Numerical models of the solar wind and coronal mass ejections (CMEs) utilize photospheric magnetic field observations to prescribe the inner boundary conditions for the plasma solutions. These magnetic field data are available to the community through various observational instruments, prepared via different methodologies and/or flux transport models. The solar wind solution driven by these maps provides the ambient plasma environment into which CMEs travel. Interaction with the surrounding solar wind impacts CME evolution and propagation in the solar corona and inner heliosphere. In this work, we use different input magnetic field maps for the same time period to drive the global Alfv & eacute;n Wave Solar atmosphere Model. We obtain the ambient solar wind conditions and compare the plasma properties and magnetic morphology in the corona domain to study the influence of the input maps. To understand how the resulting coronal solutions impact CMEs, we launch eruptions described by analytical flux ropes into these data-driven solutions and compare their evolution in the coronal domain (up to 24 R circle dot radially). The CMEs achieve varying speeds, deceleration rates, propagation directions, mass, and energies while coupling with the background solar wind. We quantify these differences to show that the different input driving maps can significantly impact the simulated CME propagation in the solar wind plasma. This also highlights the importance of understanding the uncertainties associated with data-driven modeling that become increasingly important in operational models and space weather prediction.

Strict energy conservation is, perhaps, the most basic principle in all physics, but has proven to be difficult to satisfy in numerical simulations of solar eruptions. The Alfvén Wave Solar atmosphere Model (AWSoM) is used to perform a rigorous comparison of CME simulations whose only difference is the use of a conservative vs. non-conservative scheme for the energy equation. A simple, symmetric active region is assumed for the initial magnetic field. As expected, the different numerical schemes result in very different plasma thermal energy, but surprisingly, we also find a factor >2 difference in the final kinetic energy, with the energy substantially larger in the energy-conservative scheme. The increase in thermal energy is comparable to the increase in kinetic energy in the conservative simulation. Our analysis reveals that the flare reconnection and increase of kinetic energy terminate earlier with the non-conservative scheme. We conclude that the plasma thermodynamics plays a critical role in the flare reconnection, with the thermal pressure gradient in the current sheet slowing down the reconnection. Our results imply that using strict energy-conservative numerics is critical for space weather modeling of CMEs and for understanding the CME energy budget partitioning.

Abstract We investigate the onset of magnetic reconnection in Earth's magnetotail using a coupled magnetohydrodynamic with embedded particle‐in‐cell (MHD‐EPIC) model, which self‐consistently links global magnetospheric dynamics to electron‐scale kinetic processes. Unlike standalone kinetic simulations with idealized boundary conditions, reconnection in this framework is driven by large‐scale magnetotail stretching and plasma convection. Simulations are performed in two spatial dimensions to enable a comprehensive study of grid convergence and the effects of various kinetic scaling factors and mass ratios. A systematic convergence study demonstrates that resolving the electron inertial length with approximately 1.5 grid cells is required to achieve grid‐independent reconnection onset timing and current sheet structure. The current sheet thins to a universal thickness of at onset, indicating that reconnection initiation is controlled by electron‐scale physics. In contrast, both ideal and Hall MHD simulations fail to converge, with reconnection properties remaining dependent on grid resolution. The results in this study establish quantitative criteria for resolving kinetic scales in global simulations and demonstrate that reconnection onset in the magnetotail is fundamentally governed by electron‐scale processes embedded within global dynamics.

