论文部分内容阅读
试图评估在可能前震发生后主震立即发生的概率,并根据沿日本和千岛海沟的可能前震,提出一种最优预测算法,同时还研究了该区域中前震活动的区域性变化。除了小余震外,每一个特定的地震或地震丛都被当作一次可能前震。确定可能前震的参数为:震级M_f,分区面积大小D°(纬度)×D°(经度),时间间隔T_f以及在该时空范围内所发生的地震次数N_f。但在本项研究中固定T_f为10d。将期望发生主震的空间限定在已经发生所提前震的地段,而期望时间间隔T_a为一可变参数。为了估算可能前震发生后发生震级≥M_(m0)主震的概率,可采用以下3个指标:报警率(AR),实发率(TR)和概率增益(PG)。参数N_f,M_f,D和T_a对概率的评估具有很大影响。可为我们提供完成全部预测算法的、可能前震的最佳参数值,是利用1980～1993年间日本气象厅的震源目录资料,根据前兆信息判据(PIC)来估算的。对于震级≥6的主震,估算值分别为:M_f≥5,N_f=3,T_a=5d,D=0.5°,分别给出了13％,25％,617和75的一组AR,TR,PG和PIC值;对于震级≥5的主震,估算值分别为:M_f≥4,N_f=1,T_a=1d,D=0.25°,它赋予与上述相同指标的值分别为9％,0.8％,56,220。此外,通过利用1926～1993年的资料,我们发现前震活动存在强烈的区域性交化。就是说,有4个区域显示出了前震的高活动性,且主震的震源越深,有前震作前导的主震数的百分比就越小。
An attempt was made to assess the probability of an immediate mainshock occurring immediately after a possible foreshock and to propose an optimal prediction algorithm based on possible foreshocks along the Japan and Thousand Island Trenches while studying the regional changes in foreshock activity in the area. Except for a small aftershock, each specific earthquake or earthquake cluster is considered as a possible foreshock. The parameters that determine the possible foreshocks are the magnitude M_f, the area size D ° (latitude) × D ° (longitude), the time interval T_f and the number of earthquakes N_f occurring in this space-time range. However, the fixed T_f in this study was 10 days. The space where the main shock is expected to occur is limited to the area where the advance earthquake has occurred and the expected time interval T_a is a variable parameter. In order to estimate the probability that a mainshock of magnitude ≥M_ (m0) is likely to occur after a foreshock, the following three indicators may be used: alarm rate (AR), real rate (TR) and probability gain (PG). The parameters N_f, M_f, D and T_a have a great influence on the evaluation of the probability. The best parameter values that may provide us with the possibility of a foreshock to complete the full prediction algorithm are estimated from the precursory information criterion (PIC) using the source catalog data of the Japan Meteorological Agency from 1980 to 1993. For mastiffs with magnitude ≥ 6, the estimated values are: M_f≥5, N_f = 3, T_a = 5d and D = 0.5 °, respectively, giving a set of AR, TR, PG and PIC values. For the main shock with magnitude ≥5, the estimated values are: M_f≥4, N_f = 1, T_a = 1d, D = 0.25 °. , 56,220. In addition, by using the data from 1926 to 1993, we found that there is a strong regional interdependence of foreshock activities. That is, there are 4 regions showing the high activity of the foreshocks, and the deeper the focal point of the main shock, the smaller the percentage of the main shocks with the foreshocks leading.