Ryan James Campbell

We present high-resolution multiline spectropolarimetric observations of a quiet-Sun network element obtained with the S unrise iii Chromospheric Infrared SpectroPolarimeter. The observations combine photospheric, upper-photospheric, and chromospheric diagnostics at a spatial resolution and polarimetric sensitivity that allow the transverse magnetic structure of the network boundary to be examined directly. We find that the strongest linear polarization is concentrated in a narrow ridge around the edge of the magnetic element, cospatial with enhanced transverse magnetic field inferred from multiline inversions. The magnetic azimuth exhibits a coherent, predominantly radial organization around a more vertical core, consistent with an expanding magnetic canopy. An azimuth proxy derived directly from the observed Fe i and K i linear polarization reproduces the same large-scale organization, showing that this structure is encoded in the Stokes profiles rather than imposed by the inversion. Response functions and a MURaM-based forward-synthesis test indicate that the Fe i 8468 Å linear polarization is sensitive to magnetic azimuth in the upper photosphere, with the closest proxy agreement occurring near log τ ≈ − 3 . We find no evidence for strong azimuthal shear between the Fe- and K-sensitive diagnostics. At the network boundary, we also identify localized parasitic-polarity patches associated with complex, multilobed Stokes V profiles and one case in which the Stokes V polarity reverses between photospheric Fe i and chromospheric Ca ii lines. These results demonstrate that quiet-Sun network boundaries contain organized upper-photospheric canopy fields together with a small-scale mixed-polarity structure, providing new constraints on the three-dimensional magnetic structure of network elements.

Deciphering the physical properties of flares is crucial for understanding the nature of one of the most energetic events on the Sun. We analyze high spatial and spectral resolution observations of a C-class flare, focusing on a complex light bridge in the active region. The observations show Stokes profiles with opposite Doppler shifts and asymmetries between photospheric and chromospheric spectral lines, and Ca ii infrared spectral lines in emission, tracing areas of extreme atmospheric heating. Inversion results are consistent with these spectral features, showing regions of opposite-sign plasma flows between the photosphere and chromosphere, adjacent to areas with extreme chromospheric temperatures. The latter areas show relatively weak line-of-sight velocities despite the intense heating. The magnetic field configuration shows abrupt changes with height in the regions with opposite-sign flows, while remaining nearly constant with height in the areas of extreme heating. These findings point to a scenario of small-scale magnetic reconnection in the photosphere, suggesting a flare triggering mechanism based on an avalanche of many small reconnection events.

We present DKIST spectropolarimetric observations of a sunspot penumbra revealing spatially coherent four-lobed Stokes U profiles that persist after polarimetric cross-talk correction. The observed Stokes vectors are well reproduced by stratified inversions containing gradients in velocity and magnetic parameters, including line-of-sight (LOS) variation of the magnetic azimuth. Forward synthesis experiments in which the magnetic azimuth is artificially held constant with optical depth substantially modify the Stokes U morphology while leaving Stokes I, Q, and V comparatively unchanged, indicating that azimuthal variation is an important contributor to the observed linear polarisation signatures within the tested one-component atmospheres. We explored degeneracies using both two-component atmospheres and one-component constant-azimuth models: the former generally converged toward pathological solutions, while the latter provided a plausible degenerate explanation for much of the observed four-lobed Stokes U morphology. A supervised machine learning classifier applied to Stokes V profiles shows that the associated magnetic complexity is spatially confined to the penumbra-light-bridge boundary where the four-lobed Stokes U profiles and reverse-polarity fields are observed. These results identify four-lobed Stokes U profiles as a potential spectropolarimetric signature of complex LOS magnetic structuring in sunspot penumbrae, providing new observational constraints on fine-scale penumbral magnetic topology.

Bright points (BPs) are small-scale, dynamic features that are ubiquitous across the solar disc and are often associated with the underlying magnetic field. Using broadband photospheric images obtained with the Visible Broadband Imager at the National Science Foundation’s Daniel K. Inouye Solar Telescope (DKIST), the properties of BPs have been analyzed with DKIST for the first time at the highest spatial resolutions achievable. BPs were observed to have an average lifetime of 95 ± 29 s and a mean transverse velocity of 1.60 ± 0.41 km s ⁻¹ . The BPs had a log-normal area distribution with a peak at 2300 km ² . Transverse velocities and lifetimes across the DKIST images were comparable and consistent with previous studies. The area distribution of the DKIST data peaked in areas significantly lower than those from the literature. This was explored further and was observed to be due to an overestimation of BP areas due to the merging of close features when the spatial resolution is reduced, in tandem with possible oversplitting of features in the DKIST images. Furthermore, the effect of variable seeing in the data was determined. This showed that the average spatial resolution of the data was around 0 . ″ 034 ± 0 . ″ 007 , in comparison to the theoretical diffraction limit of 0 . ″ 022 . Accounting for the influence of seeing, the peak of the area distribution of BPs in the DKIST data was estimated as 4800 km ² , which is still significantly lower than previously observed.

