Modeling the Dynamic Global Distribution of the Ring Current Oxygen Ions Using Artificial Neural Network Technique

Abstract

AbstractThe ring current is an important component of the Earth's near‐space environment, as its variations are the direct driver of geomagnetic storms that can disrupt power grids, satellite communications, and navigation systems, thereby impacting a wide range of technological and human activities. Oxygen ions (O+) are one of the major components of the ring current and play a significant role in both the enhancement and depletion of the ring current during geomagnetic storms. Although a standard statistical study can provide average global distributions of ring current ions, it can't offer insight into the short‐term dynamic variations of the global distribution. Therefore, we employed the Artificial Neural Network technique to construct a global ring current O+ ion model based on the Van Allen Probes observations. Through optimization of the combination of input geomagnetic indices and their respective time history lengths, the model can well reproduce the spatiotemporal variation of the oxygen ion flux distributions and demonstrates remarkable accuracy and minimal errors. Additionally, the model effectively reconstructs the temporal variation of ring current O+ ions for non‐training set data. Furthermore, the model provides a comprehensive and dynamic representation of global ring current O+ ion distribution. It accurately captures the dynamics of O+ ions during a geomagnetic storm with the oxygen ion fluxes enhancement and decay, and reveals distinct characteristics for different energy levels, such as injection from the plasma sheet, outflow from the ionosphere, and magnetic local time asymmetry.