This project will use a novel technique in artificial intelligence (AI) and machine learning (ML) to automatically discover the fundamental driving Partial Differential Equations (PDEs) of the Earth’s magnetically trapped high-energy electron population, the so-called radiation belts. Understanding and predicting radiation belt dynamics is essential for protecting the rapidly growing satellite fleet from damage, which has been identified in numerous governmental, and agency reports as a high national priority. Making headway in developing PDE-discovery techniques to work effectively in a demanding situation will open the doors and enable PDE discovery to work in similar challenging environments in the Earth and Space sciences (and beyond), leading to a change in paradigm in how fundamental science is done. The overarching science goal of this project is to develop a methodology that will automatically discover the Partial Differential Equations (PDEs) governing the dynamic evolution of the Phase Space Density (PSD) of radiation belt electrons and use that methodology to enable significant breakthroughs in Geoscience research, namely identifying the driving physical processes at different times and locations during geomagnetically active periods. The proposed study will use a recent multi-spacecraft PSD dataset developed in our group, which is openly available to the public, and leverage innovative approaches in Artificial Intelligence (AI). The proposed activity has the