Real-time Flare Prediction Based on Distinctions between Flaring and Non-flaring Active Region Spectra

Abstract

With machine learning entering into the awareness of the heliophysics community, solar flare prediction has become a topic of increased interest. Although machine learning models have advanced with each successive publication, the input data has remained largely fixed on magnetic features. Despite this increased model complexity, results seem to indicate that photospheric magnetic field data alone may not be a wholly sufficient source of data for flare prediction. For the first time we have extended the study of flare prediction to spectral data. In this work, we use Deep Neural Networks to monitor the changes of several features derived from the strong resonant Mg II h&k lines observed by IRIS. The features in descending order of predictive capability are: The triplet emission at 2798.77 Å, line core intensity, total continuum emission between the h&k line cores, the k/h ratio, line-width, followed by several other line features such as asymmetry and line center. Regions that are about to flare generate spectra which are distinguishable from non-flaring active region spectra. Our algorithm can correctly identify pre-flare spectra approximately 35 minutes before the start of the flare, with an AUC of 86 % and an accuracy, precision and recall of 80 %. The accuracy and AUC monotonically increases to 90 % and 97 % respectively as we move closer in time to the start of the flare. Our study indicates that spectral data alone can lead to good predictive models and should be considered as an additional source of information alongside photospheric magnetograms.

A New View of the Solar Interface Region from the Interface Region Imaging Spectrograph (IRIS)

“…Zhu et al, 2019;Graham et al, 2020), as well as new approaches to the analysis of the observations based on statistical studies and machine-learning techniques (e.g. Panos et al, 2018;Panos and Kleint, 2020;Sadykov et al, 2019). With the Sun going towards a long period of increased activity, many new coordinated observations of flares with existing ground-based telescopes (including SST, ALMA, BBSO, GREGOR) and space-based instruments (e.g.…”

“…Most attempts at flare forecasting prior to IRIS primarily used the line-of-sight magneticfield data (see, e.g., Barnes et al, 2016), but recent IRIS results show exciting promise for accurate predictions of the timing of the onset of solar flares using UV spectroscopic data. Panos and Kleint (2020) applied machine-learning techniques to show how chromospheric spectra (Mg II h/k and, in particular, triplet emission) in pre-flare regions (40 minutes before flare) are clearly distinct, and could, in principle, be used to predict flare occurrence. Woods et al (2017) studied the pre-flare phase of an X1 flare observed by IRIS and found strongly blue-shifted plasma flows in the IRIS Si IV line with velocities up to 200 km s −1 40 minutes before the eruption.…”

Exploring mutual information between IRIS spectral lines. I. Correlations between spectral lines during solar flares and within the quiet Sun

“…An alternative and, in our opinion, more forgiving approach is to isolate flaring regions based on spectral shape. Panos & Kleint (2020) showed that Mg II h&k line profiles associated with quiet Sun and flaring atmospheres could easily be distinguished using neural networks (NNs). Although their models incorporated an intensity component, features such as k/h ratio and subordinate line emission appeared to be strongly descriptive of the two classes.…”

Operational solar flare prediction model using Deep Flare Net

“…Recently, the application of supervised machine learning methods, especially deep neural networks (DNNs), to solar flare prediction has been a hot topic, and their successful application in research has been reported (Huang et al 2018;Nishizuka et al 2018;Park et al 2018;Chen et al 2019;Domijan et al 2019;Liu et al 2019;Zheng et al 2019;Bhattacharjee et al 2020;Jiao et al 2020;Li et al 2020;Panos & Kleint 2020;Yi et al 2020). However, there is insufficient discussion on how to develop the methods available to real-time operations in space weather forecasting offices, including the methods for validation and verification of the models.…”