论文部分内容阅读
VAN 地震预报方法是以地震电信号(SES)为基础的。但是,所引起的争论好象仅仅是直接针对地震资料预报统计显著性的。为此我们将逻辑回归方法引用到地震活动性研究中来,估计了希腊地震发生的概率。我们研究的主要目的在于考察是否可以找到一种精细的地震图象,其可以用来明显地提高概率估计。就我们所考察的全部情况来说,如果把地震震级阈值从5.0降到4.3,则可发现概率高达0.75,其可以与 Hamada(1993)所估计的 VAN 方法的成功率相比较。不过,在这些情况下,如此高的概率大部分来自余震资料;如果将余震从资料中删去,则概率降到0.5以下。
The VAN earthquake prediction method is based on seismic electrical signals (SES). However, the controversy that has arisen seems to be that it directly addresses the statistical significance of seismic data. For this reason, we used the method of logistic regression to study the seismicity, and estimated the probability of Greek earthquake. The main purpose of our research is to examine whether a fine seismic image can be found that can be used to significantly improve probability estimation. For all the cases we examined, the probability of finding up to 0.75 if the earthquake magnitude threshold was reduced from 5.0 to 4.3 was comparable to the success rate of the VAN method as estimated by Hamada (1993). However, in these cases, the high probability of such a high probability comes from the aftershock data; if the aftershock is omitted from the data, the probability is reduced below 0.5.