Francesco Carella
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Francesco Carella has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Stefaan Poedts
KU Leuven
4 shared publications - George Miloshevich
Centre for mathematical Plasma Astrophysics (CmPA) KULeuven Leuven Belgium
3 shared publications - Panagiotis Gonidakis
Centre for mathematical Plasma Astrophysics (CmPA) KULeuven Leuven Belgium
3 shared publications - Alessandro Bemporad
National Institute for Astrophysics, Astrophysical Observatory of Torino, Pino Torinese
2 shared publications - Ekaterina Dineva
Centre for mathematical Plasma Astrophysics, Department of Mathematics, KU Leuven, Celestijnenlaan 200B, 3001 Leuven, Belgium;
2 shared publications - Haopeng Wang
Centre for mathematical Plasma Astrophysics, Department of Mathematics, KU Leuven, Celestijnenlaan 200B, 3001 Leuven, Belgium;
2 shared publications
Latest Publications
- Open Access
- 14 March 2026
- Published by Copernicus GmbH
Probabilistic Solar Flare Forecasting via Weakly Supervised Contrastive Refinement of VAE Latent Spaces
Reliable solar flare forecasting is limited by two forms of class imbalance in active region time series: (i) the overwhelming dominance of the non-flaring, quiet state over the eruptive state, and (ii) the insufficient separability between common, physically similar event classes (e.g. C-class versus M-class flares). Although empirical parameters derived from the photospheric vector magnetic field (VMF), such as those provided by SDO/HMI SHARP products, capture aspects of active region complexity and free energy buildup, they often evolve smoothly and overlap across flare classes. Consequently, while many models can distinguish between flares and no-flares reasonably well, they struggle to distinguish flare magnitude and association with eruptive phenomena (e.g. CMEs) using photospheric information alone. This suggests that improved flare-class separation requires (a) the explicit definition of what constitutes 'similarity' between pre-flare states, and (b) parametrization that emphasizes flare-relevant structure over common active region features.We investigate a representation learning strategy that combines the parametrization of SDO/HMI SHARP VMF cutouts using a Variational Autoencoder (VAE) with a contrastive stage to reshape the resulting embedding geometry. First, a VAE is trained to encode SHARP cutouts into compact latent vectors that capture active region morphology. These vectors are then refined using a Siamese-like objective constructed from weak supervision, which uses event labels and empirical SHARP parameters as proxies for elevated flare likelihood. The contrastive stage then uses this weak supervision to encourage a latent geometry that better reflects flare-relevant evolution. This study emphasizes latent-space structure, i.e. neighborhood consistency and class-conditional clustering, and evaluates whether these properties facilitate improved probabilistic prediction across multiple forecast horizons, by training lightweight downstream models on (i) empirical parameters, (ii) VAE latents and (iii) their combined representations.
- 14 March 2026
- Published by Copernicus GmbH
Open AccessProbabilistic Solar Flare Forecasting via Weakly Supervised Contrastive Refinement of VAE Latent Spaces
Reliable solar flare forecasting is limited by two forms of class imbalance in active region time series: (i) the overwhelming dominance of the non-flaring, quiet state over the eruptive state, and (ii) the insufficient separability between common, physically similar event classes (e.g. C-class versus M-class flares). Although empirical parameters derived from the photospheric vector magnetic field (VMF), such as those provided by SDO/HMI SHARP products, capture aspects of active region complexity and free energy buildup, they often evolve smoothly and overlap across flare classes. Consequently, while many models can distinguish between flares and no-flares reasonably well, they struggle to distinguish flare magnitude and association with eruptive phenomena (e.g. CMEs) using photospheric information alone. This suggests that improved flare-class separation requires (a) the explicit definition of what constitutes 'similarity' between pre-flare states, and (b) parametrization that emphasizes flare-relevant structure over common active region features.We investigate a representation learning strategy that combines the parametrization of SDO/HMI SHARP VMF cutouts using a Variational Autoencoder (VAE) with a contrastive stage to reshape the resulting embedding geometry. First, a VAE is trained to encode SHARP cutouts into compact latent vectors that capture active region morphology. These vectors are then refined using a Siamese-like objective constructed from weak supervision, which uses event labels and empirical SHARP parameters as proxies for elevated flare likelihood. The contrastive stage then uses this weak supervision to encourage a latent geometry that better reflects flare-relevant evolution. This study emphasizes latent-space structure, i.e. neighborhood consistency and class-conditional clustering, and evaluates whether these properties facilitate improved probabilistic prediction across multiple forecast horizons, by training lightweight downstream models on (i) empirical parameters, (ii) VAE latents and (iii) their combined representations.
