenuation relationship is 100-400km, which is caused by sample distribution. In addition, the near-field attenuation relationship needs more near-field data support. ③ By comparing the attenuation relations of different components, it can be seen that the peak velocity of P-wave of the U-D component is the highest, while the S-wave is the opposite. This is consistent with the general idea, and the four components in the horizontal direction do not differ much.

 Fig. 7 The comparison of peak velocity attenuation relations of P-wave and S-wave of 5 components (a) P-wave; (b) S-wave
4.3 The Idea of Applying the Characteristics of Peak Velocity Attenuation to the Design of Experimental Observation System

In the underground medium detection or monitoring experiment with airgun as the main artificial seismic source, when the observation system is designed, the relationship between activation times and signal-to-noise ratios at different epicentral distances should be scientifically and reasonably determined according to the detection or monitoring scope. Therefore, we try to explore the relationship between activation times and signal-to-noise ratios received at different epicentral distances based on the attenuation relation of peak velocity, and this idea can be used for reference in the design of observation system. The idea is as follows: firstly, the noise level N of observation stations with different epicentral distances are determined within the detection or monitoring range. It is also assumed that under the premise of the same epicentral distances, different azimuth angles, and the same airgun signal energy, the airgun source equivalent magnitude and peak velocity attenuation relationship can be used to estimate peak velocity F at different epicentral distances. And the equivalent magnitude of the airgun source can be estimated according to the similar activation ability and activation environment, or it can be estimated by using the experimental method of 10-25 times of activation, then F/N is used as the signal-to-noise ratio of the observation station at a single activation. Combined with the relation that the SNR increases by $\sqrt n$ after n times of activation, we can obtain that the SNR after n times of activation is $\sqrt{n} \times F / N$. In this way, we can determine the number of activation times required by different stations according to the pre-set signal SNR requirements, and the maximum number of activation times in all stations is taken as the number of activation times in the experimental design.

For example, we assume that the noise level N of the vertical components of five observation stations with epicentral distance of 50, 100, 150, 200 and 250km are 200, 200, 100, 150 and 100nm respectively, and the equivalent magnitude of P-wave is 0.7. Using the attenuation relation corresponding to P (U-D) in Table 2, the relation curves of activation times and SNR at different epicentral distances can be calculated (Fig. 8). If the experiment requires that the SNR of P-wave observation signal within 250km is at least greater than 3, it can be judged from figure 8 that at least about 500 times of activation are needed.

 Fig. 8 The relation curves of activation times and SNR at different epicentral distances
5 CONCLUSIONS

Based on the observation records of airgun source activation at fixed sites in Fujian and Taiwan Strait, we analyze and discuss the estimation of the activation efficiency and peak velocity, and draw the following conclusions: within the epicentral distance of 270km, seismic wave signals activated by the airgun source can be received by all fixed seismic observation stattions; the activation efficiency increases with the enhancement of the total capacity of the airgun array. The energy of S-wave generated by the local airgun source is greater than that of P-wave, and different activation environment results in great difference in activation efficiency; Besides, the peak velocity increases with the equivalent magnitude and decreases with the epicentral distance, and the peak velocity of the vertical component of P-wave is the greatest among the three components. In contrast, S-wave has the smallest vertical component and similar horizontal components. We hope this study can provide some help for further investigation of the propagation characteristics of airgun source and broader scientific and engineering application using airgun source.

ACKNOWLEDGEMENT

Several comrades from Fujian Earthquake Agency participated in the experiment of reservoir airgun activation, Seismic Monitoring Center of FJEA provides the required continuous recording data, and anonymous reviewers provided valuable suggestions for the revision of this article, here we extend our heartfelt thanks.

