TEC Anomalies Detection for Qinghai and Yunnan Earthquakes on 21 May 2021

Abstract

Earthquake prediction by using total electron content is a commonly used seismic research method. The long short-term memory model is a kind of method to predict time series and has been used for the prediction of total electron content, and the relative power spectrum method is one of the pre-seismic infrared anomaly detection algorithms in the frequency domain. In this paper, a new method combining these two algorithms is used to extract abnormal signals; thus scientists can more easily detect anomalies of total electron content similar to those before the Qinghai and Yunnan earthquakes happened on 21 May 2021. There are pre-seismic anomalies with the high-value relative power spectrum near two epicenters. To validate the correlation between anomalies and earthquakes statistically, the spatiotemporal characteristics of TEC anomalies are analyzed based on connected region recognition. Then, the proportion of earthquake-related anomalies (the correlation rate), the proportion of earthquakes outside the predicted range (the miss rate), and the ratio of the proportion of earthquakes within the predicted range to the spatiotemporal occupancy of anomalies, which is called the probability gain, were used to assess the method. The appropriate parameters of the algorithm for the miss rate below 50% were searched. The highest probability gain is 1.91, which means anomalies of total electron content may decrease the uncertainty of earthquake prediction.

Responses to the Preparation of the 2021 M7.4 Madoi Earthquake in the Lithosphere–Atmosphere–Ionosphere System

“…Because the altitude of the CSES satellite is about 500 km, it cannot record disturbances over the entire ionospheric height like GNSS-TEC can. Yue et al [57] reported that significant TEC anomalies were observed three days before the earthquake. We have identified anomalies recorded on the two orbits that passed over the epicenter on 16 and 21 May.…”

Feasibility of Principal Component Analysis for Multi-Class Earthquake Prediction Machine Learning Model Utilizing Geomagnetic Field Data

“…EQ precursor studies have shown that many short-term precursors are non-seismic, with the ionosphere, atmosphere, and lithosphere being perturbed prior to an EQ [2]. Various methods can be employed for EQ prediction, including the study of precursor phenomena such as fluctuations in electric and magnetic fields [3], variations in the total electron content of the ionosphere [4], observations of animal behavior [5], and the use of multiple remote sensing data sources such as electron and ion density data [6,7]. Hattori et al [8] and Ouyang et al [9], in their studies, observed distinctive perturbations in the spectral density ratio between the horizontal and vertical components of Ultra-Low-Frequency (ULF) geomagnetic field measurements.…”

Anomalies in Infrared Outgoing Longwave Radiation Data before the Yangbi Ms6.4 and Luding Ms6.8 Earthquakes Based on Time Series Forecasting Models

“…Some researchers have made improvements to the LSTM algorithm. A multi-network-based hybrid long and short-term memory (N-LSTM) model was proposed by Senturk et al [15]; Yue et al [16] combined the LSTM model and the relative power spectrum method for TEC anomaly detection; Xiong et al [17] proposed an encoder-decoder extended short-term memory expansion model; all these improved models obtained better performance in their article applications.…”