Search

AI-FLARES (Artificial Intelligence for the Analysis of Solar Flares Data) is a research project funded by the Agenzia Spaziale Italiana and by the Istituto Nazionale di Astrofisica within the framework of the ``Attivit\`a di Studio per la Comunit\`a Scientifica Nazionale Sole, Sistema Solare ed Esopianeti'' program. The topic addressed by this project was the development and use of computational methods for the analysis of remote sensing space data associated to solar flare emission. This paper overviews the main results obtained by the project, with specific focus on solar flare forecasting, reconstruction of morphologies of the flaring sources, and interpretation of acceleration mechanisms triggered by solar flares.

The project CAESAR (Comprehensive spAce wEather Studies for the ASPIS prototype Realization) is aimed to tackle all the relevant aspects of Space Weather (SWE) and realize the prototype of the scientific data centre for Space Weather of the Italian Space Agency (ASI) called ASPIS (ASI SPace Weather InfraStructure). This contribution is meant to bring attention upon the first steps in the development of the CAESAR prototype for ASPIS and will focus on the activities of the Node 2000 of CAESAR, the set of Work Packages dedicated to the technical design and implementation of the CAESAR ASPIS archive prototype. The product specifications of the intended resources that will form the archive, functional and system requirements gathered as first steps to seed the design of the prototype infrastructure, and evaluation of existing frameworks, tools and standards, will be presented as well as the status of the project in its initial stage., Comment: 4 pages, 2 figures, ADASS XXXII (2022) Proceedings

Medical ultrasound imaging is the most widespread real-time non-invasive imaging system and its formulation comprises signal transmission, signal reception, and image formation. Ultrasound signal transmission modelling has been formalized over the years through different approaches by exploiting the physics of the associated wave problem. This work proposes a novel computational framework for modelling the ultrasound signal transmission step in the time-frequency domain for a linear-array probe. More specifically, from the impulse response theory defined in the time domain, we derived a parametric model in the corresponding frequency domain, with appropriate approximations for the narrowband case. To validate the model, we implemented a numerical simulator and tested it with synthetic data. Numerical experiments demonstrate that the proposed model is computationally feasible, efficient, and compatible with realistic measurements and existing state-of-the-art simulators. The formulated model can be employed for analyzing how the involved parameters affect the generated beam pattern, and ultimately for optimizing measurement settings in an automatic and systematic way.

In many contexts, customized and weighted classification scores are designed in order to evaluate the goodness of the predictions carried out by neural networks. However, there exists a discrepancy between the maximization of such scores and the minimization of the loss function in the training phase. In this paper, we provide a complete theoretical setting that formalizes weighted classification metrics and then allows the construction of losses that drive the model to optimize these metrics of interest. After a detailed theoretical analysis, we show that our framework includes as particular instances well-established approaches such as classical cost-sensitive learning, weighted cross entropy loss functions and value-weighted skill scores.

Coronal Mass Ejections (CMEs) correspond to dramatic expulsions of plasma and magnetic field from the solar corona into the heliosphere. CMEs are scientifically relevant because they are involved in the physical mechanisms characterizing the active Sun. However, more recently CMEs have attracted attention for their impact on space weather, as they are correlated to geomagnetic storms and may induce the generation of Solar Energetic Particles streams. In this space weather framework, the present paper introduces a physics-driven artificial intelligence (AI) approach to the prediction of CMEs travel time, in which the deterministic drag-based model is exploited to improve the training phase of a cascade of two neural networks fed with both remote sensing and in-situ data. This study shows that the use of physical information in the AI architecture significantly improves both the accuracy and the robustness of the travel time prediction.