Abstract Accurate forecasting of geomagnetic perturbations is essential for assessing space weather risk of geomagnetic activities. Thanks to decades of magnetometer observations from world‐wide distributed networks and upstream solar wind observations at L1, data‐driven modeling of global geomagnetic perturbations has become increasingly feasible. Such models are promising for operational use because they typically run at lower computational cost and can offer higher predictive skill than first‐principles, physics‐based simulations when observations available for training are abundant. However, this potential has not been fully realized as sparse observational coverage in the Southern Hemisphere (SH) has long limited model generalizability. In this work, we use a zeroth‐order mirror‐symmetry assumption in ionospheric electrodynamics under the joint sign reversal of magnetic latitude λ, interplanetary magnetic field By and dipole tilt θ to improve SH predictions using information from the densely observed Northern Hemisphere (NH). Rather than enforcing exact symmetry, we build on our previous GeoDGP model by introducing a symmetry‐aware kernel for a deep Gaussian process that learns the symmetry‐asymmetry balance from data. For comparison, we also experiment with an alternative approach that imposes symmetry as a hard constraint via data augmentation. We evaluate the models on 22 geomagnetic storms and present a case study of the 2024‐05‐10 Gannon extreme storm. The results show that both methods significantly improve the SH predictions, while the kernel‐based approach achieves the best performance and maintains NH accuracy comparable to GeoDGP.

We show new results of a global 3D magnetohydrodynamic (MHD) simulation of the Boston University outer heliosphere model where we used a newly developed approach that distributes the non-adiabatic shock heating among the cold protons, electrons, and pickup-ions (PUIs), while maintaining total energy conservation of all ions (cold protons and PUIs) and electrons. In our previous simulations ( E.S. Bair et al. 2025; B. van der Holst et al. 2026), all non-adiabatic shock heating was channeled to the cold protons, resulting in a too large temperature jump at the termination shock (TS) for the thermal solar wind. Using a new methodology we improved the simulation results with respect to the temperature jump observed at the TS by Voyager 2 (V2) spacecraft. Our simulations approached the observed jump conditions of a factor 10–20 in the cold solar wind temperature V2 measurements, and we obtain improvements in the simulation results relative to the data. Because we directly estimate in this way the distribution of non-adiabatic heating at the TS, we have the opportunity to study the physical process of heating cold plasma and PUIs in the TS. The results show that having almost 100% non-adiabatic shock heating going towards PUIs at the TS reproduces the jump conditions observed along the V2 trajectory. This information is key to understanding the physical processes that shape the heliosphere. As shown by M. Opher et al. (2020), PUIs significantly change the shape of the heliosphere, for example, the presence of hot PUIs results in a deflated inner heliosheath. Our work provides the architecture of how energy partitioning at shocks in kinetic simulations (J. Giacalone et al. 2021) can be utilized in global MHD models.

This study investigates the applicability of the semi-implicit particle-in-cell code FLexible Exascale Kinetic Simulator ( FLEKS ) to heliospheric shock simulations. We examine one- and two-dimensional local planar shock simulations, initialized using MHD states with upstream conditions representative of plasmas in the hypersonic, β ∼ 1 regime, for both quasi-perpendicular and quasi-parallel configurations. The refined algorithm in FLEKS proves robust, enabling accurate shock simulations with a grid resolution on the order of the electron inertial length d _e . Our simulations successfully capture key shock features, including shock structures (foot, ramp, overshoot, and undershoot), upstream and downstream waves (fast magnetosonic, whistler, Alfvén ion-cyclotron, and mirror modes), and non-Maxwellian particle distributions. Crucially, we find that at least two spatial dimensions are critical for accurately reproducing downstream-wave physics in quasi-perpendicular shocks and capturing the complex dynamics of quasi-parallel shocks, including surface rippling, shocklets, short, large-amplitude magnetic structures, magnetic reconnection, and jets. Furthermore, our parameter studies demonstrate the impact of mass ratio and grid resolution on shock physics. This work provides valuable guidance for selecting appropriate physical and numerical parameters for shock simulations using a semi-implicit PIC method, paving the way for incorporating kinetic shock processes into large-scale collisionless plasma simulations with the MHD-AEPIC model.