REFERENCES
 Chen Huifang, Lin Binhua, Jin Xing, Wu Lihua, Cai HuitengChen Huifang, Lin Binhua, Jin Xing, Wu Lihua, Cai Huiteng. An experimental study on optimization of large-volume airgun source excitation conditions in a reservoir[J]. Earthquake Research in China, 2016, 30(3): 355-363. Chen Huifang, Lin Binhua, Jin Xing, Cai Huiteng, Guo YangChen Huifang, Lin Binhua, Jin Xing, Cai Huiteng, Guo Yang. Study of the optimal excitation effect of air-gun source experiments in Nanyi reservoir, Fujian Province[J]. China Earthquake Engineering Journal, 2017, 39(4): 667-676 (in Chinese with English abstract). DOI:10.3969/j.issn.1000-0844.2017.04.0667 Chen Qifu, Li Li, Li Gang, Chen Ling, Peng Wentao, Tang Yi, Chen Yong, Wang FuyunChen Qifu, Li Li, Li Gang, Chen Ling, Peng Wentao, Tang Yi, Chen Yong, Wang Fuyun. Seismic features of vibration induced by train[J]. Acta Seismologica Sinica, 2004, 26(6): 651-659 (in Chinese with English abstract). DOI:10.3321/j.issn:0253-3782.2004.06.011 Chen Yong, Zhang Xiankang, Qiu Xuelin, Ge Hongkui, Liu Baojin, Wang BaoshanChen Yong, Zhang Xiankang, Qiu Xuelin, Ge Hongkui, Liu Baojin, Wang Baoshan. A new method to generate seismic wave on land[J]. Chinese Science Bulletin, 2007, 52(11): 1317-1321 (in Chinese). DOI:10.3321/j.issn:0023-074X.2007.11.017(inChinese) Chen Yong, Wang Baoshan, Yao JianhuaChen Yong, Wang Baoshan, Yao Jianhua. Seismic airgun exploration of continental crust structures[J]. Science China Earth Sciences, 2017, 60(10): 1739-1751. DOI:10.1007/s11430-016-9096-6 Ge Hongkui, Lin Jianmin, Wang Baoshan, Song Lili, Luo Guichun, Chen YongGe Hongkui, Lin Jianmin, Wang Baoshan, Song Lili, Luo Guichun, Chen Yong. A field experiment study of improving the seismic detectability with coded source[J]. Chinese Journal of Geophysics, 2006, 49(3): 864-870 (in Chinese with English abstract). DOI:10.3321/j.issn:0001-5733.2006.03.031 General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China, Standardization Administration of China. General Ruler for Earthquake Magnitude(GB 17740-2017)[S]. Beijing: China Standard Press, 2017 (in Chinese). Jiang Shengmiao, Wang Baoshan, Zhang Yunpeng, Chen YongJiang Shengmiao, Wang Baoshan, Zhang Yunpeng, Chen Yong. The influence of noise on the stacking effect of airgun signal and the automatic data screening method[J]. Journal of Seismological Research, 2017, 40(4): 534-542 (in Chinese with English abstract). DOI:10.3969/j.issn.1000-0666.2017.04.004 Jin Chaoyu, Yu YanxiangJin Chaoyu, Yu Yanxiang. Attenuation relationships for horizontal ground motion on rock of small-moderate earthquakes in Europe[J]. Earthquake Research in China, 2009, 25(2): 170-177 (in Chinese with English abstract). DOI:10.3969/j.issn.1001-4683.2009.02.007 Jin Xing, Ma Qiang, Li ShanyouJin Xing, Ma Qiang, Li Shanyou. Real-time simulation of ground displacement and acceleration using digital velocity record[J]. Earthquake Engineering and Engineering Vibration, 2004, 24(6): 9-14, 38 (in Chinese with English abstract). Li Hui, Dai Xuchu, Ge Hongkui, Wang Baoshan, Lin Jianmin, Chen YongLi Hui, Dai Xuchu, Ge Hongkui, Wang Baoshan, Lin Jianmin, Chen Yong. Seismic signal detection and first arrival pickup based on mutual information[J]. Chinese Journal of Geophysics, 2007, 50(4): 1190-1197 (in Chinese with English abstract). DOI:10.3321/j.issn:0001-5733.2007.04.028 Lin Binhua, Li Jun, Jin Xing, Huang Lingzhu, Wei Yongxiang, Chen Huifang, Chen ZhiyongLin Binhua, Li Jun, Jin Xing, Huang Lingzhu, Wei Yongxiang, Chen Huifang, Chen Zhiyong. Development and application of air-gun signal real-time processing system[J]. Progress in Geophysics, 2017a, 32(4): 1808-1816 (in Chinese with English abstract). Lin Binhua, Jin Xing, Li Jun, Cai Huiteng, Liao Shirong, Chen HuifangLin Binhua, Jin Xing, Li Jun, Cai Huiteng, Liao Shirong, Chen Huifang. Station network ambient noise level evaluation and its influence on air gun source excitation effect[J]. Acta Seismologica Sinica, 2017b, 39(3): 330-342 (in Chinese with English abstract). Lin Jianmin, Wang Baoshan, Ge Hongkui, Xu Ping, Chen YongLin Jianmin, Wang Baoshan, Ge Hongkui, Xu Ping, Chen Yong. Characters of large volume air-gun source excitation[J]. Chinese Journal of Geophysics, 2010, 53(2): 342-349 (in Chinese with English abstract). McIntosh K., Nakamura Y., Wang T.K., Shih R.C., Chen A., Liu C.S.McIntosh K., Nakamura Y., Wang T.K., Shih R.C., Chen