Remotely triggered seismicity in north China following the 2008 Mw 7.9 Wenchuan earthquake
© The Society of Geomagnetism and Earth, Planetary and Space Sciences (SGEPSS); The Seismological Society of Japan; The Volcanological Society of Japan; The Geodetic Society of Japan; The Japanese Society for Planetary Sciences; TERRAPUB. 2010
Received: 9 October 2008
Accepted: 9 March 2009
Published: 26 January 2011
We conduct a systematic survey of remote triggering of earthquakes in north China following the 2008 Mw 7.9 Wenchuan earthquake. We identify triggered earthquakes as impulsive seismic energies with clear P and S arrivals on 5 Hz high-pass-filtered three-component velocity seismograms during and immediately after the passage of teleseismic waves. We find clearly triggered seismic activity near the Babaoshan and Huangzhuang-Gaoliying faults southwest of Beijing, and near the aftershock zone of the 1976 MW 7.6 Tangshan earthquake. While several earthquakes occur during and immediately after the teleseismic waves in the aftershock zone of the 1975 Mw 7.0 Haicheng earthquake, the change of seismicity is not significant enough to establish the direct triggering relationship. Our results suggest that intraplate regions with active faults associated with major earthquakes during historic or recent times are susceptible to remote triggering. We note that this does not always guarantee the triggering to occur, indicating that other conditions are needed. Since none of these regions is associated with any active geothermal or volcanic activity, we infer that dynamic triggering could be ubiquitous and occur in a wide range of tectonic environments.
Key wordsDynamic triggering the 2008 Wenchuan earthquake earthquake interaction
It is well known that earthquakes interact with each other. For example, large shallow earthquakes are typically followed by increased seismic activity in nearby regions, known as “aftershocks” (e.g., Utsu et al., 1995). However, only recently seismologists started to realize that large earthquakes could also cause a significant increase of seismic activity in regions that are several hundreds to thousands of kms away (e.g., Hill et al., 1993; Brodsky et al., 2000; Gomberg et al., 2001; Hough and Kanamori, 2002; Kilb et al., 2002; Hough et al., 2003; Gomberg et al., 2004; Prejean et al., 2004; Freed, 2005; Hill and Prejean, 2007; Velasco et al., 2008). Because static stresses resulting from fault displacements fall off more rapidly and become negligible beyond several hundred kms (e.g., Freed, 2005), the elevated seismicity at large distances is generally considered as “remotely triggered” by the dynamic stresses during the passage of seismic waves generated by large earthquakes. Since the immediate cause of these triggered events is apparent and relatively easy to quantify, systematic analysis of remote triggering offers a unique opportunity for better understanding the underlying mechanism of earthquake interaction and physics of earthquake occurrence.
So far most of the remotely triggered earthquakes are identified near active plate boundary faults with abundant background seismicity (e.g., Hill et al., 1993; Gomberg et al., 2001; Prejean et al., 2004). Only a few studies have found remotely triggered earthquakes in relatively stable intraplate regions (e.g., Hough et al., 2003; Gomberg et al., 2004). Recently, Velasco et al. (2008) conducted a systematic global survey of microearthquakes triggered by many large earthquakes. They found that earthquake triggering is a ubiquitous phenomenon that is independent of the tectonic environment of the main earthquake or the triggered event. Their study motivated us to search for further evidence of remote triggering in intraplate regions to better understand the necessary conditions that favor dynamic triggering.
The 2008/05/12 MW 7.9 Wenchuan earthquake offers a unique opportunity to conduct such study. This earthquake occurred in Eastern Sichuan, China, and ruptured predominately northeastwards for ∼250 km along the Longmenshan fold-and-thrust belt that bounds the Tibetan plateau and the Sichuan basin (Burchfiel et al., 2008). The main event and its rigorous aftershock sequences were recorded by many permanent and temporary seismic stations deployed by the China Earthquake Administration (CEA).
2. Analysis Procedure
We download continuous seismic data recorded by stations in the BSN and nearby regional seismic network 6 hour before and after the origin time of the Wenchuan mainshock (2008/05/12 06:28:01.57). We scan through all the records, and remove those bad traces with no seismic data or severely clipped during the surface waves of the Wenchuan earthquake. After the selection process, we obtain a total of 61 stations for further analysis.
