Low-dimensional Convolutional Neural Network for Solar Flares GOES Time-series Classification

Space weather phenomena such as solar flares have a massive destructive power when they reach a certain magnitude. Here, we explore the deep-learning approach in order to build a solar flare-forecasting model, while examining its limitations and feature-extraction ability based on the available Geostationary Operational Environmental Satellite (GOES) X-ray time-series data. We present a multilayer 1D convolutional neural network to forecast the solar flare event probability occurrence of M- and X-class flares at 1, 3, 6, 12, 24, 48, 72, and 96 hr time frames. The forecasting models were trained and evaluated in two different scenarios: (1) random selection and (2) chronological selection, which were compared afterward in terms of common score metrics. Additionally, we also compared our results to state-of-the-art flare-forecasting models. The results indicates that (1) when X-ray time-series data are used alone, the suggested model achieves higher score results for X-class flares and similar scores for M-class as in previous studies. (2) The two different scenarios obtain opposite results for the X- and M-class flares. (3) The suggested model combined with solely X-ray time-series fails to distinguish between M- and X-class magnitude solar flare events. Furthermore, based on the suggested method, the achieved scores, obtained solely from X-ray time-series measurements, indicate that substantial information regarding the solar activity and physical processes are encapsulated in the data, and augmenting additional data sets, both spatial and temporal, may lead to better predictions, while gaining a comprehensive physical interpretation regarding solar activity. All source codes are available at https://github.com/vladlanda.

Observational Signatures of Tearing Instability in the Current Sheet of a Solar Flare

Magnetic reconnection is a fundamental physical process converting magnetic energy into not only plasma energy but also particle energy in various astrophysical phenomena. In this Letter, we show a unique data set of a solar flare where various plasmoids were formed by a continually stretched current sheet. Extreme ultraviolet images captured reconnection inflows, outflows, and particularly the recurring plasma blobs (plasmoids). X-ray images reveal nonthermal emission sources at the lower end of the current sheet, presumably as large plasmoids with a sufficiently amount of energetic electrons trapped in them. In the radio domain, an upward, slowly drifting pulsation structure, followed by a rare pair of oppositely drifting structures, was observed. These structures are supposed to map the evolution of the primary and the secondary plasmoids formed in the current sheet. Our results on plasmoids at different locations and scales shed important light on the dynamics, plasma heating, particle acceleration, and transport processes in the turbulent current sheet and provide observational evidence for the cascading magnetic reconnection process.

A Study of Solar Flare Effects on the Geomagnetic Field Components during Solar Cycles 23 and 24

In this paper, we investigated the impact of solar flares on the horizontal (H), eastward (Y) and vertical (Z) components of the geomagnetic field during solar cycles 23 and 24 (SC23/24) using data of magnetometer measurements on the sunlit side of the Earth. We examined the relation between sunspot number and solar flare occurrence of various classes during both cycles. During SC23/24, we obtained correlation coefficient of 0.93/0.97, 0.96/0.96 and 0.60/0.56 for C-class, M-class and X-class flare, respectively. The three components of the geomagnetic field reached a peak a few minutes after the solar flare occurrence. Generally, the magnetic crochet of the H component was negative between the mid-latitudes and Low-latitudes in both hemispheres and positive at low latitudes. By contrast, the analysis of the latitudinal variation of the Y and Z components showed that unlike the H component, their patterns of variations were not coherent in latitude. The peak amplitude of solar flare effect (sfe) on the various geomagnetic components depended on many factors including the local time at the observing station, the solar zenith angle, the position of the station with respect to the magnetic equator, the position of solar flare on the sun and the intensity of the flare. Thus, these peaks were stronger for the stations around the magnetic equator and very low when the geomagnetic field components were close to their nighttime values. Both cycles presented similar monthly variations with the highest sfe value (ΔHsfe = 48.82 nT for cycle 23 and ΔHsfe = 24.68 nT for cycle 24) registered in September and lowest in June for cycle 23 (ΔHsfe = 8.69 nT) and July for cycle 24 (ΔHsfe = 10.69 nT). Furthermore, the sfe was generally higher in cycle 23 than in cycle 24.

Geomagnetic Pulsations, Blackout and Low-Latitude Aurorae Associated with Power Magnetic Storm

The research analyzed the regularities of the dynamics of geomagnetic pulsation regimes in the frequency range 0.002–5 Hz, the generation of which reflects the interaction with the Earth's magnetosphere of the solar filament ejected by a powerful solar flare of 3B. We compared the dynamics of the change in the types and modes of geomagnetic pulsations with the dynamics of the atmosphere glow in two spectral lines and the total ionospheric absorption of radio waves. The study developed a possible model of the observed phenomenon.

Fine-grained Solar Flare Forecasting Based on the Hybrid Convolutional Neural Networks*

Improving the performance of solar flare forecasting is a hot topic in the solar physics research field. Deep learning has been considered a promising approach to perform solar flare forecasting in recent years. We first used the generative adversarial networks (GAN) technique augmenting sample data to balance samples with different flare classes. We then proposed a hybrid convolutional neural network (CNN) model (M) for forecasting flare eruption in a solar cycle. Based on this model, we further investigated the effects of the rising and declining phases for flare forecasting. Two CNN models, i.e., M rp and M dp, were presented to forecast solar flare eruptions in the rising phase and declining phase of solar cycle 24, respectively. A series of testing results proved the following. (1) Sample balance is critical for the stability of the CNN model. The augmented data generated by GAN effectively improved the stability of the forecast model. (2) For C-class, M-class, and X-class flare forecasting using Solar Dynamics Observatory line-of-sight magnetograms, the means of the true skill statistics (TSS) scores of M are 0.646, 0.653, and 0.762, which improved by 20.1%, 22.3%, and 38.0% compared with previous studies. (3) It is valuable to separately model the flare forecasts in the rising and declining phases of a solar cycle. Compared with model M, the means of the TSS scores for No-flare, C-class, M-class, and X-class flare forecasting of the M rp improved by 5.9%, 9.4%, 17.9%, and 13.1%, and those of the M dp improved by 1.5%, 2.6%, 11.5%, and 12.2%.