Aims. To provide a schematic mathematical description of the imaging concept of the Spectrometer/Telescope for Imaging X-rays (STIX) on board Solar Orbiter. The derived model is the fundamental starting point for both the interpretation of STIX data and the description of the data calibration process. Methods. We describe the STIX indirect imaging technique which is based on spatial modulation of the X-ray photon flux by means of tungsten grids. We show that each of 30 STIX imaging sub-collimators measures a complex Fourier component of the flaring X-ray source corresponding to a specific angular frequency. We also provide details about the count distribution model, which describes the relationship between the photon flux and the measured pixel counts. Results. We define the image reconstruction problem for STIX from both visibilities and photon counts. We provide an overview of the algorithms implemented for the solution of the imaging problem, and a comparison of the results obtained with these different methods in the case of the SOL2022-03-31T18 flaring event.

In this work we present a fast, globally convergent, iterative algorithm for computing the asymptotically stable states of nonlinear large--scale systems of quadratic autonomous Ordinary Differential Equations (ODEs) modeling, e.g., the dynamic of complex chemical reaction networks. Towards this aim, we reformulate the problem as a box--constrained optimization problem where the roots of a set of nonlinear equations need to be determined. Then, we propose to use a projected Newton's approach combined with a gradient descent algorithm so that every limit point of the sequence generated by the overall algorithm is a stationary point. More importantly, we suggest replacing the standard orthogonal projector with a novel operator that ensures the final solution to satisfy the box constraints while lowering the probability that the intermediate points reached at each iteration belong to the boundary of the box where the Jacobian of the objective function may be singular. The effectiveness of the proposed approach is shown in a practical scenario concerning a chemical reaction network modeling the signaling network of colorectal cancer cells. Specifically, in this scenario the proposed algorithm is proven to be faster and more accurate than a classical dynamical approach where the asymptotically stable states are computed as the limit points of the flux of the Cauchy problem associated with the ODEs system., Comment: 22 pages, 5 figures, 1 table

Ultrasound imaging is extensively employed in clinical settings due to its non-ionizing nature and real-time capabilities. The beamformer represents a crucial component of an ultrasound machine, playing a significant role in shaping the ultimate quality of the reconstructed image. Therefore, Transmit Beam Pattern (TBP) optimization is an important task in medical ultrasound, but state-of-the-art TBP optimization has well-known drawbacks like non-uniform beam width over depth, presence of significant side lobes, and quick energy drop out after the focal depth. To overcome these limitations, we developed a novel optimization approach for TBP by focusing the analysis on its narrowband approximation, particularly suited for Acoustic Radiation Force Impulse (ARFI) elastography, and considering transmit delays as free variables instead of linked to a specific focal depth. We formulate the problem as a non linear Least Squares problem to minimize the difference between the TBP corresponding to a set of delays and the desired one, modeled as a 2D rectangular shape elongated in the direction of the beam axis. In order to quantitatively evaluate the results, we define three quality metrics based on main lobe width, side lobe level, and central line power. Results obtained by our synthetic software simulation show that the main lobe width is considerably more intense and uniform over the whole depth range with respect to classical focalized Beam Patterns, and our optimized delay profile results in a combination of standard delay profiles at different focal depths. The application of the proposed method to ARFI elastography shows improvements in the concentration of the ultrasound energy along a desired axis., Comment: 14 pages, 14 figures

Operational flare forecasting aims at providing predictions that can be used to make decisions, typically at a daily scale, about the space weather impacts of flare occurrence. This study shows that video-based deep learning can be used for operational purposes when the training and validation sets used for the network optimization are generated while accounting for the periodicity of the solar cycle. Specifically, the paper describes an algorithm that can be applied to build up sets of active regions that are balanced according to the flare class rates associated to a specific cycle phase. These sets are used to train and validate a Long-term Recurrent Convolutional Network made of a combination of a convolutional neural network and a Long-Short Memory network. The reliability of this approach is assessed in the case of two prediction windows containing the solar storm of March 2015 and September 2017, respectively.

