基于图像信息方法的南北地震带地震预测研究

doi:10.12196/j.issn.1000-3274.2023.03.012

摘要/Abstract

摘要： 图像信息法(PI)计算过程涉及到地震活动性的归一化, 因此针对地震活动性相似的地区PI预测效能较好。以南北地震带为研究区域, 对图像信息法(PI)在南北地震带不同范围的预测效能进行了研究。采用中国地震台网中心提供的1970年以来中国地震目录, 选取5年尺度的“异常学习窗”和“预测时间窗”以及1°×1°网格, 以2017年以来发生的M_S≥6地震为目标, 通过ROC检验, 对地震进行回溯性预测研究。研究结果显示: ① PI方法对南北地震带南段、中段、北段区域进行计算的预测效能呈现逐渐降低的趋势, 但幅度较小, 显示PI方法的预测效能可能受到区域内地震活动性强度的影响。 ② PI方法对南北地震带分段计算的预测效能优于对整个南北地震带进行计算的预测效能。其中, 把南北地震带北、中段合起来作为一个计算区域时预测效能最好, 南、中、北各段预测效果次之, 把中、南段合起来作为一个计算区域时预测效能相对较低, 南北地震带整体的预测效能最低。通过地震活动性对南北地震带进行区域划分能使得预测效能提高。 ③ 计算结果显示, 滇西南永德—泸水地区、龙门山断裂南段附近存在“热点”, 未来5年内这些地区为M_S≥6地震值得关注的重点区域。

关键词: 图像信息法(PI), 南北地震带, 强震预测, ROC检验

Abstract: We study the prediction efficiency of Pattern Informatics method (PI) in different areas of the North-South Seismic Belt in this paper. The earthquake catalogue since 1970 is taken from the China Earthquake Networks Center. Both the change interval and forecast interval are fixed as 5 years and the grid size is taken as 1°×1°. We have the retrospective forecasting for 6 earthquakes above M_S6.0 in the study region since 2017. The forecasting efficiency of PI is tested by ROC (Receiver Operating Characteristic) method. The results show that: ① The prediction efficiency of the south, middle and north sections of the North-South Seismic Belt shows a trend of decreasing gradually, but the change is small. This indicates that the prediction performance of PI method is affected by the intensity of regional seismicity to a certain extent. ② The prediction efficiency of piecewise calculation for the North-South Seismic Belt is better than that of the whole calculation for the North-South Seismic Belt. In terms of prediction efficiency, the north and middle sections of the North-South Seismic Belt are the best, the south, middle and north sections are the second, the middle and south sections are relatively low, and the North-South Seismic Belt as a whole is the lowest. The analysis shows that the calculation process of PI method involves the normalization of seismicity, so PI prediction performance is better for areas with similar seismicity. Hot spots generated by PI in regions with different seismicity are mainly generated in areas with high seismicity and the prediction efficiency is low. Therefore, regional division of the North-South Seismic Belt by seismicity can improve the prediction efficiency. ③ There are persistent high probability “hot spots” in Yongde-Lushui area in southwest Yunnan and near the southern section of Longmenshan Fault. These areas should be concerned for earthquakes with M_S≥6 in the next five years.

Key words: Pattern informatics methods, North-South seismic belt, Strong earthquake prediction, ROC test

中图分类号:

P315.75

田唯熙, 张永仙. 基于图像信息方法的南北地震带地震预测研究[J]. 地震, 2023, 43(3): 159-177.

TIAN Wei-xi, ZHANG Yong-xian. Study on Earthquake Forecast in the North-South Seismic Belt by Pattern Informatics Method[J]. EARTHQUAKE, 2023, 43(3): 159-177.

