利用并行计算方法实现地震活动速率变化参数Z值的空间扫描处理及其计算效能评价
|
摘要
近30多年的研究表明,地震平静可能是中长期的地震前兆现象,而地震平静的内在特征体现了地震活动速率的变化。Z检验法是检验地震活动速率变化应用最为广泛的方法之一,其度量值即Z值。由于Z值空间扫描中采用等地震数的空间扫描窗口,每个空间扫描窗口的大小通常都是变化的,这样带来了大计算量的问题,因此,本文在自行组建的4节点PC机群上实现了Z值空间扫描的并行化处理。初步测试结果说明,并行化处理中获取了很高的加速比和并行计算效率,可以将此种方法应用于其他地震活动性参数的空间扫描处理。
Nearly 30 years of studies indicate that seismic quiescence may be the medium to long-term precursory phenomenon which contains seismicity rate changes.Z-Test is widely used for significance testing of seismicity rate changes,which is measured by Z-value.As a result of choosing spatial scanning window of fixed number of earthquakes,each window size of grid is usually different,which results in large amount of calculation.In this paper,we have realized the spatial scanning of Z-value based on the self-built cluster consisting of 4 nodes by parallel processing.The preliminary test shows that parallel program can run with high speedup ratio and prarallel efficiency,this is to say,the parallel process method of space scanning of seismicity parameters can be widely used.
Nearly 30 years of studies indicate that seismic quiescence may be the medium to long-term precursory phenomenon which contains seismicity rate changes.Z-Test is widely used for significance testing of seismicity rate changes,which is measured by Z-value.As a result of choosing spatial scanning window of fixed number of earthquakes,each window size of grid is usually different,which results in large amount of calculation.In this paper,we have realized the spatial scanning of Z-value based on the self-built cluster consisting of 4 nodes by parallel processing.The preliminary test shows that parallel program can run with high speedup ratio and prarallel efficiency,this is to say,the parallel process method of space scanning of seismicity parameters can be widely used.
引文
[1]Ogata Y.Space-time model for regional seismicity and detection of crustal stress changes[J].Journalof Geophysical Research-Solid Earth,2004,109(B3):23.
[2]Ohtake M,Matumoto T,Latham G.Seismicity gap near Oaxaca,sourthern Mexico as a probableprecursor to a large earthquake[J].Pure and Applied Geophysics,1977,115:375-385.
[3]Kisslinger C.An experiment in earthquake prediction and the 7 May 1986 Andreanof islands earth-quake[J].Bull Seism Soc Am,1988,78:218-229.
[4]Wyss M,Martirosyan A H.Seismic quiescence before the 1988M7 Spitak earthquake,Armenia[J].Geophysical Journal International,1998,134(2):329-340.
[5]Wiemer S,Wyss M.Seismic quiescence before the Landers(M=7.5)and Big Bear(M=6.5)1992earthquakes[J].Bull Seism Soc Am,1994,84:900-916.
[6]Huang Q,Sobolev G A,Nagao T.Characteristics of the seismic quiescence and activation patternsbefore theM=7.2 Kobe earthquake,January 17,1995[J].Tectonophysics,2001,337:99-116.
[7]Sobolev G A,Huangm Q,Nagao T.Phases of earthquake's preparation and by chance test of seismicquiescence anomaly[J].Journal of Geodynamics,2002,33:413-424.
[8]Yih-Min Wu,Ling-Yun Chiao.Seismic quiescence before the 1999 Chi-Chi,Taiwan,MW7.6 Earth-quake[J].Bull seism Soc Am,2006,96(1):321-327.
[9]Huang Q,Oncel A O,Sobolev G A.Precursory seismicity changes associated with theMW=7.41999 August 17 Izmit(Turkey)earthquake[J].Geophysical Journal International,2002,151:235-242.
[10]戴志阳,陈宇卫,傅容珊,等.前兆性地震平静研究[J].地球物理学进展,2006,21(1):18-24.
[11]Habermann R E.Teleseismic detection in the Aleutian Islands arc[J].J Geophys Res,1983,88:5056-5 064.
[12]Habermann R E.Man-made changes of seismicity rates[J].Bull Seism Soc Am,1987,(77):141-159.
[13]Habermann R E.Seismicity rate variations and systematic changes in magnitudes in teleseismic cata-logs[J].Tectonophysics,1991,193(4):277-289.
[14]Habermann R E.Precursory seismic quiescence:Past,present and future[J].Pure and Applied Ge-ophysics,1988,126(2~4):279-318.
[15]陈国良编著.并行算法结构、算法、编程[M].北京:高等教育出版社,2003.
[2]Ohtake M,Matumoto T,Latham G.Seismicity gap near Oaxaca,sourthern Mexico as a probableprecursor to a large earthquake[J].Pure and Applied Geophysics,1977,115:375-385.
[3]Kisslinger C.An experiment in earthquake prediction and the 7 May 1986 Andreanof islands earth-quake[J].Bull Seism Soc Am,1988,78:218-229.
[4]Wyss M,Martirosyan A H.Seismic quiescence before the 1988M7 Spitak earthquake,Armenia[J].Geophysical Journal International,1998,134(2):329-340.
[5]Wiemer S,Wyss M.Seismic quiescence before the Landers(M=7.5)and Big Bear(M=6.5)1992earthquakes[J].Bull Seism Soc Am,1994,84:900-916.
[6]Huang Q,Sobolev G A,Nagao T.Characteristics of the seismic quiescence and activation patternsbefore theM=7.2 Kobe earthquake,January 17,1995[J].Tectonophysics,2001,337:99-116.
[7]Sobolev G A,Huangm Q,Nagao T.Phases of earthquake's preparation and by chance test of seismicquiescence anomaly[J].Journal of Geodynamics,2002,33:413-424.
[8]Yih-Min Wu,Ling-Yun Chiao.Seismic quiescence before the 1999 Chi-Chi,Taiwan,MW7.6 Earth-quake[J].Bull seism Soc Am,2006,96(1):321-327.
[9]Huang Q,Oncel A O,Sobolev G A.Precursory seismicity changes associated with theMW=7.41999 August 17 Izmit(Turkey)earthquake[J].Geophysical Journal International,2002,151:235-242.
[10]戴志阳,陈宇卫,傅容珊,等.前兆性地震平静研究[J].地球物理学进展,2006,21(1):18-24.
[11]Habermann R E.Teleseismic detection in the Aleutian Islands arc[J].J Geophys Res,1983,88:5056-5 064.
[12]Habermann R E.Man-made changes of seismicity rates[J].Bull Seism Soc Am,1987,(77):141-159.
[13]Habermann R E.Seismicity rate variations and systematic changes in magnitudes in teleseismic cata-logs[J].Tectonophysics,1991,193(4):277-289.
[14]Habermann R E.Precursory seismic quiescence:Past,present and future[J].Pure and Applied Ge-ophysics,1988,126(2~4):279-318.
[15]陈国良编著.并行算法结构、算法、编程[M].北京:高等教育出版社,2003.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心