Aima. To determine to what extent the problem of forward fitting visibilities measured by the Spectrometer/Telescope Imaging X-rays (STIX) on-board Solar Orbiter is more challenging with respect to the same problem in the case of previous hard X-ray solar imaging missions; to identify an effective optimization scheme for parametric imaging for STIX. Methods. This paper introduces a Particle Swarm Optimization (PSO) algorithm for forward fitting STIX visibilities and compares its effectiveness with respect to the standard simplex-based optimization algorithm used so far for the analysis of visibilities measured by the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI). This comparison is made by considering experimental visibilities measured by both RHESSI and STIX, and synthetic visibilities generated by accounting for the STIX signal formation model. Results. We found out that the parametric imaging approach based on PSO is as reliable as the one based on the simplex method in the case of RHESSI visibilities. However, PSO is significantly more robust when applied to STIX simulated and experimental visibilities. Conclusions. Standard deterministic optimization is not effective enough for forward-fitting the few visibilities sampled by STIX in the angular frequency plane. Therefore a more sofisticated optimization scheme must be introduced for parametric imaging in the case of the Solar Orbiter X-ray telescope. The forward-fitting routine based on PSO we introduced in this paper proved to be significantly robust and reliable, and could be considered as an effective candidate tool for parametric imaging in the STIX context.

Context. The Spectrometer/Telescope for Imaging X-rays (STIX) is one of 6 remote sensing instruments on-board Solar Orbiter. It provides hard X-ray imaging spectroscopy of solar flares by sampling the Fourier transform of the incoming flux. Aims. To show that the visibility amplitude and phase calibration of 24 out of 30 STIX sub-collimators is well advanced and that a set of imaging methods is able to provide the first hard X-ray images of the flaring Sun from Solar Orbiter. Methods. We applied four visibility-based image reconstruction methods and a count-based one to calibrated STIX observations. The resulting reconstructions are compared to those provided by an optimization algorithm used for fitting the amplitudes of STIX visibilities. Results. When applied to six flares with GOES class between C4 and M4 which occurred in May 2021, the five imaging methods produce results morphologically consistent with the ones provided by the Atmospheric Imaging Assembly on-board the Solar Dynamic Observatory (SDO/AIA) in UV wavelengths. The $\chi^2$ values and the parameters of the reconstructed sources are comparable between methods, thus confirming their robustness. Conclusions. This paper shows that the current calibration of the main part of STIX sub-collimators has reached a satisfactory level for scientific data exploitation, and that the imaging algorithms already available in the STIX data analysis software provide reliable and robust reconstructions of the morphology of solar flares.

In this work we deal with parametric inverse problems, which consist in recovering a finite number of parameters describing the structure of an unknown object, from indirect measurements. State-of-the-art methods for approximating a regularizing inverse operator by using a dataset of input-output pairs of the forward model rely on deep learning techniques. In these approaches, a neural network is trained to predict the value of the sought parameters directly from the data. In this paper, we show that these methods provide suboptimal results when the topology of the parameter space is strictly coarser than the Euclidean one. To overcome this issue, we propose a two-step strategy for approximating a regularizing inverse operator by means of a neural network, which works under general topological conditions. First, we embed the parameters into a subspace of a low-dimensional Euclidean space; second, we use a neural network to approximate a homeomorphism between the subspace and the image of the parameter space through the forward operator. The parameters are then retrieved by applying the inverse of the embedding to the network predictions. The results are shown for the problem of X-ray imaging of solar flares with data from the Spectrometer/Telescope for Imaging X-rays. In this case, the parameter space is a family of Moebius strips that collapse into a point. Our simulation studies show that the use of a neural network for predicting the parameters directly from the data yields systematic errors due to the non-Euclidean topology of the parameter space. The proposed strategy overcomes the topological issues and furnishes stable and accurate reconstructions.

