Abstract: The magnetic field data of Swarm satellite are influenced by geomagnetic activity and local time. In this paper, the influence of geomagnetic activity was firstly removed by variational mode decomposition, and then we built a time-varying background field to eliminate the influence of local time on the data. Based on the established time-varying background field, we used the Swarm satellite magnetic field data to conduct a pre-earthquake anomaly analysis of the 2020 Jamaica earthquake. We calculated the amount of energy of the magnetic data inside the earthquake affected area, then compared with the threshold value of the time-varying background field. If the difference between the amount of energy of a track inside the earthquake affected area and the corresponding background value is over the threshold value, the track was considered to be pre-earthquake anomalies. The results show that the cumulative number of anomalous tracks in function of the time shows accelerated growth from 50 days to 43 days before the earthquake. In addition, we also built day-side and night-side background fields to compare with the time-varying background field. By analyzing the differences between the anomalous orbits extracted based on the two methods, we find that the high-value background fields will be pulled down by the low-value background fields, which leads to some nonanomalous orbits being wrongly identified as anomalous orbits. For the same reason, the low-value background fields will be pulled up by the high-value background fields, which leads to some abnormal orbits cannot be recognized. It is because that the day-side and night-side background fields were built by the data of multiple local times magnetic field data, which cannot identify the great variations of magnetic field at different local time. However, the time resolution of time-varying background field is high, so it can highlight the great variations of magnetic field at different local time. As a result, the establishment of time-varying background field is crucial for the accurate detection of anomalous orbits. Further, we analyzed the cumulative number of anomalies in lithosphere, atmosphere and ionosphere, and explained the time correlation of these anomalies, which proved that these anomalies may be related to the seismogeny of the Jamaica earthquake.