参考文献

[1] 康来迅. 中国南北地震带的基本特征及其形成机理之探讨[J]. 地壳形变与地震, 1991, 11(4): 76-85.
KANG Lai-xun. Discussion on the basic characteristics and formation mechanism of the South-North seismic zone in China [J]. Crustal Deformation and Earthquake, 1991, 11(4): 76-85 (in Chinese).
[2] Rundle J B, Klein W, Tiampo K F, et al. Linear pattern dynamics in nonlinear threshold systems[J]. Physical Review E Statistical Physics, Plasmas, Fluids, and Related Interdisciplinary Topics, 2000, 61(3): 2418-2431.
[3] Rundle J B, Klein W, Turcotte D L, et al. Precursory seismic activation and critical-point phenomena[J]. Pure and Applied Geophysics, 2000, 157: 2165-2182.
[4] Rundle J B, Tiampo K F, Klein W, et al. Self-organization in leaky threshold systems: The influence of near-mean field dynamics and its implications for earthquakes, neurobiology, and forecasting[J]. Proceedings of the National Academy of Sciences, 2002, 99(1): 2514-2521.
[5] Tiampo K F, Rundle J B, McGinnis S A, et al. Pattern dynamics and forecast methods in seismically active regions[J]. Pure and Applied Geophysics, 2002, 159(10): 2429-2467.
[6] Chen C C, Rundle J B, Li H C, et al. From tornadoes to earthquakes: Forecast verification for binary events applied to the 1999 Chi-Chi, Taiwan, earthquake[J]. Terrestrial, Atmospheric and Oceanic Sciences, 2006, 17(3): 503-516.
[7] 蒋长胜, 吴忠良, 马宏生, 等. PI算法用于川滇—安达曼—苏门答腊地区7.0级以上强震危险性预测的回溯性检验[J]. 地震学报, 2009, 31(3): 307-318.
JIANG Chang-sheng, WU Zhong-liang, MA Hong-sheng, et al. Sichuan-Yunnan versus Andaman-Sumatra: PI approach and retrospective forecast test[J]. Acta Seismologica Sinica, 2009, 31(3): 307-318 (in Chinese).
[8] Nanjo K Z, Rundle J B, Holliday J R, et al. Pattern informatics and its application for optimal forecasting of large earthquakes in Japan[J]. Pure and Applied Geophysics, 2006, 163: 2417-2432.
[9] Nanjo K Z, Holliday J R, Chen C C, et al. Application of a modified pattern informatics method to forecasting the locations of future large earthquakes in the central Japan[J]. Tectonophysics, 2006, 424(3-4): 351-366.
[10] Holliday J R, Rundle J B, Tiampo K F, et al. Systematic procedural and sensitivity analysis of the pattern informatics method for forecasting large (M>5) earthquake events in southern California[J]. Pure and Applied Geophysics, 2006, 163(11): 2433-2454.
[11] Mohanty W K, Mohapatra A K, Verma A K, et al. Earthquake forecasting and its verification in northeast India[J]. Geomatics, Natural Hazards and Risk, 2016, 7(1): 194-214.
[12] Radan M Y, Hamzehloo H, Peresan A, et al. Assessing performances of pattern informatics method: A retrospective analysis for Iran and Italy[J]. Natural Hazards, 2013, 68(2): 855-881.
[13] Jiang C S, Wu Z L. Retrospective forecasting test of a statistical physics model for earthquakes in Sichuan-Yunnan region[J]. Science in China Series D: Earth Sciences, 2008, 51(10): 1401-1410.
[14] Zhang Y X, Zhang X T, Wu Y J, et al. Retrospective study on the predictability of pattern informatics to the Wenchuan M8.0 and Yutian M7.3 earthquakes[J]. Pure and Applied Geophysics, 2013, 170(1-2): 197-208.
[15] Zhang Y X, Xia C Y, Song C, et al. Test of the predictability of the PI method for recent large earthquakes in and near Tibetan Plateau[J]. Pure and Applied Geophysics, 2017, 174(6): 2411-2426.
[16] Chang L Y, Chen C C, Telesca L, et al. Pattern informatics and the soup-of-groups model of earthquakes: A case study of Italian seismicity[J]. Pure and Applied Geophysics, 2020, 177(9): 4089-4096.
[17] 尼鲁帕尔·买买吐孙, 余怀忠. 利用图像信息方法研究库车M_S5.6、于田M_S6.4地震前异常演化[J]. 内陆地震, 2020, 34(3): 215-222.
Nilupaer Maimaitusun, YU Huai-zhong. Anomalous evolution before Kuche M_S5.6 and Yutian M_S6.4 earthquake by pattern informatics method[J]. Inland Earthquake, 2020, 34(3): 215-222 (in Chinese).
