TerraVirtuale: a ERC AdG funded project to model planetary magnetospheres with full electron physics. 

Abstract

The goal is to develop a model of the space environment around a planet based on a particle description of both ions and electrons. At present, models rely on a fluid description of the electrons due to algorithmic and computational challenges. We plan to use the particle in cell (PIC) model where both ions and electrons retain their nature as particles. This PIC model will allow us to investigate the critical role of energetic electrons participating in the energy and matter transfer from the solar wind to the planet inner space. The question about how solar storms impact the planet atmosphere and how planets’ atmospheres respond to the solar wind has both fundamental scientific importance and great social impacts for protecting our infrastructure from the most powerful solar storms.        What makes this goal now possible is the Energy Conserving semi implicit method (ECsim), developed by the author. The ECsim conserves energy exactly, a critical element in the investigation of energy flow from the solar wind. In addition, the energy conservation leads to enhanced numerical stability, which in turn greatly augment ECsim’s capability to simulate very large systems such as planet atmospheres while treating electrons as particles rather than fluid. TERRAVIRTUALE will start from this new development and introduce two critical innovations. First, we will implement the variable spatial and temporal resolution adaptation algorithm to ECsim for finer resolution closer to the planet and in selected areas of interest and coarser elsewhere. Second, we will implement CPU-GPU heterogenous algorithm to utilise new heterogeneous supercomputers developed by HPCEuropa. These new achievements will enhance ECsim capability to conduct PIC model by more than an order of magnitude to simulate the entire Earth’s atmosphere in their interaction with powerful solar winds.        TERRAVIRTUALE, if successful, will be the first PIC model to describe a planetary space environment where the correct particle nature of the electrons is considered with all its implication for the energy and matter transport. The challenge is to successfully augment energy conserving ECsim’s capability to model a region as big as the Earth space environment with the computers available within the next 3-5 years.