论文部分内容阅读
当信号振幅小和噪声水平高而无法测量信号时,提出了一种利用震相振幅内在信息的区域性地震识别法。该法称为二次否定证据识别法,是对Elvers(1974)提出的远震技术的发展,并推广用于区域识别。本文提出的方法是为单个地震台站开发的,利用了区域的P,与L,对比判据中的经验信息(Pomeroy,et al,1983)。我们推导出计算特定台站爆炸漏报和虚报的误识率方程。这些误识率取决于所要求的最小信噪比,且可在一定范围内调至需要的水平。我们还证明了这些方程是对错误地震识别的准确估计。我们认为现在很多估计错误地震识别的方法都过于乐观,用于某些实例偏差还相当大。对中国西部地区数据的分析[P_q与L_q(1.5～3Hz)],被广为应用的爆炸漏报率估计为20％,即认为80％的爆炸可以被准确地识别。而用二次否定证据识别法正确计算爆炸漏报率得出了爆炸可被识别的准确率为73％。
When the signal amplitude is small and the noise level is high and no signal can be measured, a regional seismic identification method using the inherent information of the amplitude of the phase and phase is proposed. This method, called the second negative evidence recognition method, is a development of the teleseismic technique proposed by Elvers (1974) and is applied to regional identification. The proposed method is developed for a single seismic station and uses empirical information from regional P, L, contrast criteria (Pomeroy, et al., 1983). We derive the error rate equation for calculating the false negatives and false positives for a particular station explosion. These misclassifications depend on the minimum signal-to-noise ratio required and can be adjusted to the required level within a certain range. We also show that these equations are accurate estimates of false earthquakes. We think that many methods for estimating erroneous earthquakes are too optimistic now, and the deviations used in some cases are still considerable. An analysis of the data in Western China [P_q and L_q (1.5 ~ 3Hz)] shows that the most widely used rate of undetected explosions is 20%, which means that 80% of the explosions can be accurately identified. However, correctly calculating the false negative rate by using the second negative evidence recognition method, the accurate rate that the explosion can be identified is 73%.