- Open Access
- 1 December 2025
- Published by American Astronomical Society in The Astrophysical Journal Supplement Series
- Vol. 281 (2) , 63
Artificial-intelligence-based Reconstruction of Solar Farside Vector Magnetograms from Multispacecraft Extreme-ultraviolet Data
In this study, we generate full-disk vector magnetic field data of the solar farside, as viewed from the Solar Terrestrial Relations Observatory-Ahead (STEREO-A), Solar Terrestrial Relations Observatory-Behind (STEREO-B), and Solar Orbiter (SolO), using a deep learning model based on the Pix2PixCC architecture. Our model takes extreme-ultraviolet (EUV) 304 and 171 Å images, together with reference magnetic field data from a surface flux transport (SFT) model, as inputs. To train and evaluate the model, we use EUV images from the Solar Dynamics Observatory (SDO)/Atmospheric Imaging Assembly and SFT-predicted magnetic field data from one solar rotation earlier as inputs, and we use vector magnetograms from the SDO/Helioseismic and Magnetic Imager (HMI) as targets. For frontside test datasets covering Solar Cycles 24 and 25, our model successfully generates all three vector magnetic field components consistent with those from SDO/HMI, showing improved performance compared with previous studies. We then generate vector magnetic field data by applying the trained model to EUV observations from STEREO-A and SolO. For the first time, we compare the artificial intelligence (AI)-generated results with corresponding SDO/HMI data obtained when STEREO-A and SolO were near inferior conjunction with SDO in 2023 and 2022, respectively. We also track active-region magnetic fields and derive vector magnetic parameters using AI-generated farside data from STEREO-A, STEREO-B, and SolO, along with frontside SDO/HMI data. These results demonstrate the potential for continuous monitoring of solar vector magnetic fields and derived parameters from the farside to the frontside using AI-generated and SDO/HMI data.
- 1 December 2025
- Published by American Astronomical Society in The Astrophysical Journal Supplement Series
- Vol. 281 (2) , 63
Open AccessArtificial-intelligence-based Reconstruction of Solar Farside Vector Magnetograms from Multispacecraft Extreme-ultraviolet Data
In this study, we generate full-disk vector magnetic field data of the solar farside, as viewed from the Solar Terrestrial Relations Observatory-Ahead (STEREO-A), Solar Terrestrial Relations Observatory-Behind (STEREO-B), and Solar Orbiter (SolO), using a deep learning model based on the Pix2PixCC architecture. Our model takes extreme-ultraviolet (EUV) 304 and 171 Å images, together with reference magnetic field data from a surface flux transport (SFT) model, as inputs. To train and evaluate the model, we use EUV images from the Solar Dynamics Observatory (SDO)/Atmospheric Imaging Assembly and SFT-predicted magnetic field data from one solar rotation earlier as inputs, and we use vector magnetograms from the SDO/Helioseismic and Magnetic Imager (HMI) as targets. For frontside test datasets covering Solar Cycles 24 and 25, our model successfully generates all three vector magnetic field components consistent with those from SDO/HMI, showing improved performance compared with previous studies. We then generate vector magnetic field data by applying the trained model to EUV observations from STEREO-A and SolO. For the first time, we compare the artificial intelligence (AI)-generated results with corresponding SDO/HMI data obtained when STEREO-A and SolO were near inferior conjunction with SDO in 2023 and 2022, respectively. We also track active-region magnetic fields and derive vector magnetic parameters using AI-generated farside data from STEREO-A, STEREO-B, and SolO, along with frontside SDO/HMI data. These results demonstrate the potential for continuous monitoring of solar vector magnetic fields and derived parameters from the farside to the frontside using AI-generated and SDO/HMI data.
