青藏高原东北缘地震活动性广义帕累托模型的全域敏感性分析

任梦依 刘哲

任梦依,刘哲. 2022. 青藏高原东北缘地震活动性广义帕累托模型的全域敏感性分析. 地震学报,44(0):1−14 doi: 10.11939/jass.20210112
引用本文: 任梦依,刘哲. 2022. 青藏高原东北缘地震活动性广义帕累托模型的全域敏感性分析. 地震学报,44(0):1−14 doi: 10.11939/jass.20210112
Ren M Y,Liu Z. 2022. Global sensitivity analysis of the generalized Pareto distribution model for seismicity in the northeast Tibetan. Acta Seismologica Sinica,44(0):1−14 doi: 10.11939/jass.20210112
Citation: Ren M Y,Liu Z. 2022. Global sensitivity analysis of the generalized Pareto distribution model for seismicity in the northeast Tibetan. Acta Seismologica Sinica44(0):1−14 doi: 10.11939/jass.20210112

青藏高原东北缘地震活动性广义帕累托模型的全域敏感性分析

doi: 10.11939/jass.20210112
详细信息
  • 中图分类号:  

Global sensitivity analysis of the generalized Pareto distribution model for seismicity in the northeast Tibetan

  • 摘要: 基于广义帕累托分布构建地震活动性模型,其输入参数取值难以避免不确定性,从而导致依据该模型所得的地震危险性估计结果具有不确定性。鉴于此,本文选取青藏高原东北缘为案例研究区,提出了基于全域敏感性分析的地震危险性估计的不确定性分析流程和方法。首先,利用地震活动性广义帕累托模型,进行研究区地震危险性估计;然后,选取地震记录的起始时间和震级阈值作为地震活动性模型的输入参数,采用具有全域敏感性分析功能的E-FAST方法,对上述两个参数的不确定性以及两参数之间的相互作用对地震危险性估计不确定性的影响进行定量分析。结果表明:地震危险性估计结果(不同重现期的震级重现水平、震级上限及相应的置信区间)对两个输入参数中的震级阈值更为敏感;不同重现期的地震危险性估计结果对震级阈值的敏感性程度不同;对不同的重现期而言,在影响地震危险性估计结果的不确定性上,两个输入参数之间存在非线性效应,且非线性效应程度不同。本文提出的不确定性分析流程和方法,可以推广应用于基于其它类型地震活动性模型的地震危险性估计不确定性分析。

     

  • 图  1  青藏高原东北缘主要活动断裂与MS≥4.0地震分布图

    Figure  1.  Map of main active faults and MS≥4.0 earthquakes distribution in northeastern Tibetan Plateau

    图  2  青藏高原东北缘1880—2020年地震的M-t图(MS≥5.0)

    Figure  2.  M-t map of the earthquakes from 1880 to 2020 in northeastern Tibetan Plateau (MS≥5.0)

    图  3  震级平均超出量分布函数图

    Figure  3.  Diagram of mean excess function of magnitude

    图  4  修饰的尺度参数(a)和形状参数(b)估计值随震级阈值选取的变化图

    Figure  4.  Variation diagram of estimators for modified scale parameter(a)and shape parameter(b) according to the magnitude threshold variation

    图  5  广义帕累托分布拟合诊断图

    (a)概率图;(b)分位数图;(c)重现水平图;(d)概率密度图

    Figure  5.  Diagnostic plots of the generalized Pareto distribution fitted to magnitude

    图  6  地震危险性估计结果对地震目录起始时间ts和震级阈值u的主效应指标Si直方图

    Figure  6.  Histogram of the first-order effect Si of seismic hazard estimation on earthquake catalogue start time ts and magnitude threshold u

    图  8  地震危险性估计结果对地震目录起始时间ts和震级阈值u的主效应指标和交互效应指标直方图

    Figure  8.  Histogram of the first-order effect and interactions on earthquake catalogue start time ts and magnitude threshold u

