Using Space-weather HMI Active Region Patch (SHARP) parameters to identify solar-flare precursors for space weather forecasting

Abstract

Abstract A solar flare results from the release of magnetic energy by magnetic-reconnection processes. It is preceded by the accumulation of energy, which is indicated by various magnetic parameters. An analysis of the magnetic parameters preceding the high energy release is therefore essential for understanding the occurrence of a solar flare. In this work, we studied 18 magnetic parameters from the Space-weather Helioseismic and Magnetic Imager (HMI) Active Region Patch (SHARP) database for X-class solar flares during the time span from May 10, 2010, to December 31, 2021, using a support vector machine. We found that the time-interval distribution between the maxima of the Space-weather HMI Active Region Patch (SHARP) parameters and solar flare events can be approximated by a Gaussian. About 67% of the maximum values of the SHARP indices occurred within a time window of 22.4 hours before a solar flare. This result strengthens the conclusion that the SHARP parameters can be employed to identify solar-flare precursors. In terms of Recall and True Skill Statistics, our support vector machine model performs better in forecasting solar flares than do previous results. By considering the significance of these 18 SHARP parameters, we found that seven parameters are the optimal number for further analysis.