- Open Access
- 17 September 2025
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 701, A212
Clustering Wind data at 1 AU to contextualize magnetic reconnection in the solar wind
Context. Magnetic reconnection events are frequently observed in the solar wind. Understanding the patterns and structures within the solar wind is crucial to put observed magnetic reconnection events into context, since their occurrence rate and properties are likely influenced by solar wind conditions. Aims. We employed unsupervised learning techniques such as self-organizing maps (SOM) and K-Means to cluster and interpret solar wind data at 1 AU for an improved understanding of the conditions that lead to magnetic reconnection in the solar wind. Methods. We collected magnetic field data and proton density, proton temperature, and solar wind speed measurements taken by the Wind spacecraft. After preprocessing the data, we trained a SOM to visualize the high-dimensional data in a lower-dimensional space and applied K-Means clustering to identify distinct clusters within the solar wind data. We then compare the results with the Xu, F., & Borovsky, J. E. (2015, J. Geophys. Res. Space Phys., 120, 70) classification of the solar wind. Results. Our analysis revealed that the reconnection events are distributed across five different clusters: (a) slow solar wind, (b) compressed slow wind, (c) highly Alfvénic wind, (d) compressed fast wind, and (e) ejecta. Compressed slow and fast wind and ejecta are clusters associated with solar wind transients such as stream interaction regions and interplanetary coronal mass ejections. The majority of the reconnection events are associated with the slow solar wind, followed by the highly Alfvénic wind, compressed slow wind, and compressed fast wind, and a small fraction of the reconnection events are associated with ejecta. Conclusions. Unsupervised learning approaches with SOM and K-Means lead to physically interpretable solar wind clusters based on their transients and allow for the contextualization of magnetic reconnection exhausts’ occurrence in the solar wind.
- 17 September 2025
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 701, A212
Open AccessClustering Wind data at 1 AU to contextualize magnetic reconnection in the solar wind
Context. Magnetic reconnection events are frequently observed in the solar wind. Understanding the patterns and structures within the solar wind is crucial to put observed magnetic reconnection events into context, since their occurrence rate and properties are likely influenced by solar wind conditions. Aims. We employed unsupervised learning techniques such as self-organizing maps (SOM) and K-Means to cluster and interpret solar wind data at 1 AU for an improved understanding of the conditions that lead to magnetic reconnection in the solar wind. Methods. We collected magnetic field data and proton density, proton temperature, and solar wind speed measurements taken by the Wind spacecraft. After preprocessing the data, we trained a SOM to visualize the high-dimensional data in a lower-dimensional space and applied K-Means clustering to identify distinct clusters within the solar wind data. We then compare the results with the Xu, F., & Borovsky, J. E. (2015, J. Geophys. Res. Space Phys., 120, 70) classification of the solar wind. Results. Our analysis revealed that the reconnection events are distributed across five different clusters: (a) slow solar wind, (b) compressed slow wind, (c) highly Alfvénic wind, (d) compressed fast wind, and (e) ejecta. Compressed slow and fast wind and ejecta are clusters associated with solar wind transients such as stream interaction regions and interplanetary coronal mass ejections. The majority of the reconnection events are associated with the slow solar wind, followed by the highly Alfvénic wind, compressed slow wind, and compressed fast wind, and a small fraction of the reconnection events are associated with ejecta. Conclusions. Unsupervised learning approaches with SOM and K-Means lead to physically interpretable solar wind clusters based on their transients and allow for the contextualization of magnetic reconnection exhausts’ occurrence in the solar wind.
