Machine learning based automatic foreshock catalog building for the 2019 MS6.0 Changning, Sichuan earthquake
-
摘要:
将深度学习到时拾取、震相关联技术与传统定位方法联系起来，构建一套连续波形自动化处理与地震目录自动构建流程，对于高效充分利用地震资料，提升微震检测能力具有十分重要的意义.我们应用最新发展的迁移学习震相识别技术、震相自动关联技术，对长宁MS6.0地震震中附近21个台站震前半个月（6月1日—6月17日）的连续记录波形进行P、S震相识别、震相自动关联和初步定位，并应用传统绝对定位和相对定位技术得到了长宁地震震前微震活动的绝对和相对定位目录.其中绝对定位目录能在较小的误差范围匹配85%的人工处理目录，其发震时刻平均误差为0.36±0.07 s，震级平均误差为0.15±0.024级，水平定位平均误差为1.45±0.028 km，其识别的1.0级以下微震数目是人工的8倍以上，将长宁地震震前微震目录的检测下限提升至ML-1左右，证明了基于深度学习到时识取和REAL（Rapid Earthquake Association and Location，快速震相关联和定位技术）震相自动关联来构建微震目录具有较好的实用性.我们的自动地震目录揭示了长宁MS6.0主震所发生的区域震前异常频繁的微震活动，以及与区域内盐矿注水井的关联性，更好地描绘了这些微震活动的时空演化特征，其空间活动性分布特征与长宁MS6.0余震序列的分布一致.
Abstract:We establish an automatic workflow for cataloging foreshocks of the 2019 MS6.0 Changning, Sichuan earthquake from continuous raw seismic data by sequentially applying the transfer learning phase picking, automatic phase association, and traditional earthquake location methods. In the workflow, we first pick seismic P and S picks using a deep-neural-network-based phase picker (PhaseNet). By applying the Rapid Earthquake Association and Location (REAL) method, seismic phases are associated into particular earthquakes, whose locations are preliminarily determined at the same time. Then we further sequentially improve earthquake absolute and relative locations using the VELEST and HYPODD software. We apply this workflow to study foreshocks of the 2019 MS6.0 Changning, Sichuan earthquake from June 1 to June 17, 2019. We automatically pick seismic picks from 28 GB continuous waveforms recorded at 21 three-component seismic stations and associate 4358 P phases and 5022 S phases into 1067 earthquakes. We apply the VELEST program to refine their locations, resulting in 870 events after applying critical event selection thresholds based on station gap and travel time residual. At last, 416 earthquakes are relatively relocated by the HYPODD software. Compared with 101 cataloged earthquakes from the Sichuan Earthquake Administration, our absolute VELEST catalog and relative HYPODD catalog can match 85% and 74% of the routine catalog, respectively. The average origin time difference of common events between the absolute VELEST catalog and the routine catalog is 0.36±0.07 seconds, the average magnitude difference is ML0.15±0.024, and the horizontal location error is 1.45±0.028 kilometers. We recover all ML>1.5 cataloged earthquakes with consistent location, except for one ML2.0 event, which was located with large error in the routine catalog. The number of events with ML < 1.0 in our VELEST catalog is more than 8 times than that in the routine catalog. Our earthquake catalog shows that plenty of small foreshocks occurred close to a nearby water injection well for salt mining and the epicenter of the Changning MS6.0 mainshock. Thus, we suggest the MS6.0 Changning mainshock and its foreshocks are most likely related to salt mining in the region, which is supported by the recent InSAR observations as well. This study demonstrates that our workflow not only enables us to automatically and efficiently detect, locate and associate foreshocks of the MS6.0 Changning earthquake, but also significantly improves the completeness magnitude and extends the detection limit to ML-1.
