【摘要】 本文在地震的发生时间服从Ｐｏｉｓｏｎ过程，而地震震级服从ＧｕｔｅｎｂｅｒｇＲｉｃｈｔｅｒ关系的前提下，对不同定义的加卸载响应比Ｙ值的随机分布进行了探讨。结果表明：当在计算窗口的地震发生的期望数目较大（＞４０）时，Ｙ１～Ｙ５值的分布基本稳定，出现高加卸载响应比的概率极低。然而当计算窗口的地震期望数目过小时，Ｙ２～Ｙ５值则变得不太稳定。也就是说，服从Ｐｏｉｓｏｎ过程的地震序列，在计算窗口的地震期望数目过小时，也可能产生Ｙ值较高的结果。为了使利用加卸载响应比预测地震更加可靠，文中给出了Ｙ１、Ｙ３在Ｐｏｉｓｏｎ模型下的９０％、９５％和９９％的置信区间，这对判别加卸载响应比异常是非常有用的。更多还原

【Abstract】 The random distribution of the loading and unloading respo nse ratios (LURR) of different definitions ( Y 1～Y 5 ) has been discussed b y using of the assumption that the earthquakes occur according to Poisson process and their magnitudes ob ey the Gutenberg Richter law. The results show that: the Y 1～Y 5 are qu ite stable or concentrated when the occurrence rate in the calculation time wind ow are relatively large ( >40 ); but when this occurrence rate becomes quite small, Y 2～Y 5 become quite variable or unstable. That is to say, a high value of LURR can be produced not only from seismicity previous to a large eart hquake, but also from a random series of earthquakes which obeys a Poisson proce ss. To check the influence of random factors in the catalogue to the LURR, the r andom distribution of the LURR under Poisson models by simulation has been calcu lated. 90%,95% and 99% confidence bands of Y 1 and Y 3 are given in this pap er, which is helpful to quantify the random influence and to determine LURR prec ursory anomalies.更多还原