    图  7  地震危险性估计结果对地震目录起始时间ts和震级阈值u的全效应指标$S^T_i $直方图

    Figure  7.  Histogram of the total effect $S^T_i $ of seismic hazard estimation on earthquake catalogue start time ts and magnitude threshold u

    表  1  不同震级地震的余震时间窗(Gardner,Knopoff,1974

    Table  1.   Aftershock time windows of different magnitudes(Gardner,Knopoff,1974

    主震震级M持续天数/天主震震级M持续天数/天
    4.0426.5790
    4.5837.0915
    5.01557.5960
    5.52908.0985
    6.05108.5985
    下载: 导出CSV

    表  2  青藏高原东北缘地震危险性估计

    Table  2.   Seismic hazard estimation for northeastern Tibetan Plateau

    重现期/年震级置信度95%的置信区间
    207.40 [ 7.10,7.69 ]
    507.80 [ 7.49,8.11 ]
    1008.03 [ 7.69,8.37 ]
    2008.22 [ 7.84,8.59 ]
    5008.41 [ 7.97,8.84 ]
    8.95 [ 8.03,9.87 ]
    下载: 导出CSV

    表  3  模型输入参数服从的概率分布

    Table  3.   Probability distribution of input factors

    模型输入参数概率分布
    地震目录起始时间tsts~ U(1880,1890)
    震级阈值uu ~ U(5.5,5.9)
    下载: 导出CSV

    表  4  地震危险性估计结果对地震目录起始时间ts和震级阈值u的主效应指标Si和全效应指标$S^T_i $

    Table  4.   Total and first-order effects of the seismic hazard estimation on earthquake catalogue start time ts and magnitude threshold u