Solar flare forecasting can be realized by means of the analysis of magnetic data through artificial intelligence techniques. The aim is to predict whether a magnetic active region (AR) will originate solar flares above a certain class within a certain amount of time. A crucial issue is concerned with the way the adopted machine learning method is implemented, since forecasting results strongly depend on the criterion with which training, validation, and test sets are populated. In this paper we propose a general paradigm to generate these sets in such a way that they are independent from each other and internally well-balanced in terms of AR flaring effectiveness. This set generation process provides a ground for comparison for the performance assessment of machine learning algorithms. Finally, we use this implementation paradigm in the case of a deep neural network, which takes as input videos of magnetograms recorded by the Helioseismic and Magnetic Imager on-board the Solar Dynamics Observatory (SDO/HMI). To our knowledge, this is the first time that the solar flare forecasting problem is addressed by means of a deep neural network for video classification, which does not require any a priori extraction of features from the HMI magnetograms.

The problem of nowcasting extreme weather events can be addressed by applying either numerical methods for the solution of dynamic model equations or data-driven artificial intelligence algorithms. Within this latter framework, the present paper illustrates how a deep learning method, exploiting videos of radar reflectivity frames as input, can be used to realize a warning machine able to sound timely alarms of possible severe thunderstorm events. From a technical viewpoint, the computational core of this approach is the use of a value-weighted skill score for both transforming the probabilistic outcomes of the deep neural network into binary classification and assessing the forecasting performances. The warning machine has been validated against weather radar data recorded in the Liguria region, in Italy

Aims: To provide the first demonstration of STIX Fourier-transform X-ray imaging using semi-calibrated (amplitude-only) visibility data acquired during the Solar Orbiter's cruise phase. Methods: We use a parametric imaging approach by which STIX visibility amplitudes are fitted by means of two non-linear optimization methods: a fast meta-heuristic technique inspired by social behavior, and a Bayesian Monte Carlo sampling method, which, although slower, provides better quantification of uncertainties. Results: When applied to a set of solar flare visibility amplitudes recorded by STIX on November 18, 2020 the two parametric methods provide very coherent results. The analysis also demonstrates the ability of STIX to reconstruct high time resolution information and, from a spectral viewpoint, shows the reliability of a double-source scenario consistent with a thermal versus nonthermal interpretation. Conclusions: In this preliminary analysis of STIX imaging based only on visibility amplitudes, we formulate the imaging problem as a non-linear parametric issue we addressed by means of two high-performance optimization techniques that both showed the ability to sample the parametric space in an effective fashion, thus avoiding local minima.

The Spectrometer/Telescope for Imaging X-rays (STIX) is the HXR instrument onboard Solar Orbiter designed to observe solar flares over a broad range of flare sizes, between 4-150 keV. We report the first STIX observations of microflares recorded during the instrument commissioning phase in order to investigate the STIX performance at its detection limit. This first result paper focuses on the temporal and spectral evolution of STIX microflares occuring in the AR12765 in June 2020, and compares the STIX measurements with GOES/XRS, SDO/AIA, and Hinode/XRT. For the observed microflares of the GOES A and B class, the STIX peak time at lowest energies is located in the impulsive phase of the flares, well before the GOES peak time. Such a behavior can either be explained by the higher sensitivity of STIX to higher temperatures compared to GOES, or due to the existence of a nonthermal component reaching down to low energies. The interpretation is inconclusive due to limited counting statistics for all but the largest flare in our sample. For this largest flare, the low-energy peak time is clearly due to thermal emission, and the nonthermal component seen at higher energies occurs even earlier. This suggests that the classic thermal explanation might also be favored for the majority of the smaller flares. In combination with EUV and SXR observations, STIX corroborates earlier findings that an isothermal assumption is of limited validity. Future diagnostic efforts should focus on multi-wavelength studies to derive differential emission measure distributions over a wide range of temperatures to accurately describe the energetics of solar flares. Commissioning observations confirm that STIX is working as designed. As a rule of thumb, STIX detects flares as small as the GOES A class. For flares above the GOES B class, detailed spectral and imaging analyses can be performed., Comment: 19 pages, 11 figures