Next, we visually identify earthquakes as impulsive seismic energies with clear P and S arrivals on 5 Hz high-pass-filtered three-component velocity seismograms, and compare events in the 6-hour time windows before and after the mainshock origin times. We also checked nearby stations to ensure that the identified earthquake signals are recorded by at least two stations. Since it is not practical to accurately determine the hypocentral locations of most events recorded at a few nearby stations, we estimate the approximate distance to the nearest station based on the travel time differences of P and S waves. We require that the S–P times of the identified events to be less than 15 s, roughly corresponding to a propagating distance of 120 km. Because this distance is much smaller than the distance to the epicenter of the Wenchuan earthquake, we use the location of the nearest station as a proxy for the location of identified earthquake.
Finally, we manually pick earthquakes on the high-pass-filtered seismograms 6 hour before and 1 hour after the Wenchuan P arrival and compute the corresponding β statistic. The analysis procedure is similar to that used in Peng et al. (2007) to identify early aftershocks, and is briefly described here. First, we take the envelope function of the 5 Hz high-pass-filtered seismogram, smooth it by the half width of 50 point, and stack the resulting three-component envelopes to obtain the envelope function for each station. Next, we identify an event by searching for clear double peaks in the envelope that correspond to the P and S arrivals. We also compute the median value of the envelope function with duration of 1 hour before the Wenchuan P arrival, and use 9 times the median absolute deviation (MAD) as an amplitude threshold. Only events with amplitude larger than such threshold are used to compute the β statistic. The threshold is chosen to ensure that we only include events with large enough amplitudes so that they are unlikely to be caused by random fluctuations in the background noise.
Overall, we find three regions that show potential earthquake activity associated with the Wenchuan earthquake. We briefly describe the characteristic of the earthquake activity in each region below, followed by a summary of observations and discussions. We find no clear evidence of earthquake activity during the 1-hour time window at other sites we examine.
3. Triggered Seismicity around Beijing
It is worth noting that none of these events were listed in the regional earthquake catalog. In addition, the seismicity rate from the catalog within 30 days of the Wenchuan earthquake does not show significant changes, with the β value of 0.35. However, the β value for the events picked from the 1-hour high-pass-filtered envelope function before and after the Wenchuan P arrival is 1.4. Since the β value depends on the duration of the time window, we also compute the β statistic using events identified from high-pass-filtered envelope functions with 6 hours before and 1 hour after the Wenchuan P arrival. The resulting β value is 2.6. Although these values are around 2, we still consider them as statistically significant, mainly because of the following reasons. First, the 4 events occurred during the arrivals of the large-amplitude surface waves, and the first event appears to be coincident with the arrival of the long-period Love waves, suggesting a possible correlation between the surface waves and the locally generated seismic events. Secondly, a direct examination of the high-pass-filtered envelope reveals that while there are a few small events that are above the threshold within 1 hour before the Wenchuan P arrival and after the passage of the surface waves, none of them has comparable amplitudes as those occurred during the surface waves of the Wenchuan event. Hence, we conclude that the 4 events are indeed triggered by the large-amplitude surface waves.
4.Triggered Seismicity around Tangshan
5. Triggered Seismicity around Haicheng
The β value from earthquakes within 30 days of the Wenchuan earthquake is 2.94, which is larger than the previous two cases, and is considered as significant (Reasenberg and Simpson, 1992; Hill and Prejean, 2007). However, the β value from the 1-hour high-pass-filtered seismograms is only −1.1. If we increase the analyzed pre-mainshock time window to 6 hours, the β value is still negative. This is mainly because there are quite significant high-frequency energies before the mainshock, some of them have comparable and even larger amplitudes than those we identified during the P and immediately after the surface waves. Because of this, we were unable to establish a direct triggering relationship between the observed seismic activity near the Haicheng aftershock zone and the Wenchuan earthquake.
