5:15 PM - 6:30 PM
[MIS08-P02] Statistical Analysis of Pre-seismic Ionospheric Electron Density Anomalies and Investigation of the Efficiency of Earthquake Forecasting
Recently, ionospheric anomalies related to earthquakes have been reported and are considered promising for realizing short-term earthquake forecasting. In the Japan region, the previous statistical study during 1958-2017 shows a significant anomaly 1-10 days before the earthquakes with M>=6, depth D<=40 km, epicenter distance R<=350 km in NmF2, which indicates the maximum electron density in the F2 layer. However, the sensitivity of NmF2 anomalies for earthquakes is not clarified. Therefore, we performed statistical analysis to investigate the magnitude, depth, and epicenter distance dependences for earthquakes in which there is a significant positive anomaly in NmF2.
This study used ionosonde data observed at Kokubunji (35.71N, 139.49E), Japan, operated by the National Institute of Information and Communications Technology (NICT). We defined the threshold of the positive anomaly as the value of median + 1.5 IQR of the NmF2 at the same hour in the previous 15 days and the anomalous day as ten or more hours of the anomalies appear in one day. We performed the Superposed Epoch Analysis (SEA) to investigate statistical significance in the correlation between NmF2 anomalies and earthquakes. To evaluate the statistical significance, we repeated the random SEA test 100,000 times.
SEA results show that there is a significant NmF2 anomaly 6-10 days before the earthquakes. NmF2 anomalies have sensitivity for earthquakes with magnitude M>=5.8, depth D<=40 km, epicenter distance R<=350 km. Furthermore, these results indicated the magnitude, depth, and epicenter distance dependences of NmF2 anomalies.
We performed Molchan’s Error Diagram (MED) analysis to evaluate the efficiency of NmF2 anomalies for earthquake forecasting. These results indicated that NmF2 anomalies are a precursor of earthquakes and it is more obvious for the shallower, the larger, and the closer earthquakes. We examined the optimum parameters for short-term earthquake forecast in lead time Δ, alarm window L, magnitude M, depth D, and epicenter distance R with NmF2 anomalies. The result indicated that the most effective forecast is given by M>=6.4, D<=20 km, R<=200 km, Δ=10, L=1. This set of parameters provides the detection of about 46% with the targeted earthquakes and the alarm rate is about 0.07.
This study used ionosonde data observed at Kokubunji (35.71N, 139.49E), Japan, operated by the National Institute of Information and Communications Technology (NICT). We defined the threshold of the positive anomaly as the value of median + 1.5 IQR of the NmF2 at the same hour in the previous 15 days and the anomalous day as ten or more hours of the anomalies appear in one day. We performed the Superposed Epoch Analysis (SEA) to investigate statistical significance in the correlation between NmF2 anomalies and earthquakes. To evaluate the statistical significance, we repeated the random SEA test 100,000 times.
SEA results show that there is a significant NmF2 anomaly 6-10 days before the earthquakes. NmF2 anomalies have sensitivity for earthquakes with magnitude M>=5.8, depth D<=40 km, epicenter distance R<=350 km. Furthermore, these results indicated the magnitude, depth, and epicenter distance dependences of NmF2 anomalies.
We performed Molchan’s Error Diagram (MED) analysis to evaluate the efficiency of NmF2 anomalies for earthquake forecasting. These results indicated that NmF2 anomalies are a precursor of earthquakes and it is more obvious for the shallower, the larger, and the closer earthquakes. We examined the optimum parameters for short-term earthquake forecast in lead time Δ, alarm window L, magnitude M, depth D, and epicenter distance R with NmF2 anomalies. The result indicated that the most effective forecast is given by M>=6.4, D<=20 km, R<=200 km, Δ=10, L=1. This set of parameters provides the detection of about 46% with the targeted earthquakes and the alarm rate is about 0.07.