Modeling Extreme Space Weather Events with MHD-AEPIC

Extreme space weather events may require particle-in-cell (PIC) modeling to capture the kinetic physics of magnetic reconnection that is not present in magnetohydrodynamic (MHD) models. The MHD with Adaptively Embedded Particle-In-Cell (MHD-AEPIC) model (Wang et al., 2022) builds on the operational Geospace space weather model by coupling the FLexible Exascale Kinetic Simulator (FLEKS) PIC code (Chen et al., 2021). The adaptive coupling selects the active PIC domain based on local criteria that identify potential reconnection sites. This saves computational cost compared to large static PIC regions while including full physics where necessary. In this study, the PIC code is activated to follow a flapping plasma sheet and adaptively select areas of potential reconnection. This eliminates the need to constantly cover a huge latitudinal extent with PIC for the entire simulation timeframe. Case studies from Wang et al., 2022 and preliminary results of the Halloween 2003 storm presented at the 2022 Geospace Environment Modeling workshop show the advantages of MHD-AEPIC in representing local particle distributions in the tail and producing global geomagnetic indices. The benefits of MHD-AEPIC over ideal MHD simulation are expected to increase during extreme driving conditions, as the numerical reconnection in ideal MHD may become increasingly inaccurate. In our NSF PREEVENTS project, we study multiple extreme events in detail, including the Halloween 2003 storm, the Bastille Day 2000 storm, and the Carrington-scale July 2012 STEREO-A near miss event. Results are compared to the Geospace model ideal MHD output. Simulations with MHD-AEPIC are robust under extreme driving conditions for these multiple events, forming a foundation for exploring hypothetical driving conditions for assessing the potential impacts of a 'worst-case' storm.