6. Summary and Discussions
Our study provides additional clear evidence of dynamic triggering within intraplate regions. Overall, we have identified three places in north China with clearly seismic activity during and immediately after the passage of teleseismic waves of the 2008 MW 7.9 Wenchuan earthquake. In the first two cases near Beijing and within the Tangshan aftershock zone, we have confirmed that the triggering effect is statistically significant by computing the β statistic from earthquakes identified from high-pass-filtered seismograms. In the last case near the Haicheng aftershock zone, the seismic activity before the mainshock was relatively high, and we did not observe clear increase of seismicity during and immediately after the Wenchuan earthquake. Hence, the observed seismic events near Haicheng during the P wave and immediately after the surface waves probably occurred by random chance, rather than triggered by the Wenchuan earthquake.
Our study region is in the northern part of the North China Plain, which is an active basin formed under a pullapart condition and crosscut by many conjugate faults in the NNE and NWW directions (Deng et al., 2003). This is consistent with the focal mechanism solutions of moderate and large earthquakes in this region (Fig. 1). Hence, our study area can be best categorized as an intraplate transtensional tectonic regime. While the GPS measurements of crustal movement in north China is on the order of a few mm/yr or less (Wang et al., 2001), elevated seismic activity during 1960s and 1970s with several large damaging earthquakes (including the 1975 Haicheng and 1976 Tangshan earthquakes) suggests that this region is seismically active. Our results are generally consistent with previous observations that remote triggering of microearthquakes is mostly observed at tectonically active extensional and transtensional regimes, although this picture could be partially biased by the uneven sampling in different tectonic regimes (e.g., Hill and Prejean, 2007).
Hough et al. (2003) conducted a survey of remote triggering in stable eastern US after the 1811–1812 New Madrid earthquake sequences, and the 1886 Charleston earthquake, and found that remotely triggered earthquakes tends to occur at sites known to be seismically active during historic times. Their results, together with our observation in the aftershock zone of the 1976 Tangshan earthquake, suggest that regions with elevated background seismicity rates (such as the aftershock zones of previous large earthquakes) are more susceptible to dynamic triggering, most likely because these areas contain nucleation points more frequently closer to failure (e.g., Hill et al., 1993; Hough et al., 2003; Hill and Prejean, 2007; Savage and Marone, 2008). However, we did not find clear triggering effects within the aftershock zone of the 1975 Haicheng earthquake, suggesting that high background seismicity would help, but is not the only condition for triggering to occur.
In all three cases, the most clear triggering effect is observed at stations NKY and DHC about 35 km SW of Beijing. This region defines a distinct lateral variation of lithospheric thickness that coincides with the major boundary between the Taihang Mountain to the west and the Bohai Bay Basin to the east (Chen et al., 2008). Several major parallel faults, such as the Babaoshan and Huangzhuang-Gaoliying faults, cut across this region. Both stations are very close to these active faults. This region is also seismically active in historic times, with several M 6–7 earthquakes occurred in the 17–18th century (e.g., a M∼6.0 event on 1658/02/03 and a M∼6.5 on 1730/09/30). However, currently this region is seismically quiet, with on average 23 M > 1 events each year since 2000. Although we were unable to locate the 4 events occurred during the large-amplitude surface waves, the short S–P times (less than 2 s) observed at station NKY suggest that these triggered earthquakes probably occurred directly on or very close to these parallel faults. Based on the network of leveling, gravity and geomagnetic observations, several studies have suggested that the Babaoshan fault is creeping (at least in the near surface), while the Huangzhuang-Gaoliying fault is currently locked (e.g., Che et al., 1997). At this stage, the relationship between the triggered activity and fault behaviors are still not clear. If the triggered earthquake is associated with the locked Huangzhuang-Gaoliying fault, our observation may suggest that this region is critically stressed and close to fail in future seismic events (e.g., Zoback and Zoback, 2002; Hough et al., 2003).