Traditional shock-capturing schemes inherently diffuse discontinuities over multiple grid cells. This ”numerical smearing” is particularly problematic for converging shocks approaching an axis, where the resulting uncertainty in shock position and strength can obscure critical physical insights near the singularity. In this study, we present a generalized shock-fitting methodology that eliminates numerical diffusion at the shock front by treating the shock as a sharp discontinuity. By transforming the hyperbolic conservation laws into a non-inertial frame of reference, the accelerating shock is rendered stationary at a fixed coordinate. A shock acceleration relation is derived by projecting the Rankine-Hugoniot jump conditions into the characteristic space that governs shock acceleration. Our results demonstrate that this method accurately tracks the shock trajectory, produces flow profiles consistent with shock-capturing methods using significantly fewer grid cells, and recovers the Guderley self-similarity exponent for converging cylindrical hydrodynamic shocks without the dissipative errors typical of shock-capturing approaches.

Spacecraft measurements of Mercury indicate that it has a core dynamo with a surface field of 200-800 nT. These data also indicate that the northern hemisphere crust contains remanent magnetization likely produced by an ancient magnetic field. The inferred magnetization intensity is consistent with a wide range of paleofield strengths (0.2-50 mu T), possibly indicating that Mercury once had a dynamo field much stronger than today. Recent modeling of ancient impacts on the Moon has demonstrated that plasma generated during basin-formation can transiently amplify a planetary dynamo field near the surface. Simultaneously, impact-induced pressure waves can then record these fields as a crustal shock remanent magnetization (SRM). Here, we present impact hydrocode and magnetohydrodynamic simulations of a Caloris-sized basin (similar to 1,550 km diameter) formation event. Our results demonstrate that the ancient magnetospheric field (similar to 0.5-0.9 mu T) created by the interaction of the ancient interplanetary magnetic field and Mercury's dynamo field can be amplified by the plasma up to similar to 13 mu T and, via impact pressure waves, recorded as SRM at the basin antipode. Such magnetization could produce similar to 5 nT crustal fields at 20-km altitude antipodal to Caloris detectable by future spacecraft like BepiColombo. Furthermore, impacts in the southern hemisphere that formed similar to 1,000 km diameter basins (e.g., Andal-Coleridge, Matisse-Repin, Eitkou-Milton, and Sadi-Scopus) could impart crustal magnetization in the northern hemisphere, contributing to the overall remanent field measured by MESSENGER. Overall, the impact plasma amplification process can contribute to crustal magnetization on airless bodies and should be considered when reconstructing dynamo history from crustal anomaly measurements.

Abstract We study the propagation and impacts of a strong density pulse through the magnetosphere during 11:40–12:00 UT near the peak of a geomagnetic storm on 3 February 2022. Using the Space Weather Modeling Framework Geospace global simulation, we show how the energy transport in the magnetosphere is enhanced promptly after the pulse arrival at the dayside magnetosphere, causing enhanced flows in the magnetotail prior to the compression reaching the tail region. The flows are shown to transport energy into the inner magnetosphere and couple to the dawn sector ionosphere, where the magnetic field variations caused large geomagnetically induced currents. We discuss the characteristics of pulse‐driven disturbances in the magnetosphere—ionosphere system, and argue that despite some similarities, the activity is distinct from substorms.

Mercury's small magnetosphere and strong solar wind driving result in the rapid formation of planetward and tailward-moving flux ropes (FRs) and dipolarization fronts (DFs) in its magnetotail. These are characterized by large variations in the dipole-aligned magnetic field component over s timescales and are often associated with fast planetward or tailward plasma flows. At present, the dynamic evolution of DFs and FRs and the relationship between them remains poorly understood at Mercury. To contextualize single-point observations of these events by the MESSENGER spacecraft, we present coupled fluid-kinetic simulations of Mercury's magnetosphere using the Magnetohydrodynamics with Adaptively Embedded Particle-in-Cell (MHD-AEPIC) code. Driven by steady southward interplanetary magnetic field and nominal solar wind conditions about aphelion, the simulation produces recurrent DFs in the tail with peak occurrence rates of DFs per minute and jumps in magnetic field ranging from nT, mirroring those observed by MESSENGER. By tracking the 3D, time-resolved propagation of the simulated DFs, we find that of events originate directly through reconnection at the nearest x-line to the planet, while the remainder originate from FRs that undergo secondary reconnection or "re-reconnection" closer to the planet to create DF-like signatures. Our results suggest that this secondary reconnection process leads to localized heating of electrons along the reconnecting flux tube up to temperatures of 4 keV, which may help account for the suprathermal electrons observed in Mercury's low-altitude current sheet.

