Abstract Solar flares are a primary driver of space weather, and forecasting their occurrence remains a significant challenge. This paper presents a novel flare prediction model based on topologically derived photospheric magnetic parameters. We employ the ARTop framework to compute the time-dependent input rates of magnetic winding and helicity across >10 5 active region observations, decomposing them into current-carrying and potential components to reduce sensitivity to optical flow methods. An XGBoost machine learning model is trained on these time series, alongside engineered features including rolling statistics, kurtosis, and flare history, to predict the probability of ≥M1.0 class flares within the next 24 hr. The model demonstrates strong performance on a validation set, with a true skill statistic (TSS) of 0.804 for once-daily operational region forecasts. When applied to a fully independent dataset, the model achieves a TSS of 0.524. A SHAP analysis confirms the model’s physical interpretability, identifying flare history and accumulated current-carrying winding and helicity as the most important features. The main challenges identified are false positives arising from active regions with frequent C-class flaring and systematic errors via projection effects when active regions are near the limb. Excluding limb-affected data yields no improvement in the holdout set TSS (0.521 versus 0.524), due to the overall decreased number of flares. However, our per-region analysis indicates that mitigating these projection effects is crucial for future operational deployment. This work establishes magnetic topology, particularly its current-carrying components, as a highly effective and physically meaningful set of predictors for solar flare forecasting.
Williams et al. (Wed,) studied this question.