Solar flares emanate from solar active regions hosting complex and strong bipolar magnetic fluxes. Estimating the probability of an active region to flare and defining reliable precursors of intense flares is an extremely challenging task in the space weather field. In this work, we focus on two metrics as flare precursors, the unsigned flux R, tested on MDI/SOHO data and one of the most used parameters for flare forecasting applications, and a novel topological parameter D representing the complexity of a solar active region. More in detail, we propose an algorithm for the computation of the R value which exploits the higher spatial resolution of HMI maps. This algorithm leads to a differently computed R value, whose functionality is tested on a set of cycle 24th solar flares. Furthermore, we introduce a topological parameter based on the automatic recognition of magnetic polarity-inversion lines in identified active regions, and able to evaluate its magnetic topological complexity. We use both a heuristic approach and a supervised machine learning method to validate the effectiveness of these two descriptors to predict the occurrence of X- or M- class flares in a given solar active region during the following 24 hours period. Our feature ranking analysis shows that both parameters play a significant role in prediction performances. Moreover, the analysis demonstrates that the new topological parameter D is the only one, among 173 overall predictors, which is always present for all test subsets and is systematically ranked within the top-ten positions in all tests concerning the computation of the weighs with which each predictor impacts the flare forecasting., Comment: 16 pages, 8 figures, accepted in the Astrophysical Journal

In this paper we propose a novel approach to realize forecast verification. Specifically, we introduce a strategy for assessing the severity of forecast errors based on the evidence that, on the one hand, a false alarm just anticipating an occurring event is better than one in the middle of consecutive non-occurring events, and that, on the other hand, a miss of an isolated event has a worse impact than a miss of a single event, which is part of several consecutive occurrences. Relying on this idea, we introduce a novel definition of confusion matrix and skill scores giving greater importance to the value of the prediction rather than to its quality. Then, we introduce a deep ensemble learning procedure for binary classification, in which the probabilistic outcomes of a neural network are clustered via optimization of these value-weighted skill scores. We finally show the performances of this approach in the case of three applications concerned with pollution, space weather and stock prize forecasting.

In this work, we introduce a novel estimator of the predictive risk with Poisson data, when the loss function is the Kullback-Leibler divergence, in order to define a regularization parameter's choice rule for the Expectation Maximization (EM) algorithm. To this aim, we prove a Poisson counterpart of the Stein's Lemma for Gaussian variables, and from this result we derive the proposed estimator showing its analogies with the well-known Stein's Unbiased Risk Estimator valid for a quadratic loss. We prove that the proposed estimator is asymptotically unbiased with increasing number of measured counts, under certain mild conditions on the regularization method. We show that these conditions are satisfied by the EM algorithm and then we apply this estimator to select its optimal reconstruction. We present some numerical tests in the case of image deconvolution, comparing the performances of the proposed estimator with other methods available in the literature, both in the inverse crime and non-inverse crime setting.

Machine learning is nowadays the methodology of choice for flare forecasting and supervised techniques, in both their traditional and deep versions, are becoming the most frequently used ones for prediction in this area of space weather. Yet, machine learning has not been able so far to realize an operating warning system for flaring storms and the scientific literature of the last decade suggests that its performances in the prediction of intense solar flares are not optimal. The main difficulties related to forecasting solar flaring storms are probably two. First, most methods are conceived to provide probabilistic predictions and not to send binary yes/no indications on the consecutive occurrence of flares along an extended time range. Second, flaring storms are typically characterized by the explosion of high energy events, which are seldom recorded in the databases of space missions; as a consequence, supervised methods are trained on very imbalanced historical sets, which makes them particularly ineffective for the forecasting of intense flares. Yet, in this study we show that supervised machine learning could be utilized in a way to send timely warnings about the most violent and most unexpected flaring event of the last decade, and even to predict with some accuracy the energy budget daily released by magnetic reconnection during the whole time course of the storm. Further, we show that the combination of sparsity-enhancing machine learning and feature ranking could allow the identification of the prominent role that energy played as an Active Region property in the forecasting process.