- Open Access
- 20 August 2025
- Vol. 989 (2)
Prediction of Time-evolving Radial Magnetic Fields on the Solar Surface Using Deep Learning
- 20 August 2025
- Vol. 989 (2)
Open AccessPrediction of Time-evolving Radial Magnetic Fields on the Solar Surface Using Deep Learning
- Open Access
- 18 March 2025
- Published by Copernicus GmbH
Identification of fast solar wind flows and CMEs in the in situ data using Self-Organizing Maps and clustering techniques
The identification and characterization of the coronal mass ejections (CMEs) and fast solar wind flows in the in situ data are important for understanding dynamics of these phenomena and consequently for space weather forecasting. In this study, we apply Self-Organizing Maps (SOMs) and clustering techniques to analyze in situ solar wind observations. SOMs (Kohonen, T, 1982) [1] an unsupervised learning technique, is employed to project high-dimensional interplanetary plasma parameters such as velocity, density, temperature, and magnetic field onto a lower-dimensional representation, preserving the topological structure of the data. Clustering algorithms, such as k-means, are then applied to the SOM output to distinguish between ICME events, fast and slow solar wind flows. Our approach is validated using a few months long interval of the ACE and Wind in situ observations, with labeled CME intervals from Richardson and Cane [2] as a benchmark. This combination of SOMs and clustering provides a framework for automated identification of interplanetary plasma structures, important for space weather studies but also for operational services. [1] T. Kohonen, ‘Self-organized formation of topologically correct feature maps’, Biol. Cybern., vol. 43, no. 1, pp. 59–69, Jan. 1982, doi: 10.1007/BF00337288[2] Richardson, Ian; Cane, Hilary, 2024, "Near-Earth Interplanetary Coronal Mass Ejections Since January 1996"https://doi.org/10.7910/DVN/C2MHTH
- 18 March 2025
- Published by Copernicus GmbH
Open AccessIdentification of fast solar wind flows and CMEs in the in situ data using Self-Organizing Maps and clustering techniques
The identification and characterization of the coronal mass ejections (CMEs) and fast solar wind flows in the in situ data are important for understanding dynamics of these phenomena and consequently for space weather forecasting. In this study, we apply Self-Organizing Maps (SOMs) and clustering techniques to analyze in situ solar wind observations. SOMs (Kohonen, T, 1982) [1] an unsupervised learning technique, is employed to project high-dimensional interplanetary plasma parameters such as velocity, density, temperature, and magnetic field onto a lower-dimensional representation, preserving the topological structure of the data. Clustering algorithms, such as k-means, are then applied to the SOM output to distinguish between ICME events, fast and slow solar wind flows. Our approach is validated using a few months long interval of the ACE and Wind in situ observations, with labeled CME intervals from Richardson and Cane [2] as a benchmark. This combination of SOMs and clustering provides a framework for automated identification of interplanetary plasma structures, important for space weather studies but also for operational services. [1] T. Kohonen, ‘Self-organized formation of topologically correct feature maps’, Biol. Cybern., vol. 43, no. 1, pp. 59–69, Jan. 1982, doi: 10.1007/BF00337288[2] Richardson, Ian; Cane, Hilary, 2024, "Near-Earth Interplanetary Coronal Mass Ejections Since January 1996"https://doi.org/10.7910/DVN/C2MHTH
- Open Access
- 18 March 2025
- Published by Copernicus GmbH
Efficient Segmentation and Clustering of Solar Coronal Structures: A Comparison of U-Net and Classical Computer Vision Techniques Using SDO Data
Segmentation and characterization of solar coronal structures are essential for advancing our understanding of solar atmosphere and accurately identifying key regions, such as active regions and coronal holes, which are precursors to phenomena like solar flares and coronal mass ejections (CMEs). In this study, we investigate two complementary approaches to automate this process. First, we employ a previously presented deep learning-based U-Net architecture tailored for segmenting and characterizing solar coronal structures. Second, we develop a lightweight algorithm aimed at optimizing resource efficiency, consisting of classical computer vision techniques, which include thresholding and morphological filtering. The approach that best balances segmentation performance and computational efficiency will be selected for integration