Many physical models have been proposed to explain how relatively small-amplitude seismic waves can triggered earthquakes at teleseismic distances (Hill and Prejean, 2007; and references therein). We are unable to identify or favor any physical model over others in this study, due to the relatively small number and short duration of observations. We note that our study area is not associated any geothermal or volcanic activities, suggesting that the triggering mechanism does not require conditions that is unique to these areas (e.g., movement of magmas). This is consistent with the recent results by Gomberg et al. (2004) and Velasco et al. (2008), suggesting that dynamic triggering is a ubiquitous phenomenon and can occur in a wide range of tectonic environments. However, we observed only a few triggered earthquakes during and immediately after the passage of teleseismic waves, as compared with swarm-like character of triggered events in geothermal and volcanic regions (e.g., Hill et al., 1993; Prejean et al., 2004). Hence, if the triggering mechanisms and necessary conditions were similar in all these regions, the conditions would be more abundant in geothermal and volcanic regions. Possible candidates include elevated fluid pressures, intensive faults and fractures, elevated background seismicity, or a combination of them.
If we assume plane wave propagation for teleseismic waves, the peak dynamic stress σd is proportional to (Jaeger and Cook, 1979), where G is the shear modulus, is the peak ground velocity, and υs is the phase velocity. The peak ground velocity measured from the instrument corrected vertical seismograms in this region is 1–2 cm/s. Using a nominal G value of 30 GPa, and υs = 3.5 km/s for the Rayleigh waves, we estimate the dynamic stress to be 0.085–0.17 MPa. This is within the range of the dynamic stresses (roughly 0.01 to 0.3 MPa) obtained from previous studies of remote triggering of microearthquakes (e.g., Gomberg et al., 2004; Prejean et al., 2004).
It is interesting to note that none of the identified events are listed in the earthquake catalog compiled by the regional seismic network around the study region, and there is no clear change in seismicity rate within 30 days before and after the Wenchuan earthquake. Hence, based on catalog information alone, one would conclude that no earthquakes are triggered by the Wenchuan earthquake in this region. Similar patterns are observed in California and Pacific Northwest after the 2002 MW 7.8 Denali Fault earthquake (Gomberg et al., 2004; Prejean et al., 2004), suggesting the importance of analyzing continuous seismic recordings for studying remotely triggered earthquakes.
In light of recent discovery of triggered non-volcanic tremor along the subduction zones in southwest Japan (e.g., Miyazawa and Mori, 2006) and Pacific northwest (Rubinstein et al., 2007), and along the San Andreas fault system in California (Gomberg et al., 2008; Peng et al., 2008), we also examined all the data in our study region for possible evidence of triggered tremor. While we saw some elevated tremor-like signals during and immediately after the passage of teleseismic waves, the signals were not coherent among nearby stations and sometimes existed before the P waves of the Wenchuan earthquake. So far we have not find any clear evidence of triggered tremor in this region. We are also in the process of conducting a global survey of triggered earthquakes and non-volcanic tremor associated with the Wenchuan earthquake (Jiang et al., 2008), and the results will be reported elsewhere.
We thank the CEA network center for providing data used in this study. ZP is supported by the National Science Foundation (grant EAR-0809834) and Georgia Tech College of Science Faculty Research Development Grant. The manuscript benefited from valuable comments by the guest editor Dr. Chuanzhen Zhu and the anonymous reviewers. WW and QFC are supported by National Natural Science Foundation of China (grant 40504007). TJ is supported by the “China Young Faculty Scholarship” from the China Scholarship Council.
- Brodsky, E. E., V. Karakostas, and H. Kanamori, A new observation of dynamically triggered regional seismicity: earthquakes in Greece following the August, 1999 Izmit, Turkey earthquake, Geophys. Res. Lett., 27, 2741–2744, 2000.View ArticleGoogle Scholar
- Burchfiel, B. C., L. H. Royden, R. D. van der Hilst, B. H. Hager, Z. Chen, R. W. Kingc, C. Li, J. Lüc, H. Yao, and E. Kirby, A geological and geophysical context for the Wenchuan earthquake of 12 May 2008, Sichuan, People’s Republic of China, GSA Today, 18, 4–11, 2008.View ArticleGoogle Scholar
- Che, Z., Y. Gong, S. Liu, T. Liu, and G. Meng, Comprehensive study of fault activity of the Huangzhuang-Gaoliying and Babaoshan Fault in the middle section, Earthquake Res. China, 13, 330–337, 1997.Google Scholar
- Chen, Q.-F., Y. Chen, and L. Li, China digital seismic network improves coverage and quality, Eos Trans. AGU, 87 (30), doi:https://doi.org/10.1029/2006EO300003, 2006.