[18] 余怀忠, 于晨, 张小涛. 2020年6月26日于田M_S6.4地震前的多重地震参数异常研究[J]. 内陆地震, 2020, 34(3): 207-214.
YU Huai-zhong, YU Chen, ZHANG Xiao-tao. Study on multiple seismic parameter anomalies before Yutian M_S6.4 earthquake on June 26^th, 2020[J]. Inland Earthquake, 2020, 34(3): 207-214 (in Chinese).
[19] 蒋长胜, 吴忠良. 对地震预测的一个统计物理算法在川滇地区的回溯性预测检验[J]. 中国科学D辑: 地球科学, 2008, 38(7): 852-861.
JIANG Chang-sheng, WU Zhong-liang. Retrospective forecasting test of a statistical physics model for earthquakes in Sichuan-Yunnan region[J]. Science in China Series D: Earth Sciences,2008, 38(7): 852-861 (in Chinese).
[20] Zhao Y Z, Wu Z L, Jiang C S, et al. Reverse tracing of precursors applied to the annual earthquake forecast: Retrospective test of the annual consultation in the Sichuan-Yunnan region of southwest China[J]. Pure and Applied Geophysics, 2010, 167(6-7): 783-800.
[21] 张小涛, 张永仙, 夏彩韵, 等. 利用图像信息方法研究芦山M_S7.0地震前川滇及附近地区的图像异常[J]. 地震学报, 2014, 36(5): 780-789.
ZHANG Xiao-tao, ZHANG Yong-xian, XIA Cai-yun, et al. Anomalous seismic activities in the Sichuan-Yunnan region and its adjacent areas before the Lushan M_S7.0 earthquake by the pattern informatics method[J]. Acta Seismologica Sinica, 2014, 36(5): 780-789 (in Chinese).
[22] 张盛峰. “地震可预测性合作研究”计划(CSEP)南北地震带试验区地震活动的一些统计地震学特征及其在地震预测中的应用[D]. 北京: 中国地震局地球物理研究所, 2015.
ZHANG Sheng-feng. Statistical-seismological features of seismicity in the CSEP testing region of the central China South-North seismic zone with implications for earthquake predictability[D]. Beijing: Institute of Geophysics, China Earthquake Administration, 2015 (in Chinese).
[23] 袁伏全, 孙世瑞, 郭鹏, 等. PI算法在2021年玛多M_S7.4和比如M_S6.1地震回溯性检验中的应用[J]. 地震地磁观测与研究, 2021, 42(S1): 136-138.
YUAN Fu-quan, SUN Shi-rui, GUO Peng, et al. Retrospective test of the PI algorithm: Case studies of the Maduo M_S7.4 and Biru M_S6.1 earthquakes in 2021[J]. Seismological and Geomagnetic Observation and Research, 2021, 42(S1): 136-138 (in Chinese).
[24] 袁伏全, 孙世瑞, 王小玲. PI算法用于青海地区中强震危险性预测的回溯性检验研究[J]. 地震研究, 2016, 39(S1): 76-82+133.
YUAN Fu-quan, SUN Shi-rui, WANG Xiao-ling. Retrospective forecast test study in seismic risk prediction of medium-strong earthquakes in Qinghai Region by PI Algorithm[J]. Journal of Seismological Research, 2016, 39(S1): 76-82+133 (in Chinese).
[25] 余娜, 张晓清, 袁伏全, 等. 基于PI算法的青藏块体强震回溯性研究[J]. 地震地磁观测与研究, 2021, 42(1): 35-41.
YU Na, ZHANG Xiao-qing, YUAN Fu-quan, et al. A retrospective study on strong earthquakes of Qinghai-Tibet block by the Pattern Informatics[J]. Seismological and Geomagnetic Observation and Research, 2021, 42(1): 35-41 (in Chinese).
[26] Swets J A. The relative operating characteristic in psychology: A technique for isolating effects of response bias finds wide use in the study of perception and cognition[J]. Science, 1973, 182(4116): 990-1000.
[27] Molchan G M. Earthquake prediction as a decision-making problem[J]. Pure and Applied Geophysics, 1997, 149(1): 233-247.
[28] Shaw B E, Carlson J M, Langer J S. Patterns of seismic activity preceding large earthquakes[J]. Journal of Geophysical Research, 1992, 97(B1): 479-488.
[29] Jaumé S C, Sykes L R. Changes in state of stress on the southern San Andreas fault resulting from the California earthquake sequence of April to June 1992[J]. Science, 1992, 258(5046): 1325-1328.
[30] Rundle J B, Klein W. New ideas about the physics of earthquakes[J]. Reviews of Geophysics, 1995, 33(S1): 283-286.
[31] Wu Y H, Chen C C, Rundle J B. Precursory small earthquake migration patterns[J]. Terra Nova, 2011, 23(6): 369-374.
[32] 汪一鹏, 马瑾, 李传友. 南北地震带强震迁移特征及其与南亚地震带的联系[J]. 地震地质, 2007, 29(1): 1-14.
WANG Yi-peng, MA Jin, LI Chuan-you. The migration characteristics of strong earthquakes on the North-South seismic belt and its relation with the South Asia seismic belt[J]. Seismology and Geology, 2007, 29(1): 1-14 (in Chinese).
[33] 杨兴悦, 杨立明, 陈继峰, 等. 南北地震带南北段地震迁移现象[J]. 地震地磁观测与研究, 2012, 33(5-6): 7-12.