Interstellar dust has been detected in situ flowing through the heliosphere. Understanding the implications of this dust for the nature of interstellar dust in the very local interstellar medium requires modeling the transport of the grains as they interact with the solar wind magnetic field. The magnetic field in the sector region (SR) that contains the heliospheric current sheet is substantially different from that in the monopolar solar wind. The rapid polarity flips that occur in the SR can present an effectively very low averaged field strength to grains that have gyroradii of tens of au. We present new calculations of dust transport through the heliosphere using a model that includes the SR. We show that the SR can act as a window allowing even relatively small grains to penetrate deep into the heliosphere. The presence of the SR reduces the variation in dust density with the solar cycle (as compared to models without it), with very little concentration or dilution of the dust for grains larger than ∼ 0.1 μm for most of the solar cycle (except for a focusing overall polarity of the field at solar minimum.) While the lack of time dependence of the magnetic field during transport of grains through the heliosphere is a limitation of the model, the relative lack of variation as a function of the point in the solar cycle of the grain density in the inner heliosphere suggests that our results will not deviate dramatically from a model that fully incorporates time dependence.

We present a community effort to assess how open science can advance heliophysics and space weather modeling. Open science has the potential to enhance the quality and pace of scientific discovery, but its application to scientific modeling requires more careful consideration with respect to open data and open software guidelines, as complex scientific models are not ordinary software. We gathered feedback from modeling teams worldwide through a living survey and discussion sessions at the Open Science Workshop in College Park, USA, in 2024, and the COSPAR ISWAT Initiative Working Meeting in Cape Canaveral, USA, in 2025. We complement these findings with lessons learned from almost 25 years of experience at the Community Coordinated Modeling Center in enabling open use of models. We identify key roadblocks in current open science practices and guidelines and offer recommendations for future progress. Our findings are organized into four overlapping themes: open use of models and simulation results, open validation, open development, and open collaboration. An essential outcome of the discussion is the need for model developers and users to speak with a united voice and promote the role of models in future open science efforts. We introduce a new cross-domain community initiative called Heliophysics Open Modeling Environment (HOME), which will be integrated as an overarching activity within the COSPAR ISWAT Initiative. HOME will serve as a platform for modelers and model users to work together, facilitate community modeling, improve the scientific return on modeling investment, and advance innovation in heliophysics and space weather.

Solar eruptive events are generally believed to involve magnetic flux ropes (MFR), formed either in the pre-eruptive phase of the event or during the eruption itself. These MFR eruptions exhibit significant complexity and variations due to the interplay of the physical mechanisms involved, in particular magnetic reconnection and ideal instabilities. This work considers the effect of the background magnetic field on the nature of eruptions with pre-existing MFRs. We used a new MHD model to simulate the whole MFR eruption process, including the pre-eruptive stage and the initiation. Three simulations were performed, all of which used an identical bipolar active region, but with different background magnetic fields in the three cases. The simulations resulted in two successful eruptions (CMEs) and one failed eruption (a confined flare). We analyzed the energetics and the acceleration of the MFR in detail, and found a transition to a rapid exponential rise phase in two of the simulations. We also calculated the criterion for the torus instability and the timing of the breakout and flare reconnections. Our results show that the rapid exponential rise phase is likely due to breakout reconnection. We conclude that a background field antiparallel to the active-region field lowers the magnetic free-energy threshold for eruption; but, does not guarantee a successful eruption. We also found that an antiparallel background field leads to faster flare reconnection, but of shorter duration. Our findings underscore the importance of the background magnetic field in understanding CMEs.

