The ratio of radially to tangentially polarized Thomson-scattered white light provides a powerful tool for locating the 3D position of compact structures in the solar corona and inner heliosphere, and the Polarimeter to Unify the Corona and Heliosphere (PUNCH) has been designed to take full advantage of this diagnostic capability. Interestingly, this same observable that establishes the position of transient blob-like structures becomes a local measure of the slope of the global falloff of density in the background solar wind. It is thus important to characterize the extent along the line of sight of structures being studied, in order to determine whether they are sufficiently compact for 3D positioning. In this paper, we build from analyses of individual lines of sight to three-dimensional models of coronal mass ejections (CMEs), allowing us to consider how accurately polarization properties of the transient and quiescent solar wind are diagnosed. In this way, we demonstrate the challenges and opportunities presented by PUNCH polarization data for various quantitative diagnostics.

The Polarimeter to Unify the Corona and Heliosphere (PUNCH) mission is a NASA Small Explorer to determine the cross-scale processes that unify the solar corona and heliosphere. PUNCH has two science objectives: (1) understand how coronal structures become the ambient solar wind, and (2) understand the dynamic evolution of transient structures, such as coronal mass ejections, in the young solar wind. To address these objectives, PUNCH uses a constellation of four small spacecraft in Sun-synchronous low Earth orbit, to collect linearly polarized images of the K corona and young solar wind. The four spacecraft each carry one visible-light imager in a 1 + 3 configuration: a single Narrow Field Imager solar coronagraph captures images of the outer corona at all position angles, and at solar elongations from 1.5° (6 R⊙) to 8° (32 R⊙); and three separate Wide Field Imager heliospheric imagers together capture views of the entire inner solar system, at solar elongations from 3° (12 R⊙) to 45° (180 R⊙) from the Sun. PUNCH images include linear-polarization data, to enable inferring the three-dimensional structure of visible features without stereoscopy. The instruments are matched in wavelength passband, support overlapping instantaneous fields of view, and are operated synchronously, to act as a single “virtual instrument” with a 90∘ wide field of view, centered on the Sun. PUNCH launched in March of 2025 and began science operations in June of 2025. PUNCH has an open data policy with no proprietary period, and PUNCH Science Team Meetings are open to all.

Deep generative models have shown immense potential in generating unseen data that has properties of real data. These models learn complex data-generating distributions starting from a smaller set of latent dimensions. However, generative models have encountered great skepticism in scientific domains due to the disconnection between generative latent vectors and scientifically relevant quantities. In this study, we integrate three types of machine learning models to generate solar magnetic patches in a physically interpretable manner and use those as a query to find matching patches in real observations. We use the magnetic field measurements from Space-weather HMI Active Region Patches (SHARPs) to train a Generative Adversarial Network (GAN). We connect the physical properties of GAN-generated images with their latent vectors to train Support Vector Machines (SVMs) that do mapping between physical and latent spaces. These produce directions in the GAN latent space along which known physical parameters of the SHARPs change. We train a self-supervised learner (SSL) to make queries with generated images and find matches from real data. We find that the GAN-SVM combination enables users to produce high-quality patches that change smoothly only with a prescribed physical quantity, making generative models physically interpretable. We also show that GAN outputs can be used to retrieve real data that shares the same physical properties as the generated query. This elevates Generative Artificial Intelligence (AI) from a means-to-produce artificial data to a novel tool for scientific data interrogation, supporting its applicability beyond the domain of heliophysics.

