Authors: Aatiya Ali (Georgia State University), Viacheslav Sadykov (Georgia State University)

Solar energetic particles (SEPs) travel through the heliosphere and interact with Earth’s environment, and can result in unfavorable consequences to numerous facets of life and technology. A subclass of SEP events known as solar proton events (SPEs) are detected as increases in proton flux with energies ≥10 MeV, and are capable of enhancing radiation levels across space, posing risks to astronaut health and impacting satellite calibration and operations. These significant consequences underscore the need for accurate predictions of these events. Building upon our previous work (Ali et al. 2024), which utilized GOES flux data to develop a machine learning-based SPE predictive model, this study advances our methodologies to the cis-lunar environment. Now utilizing data from SOHO EPHIN at the L1 point— currently, the most representative of the lunar environment— this analysis compares SPE characteristics between cis-lunar and geostationary orbits. We hypothesize that variations in particle dynamics between these locations —such as differences in radiation and the absence of magnetic shielding— will manifest as discrepancies in proton flux data when SPEs are detected by both spacecraft. Any disparities observed will need to be taken into account when making SPE predictions, as they could alter factors such as peak fluxes, arrival times, and other parameters as SEPs travel through space. Lastly, to advance our prediction model with a physics-based approach, our input will incorporate plasma temperatures, emission measures, and other potential flare- and CME-related precursors. Our findings will aid SPE forecasting for lunar exploration, potentially benefiting missions like Artemis, and contribute to a comprehensive understanding of SEP behaviors across varying space environments.