Traditional adaptive optics (AO) systems enable diffraction-limited imaging in astronomical observations, yet suffer under strong turbulence and faint guide stars. Focal-plane wavefront sensing (FPWFS) offers a promising alternative but is hindered by slow convergence and phase ambiguity in iteration or deep learning implementations, making it impractical for astronomical observations. Building upon our previously proposed nondeterministic estimation (ND-estimation) method that resolves the phase ambiguity in focal-plane wavefront estimation, we develop a new control strategy—the Single-frame Focal-plane Wavefront Correction ( SFWC ) algorithm—integrating ND-estimation into a deep learning–based closed-loop AO architecture. The SFWC algorithm introduces a direction-resolved feedback mechanism that converts the ambiguity in focal-plane estimation into a deterministic control process through bidirectional perturbation. Furthermore, an adaptive filter is incorporated to predict the wavefront direction of the subsequent frame based on previously determined wavefronts. By integrating this predicted direction with the estimated wavefront of the network, correction can be achieved using only a single measurement. Laboratory experiments demonstrate robust static aberration correction across various turbulence strengths, achieving an average Strehl ratio of 0.93 with only ∼8 iterations. In addition, to the best of our knowledge, this work provides the first experimental demonstration of real-time closed-loop correction at 1200 Hz for dynamic aberrations using only focal-plane images. This work transforms ND-estimation from a static estimation into an operational tool for high-speed AO control, and it is easy to implement without modifying the optical path, thereby establishing a direct path toward FPWFS application in astronomical telescopes.
Deep learning has proven to be a vital advancement in many fields that present difficult classification and prediction tasks. This includes astrophysics, particularly space weather prediction, where accurate modeling has become essential for protecting satellite and ground infrastructure from the impacts of solar flares and solar energetic particle events. Despite their strong predictive ability, one issue with deep learning as a method for space weather prediction is the poor interpretability of these “black box” methods. A possible solution to this comes in the form of a structured deep learning framework named Amortized Causal Discovery (ACD), which aims to enforce an easily interpretable layer in the neural network that represents the causal summary graph. We apply this method to the Space Weather Analytics for Solar Flares dataset, a rich source of multivariate time series (MVTS) data of solar active regions, to demonstrate that high-intensity solar flares can be detected ahead of time by comparing their causal summary graphs with those of other flare events. In this work, we use graph metric statistics as well as objective performance tests against state-of-the-art classifiers to demonstrate the depth of insight that ACD can produce into differences between MVTS samples in the context of solar flare prediction.
S. T. Bramwell et al. (1998) showed that the fluctuations of turbulence have the same statistical description as those of the order parameter in equilibrium-critical phenomena like the magnetization of ferromagnets. Later, it was shown that various other critical systems exhibit similar fluctuation statistics of the variance κ _1 of natural time establishing a universal way of defining an order parameter. These systems include avalanches in 3D rice piles, the magnetic flux penetration in type II superconductors, the occurrence of earthquakes in the Earth’s solid crust, acoustic emission before fracture, as well as avalanches generated by the Burridge–Knopoff or Olami–Feder–Christensen models. Here, we show by the most recent measurements of the Interstellar Mapping and Acceleration Probe (IMAP) that the interplanetary magnetic field components under steady solar wind conditions at 1 au when analyzed in natural time share the same statistical description of fluctuations as in critical systems. This property can identify early-warning signatures of episodic events interwoven with the magnetic field.
We present two complementary machine-readable catalogs for quasar (QSO) candidate selection in the Kilo-Degree Survey (KiDS) DR5. The KiDS-only catalog covers the full survey source sample and provides optical/near-infrared-based classifications together with QSO photometric redshift estimates derived without Wide-field Infrared Survey Explorer (WISE) information. The KiDS–AllWISE catalog is restricted to sources with AllWISE counterparts and provides WISE-assisted classifications and photometric redshifts. Separate stacking ensembles are trained to distinguish galaxies, stars, and QSOs. On a stratified held-out test set, the WISE-assisted classifier achieves an overall accuracy of 98.34%, a macroaveraged area under the receiver operating characteristic curve (ROC AUC) of 99.78%, and a QSO F1-score of 97.09%. Spatially disjoint R.A.-block validation yields QSO F1-scores of 95.87%–96.31%, indicating only modest spatial degradation. Feature-ablation tests show that W1/W2-based mid-infrared information provides the largest gain in QSO recovery, whereas Bayesian photometric redshift (BPZ)-derived quantities have only a minor effect. For QSO photometric redshift estimation, we train a LightGBM–XGBoost ensemble on Sloan Digital Sky Survey (SDSS) DR16 Quasar Catalog (DR16Q) quasars with valid optical and WISE photometry. On 94,057 held-out quasars, the WISE-assisted model achieves σ _NMAD = 0.0633 and a catastrophic-outlier fraction of η = 12.0%; excluding WISE features degrades these values to σ _NMAD = 0.0948 and η = 26.8%. Both catalogs provide probability-ranked QSO candidates and thresholded p _Q samples for different follow-up priorities. As the candidates are not spectroscopically confirmed, statistical applications require independent calibration of the corresponding selection functions.
The diagnostic of the green coronal emission line at Fe XIV 5303 Å is very suited for studying the physical conditions and dynamics of the solar corona. The Visible Emission Line Coronagraph (VELC) on board Aditya-L1 has the facility to make spectroscopic observations of this emission line both in raster-scan and sit-and-stare modes. To obtain reliable physical parameters, we have developed a data-processing pipeline that derives the emission peak intensity, line width (full width at half-maximum, FWHM), Doppler velocity, and their associated uncertainties. The data pipeline includes dark subtraction, flat-field correction, spectral curvature correction, spectral drift correction, compensation for filter transmission effects, removal of blending due to an absorption line at the edge of the emission profile, and Gaussian fitting to the corrected emission profile. Instrumental broadening is removed from the observed line width to determine the intrinsic emission-line width. The final outputs—peak intensity, line width, and Doppler velocity—are stored as multilayer FITS cubes containing parameter maps and corresponding uncertainty maps, along with updated header metadata. In this paper, we describe the complete methodology and calibration procedures used in the pipeline. The resulting data products are suitable for coronal diagnostics, Doppler flow measurements, and investigations of temporal and spatial evolution of coronal structures.