Mercury’s small magnetosphere and strong solar wind driving results in short-lived, highly dynamic substorms where large amounts of magnetic flux is processed in the magnetotail through nightside reconnection. This processing is realized through the rapid formation of planetward and tailward-moving flux ropes and dipolarization fronts, which lead to plasma heating and flux transport in the low-altitude plasma sheet. The MESSENGER spacecraft observed dipolarization fronts and flux ropes during its orbital campaign from 2011-2015, but open questions about their dynamics, relationship to each other, and role in broader magnetospheric processes persist. To contextualize these observations, we present coupled fluid-kinetic simulations of Mercury's magnetosphere using the MHD-AEPIC code, implemented through the Space Weather Modeling Framework. This model utilizes a Hall-MHD solver for the global magnetosphere coupled to an embedded particle-in-cell code, which simulates the magnetotail dynamics. By tracking the 3D, time-resolved propagation of dipolarization fronts, we find that only ~60 of events originate directly through single x-line reconnection, while the remainder originate from flux ropes that undergo secondary reconnection closer to the planet to create DF-like signatures. We present case studies of both event types, finding that the secondary reconnection process leads to localized heating of the electron fluid along the reconnecting flux tube to temperatures of >4 keV. We compare these characteristics to dipolarization fronts detected by MESSENGER, finding that this model may help account for some of the observed magnetic signatures, associated electron injections, and dawn-dusk distribution asymmetries. Future observations by BepiColombo will be important for further characterizing the frequency and impact of this process within Mercury's magnetotail.

We implement a novel approach to performing Sun-to-Earth coronal mass ejection (CME) simulations and test it on three geo-effective space weather events. Using a vector magnetogram observed prior to the CME as the boundary condition, we reconstruct non-linear force free field (NLFFF) solutions in solar active regions with an established magneto-frictional method. We find a pre-eruption solar corona containing the NLFFF in the AWSoM model, which then spontaneously erupts. We apply STITCH, a photospheric driving method, when needed, to increase the strength of the CME shock. The eruptions successfully produce magnetic flux ropes (MFRs) that propagate to 1 au in the full MHD simulation. The synthetic white light images of the simulated CMEs share a striking resemblance in shape to observations. The interplanetary MFRs (IMFRs) arrive at 1 au with a 1.5- to 9-hour error. A comparison of simulated solar wind plasma with in-situ measurements shows that IMFR crossing can reproduce a southward B_z and often its magnitudes, which determine the geo-effectiveness of the event.

Magnetic reconnection in coronal current sheet(s) is widely believed to be the main energy release process powering solar eruptive events, such as flares, coronal mass ejections (CME), and coronal jets. Modeling this process and determining the channels for the energy release, mass motions and heating, has long been a major goal in space science. We present results from a two-fluid MHD simulation of an eruptive flare/CME using a newly developed Strategic Capability, SCEPTER, which is based on the well-validated and widely used Space Weather Modeling Framework. SCEPTER incorporates two major advances in numerical capability. First, we use the STITCH formalism for the energy buildup, so that we start with a potential-field minimum-energy state and slowly form a sheared filament channel over a polarity inversion line as is observed on the Sun. Second, we use a new formulation of the plasma energetics that is explicitly energy conserving while calculating separate electron and ion temperatures and separate parallel and perpendicular pressures, as desired. For this first simulation with our new model, we opted for the non-adiabatic heating to go solely into the protons and for an isotropic pressure. We discuss the resulting energetics of the reconnection and, in particular, the plasma heating in the reconnecting current sheets, mass acceleration, and shock formation. We also discuss the implications of our results for understanding solar eruptions, in general. This work was supported by the NASA Living With a Star Program.

