应用单台垂向记录进行区域地震事件实时检测和直达P波初动自动识别

Real-time detection of regional events and automatic P-phase identification from the vertical component of a single station record

提出一种基于直达P波信号和其它背景噪声在能量、非高斯性、非线性和偏振特性的不同而进行区域地震事件实时检测的新方法信噪综合差异特征量方法(简写为EFGLP方法),同时对比分析了应用信号的不同统计特性来精细识别震相初至的3种有效方法,其中的TOC-AIC方法是新提出的.应用山东数字地震波资料处理的结果表明:①与常规的STA/LTA地震事件触发算法相比,EFGLP方法能够有效降低地震事件的错误报警率和漏报率;②与人机交互震相识别结果相比,当信噪比比较低、震相初至比较模糊时,3种震相精细识别方法中的TOC-AIC方法识别精度最高;当信噪比比较高、震相初至比较清晰时,基于VAR-AIC和TOC-AIC方法所测量得到的震相初至识别基本一致.

In this paper, we present a new method for detecting and identifying regional seismic events in real time based on recognizing the different content between direct P-wave signal and background noise in energy, non-Gaussian characteristics, non-linearity and polarization of P-wave （we called as energy fil- ter, Gauss linearity and polarization method, abbreviated to EFGLP method ）. We use AR-AIC, VAR-AIC and TOC-AIC methods to identify earthquake phases. Here TOC-AIC is a new algorithm. We have found that EFGLP method has lower misinformation ratio and error report ratio than conventional STA/ LTA technique as a seismic event trigger. When signal-to-noise ratio （SNR） is low and the arrival is not from an event, the TOC-AIC method performs well and is the best one among the three methods for direct P-wave onset picking. When SNR is high, all three methods work well.