A., Liu C.S. Crustal-scale seismic profiles across Taiwan and the western Philippine Sea[J]. Tectonophysics, 2005, 401(1/2): 23-54. Melhuish A., Holbrook W.S., Davey F., Okaya D.A., Stern T.Melhuish A., Holbrook W.S., Davey F., Okaya D.A., Stern T. Crustal and upper mantle seismic structure of the Australian Plate, South Island, New Zealand[J]. Tectonophysics, 2005, 395(1/2): 113-135. Nazareth J.J., Clayton R.W.Nazareth J.J., Clayton R.W. Crustal structure of the borderland-continent transition zone of southern California adjacent to Los Angeles[J]. Journal of Geophysical Research, 2003, 108(B8): 2404. DOI:10.1029/2001JB000223 Qiu Xuelin, Chen Yong, Zhu Rixiang, Xu Huilong, Shi Xiaobin, Ye Chunming, Zhao Minghui, Xia ShaohongQiu Xuelin, Chen Yong, Zhu Rixiang, Xu Huilong, Shi Xiaobin, Ye Chunming, Zhao Minghui, Xia Shaohong. The application of large volume airgun sources to the onshore-offshore seismic surveys:implication of the experimental results in northern South China Sea[J]. Chinese Science Bulletin, 2007, 52(4): 553-560. DOI:10.1007/s11434-007-0051-1 Van Avendonk H.J.A., Holbrook W.S., Okaya D., Austin J.K., Davey F., Stern T.Van Avendonk H.J.A., Holbrook W.S., Okaya D., Austin J.K., Davey F., Stern T. Continental crust under compression:a seismic refraction study of South Island Geophysical TransectⅠ, South Island, New Zealand[J]. Journal of Geophysical Research, 2004, 109(B6): B06302. DOI:10.1029/2003JB002790 Wang Baoshan, Zhu Ping, Chen Yong, Niu Fenglin, Wang BinWang Baoshan, Zhu Ping, Chen Yong, Niu Fenglin, Wang Bin. Continuous subsurface velocity measurement with coda wave interferometry[J]. Journal of Geophysical Research, 2008, 113(B12): B12313. DOI:10.1029/2007JB005023 Wang Baoshan, Yang Wei, Yuan Songyong, Guo Shijun, Ge Hongkui, Xu Ping, Chen YongWang Baoshan, Yang Wei, Yuan Songyong, Guo Shijun, Ge Hongkui, Xu Ping, Chen Yong. An experimental study on the excitation of large volume airguns in a small volume body of water[J]. Journal of Geophysics and Engineering, 2010, 7(4): 388-394. DOI:10.1088/1742-2132/7/4/005 Wang Baoshan, Ge Hongkui, Yang Wei, Wang Haitao, Wang Bin, Wu Guohua, Su YoujinWang Baoshan, Ge Hongkui, Yang Wei, Wang Haitao, Wang Bin, Wu Guohua, Su Youjin. Transmitting seismic station monitors fault zone at depth[J]. EOS, Transactions American Geophysical Union, 2012, 93(5): 49-50. DOI:10.1029/2012EO050001 Wei Bin, Su Jinbo, Wang Haitao, Zheng Liming, Wang Qiong, Zhang Wenlai, Yuan Shun, Wei Yunyun, Chen HaoWei Bin, 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). DOI:10.3969/j.issn.1001-4683.2016.02.008 Xia Ji, Jin Xing, Cai Huiteng, Xu JiajunXia Ji, Jin Xing, Cai Huiteng, Xu Jiajun. The time-frequency characteristic of a large volume airgun source wavelet and its influencing factors[J]. Earthquake Research in China, 2016, 30(3): 364-379. Xu Jiajun, Cai Huiteng, Jin Xing, Wang Shanxiong, Xia Ji, Li PeiXu Jiajun, Cai Huiteng, Jin Xing, Wang Shanxiong, Xia Ji, Li Pei. Signal detection of large volume airgun source excitation in the fixed field of the Yangtze River[J]. Earthquake Research in China, 2016, 30(3): 418-429. Zhang Wei, Dai Hengchang, Li Xiangyang, Chen Qifu, Chen Hanlin, Chen YongZhang Wei, Dai Hengchang, Li Xiangyang, Chen Qifu, Chen Hanlin, Chen Yong. Shear wave generated by airguns in a reservoir and its splitting[J]. Seismology and Geology, 2009, 31(1): 22-33 (in Chinese). Zhang Yuansheng, Guo Xiao, Qin Manzhong, Liu Xuzhou, Wei Congxin, Shen Xuzhang, Yan Wenhua, Zou Rui, Yin Liang, Wang Yahong, Sun dianfeng, Feng Hongwu, Zhang Lifeng, Guo YingxiaZhang Yuansheng, Guo Xiao, Qin Manzhong, Liu Xuzhou, Wei Congxin, Shen Xuzhang, Yan Wenhua, Zou Rui, Yin Liang, Wang Yahong, Sun dianfeng, Feng Hongwu, Zhang Lifeng, Guo Yingxia. The construction of active source repeated monitoring in the Qilian Mountains of Gansu Province[J]. Earthquake Research in China, 2016, 32(2): 209-215 (in Chinese with English abstract). Zhao Minghui, Qiu Xuelin, Xia Shaohong, Wang Ping, Xu Huilong, Ye Chunming, Kang YingZhao Minghui, Qiu Xuelin, Xia Shaohong, Wang Ping, Xu Huilong, Ye Chunming, Kang Ying. Large volume air-gun sources and its seismic waveform characters[J]. Chinese Journal of Geophysics, 2008, 51(2): 558-565 (in Chinese with English abstract). Zhou Yun. Civil Engineering Disaster Prevention and Mitigation.Guangzhou: South China University of Technology Press, 2005: 11-12.