Maximum Entropy is an image reconstruction method conceived to image a sparsely occupied field of view and therefore particularly appropriate to achieve super-resolution effects. Although widely used in image deconvolution, this method has been formulated in radio astronomy for the analysis of observations in the spatial frequency domain, and an Interactive Data Language (IDL) code has been implemented for image reconstruction from solar X-ray Fourier data. However, this code relies on a non-convex formulation of the constrained optimization problem addressed by the Maximum Entropy approach and this sometimes results in unreliable reconstructions characterized by unphysical shrinking effects. This paper introduces a new approach to Maximum Entropy based on the constrained minimization of a convex functional. In the case of observations recorded by the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI), the resulting code provides the same super-resolution effects of the previous algorithm, while working properly also when that code produces unphysical reconstructions. Results are also provided of testing the algorithm with synthetic data simulating observations of the Spectrometer/Telescope for Imaging X-rays (STIX) in Solar Orbiter. The new code is available in the {\em{HESSI}} folder of the Solar SoftWare (SSW)tree.

In this work, we propose a new criterion for choosing the regularization parameter in Tikhonov regularization when the noise is white Gaussian. The criterion minimizes a lower bound of the predictive risk, when both data norm and noise variance are known, and the parameter choice involves minimizing a function whose solution depends only on the signal-to-noise ratio. Moreover, when neither noise variance nor data norm is given, we propose an iterative algorithm which alternates between a minimization step of finding the regularization parameter and an estimation step of estimating signal-to-noise ratio. Simulation studies on both small- and large-scale datasets suggest that the approach can provide very accurate and stable regularized inverse solutions and, for small sized samples, it outperforms discrepancy principle, balancing principle, unbiased predictive risk estimator, L-curve method generalized cross validation, and quasi-optimality criterion, and achieves excellent stability hitherto unavailable., Comment: 21 pages, 8 figures, 3 tables

Solar flares originate from magnetically active regions but not all solar active regions give rise to a flare. Therefore, the challenge of solar flare prediction benefits by an intelligent computational analysis of physics-based properties extracted from active region observables, most commonly line-of-sight or vector magnetograms of the active-region photosphere. For the purpose of flare forecasting, this study utilizes an unprecedented 171 flare-predictive active region properties, mainly inferred by the Helioseismic and Magnetic Imager onboard the Solar Dynamics Observatory (SDO/HMI) in the course of the European Union Horizon 2020 FLARECAST project. Using two different supervised machine learning methods that allow feature ranking as a function of predictive capability, we show that: i) an objective training and testing process is paramount for the performance of every supervised machine learning method; ii) most properties include overlapping information and are therefore highly redundant for flare prediction; iii) solar flare prediction is still - and will likely remain - a predominantly probabilistic challenge.

Image saturation has been an issue for several instruments in solar astronomy, mainly at EUV wavelengths. However, with the launch of the Atmospheric Imaging Assembly (AIA) as part of the payload of the Solar Dynamic Observatory (SDO) image saturation has become a big data issue, involving around 10^$ frames of the impressive dataset this beautiful telescope has been providing every year since February 2010. This paper introduces a novel desaturation method, which is able to recover the signal in the saturated region of any AIA image by exploiting no other information but the one contained in the image itself. This peculiar methodological property, jointly with the unprecedented statistical reliability of the desaturated images, could make this algorithm the perfect tool for the realization of a reconstruction pipeline for AIA data, able to work properly even in the case of long-lasting, very energetic flaring events.