Solar spectropolarimetric inversion—inferring atmospheric conditions from the Stokes vector—is a key diagnostic tool for understanding solar magnetism, but traditional inversion methods are computationally expensive and sensitive to local minima. Advances in artificial intelligence offer faster solutions, but are often restricted to shallow models or a few spectral lines. We present a proof-of-concept study using a transformer machine learning model for multiline, full-Stokes inversion, to infer stratified parameters from synthetic spectra produced from 3D magnetohydrodynamic simulations. We synthesize a large set of Stokes vectors using forward modeling across 15 spectral lines spanning the deep photosphere toward the chromosphere. The model maps full-Stokes input to temperature, magnetic field strength, inclination, azimuth (encoded as sin 2 ϕ , cos 2 ϕ ), and line-of-sight velocity as a function of optical depth. The transformer incorporates an attention mechanism that allows the model to focus on the most informative regions of the spectrum for each inferred parameter, and uses positional embedding to encode wavelength and depth order. We benchmark it against a multilayer perceptron (MLP), test robustness to noise, and assess generalization. The transformer outperforms the MLP, especially in the higher layers and for magnetic parameters, yielding higher correlations and more regularized stratifications. The model retains strong performance across a range of noise levels typical for real observations, with magnetic parameter inference degrading predictably while temperature and velocity remain stable. We explore attention maps, linking the transformer’s learned behaviour to line-formation physics.

Since its launch in 2010, the Solar Dynamics Observatory (SDO) has provided continuous high-cadence, multi-wavelength observations of the Sun, capturing thousands of solar flares and offering new insights into coronal dynamics. Among the discoveries enabled by SDO is the EUV late-phase (ELP), characterised by a secondary enhancement in warm coronal emission occurring tens of minutes after the main flare. While recent work has demonstrated the relevance of the ELP for space weather, its statistical behaviour and physical origin remain poorly constrained. Here, we present the most comprehensive review of the ELP to date, based on 15 years of Fe XVI observations from the Atmospheric Imaging Assembly onboard SDO (SDO/AIA). From a sample of 5335 isolated flares between 2010 and 2025, we identify and validate 467 ELP events. The overall ELP occurrence rate is 8 percent, with no significant dependence on the solar cycle and only a modest enhancement in the low-to-mid M-class range. The ELP typically exhibits and onset delay of 19 minutes, a peak-to-peak delay of 88 minutes, and a duration of 93 minutes. Strong correlations are found between ELP rise and decay rates (p=0.76), and between flare and ELP impulsivity (p=0.61), while no significant correlation is observed between flare and ELP phases. A Principal Component Analysis revealed three dominant axes of variation, corresponding to a timescale component, an energy-release intensity axis, and a partitioning of energy between the flare and ELP. These results suggest that ELP evolution is governed by both flare loop properties and reconnection-driven energetics, likely modulated by a finite magnetic energy budget, and highlight the importance of SDO's long-term observations for understanding flare evolution and the Sun-Earth connection.

The morphology of circular polarization profiles from solar spectropolarimetric observations encodes information about the magnetic field strength, inclination, and line-of-sight velocity gradients. Previous studies used manual methods or unsupervised machine learning (ML) to classify the shapes of circular polarization profiles. We trained a multilayer perceptron comparing classifications with unsupervised ML. The method was tested on quiet Sun data sets from Daniel K. Inouye Solar Telescope (DKIST), Hinode, and GREGOR, as well as simulations of granulation and a sunspot. We achieve validation metrics typically close to or above 90%. We also present the first statistical analysis of quiet Sun DKIST/ViSP data using inversions and our supervised classifier. We demonstrate that classifications with unsupervised ML alone can introduce systemic errors that could compromise statistical comparisons. DKIST and Hinode classifications in the quiet Sun are similar, despite our modeling indicating spatial resolution differences should alter the shapes of circular polarization signals. Asymmetrical (symmetrical) profiles are less (more) common in GREGOR than DKIST or Hinode data, consistent with narrower response functions in the 1564.85 nm line. Single-lobed profiles are extremely rare in GREGOR data. In the sunspot simulation, the 630.25 nm line produces “double” profiles in the penumbra, likely a manifestation of magneto-optical effects in horizontal fields; these are rarer in the 1564.85 nm line. We find the 1564.85 nm line detects more reverse polarity magnetic fields in the penumbra, in contradiction to observations. We detect mixed-polarity profiles in nearly one fifth of the penumbra. Supervised ML robustly classifies solar spectropolarimetric data, enabling detailed statistical analyses of magnetic fields.

The strongly coupled hydrodynamic, magnetic, and radiation properties of the plasma in the solar chromosphere make it a region of the Sun's atmosphere that is poorly understood. We use data obtained with the high-resolution Visible Broadband Imager (VBI) equipped with an Hβ filter and the Visible Spectro-Polarimeter (ViSP) on the Daniel K. Inouye Solar Telescope to investigate the fine-scale structure of the plage chromosphere. To aid in the interpretation of the VBI imaging data, we also analyze spectra from the CHROMospheric Imaging Spectrometer on the Swedish Solar Telescope. The analysis of spectral properties, such as enhanced line widths and line depths, explains the high contrast of the fibrils relative to the background atmosphere demonstrating that Hβ is an excellent diagnostic for the enigmatic fine-scale structure of the chromosphere. A correlation between the parameters of the Hβ line indicates that opacity broadening created by overdense fibrils could be the main reason for the spectral line broadening frequently observed in chromospheric fine-scale structures. Spectropolarimetric inversions of the ViSP data in the Ca ii 8542 Å and Fe i 6301/6302 Å lines are used to construct semiempirical models of the plage atmosphere. Inversion outputs indicate the existence of dense fibrils in the Ca ii 8542 Å line. The analyses of the ViSP data show that the morphological characteristics, such as orientation, inclination, and length of fibrils, are defined by the topology of the magnetic field in the photosphere. Chromospheric maps reveal a prominent magnetic canopy in the area where fibrils are directed toward the observer.