    重现期/年参数震级置信度95%的置信区间下端点置信度95%的置信区间上端点
    Si$S^T_i $Si$S^T_i $Si$S^T_i $
    20年 u 0.742 4 0.937 8 0.779 1 0.968 4 0.681 3 0.880 5
    ts 0.059 5 0.184 0 0.030 2 0.158 9 0.112 5 0.229 4
    50年 u 0.749 5 0.868 0 0.740 6 0.982 8 0.402 3 0.797 9
    ts 0.117 8 0.178 2 0.015 3 0.182 4 0.138 9 0.502 4
    100年 u 0.374 1 0.678 8 0.709 5 0.987 7 0.570 5 0.964 7
    ts 0.228 6 0.522 3 0.008 8 0.204 5 0.011 8 0.329 7
    200年 u 0.487 6 0.936 6 0.685 8 0.988 9 0.600 7 0.978 9
    ts 0.024 9 0.405 1 0.005 6 0.223 1 0.004 3 0.300 2
    500年 u 0.568 6 0.975 0 0.661 6 0.987 6 0.609 4 0.982 2
    ts 0.004 8 0.329 7 0.004 0 0.244 3 0.003 3 0.292 1
    u 0.460 0 0.958 8 0.345 0 0.929 2 0.386 0 0.940 2
    ts 0.015 2 0.473 8 0.029 5 0.623 5 0.024 2 0.573 0
    下载: 导出CSV
  • 顾功叙. 1983. 中国地震目录: 公元前1831—公元1969年[M]. 北京: 科学出版社: 773–791.
    Gu G X. 1983. Catalogue of Earthquakes in China: 1831BC-1969AD[M]. Beijing: Science Press: 773–791 (in Chinese).
    国家地震局震害防御司. 1995. 中国历史强震目录[M]. 北京: 地震出版社: 496–499.
    State Seismological Bureau. 1995. Catalogue of Historical Strong Earthquakes in China[M]. Beijing: Seismological Press: 496–499 (in Chinese).
    洪明理,任鲁川,霍振香. 2014. 基于E-FAST法分析海啸波高对潜在海啸源参数的敏感性[J]. 地震学报,36(2):252–260.
    Hong M L,Ren L C,Huo Z X. 2014. Sensitivity analysis on maximum tsunami wave heights to the potential tsunami source parameters based on extended FAST method[J]. Acta Seismologica Sinica,36(2):252–260 (in Chinese).
    胡聿贤. 1999. 地震安全性评价技术教程[M]. 北京: 地震出版社: 227–228.
    Hu Y X. 1999. Seismic Safety Evaluation Technology Tutorial[M]. Beijing: Seismological Press: 227–228 (in Chinese).
    胡聿贤, 鹿林. 1990. 地震活动性估计的不确定性[M]//地震危险性分析中的综合概率法. 北京: 地震出版社: 176–177.
    Hu Y X, Lu L. 1990. Uncertainty in seismicity estimation[M]//Synthetic Probability Method in Seismic Hazard Analysis. Beijing: Seismological Press: 176–177 (in Chinese).
    黄玮琼,李文香,曹学锋. 1994. 中国大陆地震资料完整性研究之二:分区地震资料基本完整的起始年分布图象[J]. 地震学报,16(4):423–432.
    Huang W Q,Li W X,Cao X F. 1994. Study on the integrity of seismic data in Mainland China Ⅱ:The initial year distribution images of the complete data on each seismic zone[J]. Acta Seismologica Sinica,16(4):423–432 (in Chinese).
    蒋溥, 戴丽思. 1993. 工程地震学概论[M]. 北京: 地震出版社: 110.
    Jiang P, Dai L S. 1993. Introduction to Engineering Seismology[M]. Beijing: Seismological Press: 110 (in Chinese).
    刘杰,陈棋福,陈顒. 1996. 华北地区地震目录完全性分析[J]. 地震,16(1):59–67.
    Liu J,Chen Q F,Chen Y. 1996. Completeness analysis of the seismic catalog in North China region[J]. Earthquake,16(1):59–67 (in Chinese).
    潘华,李金臣. 2016. 新一代地震区划图的地震活动性模型[J]. 城市与减灾,(3):28–33. doi: 10.3969/j.issn.1671-0495.2016.03.008
    Pan H,Li J C. 2016. A seismicity model for a new generation of seismic zoning maps[J]. City and Disaster Reduction,(3):28–33 (in Chinese).
    钱小仕,王福昌,盛书中. 2013a. 基于广义帕累托分布的地震震级分布尾部特征分析[J]. 地震学报,35(3):341–350.
    Qian X S,Wang F C,Sheng S Z. 2013a. Characterization of tail distribution of earthquake magnitudes via generalized Pareto distribution[J]. Acta Seismologica Sinica,35(3):341–350 (in Chinese).
    钱小仕,蔡晓光,任晴晴. 2013b. 中国大陆活动地块边界带强震震级分布特征研究[J]. 地震工程与工程振动,33(1):212–220.
    Qian X S,Cai X G,Ren Q Q. 2013b. Characteristics of the great earthquake magnitude distributions for active tectonic boundaries in Chinese mainland[J]. Earthquake Engineering and Engineering Vibration,33(1):212–220 (in Chinese).
