Machine learning reveals systematic accumulation of electric current in lead-up to solar flares

Significance Reliable flare forecasting is essential for improving preparedness for severe space weather consequences. Flares also serve as probes of solar magnetic processes and the emergence of flux at the solar surface. Training machine-learning (ML) algorithms using magnetic-field observations for improving flare forecasting has been extensively studied in prior literature. Instead, here we use ML to understand the underlying mechanisms governing flares. We train ML algorithms to classify flaring and nonflaring active regions (ARs) with high fidelity and report statistical trends for AR evolution days before and after M- and X-class flares. These trends are interpreted in terms of existing models of subsurface magnetic field and flux emergence. Our results also provide hypotheses for achieving reliable flare forecasting. Solar flares—bursts of high-energy radiation responsible for severe space weather effects—are a consequence of the occasional destabilization of magnetic fields rooted in active regions (ARs). The complexity of AR evolution is a barrier to a comprehensive understanding of flaring processes and accurate prediction. Although machine learning (ML) has been used to improve flare predictions, the potential for revealing precursors and associated physics has been underexploited. Here, we train ML algorithms to classify between vector–magnetic-field observations from flaring ARs, producing at least one M-/X-class flare, and nonflaring ARs. Analysis of magnetic-field observations accurately classified by the machine presents statistical evidence for (i) ARs persisting in flare-productive states—characterized by AR area—for days, before and after M- and X-class flare events; (ii) systematic preflare buildup of free energy in the form of electric currents, suggesting that the associated subsurface magnetic field is twisted; and (iii) intensification of Maxwell stresses in the corona above newly emerging ARs, days before first flares. These results provide insights into flare physics and improving flare forecasting.