【摘要】 为研究大震前地震空间相关长度的变化特征,分析了青藏块体东北缘6次中强地震前的地震活动的时空演变.在计算空间相关长度过程中,采用长轴与孕震区主要断裂走向一致的椭圆形空间窗,并设计了优化程序.比较椭圆形和圆形空间窗的结果,表明对所研究的6次中强地震前都观测到了增长的地震空间相关长度.椭圆形空间窗的结果比圆形的结果更好,相关长度增长幅度较大,曲率参数值较小.利用3种合成地震目录对结果进行了显著性检验.3种随机数据的检验结果都否定了零假设,表明观察到的模型不是由随机数据干扰造成的,这一观点的置信水平被证实为99%以上.最后得出主震震级与临界区的尺度(等效半径)关系为logRe～0.46(±0.02)M,与理论值符合得较好.更多还原

【Abstract】 For studying the characteristics of seismic correlation length before great earthquakes,the temporal-spatial change of seismicity prior to 6 moderate strong earthquakes(ML>5.5)in the boundaries along the northeast margin of Tibet block have been analyzed.We use ellipse,whose long axis is coincident with the strike of main fault belt in the source area,as the spatial window to analyze growing correlation length prior to main-shock and design the corresponding optimization procedure.Comparison between the results from elliptical and circular spatial windows shows that the growing correlation length prior to 6 moderate strong earthquakes studied have been observed in both two spatial windows.The results from ellipse are better than that from circular,and the growth ranges of correlation length in former are larger than those in later yet the smaller curvature parameter values of C are obtained in elliptical spatial windows.Three types of random synthetic catalog are used to test the significance of the results.The results reject the null hypothesis for three types of random synthetic catalog and show that the observed patterns are not consequences of random fluctuations.That theory is found to be of above 99% confidence levels.The scaling relation we found,log Re～0.46M between the main-shock magnitude M and the equivalent radius Re of the critical region is in good agreement with theoretical values.更多还原