- 1University of California, Los Angeles, Department of Atmospheric and Oceanic Sciences, Los Angeles, United States of America (dkondras@atmos.ucla.edu)
- 2The Johns Hopkins University Applied Physics Laboratory, Laurel, Maryland, United States of America (Anthony.Sciola@jhuapl.edu)
Numerical magnetohydrodynamic models (MHD) are often used to simulate the global interaction between the solar wind and the magnetosphere system. Increasingly, such MHD models require very computationally expensive, high numerical resolutions for realistic global magnetosphere simulations of multiscale turbulent plasma flows. To address this problem, we investigate and compare several ML-based approaches for subgrid-scale (SGS) parameterizations in the coarse-scale Grid Agnostic MHD for Extended Research Applications (GAMERA) model, starting with the Large-Eddy Simulation (LES) formalism. We use a 2D simulation of MHD turbulence in the Orszag-Tang vortex as a testbed to diagnose from benchmark high-resolution GAMERA solutions the distributions of subgrid-scale (SGS) and large-scale (LS) fields, and model subgrid-scale (SGS) forcing that encapsulates induced feedbacks on the LS fields.
How to cite: Kondrashov, D. and Sciola, A.: Subgrid-scale modeling of MHD turbulence, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-14724, https://doi.org/10.5194/egusphere-egu25-14724, 2025.