SWMF/AWSoM_R

The Alfvén Wave Solar Atmosphere Model (AWSoM) is part of the Space Weather Modeling Framework (SWMF), developed by the University of Michigan. The current version allows users to choose between simulating the steady-state ambient solar wind and simulating the propagation of coronal mass ejections (CMEs) in the solar corona and inner heliosphere, along with the particle acceleration processes in the CME-driven shock.

AWSoM

AWSoM is a global magnetohydrodynamic (MHD) model that simulates the solar atmosphere from the upper chromosphere, through the solar corona, and out into the heliosphere. It solves extended MHD equations, including separate temperatures for electrons and ions, anisotropic electron heat conduction, and accounts for kinetic effects like ion-cyclotron, mirror, and firehose instabilities to ensure plasma stability. The model drives coronal heating and solar wind acceleration via nonlinear turbulent dissipation of Alfvén waves. Waves launched from the chromosphere partially reflect due to Alfvén speed gradients and vorticity, generating counter-propagating waves that interact to create a turbulent cascade, dissipating energy into electron and ion heating. This unified modeling framework captures the natural variation of heating rates in open and closed field regions, producing realistic coronal temperatures, densities, and solar wind properties without relying on empirical heating functions.

AWSoM-R

AWSoM-R (AWSoM-realtime) uses the "threaded-field-line" model to accurately simulate the lower solar corona by treating each magnetic field line as an independent 1D threaded structure within the global environment. This approach enables resolving steep gradients in temperature and density along individual magnetic field lines, especially near the transition region between the chromosphere and corona. By combining this threaded architecture with a fully 3D MHD background, the model captures the fine-scale thermodynamic dynamics of each field line without losing the global magnetic connectivity essential for realistic coronal simulation results. AWSoM-R incorporates Alfvén-wave turbulence following the AWSoM framework: waves propagate upward, reflect, and undergo nonlinear interactions, depositing energy as heat into the plasma. AWSoM-R provides a computationally efficient method for modeling low coronal heating and plasma flow processes, bridging the gap between detailed physics and global-scale solar simulations.

Stream-Aligned AWSoM-R

The Stream-aligned AWSoM-R approach enforces a perfect alignment of the plasma velocity (u) and magnetic field (B) in the co-rotating frame, effectively nullifying the electric field E = −u × B = 0 and Poynting flux in the induction and energy equations, while still including the Lorentz (Ampère) force in the momentum equation. This simplification preserves physically valid steady-state solutions—such as Parker spiral structures—without loss of generality compared to full MHD, yet avoids misalignment issues like the “u-shape” or “v-shape” magnetic field lines due to numerical diffusion. Stream-aligned AWSoM-R is crucial for tracing magnetic connectivity from spacecraft back to the Sun and for modeling energetic particle transport and space weather forecasting. It offers the option to generate a steady-state background solar wind for CME and SEP simulation runs.

The current framework offers two core simulation modes including Steady-State Ambient Solar Wind and Real-time CME and SEP Simulation, along with a Shock Capturing Tool for identifying and characterizing interplanetary shocks.

1) Steady-State Ambient Solar Wind

The steady state ambient AWSoM-R solar wind model produces a solution of the ambient corona for a magnetogram selected by the user. The computational domain begins at the top of the chromosphere and includes the transition region, corona, and inner heliosphere. The inner boundary conditions for the magnetic field are derived from a synoptic magnetogram (GONG/MDI/HMI). The initial conditions for the solar wind plasma are based on a Parker solution, while the initial magnetic field is based on the Potential Field Source Surface (PFSS) Model.

The coronal part (1.1 Rs - 24 Rs) iterates for 60,000 steps until the solution has settled into a steady state. The number of steps may change as we study the number of iterations that the solar coronal component needs to converge. Some cases may not reach equilibrium. The user should check the convergence plots provided with the output to confirm convergence to an equilibrium, and provide feedback to model developers. Then the heliospheric component (24 Rs - 1 AU) evolves a solution consistent with the coronal model for 5,000 iterations.

Automatic refinement is performed several times to increase the grid resolution within the current sheet. If the model is used with a CME, an extra refinement is performed for the active region from which the CME originates.

Below are the synthetic outputs of the steady-state solar wind. The figures display both measured and simulated in-situ plasma properties and compare the simulation results with remote sensing EUV imagery.

Figure: Comparison of in-situ plasma properties with ACE spacecraft measurements.

Figure: Remote-sensing visualizations from the model output compared with SDO/AIA observations.

2) Real-time CME and SEP Simulation.

The real time AWSoM-R model provides the capability to simulate the propagation of CME inside the solar corona and interplanetary space, together with the particle acceleration processes in the CME-driven shock and the subsequent transport process in the inner heliosphere. The real time CME/SEP simulation starts from the ambient solar wind solution. The CME generation mechanism is provided through the Eruptive Event Generator (EEG) which is available at CCMC. In EEG, three options are provided to the user: the Gibson-Low (GL), Titov-Démoulin (TD), and STITCH models. For the real-time CME/SEP run, AWSoM-R will simulate the entire propagation process of CME through the solar corona and within the inner heliosphere (Mars orbit and beyond). About four days of simulations will be performed. A series of synthetic observables, including white light movies, in-situ plasma properties, and 2D cuts of plasma properties with magnetic field lines, will be calculated and provided to the user. SEP events can be modeled with the M-FLAMPA Model.

Example simulation outputs including synthetic observables can be found here.

3) Shock Capturing Tool

The shock capturing tool is a computational method designed to identify and characterize CME-driven shocks within AWSoM(-R) CME simulations. It detects shocks by evaluating plasma and magnetic field discontinuities in the simulation domain, and reconstructs the three-dimensional shock surface geometry with high fidelity. The tool calculates key shock parameters such as the shock normal direction, compression ratio, and shock speed at each surface element. This detailed reconstruction enables accurate coupling of the shock structure with the kinetic particle transport and acceleration module, ensuring that the simulated energetic particle injection and acceleration processes are physically consistent with the evolving shock properties throughout the heliosphere. The 2D and 3D visualizations of the shock are included as standard model outputs. The figure below shows the 3D visualization of the shock with the compression ratio and upstream shock angle overplotted on the shock surface. Two views are provided, one from the front side of the active region and one from the backside of the active region.

GONG synoptic maps (magnetograms) and CME parameters derived from EEGGL tool. The user can also specify the cadence of the synthetic observables.

Outputs include the MHD plasma parameters (atomic mass unit density N, pressure P, velocity V_x, V_y, V_z, magnetic field B_x, B_y, B_z, electric currents, J_x, J_y, J_z, Alfven wave energy densities I01, I02 and dissipation rates GammaLperp, GammaCref. The above quantities are available in full 3D, in 2D slices, as well as along satellite orbits. Visualization of plasma speed and magnetic field lines in 2D slices is produced.

The model also produces synthetic observables, including the white light movies of the CME in the solar corona, and the in-situ plasma properties.

The model will also produce the geometry of the CME-driven shock and its properties. The visualization of both 2D and 3D shock surfaces is produced.

Co-author Statement: In addition, papers using the EEG tool should include three EEG developers as co-authors: M. Jin, W. Manchester, and I. Sokolov.