Pre-earthquake Anomaly Detection And Statistical Analysis Of Swarm Satellite Magnetic Field Based On Time-varying Background Field

China is one of the countries with the most serious earthquake disasters in the world.Major earthquakes such as Wenchuan and Lushan have brought huge losses to people's lives and properties.At present,China has established a multidisciplinary earthquake precursor observation network,and a great number of research on earthquake precursor observation shows that the seismic electromagnetic anomaly is one of the most sensitive short-term earthquake precursors.With the development of satellite observation technology,seismic electromagnetic observation has been extended to the ionosphere.Especially after the successful launch of China Seismoelectromagnetic Satellite,the research on ionospheric seismic electromagnetic anomalies has become a research hotspot.Seismic electromagnetic anomalies have the characteristics of strong interference background and weak change information.The establishment of strong interference background is conducive to accurately identifying and extracting weak change information caused by earthquakes,which is of great significance for the study of earthquake electromagnetic anomalies.The ionospheric magnetic field data varies with the season,local time,and spatial location.Therefore,the establishment of the magnetic field background needs to consider these factors.For the sun-synchronous orbit satellite with only two fixed local times,considering the different seasons,usually the day and night background field is established in all the grids of the earthquake affected area.However,the Swarm satellite is a non-sun-synchronous orbit satellite,and the local time is constantly changing when the satellite flies.The traditional day and night side background field cannot reflect the real time-varying background.In order to remove the influence of local time,the current Swarm satellite data analysis is based on the superposition of single-orbit data with small local time changes to obtain the background of the orbit(100° latitude span).By this way,the change of local time is considered,but the space resolution of the background field is reduced.Since Swarm satellite has the characteristics of local time variation and the single-orbit analysis introduces the reduction of space resolution,we propose to use the magnetic field data of Swarm satellite to establish a seasonal time-varying background field in earthquake affected area,which lays a background data foundation for the extraction of seismic anomalies.SWARM mission measures the geomagnetic field,which mainly includes the main magnetic field,the lithosphere magnetic field and the changing magnetic field.In order to detect the weak magnetic changes caused by the earthquake,we remove the model magnetic field by calculating the CHAOS-7 model,and remove the influence of geomagnetic activity by variational modal decomposition.Then we perform space-time gridding on the earthquake-affected area,and the median value of the Swarm satellite magnetic field data' energy in each space-time grid is used as the background value,so the time-varying background field with the time resolution of 1 hour is established.For the 2020 Jamaica earthquake,we detect the pre-earthquake anomalies of Swarm satellite magnetic field based on the time-varying background field.We set the threshold by sliding quartiles,and compare the thresholds of the time-varying background field and the day and night background field,the result shows that the timevarying background field can extract anomalous orbits more accurately.The cumulative number of anomalous tracks shows accelerated growth from 50 days before the earthquake.In order to exclude the accidentality of the result,we compare the result with the random earthquakes,and the cumulative number of anomalous tracks of random earthquakes indicates a linear increase,which verify the effectiveness of pre-earthquake anomalies.In order to further analyze the pre-earthquake anomalies,we do the comprehensive comparison of the cumulative results of multi-parameter anomalies in the lithosphere-atmosphere-ionosphere,the results show that the time of the anomalies occur in the three spheres is asynchronous.We speculate that ULF electromagnetic disturbance generated by the micro-rupture before the main shock may cause the ionosphere anomaly,and the micro-rupture may generate the subsurface movement of the fluid at the same time,which leads to the anomalies of the atmosphere,besides,the anomalies of the lithosphere may be caused by the twice instability in the stick-slip process of the fault.The anomalies of multi-parameters may be related to the Jamaica earthquake.In order to analyze the pre-earthquake anomalies characteristics of the global earthquake,we take the day of the earthquake as the time zero point,the epicenter position as the spatial zero point,then we perform the Superposed Epoch Analysis method to analyze the large earthquakes greater than magnitude 7 occurred all over the world from January 2017 to June 2020,and then we do the significance test.The results are:(1)54 days,41 days and 20 days before the earthquake,the number of anomalies increases at the significant level of 1%.(2)the number of anomalies per square kilometer is almost inversely proportional to the distance from the epicenter.This result is consistent with the conclusion that the closer to the epicenter,the greater the ionospheric disturbance caused by the earthquake,which further proves the effectiveness of extracting pre-earthquake anomaly information based on the timevarying background field.