Coronagraphic observations enable direct monitoring of coronal mass ejections (CMEs) through scattered light from free electrons, but determining the 3D plasma distribution from 2D imaging data is challenging due to the optically thin plasma and the complex image formation. We introduce Sun Neural Radiance Field for CMEs (SuNeRF-CME), a framework for 3D tomographic reconstructions of the heliosphere using multiviewpoint coronagraphic observations. The method uses a neural radiance-field to estimate the electron density in the heliosphere through ray tracing, while accounting for the underlying Thomson scattering. The model is optimized by iteratively fitting the time-dependent observational data. In addition, we apply physical constraints in terms of continuity, propagation direction, and speed of the heliospheric plasma to overcome limitations imposed by the sparse number of viewpoints. We utilize synthetic observations of a CME simulation to quantify the model's performance for different viewpoint configurations. Within this controlled synthetic setting, the results demonstrate that our method can reliably estimate the CME parameters from only two viewpoints, with a mean velocity error of 3.01% +/- 1.94% and propagation direction errors of 3.degrees 39 +/- 1.degrees 94 in latitude and 1.degrees 76 +/- 0.degrees 79 in longitude. We further show that our approach can achieve a 3D reconstruction of the simulated CME from two viewpoints, where we correctly model the three-part structure, deformed CME front, and internal plasma variations. Additional viewpoints can be seamlessly integrated, directly enhancing the reconstruction of the plasma distribution in the heliosphere. These results demonstrate the potential of physics-informed radiance-field methods for CME tomography, paving the way for future extensions toward observational data and space weather applications.

Understanding mesoscale structures in the solar wind background and coronal mass ejections (CMEs) is one of the science objectives of the PUNCH mission to be launched in early 2025. We do not fully understand what processes form these structures and where as well as how they evolve from the outer solar corona through the heliosphere. In anticipation of the detailed high-sensitive large field-of-view PUNCH imaging, MHD simulations capable of modeling the global inner heliosphere while simultaneously resolving structures at mesoscales can help predict what structures we can expect to form in certain CME-solar wind interaction scenarios. We use an efficiently parallelized and scalable physics-based MHD model with numerical algorithms featuring high resolving power to perform global inner heliosphere simulations with CMEs with a high resolution. Using the GAMERA-Helio inner heliosphere model coupled with the Gibson-Low CME model, we model the evolution of a wide and fast CME flux rope through a realistic solar wind background. The simulation resolves spatial scales down to ~0.1 solar radii (~10 Earth radii), enabling, to study mesoscale structures that form in the CME-solar wind interaction in a global context. We discuss the development of ripples and irregularities at the CME shock, compressions, and magnetic field fluctuations in the CME-driven sheath, and connect these structures with the interaction between the CME and background solar wind flows. By computing the total and polarized white light brightness from high-resolution GAMERA MHD simulations, we show how mesoscale structures that form at the CME-solar wind interface appear in synthetic images in the FOV of the PUNCH mission.

In preparation for the PUNCH mission which is planned to be launched in 2025, we construct synthetic white light (WL) images of the inner heliosphere based on GAMERA 3D MHD output. GAMERA is a 3D MHD code whose capabilities have been extended to perform time-dependent solar wind simulations by frequently updating the input magnetograms and thus the boundary conditions at the inner boundary of the code, offering a much more realistic reconstruction of the solar wind propagation and evolution. By employing this capability, we explain the phenomena and structures of the solar wind we see in the synthetic WL images from multiple view points (ACE, STEREO, PUNCH) and compare with the traditional steady state MHD approach.

Variations of the magnetic field within a solar coronal mass ejection (CME) in the heliosphere depend on the CME's magnetic structure as it leaves the solar corona and its interplanetary evolution. To account for this evolution, we developed a new numerical model of the inner heliosphere that simulates the propagation of a CME through a realistic solar wind background and allows various CME magnetic topologies. To this end, we incorporate the Gibson-Low CME model within our global MHD model of the inner heliosphere, GAMERA-Helio. We apply the model to study the propagation of the geoeffective CME that erupted on 2010 April 3, aiming to reproduce the temporal variations of the magnetic field vector during the CME's passage by Earth. Parameters of the Gibson-Low CME are informed by STEREO white-light observations near the Sun. The magnetic topology for this CME-the tethered flux rope-is informed by in situ magnetic field observations near Earth. We performed two simulations testing different CME propagation directions. For an in-ecliptic direction, the simulation shows a rotation of all three magnetic field components within the CME, as seen at Earth, similar to that observed. However, the magnitudes of the components, particularly at the back of the CME, are underestimated by the model. With a southward direction, suggested by coronal imaging observations, the B x component lacks the observed change from negative to positive. In both cases, the model favors the east-west orientation of the flux rope, consistent with the orientation previously inferred from the images from STEREO/Heliospheric Imager.