-
Key words:
- Deep learning /
- Foreshock /
- Changning earthquake /
- VELEST catalog /
- HYPODD catalog
-
-
图 1
(a) 研究区域及台站分布.方框为研究区域，也是长宁地震及其余震序列发生地点；白色圆圈表示所用台站的震中距最大范围，即以方框中心为圆心半径75 km范围内；蓝线为人工勘察断层(雷兴林等(Lei et al., 2019))，绿线为构造线；(b)图(a)方框的放大图，圆点为我们的HYPODD重定位目录，其颜色深浅代表事件的发生时间(从2019年6月1日至17日)，红色同心圆为盐矿注水井，图中两个红色虚线方框为地震集中区域，标记为1号和2号区域以便于正文中讨论.
Figure 1.
(a) Study area and station distribution. Black rectangle represents the zoomed-in area in (b), as well as the seismicity zone of the Changning earthquake sequence; White circle with radius 75 km denotes the largest station distance; Blue lines and green lines are the surveyed faults (Lei et al., 2019) and mapped tectonic lines, respectively; (b) Zoomed-in region in (a). Dots denote earthquake locations in our HYPODD catalog, which are colored by their origin time (from June 1 to 17, 2019). Red concentric circle indicates the water injection well for salt mining. Red rectangles represent two regions with large number of earthquakes, which will be discussed in the main text as regions #1 and #2.
图 2
不同算法(从上到下依次为：迁移学习PhaseNet模型，北加州PhaseNet模型，RNN模型，STA/LTA)的自动震相拾取误差分布图，左边四图为P波，右边四图为S波
Figure 2.
Performances of different picking algorithms (from up to down: Transfer learning PhaseNet model, PhaseNet model-Northern California, RNN model, STA/LTA): the left four figures are for P picks and the right four are for S picks
图 3
一个ML1.1地震(2019-06-17T02:44:35.1)的PhaseNet震相识别与REAL关联结果示意图
Figure 3.
An example of PhaseNet phase picking and REAL phase association(Event: 2019-06-17T02:44:35.1, ML1.1)
图 4
(a) P波走时-震中距关系图，其中红点表示REAL关联震相，蓝点表示人工关联震相；(b) S波走时-震中距关系图；(1)(2)为(a)(b)中虚线圈标注的“离群点”所对应的波形，其中蓝色代表P到时，红色代表S到时，实线为机器拾取，虚线为人工拾取
Figure 4.
(a) Travel time to Hypocenter distance curves for the associated P phases. Red and blue dots indicate associated P phase picks in our REAL catalog and the routine catalog, respectively; (b) Same as (a), except for the S picks; (1)(2) are the waveform examples corresponding to "outliers" marked in (a)(b), where blue represents P arrival, red represents S arrival, the solid line is machine learning picks, and the dashed line is manual picks
图 5
(a) 人工地震目录，(b) REAL初步定位目录，(c) VELEST目录和(d)HYPODD目录
Figure 5.
(a) Routine catalog, (b) REAL catalog, (c) VELEST catalog, and (d) HYPODD catalog
图 6
人工目录与VELEST目录的震源参数对比图
Figure 6.
The comparison of source parameters between routine catalog and the VELEST catalog
图 7
VELEST和人工目录中差异较大的共同事件的波形图
Figure 7.
Waveforms of common events with large differences between the VELEST and routine catalog
-
Bergen K J, Johnson P A, de Hoop M V, et al. 2019. Machine learning for data-driven discovery in solid Earth geoscience. Science, 363(6433):eaau0323.