into a prototype designed to support future space exploration missions.To characterize the segmented regions, we propose a set of carefully designed hand-crafted features to represent and characterize the resulting segmentations. These representations are analyzed using unsupervised clustering techniques, such as K-means and t-SNE, to distinguish solar coronal structures, including active regions, coronal holes and bright points.Our dataset spans multiple layers of the solar atmosphere, incorporating HMI magnetograms (photosphere) and AIA wavelengths—94 Å (flaring regions), 171 Å (quiet Sun), 193 Å (coronal structures), and 304 Å (chromosphere). The performance of both segmentation approaches is thoroughly evaluated using metrics such as Dice score and Intersection over Union (IoU), with comparisons made against state-of-the-art methods.Future work will focus on developing feature encoding techniques to better understand and predict solar phenomena, such as solar flare emissions, while investigating the impact of different feature extraction strategies on model performance.References:Galvez, Richard, et al. "A machine-learning data set prepared from the NASA solar dynamics observatory mission." The Astrophysical Journal Supplement Series 242.1 (2019): 7. Šimon Mackovjak et al. “SCSS-Net: solar corona structures segmentation by deep learning”, Monthly Notices of the Royal Astronomical Society, Volume 508, Issue 3, December 2021, Pages 3111–3124, https://doi.org/10.1093/mnras/stab2536 Gonidakis, Panagiotis & Sóñora-Mengana, Alexander & Jansen, Bart & Vandemeulebroucke, Jef. (2023). Handcrafted Features Can Boost Performance and Data-Efficiency for Deep Detection of Lung Nodules From CT Imaging. IEEE Access. PP. 1-1. 10.1109/ACCESS.2023.3331315.
- 18 March 2025
- Published by Copernicus GmbH
Open AccessEfficient Segmentation and Clustering of Solar Coronal Structures: A Comparison of U-Net and Classical Computer Vision Techniques Using SDO Data
Segmentation and characterization of solar coronal structures are essential for advancing our understanding of solar atmosphere and accurately identifying key regions, such as active regions and coronal holes, which are precursors to phenomena like solar flares and coronal mass ejections (CMEs). In this study, we investigate two complementary approaches to automate this process. First, we employ a previously presented deep learning-based U-Net architecture tailored for segmenting and characterizing solar coronal structures. Second, we develop a lightweight algorithm aimed at optimizing resource efficiency, consisting of classical computer vision techniques, which include thresholding and morphological filtering. The approach that best balances segmentation performance and computational efficiency will be selected for integration into a prototype designed to support future space exploration missions.To characterize the segmented regions, we propose a set of carefully designed hand-crafted features to represent and characterize the resulting segmentations. These representations are analyzed using unsupervised clustering techniques, such as K-means and t-SNE, to distinguish solar coronal structures, including active regions, coronal holes and bright points.Our dataset spans multiple layers of the solar atmosphere, incorporating HMI magnetograms (photosphere) and AIA wavelengths—94 Å (flaring regions), 171 Å (quiet Sun), 193 Å (coronal structures), and 304 Å (chromosphere). The performance of both segmentation approaches is thoroughly evaluated using metrics such as Dice score and Intersection over Union (IoU), with comparisons made against state-of-the-art methods.Future work will focus on developing feature encoding techniques to better understand and predict solar phenomena, such as solar flare emissions, while investigating the impact of different feature extraction strategies on model performance.References:Galvez, Richard, et al. "A machine-learning data set prepared from the NASA solar dynamics observatory mission." The Astrophysical Journal Supplement Series 242.1 (2019): 7. Šimon Mackovjak et al. “SCSS-Net: solar corona structures segmentation by deep learning”, Monthly Notices of the Royal Astronomical Society, Volume 508, Issue 3, December 2021, Pages 3111–3124, https://doi.org/10.1093/mnras/stab2536 Gonidakis, Panagiotis & Sóñora-Mengana, Alexander & Jansen, Bart & Vandemeulebroucke, Jef. (2023). Handcrafted Features Can Boost Performance and Data-Efficiency for Deep Detection of Lung Nodules From CT Imaging. IEEE Access. PP. 1-1. 10.1109/ACCESS.2023.3331315.