The 2024 April 8 total solar eclipse provides a unique opportunity to study the solar corona. This work presents our simulations of the solar corona at the time of the eclipse based on magnetohydrodynamic modeling performed with the Alfvén Wave Solar atmosphere Model in the Space Weather Modeling Framework, developed at the University of Michigan. We performed multiple simulations based on photospheric magnetic maps from four sources, i.e., ADAPT-GONG, Lockheed Martin ESFAM-HMI, HipFT-HMI, and NSO-NRT-HMI maps. Our study focuses on how differences in the magnetic field maps affect the coronal magnetic field structure and coronal heating properties in the simulation. The synthesized observables show remarkable differences due to the distinct magnetic coronal topologies, which stem from the different local magnetic flux distributions. We analyze the properties of the open magnetic flux regions of the models. We also study the coronal heating rate in the models. The total volume integrated heating rate yields a difference of 20% across the models. The results also show that the differential emission measure in the high-temperature regions is sensitive to the magnetic field maps. Our findings underscore the importance of comprehensive photospheric magnetic field data in improving future solar coronal models.

Simulations have played a critical role in the advancement of our knowledge of magnetic reconnection. However, due to the inherently multiscale nature of reconnection, it is impossible to simulate all physics at all scales. For this reason, a wide range of simulation methods have been crafted to study particular aspects and consequences of magnetic reconnection. This article reviews many of these methods, laying out critical assumptions, numerical techniques, and giving examples of scientific results. Plasma models described include magnetohydrodynamics (MHD), Hall MHD, Hybrid, kinetic particle-in-cell (PIC), kinetic Vlasov, Fluid models with embedded PIC, Fluid models with direct feedback from energetic populations, and the Rice Convection Model (RCM).

We have developed a new Adaptive Mesh Refinement (AMR) version of the Gauss-Law satisfying Energy Conserving Semi-Implicit Method (GL-ECSIM) and implemented it into the Flexible Exascale Kinetic Simulator (FLEKS). The semi-implicit Particle-In-Cell (PIC) method is particularly well suited for AMR, because, unlike in explicit PIC, the cell size does not have to resolve the Debye length for stability. In contrast with the earlier Multi-Level-Multi-Domain semi-implicit PIC algorithm developed by Innocenti+, the new algorithm uses a single set of particles over the whole domain. Particles are split and merged as needed by efficient and accurate methods. The coarser level receives both the field information and the phase space distribution (through the particles) from the fine level. The fine level uses the coarse level as boundary condition. The new algorithm satisfies Gauss Law on the entire domain, including grid resolution changes. We show various tests confirming the accuracy and robustness of the new algorithm. In particular, we simulate magnetic reconnection with an ion-electron mass ratio of 64. The AMR resolves the electron scales near the reconnection site, while the grid is eight times coarser elsewhere matching the ion scales. The overall speed up is at least tenfold compared to a uniformly fine grid simulation.

Block-Adaptive-Tree Solar-wind Roe-type Upwind Scheme (BATSRUS), our state-of-the-art extended magnetohydrodynamic code, is the most used and one of the most resource-consuming models in the Space Weather Modeling Framework. It has always been our objective to improve its efficiency and speed with emerging techniques, such as GPU acceleration. To utilize the GPU nodes on modern supercomputers, we port BATSRUS to GPUs with the OpenACC API. Porting the code to a single GPU requires rewriting and optimizing the most used functionalities of the original code into a new solver, which accounts for around 1% of the entire program in length. To port it to multiple GPUs, we implement a new message-passing algorithm to support its unique block-adaptive grid feature. We conduct weak scaling tests on as many as 256 GPUs and find good performance. The program has 50%–60% parallel efficiency on up to 256 GPUs and up to 95% efficiency within a single node (four GPUs). Running large problems on more than one node has reduced efficiency due to hardware bottlenecks. We also demonstrate our ability to run representative magnetospheric simulations on GPUs. The performance for a single A100 GPU is about the same as 270 AMD “Rome” CPU cores (2.1 128-core nodes), and it runs 3.6 times faster than real time. The simulation can run 6.9 times faster than real time on four A100 GPUs.