- Chen, L., T. Wang, L. Zhao, and T. Y. Zheng, Distinct lateral variation of lithospheric thickness in the northeastern North China Craton, Earth Planet, Sci. Lett., 267, 56–68, 2008.View ArticleGoogle Scholar
- Deng, Q., P. Zhang, Y. Ran, X. Yang, W. Min, and Q. Chu, Basic characteristics of active tectonics of China, Sci. China Ser. D-Earth Sci., 46, 356–372, 2003.Google Scholar
- Freed, A. M., Earthquake triggering by static, dynamic, and postseismic stress transfer, Ann. Rev. Earth Planet. Sci., 33, 335–367, 2005.View ArticleGoogle Scholar
- Gomberg, J., P. A. Reasenberg, P. Bodin, and R. Harris, Earthquakes triggering by seismic waves following the Lander and Hector Mine earthquakes, Nature, 411, 462–465, 2001.View ArticleGoogle Scholar
- Gomberg, J., P. Bodin, K. Larson, and H. Dragert Earthquakes nucleated by transient deformations caused by the M 7:9 Denali, Alaska, earthquake, Nature, 427, 621–624, 2004.View ArticleGoogle Scholar
- Gomberg, J., L. Rubinstein, Z. Peng, K. C. Creager, and J. E. Vidale, Widespread triggering of non-volcanic tremor in California, Science, 319, 173, doi:https://doi.org/10.1126/science.1149164, 2008.View ArticleGoogle Scholar
- Hill, D. P. and S. G. Prejean, Dynamic triggering, in Treatise on Geophysics, 257–292, edited by Schubert, G., Vol. 4, Earthquake Seismology, edited by Kanamori, H., Elsevier, Amsterdam, 2007.View ArticleGoogle Scholar
- Hill, D. et al., Seismicity remotely triggered by the magnitude 7.3 Landers, California, earthquake, Science, 260, 1617–1623, 1993.View ArticleGoogle Scholar
- Hough, S. E. and H. Kanamori, Source properties of earthquakes near the Salton Sea triggered by the 16 October, 1999 M 7.1 Hector Mine earthquake, Bull. Seismol. Soc. Am., 92, 1281–1289, 2002.View ArticleGoogle Scholar
- Hough, S. E., L. Seeber, and J. G. Armbruster, Intraplate triggered earthquakes; observations and interpretation, Bull. Seismol. Soc. Am., 93, 2212–2221, doi:https://doi.org/10.1785/0120020055, 2003.View ArticleGoogle Scholar
- Jaeger, J. C. and N. G. W. Cook, Fundamentals of rock mechanics, 3rd ed., Chapman and Hall, New York, 1979.Google Scholar
- Jiang, T., Z. Peng, W. Wang, and Q.-F. Chen, Global survey of earthquakes and non-volcanic tremor triggered by the 2008 Mw 7.9 Wenchuan earthquake, Eos Trans. AGU, 89 (53), Fall Meet. Suppl., Abstract S23E-08, 2008.Google Scholar
- Kilb, D., J. Gomberg, and P. Bodin, Aftershock triggering by complete Coulomb stress changes, J. Geophys. Res., 107 (B4), 2060, doi:https://doi.org/10.1029/2001JB000202, 2002.View ArticleGoogle Scholar
- Li, L., Q.-F. Chen, and Y. Chen, Relocated seismicity in big Beijing area and its tectonic implication, Prog. Geophys., 2, 24–34, 2007.Google Scholar
- Liu, M., Y. Yang, Z.-K. Shen, S. Wang, M. Wang, and Y. Wan, Active tectonics and intracontinental earthquakes in China: The kinematics and geodynamics, in Continental Intraplate Earthquakes: Science, Hazard, and Policy Issues, edited by S. Stein and S. Mazzotti, 209–318, doi:210.1130/2007.2425 (19), Geol. Soc Am. Special Papers, Boulder, CO, 2007.Google Scholar
- Matthews, M. V. and P. A. Reasenberg, Statistical methods for investigating quiescence and other temporal seismicity patterns, Pure Appl. Geophys., 126, 357–372, 1988.View ArticleGoogle Scholar