YANG Xing-yue, YANG Li-ming, CHEN Ji-feng, et al. Discussion on earthquake migration phenomenon of northern and southern zone in north-south seismic belt[J]. Seismological and Geomagnetic Observation and Research, 2012, 33(5-6): 7-12 (in Chinese).
[34] 阚荣举, 张四昌, 晏凤桐, 等. 我国西南地区现代构造应力场与现代构造活动特征的探讨[J]. 地球物理学报, 1977, 20(2): 96-109.
KAN Rong-ju, ZHANG Si-chang, YAN Feng-tong, et al. Present tectonic stress field and its relation to the characteristics of recent tectonic activity in Southwestern China[J]. Chinese Journal of Geophysics, 1977, 20(2): 96-109 (in Chinese).
[35] 徐锡伟, 张培震, 闻学泽, 等. 川西及其邻近地区活动构造基本特征与强震复发模型[J]. 地震地质, 2005, 27(3): 446-461.
XU Xi-wei, ZHANG Pei-zhen, WEN Xue-ze, et al. Features of active tectonics and recurrence behaviors of strong earthquakes in the western Sichuan Province and its adjacent regions[J]. Seismology and Geology, 2005, 27(3): 446-461 (in Chinese).
[36] 张盛峰. 中国地震科学实验场暨CSEP-CN计划的若干统计地震学问题[D]. 北京: 中国地震局地球物理研究所, 2019.
ZHANG Sheng-feng. Statistical seismological problems associated with CSEP-CN testing region in the context of CSES[D]. Beijing: Institute of Geophysics, China Earthquake Administration, 2019 (in Chinese).
[37] 吴忠良, 丁志峰, 张晓东, 等. 中国地震科学实验场: 历史与未来[J]. 地球与行星物理论评, 2021, 52(2): 234-238.
WU Zhong-liang, DING Zhi-feng, ZHANG Xiao-dong, et al. China seismic experimental site: Retrospective and prospective[J]. Reviews of Geophysics and Planetary Physics, 2021, 52(2): 234-238 (in Chinese).
[38] M7专项工作组. 中国大陆大地震中-长期危险性研究[M]. 北京: 地震出版社, 2012.
M7 Special Working Group. Study on the mid-to long-term potential of large earthquakes on the Chinese content[M]. Beijing: Seismological Press, 2012 (in Chinese).
[39] 况春利, 唐新功, 于恒达, 等. 南北地震带地震活动特征[J]. 国际地震动态, 2019(7): 2-8.
KUANG Chun-li, TANG Xin-gong, YU Heng-da, et al. Study on the seismic distribution and migration law of North-South seismic belt in China [J]. Recent Developments in World Seismology, 2019(7): 2-8 (in Chinese).
[40]Holliday J R, Nanjo K Z, Tiampo K F, et al. Earthquake forecasting and its verification[J]. Nonlinear Processes in Geophysics, 2005, 12(6): 965-977.
[41] Chen C C, Rundle J B, Holliday J R, et al. The 1999 Chi-Chi, Taiwan, earthquake as a typical example of seismic activation and quiescence[J]. Geophysical Research Letters, 2005, 32(22): L22315.
[42] 夏彩韵, 张永仙, 张小涛, 等. 利用两次新疆于田M_S7.3地震对图像信息方法可预测性的检验[J]. 地震学报, 2015, 37(2): 312-322.
XIA Cai-yun, ZHANG Yong-xian, ZHANG Xiao-tao, et al. Predictability test for pattern information method by two M_S7.3 Yutian, Xinjiang, earthquakes[J]. Acta Seismologica Sinica, 2015, 37(2): 312-322 (in Chinese).
[43] 宋程, 张永仙, 夏彩韵, 等. 日本东北M_W9.0地震的PI模型参数设置与预测效能回溯性检验[J]. 地震学报, 2018, 40(4): 491-505.
SONG Cheng, ZHANG Yong-xian, XIA Cai-yun, et al. Retrospective test on the forecasting efficacy of PI models with different parameters for the Tohoku-Oki M_w9.0 earthquake[J]. Acta Seismologica Sinica, 2018, 40(4): 491-505 (in Chinese).
[44] 龙锋, 闻学泽, 徐锡伟. 华北地区地震活断层的震级-破裂长度、破裂面积的经验关系[J]. 地震地质, 2006, 28(4): 511-535.
LONG Feng, WEN Xue-ze, XU Xi-wei. Empirical relationships between magnitude and rupture length, and rupture area, for seismogenic active faults in North China [J]. Seismology and Geology, 2006, 28(4): 511-535 (in Chinese).
[45] 耿冠世, 俞言祥. 中国西部地区震源破裂尺度与震级的经验关系[J]. 震灾防御技术, 2015, 10(1): 68-76.
GENG Guan-shi, YU Yan-xiang. The empirical relationship between subsurface rupture length and earthquake magnitude in western China[J]. Technology for Earthquake Disaster Prevention, 2015, 10(1): 68-76 (in Chinese).
[46] 白超英. 中强地震密集区与未来强震三要素关系的研究[J]. 西北地震学报, 1999, 21(1): 48-54.
BAI Chao-ying. Study on the relationship between concentrated area of moderate and strong earthquakes and three elements of future strong earthquakes[J]. Northwestern Seismological Journal, 1999, 21(1): 48-54 (in Chinese).