The Spectrometer/Telescope for Imaging X-rays (STIX) will look at solar flares across the hard X-ray window provided by the Solar Orbiter cluster. Similarly to the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI), STIX is a visibility-based imaging instrument, which will ask for Fourier-based image reconstruction methods. However, in this paper we show that, as for RHESSI, also for STIX count-based imaging is possible. Specifically, here we introduce and illustrate a mathematical model that mimics the STIX data formation process as a projection from the incoming photon flux into a vector made of 120 count components. Then we test the reliability of Expectation Maximization for image reconstruction in the case of several simulated configurations typical of flare morphology., Comment: submitted to A&A

We describe here the application of a machine learning method for flare forecasting using vectors of properties extracted from images provided by the Helioseismic and Magnetic Imager in the Solar Dynamics Observatory (SDO/HMI). We also discuss how the method can be used to quantitatively assess the impact of such properties on the prediction process., Comment: submitted to 'Nuovo Cimento' as proceeding of SOHE3

In this paper we investigate the connection between supervised learning and linear inverse problems. We first show that a linear inverse problem can be view as a function approximation problem in a reproducing kernel Hilbert space (RKHS) and then we prove that to each of these approximation problems corresponds a class of inverse problems. Analogously, we show that Tikhonov solutions of this class correspond to the Tikhonov solution of the approximation problem. Thanks to this correspondence, we show that supervised learning and linear discrete inverse problems can be thought of as two instances of the approximation problem in a RKHS. These instances are formalized by means of a sampling operator which takes into account both deterministic and random samples and leads to discretized problems. We then analyze the discretized problems and we study the convergence of their solutions to the ones of the approximation problem in a RKHS, both in the deterministic and statistical framework. Finally, we prove there exists a relation between the convergence rates computed with respect to the noise level and the ones computed with respect to the number of samples. This allows us to compare upper and lower bounds given in the statistical learning and in the deterministic infinite dimensional inverse problems theory., Comment: 32 pages

'Flare Likelihood and Region Eruption Forecasting (FLARECAST)' is a Horizon 2020 project, which realized a technological platform for machine learning algorithms, with the objective of providing the space weather community with a prediction service for solar flares. This paper describes the FLARECAST service and shows how the methods implemented in the platform allow both flare prediction and a quantitative assessment of how the information contained in the space data utilized in the analysis may impact the forecasting process., Comment: IEEE Italy Session - 4th International Forum on Research and Technologies for Society and Industry

We propose a forecasting approach for solar flares based on data from Solar Cycle 24, taken by the Helioseismic and Magnetic Imager (HMI) on board the Solar Dynamics Observatory (SDO) mission. In particular, we use the Space-weather HMI Active Region Patches (SHARP) product that facilitates cut-out magnetograms of solar active regions (AR) in the Sun in near-real-time (NRT), taken over a five-year interval (2012 - 2016). Our approach utilizes a set of thirteen predictors, which are not included in the SHARP metadata, extracted from line-of-sight and vector photospheric magnetograms. We exploit several Machine Learning (ML) and Conventional Statistics techniques to predict flares of peak magnitude >M1 and >C1, within a 24 h forecast window. The ML methods used are multi-layer perceptrons (MLP), support vector machines (SVM) and random forests (RF). We conclude that random forests could be the prediction technique of choice for our sample, with the second best method being multi-layer perceptrons, subject to an entropy objective function. A Monte Carlo simulation showed that the best performing method gives accuracy ACC=0.93(0.00), true skill statistic TSS=0.74(0.02) and Heidke skill score HSS=0.49(0.01) for >M1 flare prediction with probability threshold 15% and ACC=0.84(0.00), TSS=0.60(0.01) and HSS=0.59(0.01) for >C1 flare prediction with probability threshold 35%., Comment: Accepted for publication by Solar Physics

We introduce a hybrid approach to solar flare prediction, whereby a supervised regularization method is used to realize feature importance and an unsupervised clustering method is used to realize the binary flare/no-flare decision. The approach is validated against NOAA SWPC data.

The Atmospheric Imaging Assembly in the Solar Dynamics Observatory provides full Sun images every 1 seconds in each of 7 Extreme Ultraviolet passbands. However, for a significant amount of these images, saturation affects their most intense core, preventing scientists from a full exploitation of their physical meaning. In this paper we describe a mathematical and automatic procedure for the recovery of information in the primary saturation region based on a correlation/inversion analysis of the diffraction pattern associated to the telescope observations. Further, we suggest an interpolation-based method for determining the image background that allows the recovery of information also in the region of secondary saturation (blooming).