Cutting edge research of the region between the solar photosphere and Alfvén surface, where the solar wind disconnects from the solar surface, increasingly relies on largescale computational modeling.This is true not just for numerical experiments that only involve simulations, but also for interpreting the ever-more-intricate observations coming out of new facilities.A paradigm shift is currently underway in the field, away from ad hoc and ab initio models of the atmosphere and towards simulations that are directly driven by observations.While this topic is currently at the level of basic research into the techniques, the intent is for it to serve both basic research and operational space-weather needs.We must advance data driven simulations and their supporting infrastructure to the point where community members can use time dependent dynamic models on a regular basis as a tool to interpret observations and test physical theories.Because of the complexity of the task, the immense computational resources required, and the required longevity of such a project, we argue that this effort should be strategically funded.Such an effort is required to make the most out of current and upcoming multi-mission, multi-instrument, heterogeneous observational data.

Active region EUV loops are believed to trace a subset of magnetic field lines through the corona. Malanushenko et al. proposed a method, using loop images and line-of-sight photospheric magnetograms, to infer the 3D shape and field strength along each loop. McCarthy et al. used this novel method to compute the total magnetic flux interconnecting a pair of active regions observed by SDO/AIA. They adopted the common assumption that each loop had a circular cross section. The accuracy of inferred shape and circularity of cross sections can both be tested using observations of the same loops from additional vantage points as provided by STEREO/EUVI. Here we use multiple viewing angles to confirm the 3D structure of loops. Of 151 viable cases, 105 (69.5%) matched some form of visible coronal structure when viewed approximately in quadrature. A loop with a circular cross section should appear of a similar width in different perspectives. In contradiction to this, we find a puzzling lack of correlation between loop diameters seen from different perspectives, even an anticorrelation in some cases. Features identified as monolithic loops in AIA may, in fact, be more complex density enhancements. The 30.5% of reconstructions from AIA that did not match any feature in EUVI might be such enhancements. Others may be genuine loop structures, but with elliptical cross sections. We observe an anticorrelation between diameter and brightness, lending support to the latter hypothesis. Of 13 loops suitable for width analysis, 4 are consistent with noncircular cross sections, where we find anticorrelation in both comparisons.

In the solar corona, the free energy, i.e., the excess in magnetic energy over a ground-state potential field, forms the reservoir of energy that can be released during solar flares and coronal mass ejections. Such free energy provides a measure of the magnetic field nonpotentiality. Recent theoretical and observational studies indicate that the presence of nonpotential magnetic fields is imprinted into the structures of infrared, off-limb, coronal polarization. In this paper, we investigate the possibility of exploiting such observations for mapping and studying the accumulation and release of coronal free magnetic energy, with the goal of developing a new tool for identifying “hot spots” of coronal free energy such as those associated with twisted and/or sheared coronal magnetic fields. We applied forward modeling of infrared coronal polarimetry to three-dimensional models of nonpotential and potential magnetic fields. From these we defined a quantitative diagnostic of nonpotentiality that in the future could be calculated from a comparison of infrared, off-limb, coronal polarization observations from, e.g., the Coronal Multi-channel Polarimeter or the Daniel K. Inouye Solar Telescope, and the corresponding polarization signal forward-modeled from a potential field extrapolated from photospheric magnetograms. We considered the relative diagnostic potential of linear and circular polarization, and the sensitivities of these diagnostics to coronal density distributions and assumed boundary conditions of the potential field. Our work confirms the capacity of polarization measurements for diagnosing nonpotentiality and free energy in the solar corona.