Gutenberg B, Richter C F. 1944. Frequency of earthquakes in California. Bulletin of the Seismological Society of America, 34(4):185-188. doi: 10.1038/156371a0
Kissling E, Ellsworth W L, Eberhart-Phillips D, et al. 1994. Initial reference models in local earthquake tomography. Journal of Geophysical Research:Solid Earth, 99(B10):19635-19646. doi: 10.1029/93JB03138
Lei X L, Wang Z W, Su J R. 2019. The December 2018 ML5.7 and January 2019 ML5.3 earthquakes in South Sichuan basin induced by shale gas hydraulic fracturing. Seismological Research Letters, 90(3):1099-1110. doi: 10.1785/0220190182
Lei X L, Su J R, Wang Z W. 2020. Growing seismicity in the Sichuan Basin and its association with industrial activities. Science China Earth Sciences, 63(11):1633-1660. doi: 10.1007/s11430-020-9646-x
Liu F, Jiang Y R, Ning J Y, et al. 2020. An array-assisted deep learning approach to seismic phase-picking. Chinese Science Bulletin (in Chinese), 65(11):1016-1026, https://doi.org/10.1360/TB-2019-0608. doi: 10.1360/TB-2019-0608
Liu M, Zhang M, Zhu W Q, et al. 2020. Rapid characterization of the July 2019 Ridgecrest, California, earthquake sequence from raw seismic data using machine-learning phase picker. Geophysical Research Letters, 47(4):e2019GL086189, doi:10.1029/2019GL086189.
Park Y, Mousavi S M, Zhu W Q, et al. 2020. Machine-learning-based analysis of the guy-greenbrier, arkansas earthquakes:a tale of two sequences. Geophysical Research Letters, 47(6):e2020GL087032, doi:10.1029/2020GL087032.
Peng Z G, Zhao P. 2009. Migration of early aftershocks following the 2004 Parkfield earthquake. Nature Geoscience, 2(12):877-881. doi: 10.1038/ngeo697
Perol T, Gharbi M, Denolle M. 2018. Convolutional neural network for earthquake detection and location. Science Advances, 4(2):e1700578. doi: 10.1126/sciadv.1700578
Ross Z E, Meier M A, Hauksson E. 2018. P wave arrival picking and first-motion polarity determination with deep learning. Journal of Geophysical Research:Solid Earth, 123(6):5120-5129. doi: 10.1029/2017JB015251
Ross Z E, Trugman D T, Hauksson E, et al. 2019a. Searching for hidden earthquakes in Southern California. Science, 364(6442):767-771. doi: 10.1126/science.aaw6888
Ross Z E, Yue Y S, Meier M A, et al. 2019b. PhaseLink:A deep learning approach to seismic phase association. Journal of Geophysical Research:Solid Earth, 124(1):856-869. doi: 10.1029/2018JB016674
Shelly D R, Beroza G C, Ide S. 2007. Non-volcanic tremor and low-frequency earthquake swarms. Nature, 446(7133):305-307. doi: 10.1038/nature05666
Tang L, Qi G L, Su J R, et al. 2018. Comparison of the new national standard magnitude scales with the traditional magnitude scales. Acta Seismologica Sinica (in Chinese), 40(2):121-131, doi:10.11939/jass.20170075.
Waldhauser F, Ellsworth W L. 2000. A double-difference earthquake location algorithm:Method and application to the northern Hayward fault, California. Bulletin of the Seismological Society of America, 90(6):1353-1368. doi: 10.1785/0120000006
Wang J, Xiao Z W, Liu C, et al. 2019. Deep learning for picking seismic arrival times. Journal of Geophysical Research:Solid Earth, 124(7):6612-6624. doi: 10.1029/2019JB017536
Wang R J, Schmandt B, Zhang M, et al. 2020a. Injection-induced earthquakes on complex fault zones of the Raton Basin illuminated by machine-learning phase picker and dense nodal array. Geophysical Research Letters, 47(14):e2020GL088168, doi:10.1029/2020GL088168.
Wang S, Jiang G Y, Weingarten M, et al. 2020b. InSAR evidence indicates a link between fluid injection for salt mining and the 2019 Changning (China) earthquake sequence. Geophysical Research Letters, 47(16):e2020GL087603, doi:10.1029/2020GL087603.