    任梦依. 2018. 龙门山地区的地震活动性广义帕累托模型构建[J]. 地震研究,41(2):226–232. doi: 10.3969/j.issn.1000-0666.2018.02.010
    Ren M Y. 2018. The establishment of generalized Pareto distribution model of seismicity in Longmenshan region[J]. Journal of Seismological Research,41(2):226–232 (in Chinese).
    任雪梅,高孟潭,俞言祥. 2012a. 基于MGR模型修正我国大陆中强以上地震的震级-频度关系和确定震级极限值[J]. 地震学报,34(3):331–338.
    Ren X M,Gao M T,Yu Y X. 2012a. Modification of magnitude-frequency relation and magnitude limit determination based on MGR model for moderate-strong earthquakes in Chinese mainland[J]. Acta Seismologica Sinica,34(3):331–338 (in Chinese).
    任雪梅,高孟潭,张纳莉. 2012b. 基于MGR模型的我国大陆地区各地震带1970年以来震级·频度关系和震级上限[J]. 中国地震,28(3):320–327.
    Ren X M,Gao M T,Zhang N L. 2012b. Magnitude-frequency relation and magnitude limit of seismic zones based on the MGR model in Chinese mainland since 1970[J]. Earthquake Research in China,28(3):320–327 (in Chinese).
    史道济. 2006. 实用极值统计方法[M]. 天津: 天津科学技术出版社: 28–32.
    Shi D Q. 2006. Practical Extremum Statistical Methods[M]. Tianjin: Tianjin Science and Technology Press: 28–32 (in Chinese).
    宋明丹,冯浩,李正鹏,高建恩. 2014. 基于Morris和EFAST的CERES-Wheat模型敏感性分析[J]. 农业机械学报,45(10):124–131. doi: 10.6041/j.issn.1000-1298.2014.10.020
    Song M D,Feng H,Li Z P,Gao J E. 2014. Global sensitivity analyses of DSSAT-CERES-Wheat model using Morris and EFAST methods[J]. Transactions of the Chinese Society for Agricultural Machinery,45(10):124–131 (in Chinese).
    田建伟,刘哲,任鲁川. 2017. 基于广义帕累托分布的马尼拉海沟俯冲带地震危险性估计[J]. 地震,37(1):158–165.
    Tian J W,Liu Z,Ren L C. 2017. Seismic hazard estimation of the Manila trench subduction zone based on generalized Pareto distribution[J]. Earthquake,37(1):158–165 (in Chinese).
    王健,高孟潭. 1996. 地震危险性分析中的参数敏感性研究[J]. 地震学报,18(4):489–493.
    Wang J,Gao M T. 1996. Study on parameter sensitivity in seismic hazard analysis[J]. Acta Seismologica Sinica,18(4):489–493 (in Chinese).
    徐化超. 2018. 青藏高原东北缘地区主要活动断裂带的运动学研究[D]. 北京: 中国地震局地震预测研究所.
    Xu H C. 2008. Kinematics Study of Main Active Fault Zones in the Northeastern Qinghai-Tibet Plateau[D]. Beijing: Institute of Earthquake Forecasting, CEA (in Chinese).
    徐伟进. 2012. 地震危险性分析中地震时空统计分布模型研究[D]. 北京: 中国地震局地球物理研究所.
    Xu W J. 2012. Study on Space-Time Statistical Distribution Models of Earthquakes in Seismic Hazard Analysis[D]. Beijing: Institute of Earthquake Forecasting, CEA (in Chinese).
    Coles S. 2001. An Introduction to Statistical Modeling of Extreme Values[M]. London: Springer-Verlag: 74–91.
    Cramer C H,Petersen M D,Reichle M S. 1996. A Monte Carlo approach in estimating uncertainty for a seismic hazard assessment of Los Angeles,Ventura,and Orange counties,California[J]. Bull Seismol Soc Am,86(6):1681–1691. doi: 10.1785/BSSA0860061681
    Cukier R I,Fortuin C M,Shuler K E. 1973. Study of the sensitivity of coupled reaction systems to uncertainties in rate coefficients I:Theory[J]. J Chem Phys,59(8):3873–3878. doi: 10.1063/1.1680571
    Cukier R I,Levine H B,Shuler K E. 1978. Nonlinear sensitivity analysis of multiparameter model systems[J]. J Comput Phys,26(1):1–42. doi: 10.1016/0021-9991(78)90097-9
    Dutfoy A. 2019. Estimation of tail distribution of the annual maximum earthquake magnitude using extreme value theory[J]. Pure Appl Geophys,176(2):527–540. doi: 10.1007/s00024-018-2029-0