This paper is concerned with the image reconstruction problem when the measured data are solar hard X-ray modulation profiles obtained from the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI)} instrument. Our goal is to demonstrate that a statistical iterative method classically applied to the image deconvolution problem is very effective when utilized for the analysis of count modulation profiles in solar hard X-ray imaging based on Rotating Modulation Collimators. The algorithm described in this paper solves the maximum likelihood problem iteratively and encoding a positivity constraint into the iterative optimization scheme. The result is therefore a classical Expectation Maximization method this time applied not to an image deconvolution problem but to image reconstruction from count modulation profiles. The technical reason that makes our implementation particularly effective in this application is the use of a very reliable stopping rule which is able to regularize the solution providing, at the same time, a very satisfactory Cash-statistic (C-statistic). The method is applied to both reproduce synthetic flaring configurations and reconstruct images from experimental data corresponding to three real events. In this second case, the performance of Expectation Maximization, when compared to Pixon image reconstruction, shows a comparable accuracy and a notably reduced computational burden; when compared to CLEAN, shows a better fidelity with respect to the measurements with a comparable computational effectiveness. If optimally stopped, Expectation Maximization represents a very reliable method for image reconstruction in the RHESSI context when count modulation profiles are used as input data.

In this paper we propose a new statistical stopping rule for constrained maximum likelihood iterative algorithms applied to ill-posed inverse problems. To this aim we extend the definition of Tikhonov regularization in a statistical framework and prove that the application of the proposed stopping rule to the Iterative Space Reconstruction Algorithm (ISRA) in the Gaussian case and Expectation Maximization (EM) in the Poisson case leads to well defined regularization methods according to the given definition. We also prove that, if an inverse problem is genuinely ill-posed in the sense of Tikhonov, the same definition is not satisfied when ISRA and EM are optimized by classical stopping rule like Morozov's discrepancy principle, Pearson's test and Poisson discrepancy principle. The stopping rule is illustrated in the case of image reconstruction from data recorded by the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI). First, by using a simulated image consisting of structures analogous to those of a real solar flare we validate the fidelity and accuracy with which the proposed stopping rule recovers the input image. Second, the robustness of the method is compared with the other classical stopping rules and its advantages are shown in the case of real data recorded by RHESSI during two different flaring events., Comment: 20 pages

In order to study the acceleration and propagation of bremsstrahlung-producing electrons in solar flares, we analyze the evolution of the flare loop size with respect to energy at a variety of times. A GOES M3.7 loop-structured flare starting around 23:55 on 2002 April 14 is studied in detail using \textit{Ramaty High Energy Solar Spectroscopic Imager} (\textit{RHESSI}) observations. We construct photon and mean-electron-flux maps in 2-keV energy bins by processing observationally-deduced photon and electron visibilities, respectively, through several image-processing methods: a visibility-based forward-fit (FWD) algorithm, a maximum entropy (MEM) procedure and the uv-smooth (UVS) approach. We estimate the sizes of elongated flares (i.e., the length and width of flaring loops) by calculating the second normalized moments of the intensity in any given map. Employing a collisional model with an extended acceleration region, we fit the loop lengths as a function of energy in both the photon and electron domains. The resulting fitting parameters allow us to estimate the extent of the acceleration region which is between $\sim 13 \rm{arcsec}$ and $\sim 19 \rm{arcsec}$. Both forward-fit and uv-smooth algorithms provide substantially similar results with a systematically better fit in the electron domain.The consistency of the estimates from these methods provides strong support that the model can reliably determine geometric parameters of the acceleration region. The acceleration region is estimated to be a substantial fraction ($\sim 1/2$) of the loop extent, indicating that this dense flaring loop incorporates both acceleration and transport of electrons, with concurrent thick-target bremsstrahlung emission., Comment: 8 pages, 5 figures, accepted to Astronomy and Astrophysics journal