基于残差修正的ARMA模型探测门源MS6.4地震前电离层TEC异常

为提高震前电离层总电子含量(TEC)异常探测的精度，提出利用ARMA模型对预测数据进行残差修正得到TEC参考背景值，并设定相对误差和绝对误差相互制约的限差判决门限。通过对2016-01-21青海门源M_S6.4地震进行研究，验证该方法的有效性。结果表明，震前第5天存在明显的TEC异常现象，震中及附近更大区域内会形成异常范围集中的电离层异常区域，TEC异常峰值点位于震中偏南方向；震中位置与异常区域的中心点并不重合，震中位于负异常区偏东方向、正异常区偏南方向。

ARMA Residual Correction Model for Detecting Ionospheric TEC Anomalies before Menyuan Ms6.4 Earthquake

To improve the detection accuracy of the ionospheric total electronic content(TEC)anomaly,we propose ARMA model to modify the predicted data to obtain the TEC reference background value.The TEC anomaly detection before the earthquake is realized by setting the threshold of relative error and absolute error restricting each other.We analyze the 2016-01-21 Menyuan M_S6.4 earthquake to verify the validity of the method.The results show that there are obvious TEC anomalies at 5 days prior to earthquake.The ionospheric anomaly area with concentrated range is formed in the epicenter and nearby areas.The maximum TEC anomalies appear at the south of the epicenter.Meanwhile,the TEC anomaly areas are not consistent with the epicenter,which locates east of the negative anomaly area or south of the positive anomaly area.