Yang H F, Zhu L P, Chu R S. 2009. Fault-plane determination of the 18 April 2008 Mount Carmel, Illinois, earthquake by detecting and relocating aftershocks. Bulletin of the Seismological Society of America, 99(6):3413-3420. doi: 10.1785/0120090038
Yang S B, Hu J, Zhang H J, et al. 2020. Simultaneous earthquake detection on multiple stations via a convolutional neural network. Seismological Research Letters, 2020, doi: 10.1785/0220200137.
Yi G X, Long F, Liang M J, et al. 2019. Focal mechanism solutions and seismogenic structure of the 17 June 2019 MS6.0 Sichuan Changning earthquake sequence. Chinese Journal of Geophysics (in Chinese), 62(9):3432-3447, doi:10.6038/cjg2019N0297.
Zhang M, Wen L X. 2015. An effective method for small event detection:match and locate (M&L). Geophysical Journal International, 200(3):1523-1537. http://cpfd.cnki.com.cn/Article/CPFDTOTAL-ZGDW201410002016.htm
Zhang M, Ellsworth W L, Beroza G C. 2019. Rapid earthquake association and location. Seismological Research Letters, 90(6):2276-2284. doi: 10.1785/0220190052
Zhang X, Zhang J, Yuan C C, et al. 2020. Locating induced earthquakes with a network of seismic stations in Oklahoma via a deep learning method. Scientific Reports, 10:1941. doi: 10.1038/s41598-020-58908-5
Zhao M, Chen S, Yuen D. 2019a. Waveform classification and seismic recognition by convolution neural network. Chinese Journal of Geophysics (in Chinese), 62(1):374-382, doi:10.6038/cjg2019M0151.
Zhao M, Chen S, Fang L H, et al. 2019b. Earthquake phase arrival auto-picking based on U-shaped convolutional neural network. Chinese Journal of Geophysics (in Chinese), 62(8):3034-3042, doi:10.6038/cjg2019M0495.
Zhou Y J, Yue H, Kong Q K, et al. 2019. Hybrid event detection and phase-picking algorithm using convolutional and recurrent neural networks. Seismological Research Letters, 90(3):1079-1087. doi: 10.1785/0220180319
Zhu W Q, Beroza G C. 2019. PhaseNet:a deep-neural-network-based seismic arrival-time picking method. Geophysical Journal International, 216(1):261-273. doi: 10.1093/gji/ggy423
Zhu W Q, Mousavi S M, Beroza G C. 2020. Chapter four-Seismic signal augmentation to improve generalization of deep neural networks.//Moseley B, Krischer L ed. Advances in Geophysics. Elsevier, 151-177.
刘芳, 蒋一然, 宁杰远等. 2020.结合台阵策略的震相拾取深度学习方法.科学通报, 65(11):1016-1026, https://doi.org/10.1360/TB-2019-0608. http://www.cnki.com.cn/Article/CJFDTotal-KXTB202011007.htm
唐淋, 祁国亮, 苏金蓉等. 2018.新国家标准震级标度与传统震级标度对比研究.地震学报, 40(2):121-131, doi:10.11939/jass.20170075.
易桂喜, 龙锋, 梁明剑等.2019.2019年6月17日四川长宁MS6.0地震序列震源机制解与发震构造分析.地球物理学报, 62(9):3432-3447, doi:10.6038/cjg2019N0297. http://www.geophy.cn//CN/abstract/abstract15157.shtml
赵明, 陈石, Dave Yuen.2019a.基于深度学习卷积神经网络的地震波形自动分类与识别.地球物理学报, 62(1):374-382, doi:10.6038/cjg2019M0151. http://www.geophy.cn//CN/abstract/abstract14847.shtml
赵明, 陈石, 房立华等.2019b.基于U形卷积神经网络的震相识别与到时拾取方法研究.地球物理学报, 62(8):3034-3042, doi:10.6038/cjg2019M0495. http://www.geophy.cn//CN/abstract/abstract15124.shtml
-
下载:
图(8)
计量
- 文章访问数: 4054
- PDF下载数: 1057
- 施引文献: 0