    Fisher R A,Tippett L H C. 1928. Limiting forms of the frequency distribution of the largest or smallest member of a sample[J]. Math Proc Camb Philos Soc,24(2):180–190. doi: 10.1017/S0305004100015681
    Gan W J,Zhang P Z,Shen Z K,Niu Z J,Wang M,Wan Y G,Zhou D M,Cheng J. 2007. Present-day crustal motion within the Tibetan Plateau inferred from GPS measurements[J]. J Geophys Res:Solid Earth,112(B8):B08416.
    Gardner J K, Knopoff L. 1974. Is the sequence of earthquakes in southern California, with aftershocks removed, Poissonian?[J] Bull Seismol Soc Am, 64(5): 1363-1367.
    Knopoff L,Gardner J K. 1972. Higher seismic activity during local night on the raw worldwide earthquake catalogue[J]. Geophys J Int,28(3):311–313. doi: 10.1111/j.1365-246X.1972.tb06133.x
    Lawrence Livermore National Laboratory (LLNL). 1989. Seismic hazard characterization of 69 nuclear plant sites East of the Rocky Mountains: Methodology, input data and comparisons to previous results for ten test site[J]. NUREG/CR-5250-V1.
    Li Y H,Wang X C,Zhang R Q,Wu Q J,Ding Z F. 2017. Crustal structure across the NE Tibetan Plateau and Ordos block from the joint inversion of receiver functions and Rayleigh-wave dispersions[J]. Tectonophysics,705:33–41. doi: 10.1016/j.tecto.2017.03.020
    McGuire R K,Shedlock K M. 1981. Statistical uncertainties in seismic hazard evaluations in the United States[J]. Bull Seismol Soc Am,71(4):1287–1308.
    McGuire R K. 1987. Seismic hazard uncertainty and its effects on design earthquake ground motions[Z]. Proceeding of International Seminar on Seismic Zonation, 351–359.
    Petersen M D, Frankel A D, Harmsen S C, Mueller C S, Haller K M, Wheeler R L, Wesson R L, Zeng Y H, Boyd O S, Perkins D M, Luco N, Field E H, Wills C J, Rukstales K S. 2008. Documentation for the 2008 Update of the United States National Seismic Hazard Maps[R]. USGS Open-File Report 2008-1128.
    Pisarenko V F,Sornette D. 2003. Characterization of the frequency of extreme earthquake events by the generalized Pareto distribution[J]. Pure Appl Geophys,160(12):2343–2364. doi: 10.1007/s00024-003-2397-x
    Pisarenko V F,Sornette A,Sornette D,Rodkin M V. 2014. Characterization of the tail of the distribution of earthquake magnitudes by combining the GEV and GPD descriptions of extreme value theory[J]. Pure Appl Geophys,171(8):1599–1624. doi: 10.1007/s00024-014-0882-z
    Saltelli A,Tarantola S,Chan K P S. 1999. A quantitative model-independent method for global sensitivity analysis of model output[J]. Technometrics,41(1):39–56. doi: 10.1080/00401706.1999.10485594
    Sobol I M. 1993. Sensitivity estimates for nonlinear mathematical models[J]. Math Model Comput Exp,1(4):407–414.
    Taylor M,Yin A. 2009. Active structures of the Himalayan-Tibetan orogen and their relationships to earthquake distribution,contemporary strain field,and Cenozoic volcanism[J]. Geosphere,5(3):199–214. doi: 10.1130/GES00217.1
    Yin A,Harrison T M. 2000. Geologic evolution of the Himalayan-Tibetan orogen[J]. Annu Rev Earth Planet Sci,28:211–280. doi: 10.1146/annurev.earth.28.1.211
  • 加载中
图(8) / 表(4)
计量
  • 文章访问数:  42
  • HTML全文浏览量:  10
  • PDF下载量:  13
  • 被引次数: 0
出版历程
  • 网络出版日期:  2022-10-21

目录

    /

    返回文章
    返回