Coronal loops observed in soft X-rays and extreme ultraviolet imaging data offer direct evidence that coronal plasma is heated by some mechanism. That mechanism appears to energize a particular bundle of field lines somehow selected from the magnetized coronal volume. Magnetic reconnection localized to a patch within a coronal current sheet is one mechanism that would select a flux bundle at the same time it energized it. Since magnetic reconnection occurs preferentially at topological boundaries, we would expect to find coronal loops concentrated there if it were at work. We explore this hypothesis using a data set, previously compiled by McCarthy et al., consisting of 301 coronal loops interconnecting a pair of active regions over a 48 hr period. That work computed the three-dimensional geometries and magnetic field strengths for most of the loops. This revealed many bright loops lying at the periphery of the interconnecting flux domain, possibly created and energized by the reconnection that created the interconnecting flux. There were, however, many loops well inside the domain which would be difficult to attribute to that mode of reconnection. Here we use detailed magnetic models of the interconnecting domain to show that these internal loops tend to occur along internal boundaries: separatrices. This offers a novel form of evidence that coronal loops are the products of patchy reconnection even under quiescent conditions.

In this paper, we explore the potential of neural networks for making space weather predictions based on near-Sun observations. Our second goal is to determine the extent to which coronal polarimetric observations of erupting structures near the Sun encode sufficient information to predict the impact these structures will have on Earth. In particular, we focus on predicting the maximal southward component of the magnetic field ("-B-z") inside an interplanetary coronal mass ejection (ICME) as it impacts the Earth. We use Gibson & Low (G&L) self-similarly expanding flux rope model (Gibson and Low,1998) as a first test for the project, which allows to consider CMEs with varying location, orientation, size, and morphology. We vary 5 parameters of the model to alter these CME properties, and generate a large database of synthetic CMEs (over 36k synthetic events). For each model CME, we synthesize near-Sun observations, as seen from an observer in quadrature (assuming the CME is directed earthwards), of either three components of the vector magnetic field, (B-x,B-y,B-z) ("Experiment 1"), or of synthetic Stokes images, (L/I, Az, V/I) ("Experiment 2"). We then allow the flux rope to expand and record max{-B-z} as the ICME passes 1AU. We further conduct two separate machine learning experiments and develop two different regression-based deep convolutional neural networks (CNNs) to predict max{-B-z} based on these two kinds of the near-Sun input data. Experiment 1 is a test which we do as a proof of concept, to see if a 3-channel CNN (hereafter CNN1), similar to those used in RGB image recognition, can reproduce the results of the self-similar (i.e., scale-invariant) expansion of the G&L model. Experiment 2 is less trivial, as Stokes vector is not linearly related toB, and the line-of-sight integration in the optically thin corona presents additional difficulties for interpreting the signal. This second CNN (hereafter CNN2), although resembling CNN(1)in Experiment 1, will have a different number of layers and set of hyperparameters due to a much more complicated mapping between the input and output data. We find that, given three components ofB, CNN(1)can predict max{-B-z} with 97% accuracy, and for three components of the Stokes vector as input, CNN(2)can predict max{-B-z} with 95%, both measured in the relative root square error.

Magnetic reconnection occurs when new flux emerges into the corona and becomes incorporated into the existing coronal field. A new active region (AR) emerging in the vicinity of an existing AR provides a convenient laboratory in which reconnection of this kind can be quantified. We use high time-cadence 171 $\AA$ data from SDO/AIA focused on new/old active region pair 11147/11149, to quantify reconnection. We identify new loops as brightenings within a strip of pixels between the regions. This strategy is premised on the assumption that the energy brightening a loop originates in magnetic reconnection. We catalog 301 loops observed in the 48-hour time period beginning with the emergence of AR 11149. The rate at which these loops appear between the two ARs is used to calculate the reconnection rate between them. We then fit these loops with magnetic field, solving for each loop's field strength, geometry, and twist (via its proxy, coronal $\alpha$). We find the rate of newly-brightened flux overestimates the flux which could be undergoing reconnection. This excess can be explained by our finding that the interconnecting region is not at its lowest energy (constant-$\alpha$) state; the extrapolations exhibit loop-to-loop variation in $\alpha$. This flux overestimate may result from the slow emergence of AR 11149, allowing time for Taylor relaxation internal to the domain of the reconnected flux to bring the $\alpha$ distribution towards a single value which provides another mechanism for brightening loops after they are first created.