- Miyazawa, M. and J. Mori, Evidence suggesting fluid flow beneath Japan due to periodic seismic triggering from the 2004 Sumatra-Andaman earthquake, Geophys. Res. Lett., 33, L05303, doi:https://doi.org/10.1029/2005GL025087, 2006.View ArticleGoogle Scholar
- Peng, Z., J. E. Vidale, M. Ishii, and A. Helmstetter, Seismicity rate immediately before and after main shock rupture from high-frequency waveforms in Japan, J. Geophys. Res., 112, B03306, doi:https://doi.org/10.1029/2006JB004386, 2007.Google Scholar
- Peng, Z., J. E. Vidale, K. C. Creager, J. L. Rubinstein, J. Gomberg, and P. Bodin, Strong tremor near Parkfield, CA excited by the 2002 Denali Fault earthquake, Geophys. Res. Lett., 35, L23305, doi:https://doi.org/10.1029/2008GL036080, 2008.View ArticleGoogle Scholar
- Prejean, S. G. et al., Remotely triggered seismicity on the United States West Coast following the Mw 7.9 Denali Fault earthquake, Bull. Seismol. Soc. Am., 94, S348–S359, 2004.View ArticleGoogle Scholar
- Reasenberg, P. A. and R. W. Simpson, Response of regional seismicity to the static stress change produced by the Loma Prieta earthquake, Science, 255, 1687–1690, 1992.View ArticleGoogle Scholar
- Rubinstein, J. L., J. E. Vidale, J. Gomberg, P. Bodin, K. C. Creager, and S. D. Malone, Non-volcanic tremor driven by large transient shear stresses, Nature, 448, 579–582, 2007.View ArticleGoogle Scholar
- Savage, H. M. and C. Marone, Potential for earthquake triggering from transient deformations, J. Geophys. Res., 113, B05302, doi:https://doi.org/10.1029/2007JB005277, 2008.Google Scholar
- Shedlock, K. M., J. Baranowski, W. Xiao, and X. L. Hu, The Tangshan aftershock sequence, J. Geophys. Res., 92, 2791–2803, 1987.View ArticleGoogle Scholar
- Utsu, T., Y. Ogata, and R. S. Matsu’ura, The centenary of the Omori formula for a decay law of aftershock activity, J. Phys. Earth, 43 (1), 1–33, 1995.View ArticleGoogle Scholar
- Velasco, A. A., S. Hernandez, T. Parsons, and K. Pankow, Global ubiquity of dynamic earthquake triggering, Nature Geosci., 1, 375–379, doi:https://doi.org/10.1038/ngeo204, 2008.View ArticleGoogle Scholar
- Wang, K., Q.-F. Chen, S. Sun, and A. Wang, Predicting the 1975 Haicheng earthquake, Bull. Seismol. Soc. Am., 96 (3), 757–795, doi:https://doi.org/10.1785/0120050191, 2006.View ArticleGoogle Scholar
- Wang, Q., P.-Z. Zhang, J. T. Freymueller, R. Bilham, K. M. Larson, X. Lai, X. You, Z. Niu, J. Wu, Y. Li, J. Liu, Z. Yang, and Q. Chen, Present-day crustal deformation in China constrained by Global Positioning System (GPS) measurements, Science, 294, 574–577, doi:https://doi.org/10.1126/science.1063647, 2001.View ArticleGoogle Scholar
- Woessner, J. and S. Wiemer, Assessing the quality of earthquake catalogues: estimating the magnitude of completeness and its uncertainty, Bull. Seismol. Soc. Am., 95, 684–698, 2005.View ArticleGoogle Scholar
- Zoback, M. D. and M. L. Zoback, State of stress in the Earth’s lithosphere, in International Handbook of Earthquake and Engineering Seismology, Part A, edited by W. H. K. Lee, H. Kanamori, P. C. Jennings, and C. Kisslinger, 559–568, Amsterdam, Academic Press, 2002.View ArticleGoogle Scholar