SafeNet: SwArm for Earthquake Perturbations Identification Using Deep Learning Networks

Low Earth orbit satellites collect and study information on changes in the ionosphere, which contributes to the identification of earthquake precursors. Swarm, the European Space Agency three-satellite mission, has been launched to monitor the Earth geomagnetic field, and has successfully shown that in some cases it is able to observe many several ionospheric perturbations that occurred as a result of large earthquake activity. This paper proposes the SafeNet deep learning framework for detecting pre-earthquake ionospheric perturbations. We trained the proposed model using 9017 recent (2014–2020) independent earthquakes of magnitude 4.8 or greater, as well as the corresponding 7-year plasma and magnetic field data from the Swarm A satellite, and excellent performance has been achieved. In addition, the influence of different model inputs and spatial window sizes, earthquake magnitudes, and daytime or nighttime was explored. The results showed that for electromagnetic pre-earthquake data collected within a circular region of the epicenter and with a Dobrovolsky-defined radius and input window size of 70 consecutive data points, nighttime data provided the highest performance in discriminating pre-earthquake perturbations, yielding an F1 score of 0.846 and a Matthews correlation coefficient of 0.717. Moreover, SafeNet performed well in identifying pre-seismic ionospheric anomalies with increasing earthquake magnitude and unbalanced datasets. Hypotheses on the physical causes of earthquake-induced ionospheric perturbations are also provided. Our results suggest that the performance of pre-earthquake ionospheric perturbation identification can be significantly improved by utilizing SafeNet, which is capable of detecting precursor effects within electromagnetic satellite data.

EARTHQUAKE PREDICTION EVALUATION BASED ON VLF DATA USING A NOVEL INTERSECTION-UNION METHOD

Electromagnetic phenomena, especially those in the Very Low Frequency/Low Frequency (VLF/LF) bands are promising for short-term earthquake prediction. Seismo-ionospheric perturbations cause a variety of changes in different receiver-transmitter VLF/LF signal paths. Therefore, independent and simultaneous observations at different points thus in different VLF/LF signal propagation paths are necessary to better predict the earthquake. Most of the previous research on VLF data have been based on one path or limited number of paths which examined perturbations in the time domain and less attention has been paid to estimate the location of the earthquake. In the present research, using wavelet analysis, the temporal variations of seismo-ionospheric perturbations and the approximate time of earthquake are predicted. Clear disturbances are observed two weeks before the Kumamoto earthquake happened in Japan in 2016. The novelty of this study is to present an approach called Intersection-Union method to predict earthquake location. Based on the geometry of a VLF/LF network, the Intersection-Union method was introduced to estimate the earthquake epicenter. This method is based on the overlay of earthquake occurrence probable areas. With simultaneous use of different propagation paths by the Intersection-Union method, an area with a radius of about 300 km was determined as the probable location of the earthquake epicenter. The accuracy of the proposed method is 300 km compared with 1000 km accuracy of other earthquake location prediction scenarios.