Authors: Talwinder Singh (Department of Physics & Astronomy, Georgia State University, Atlanta, GA 30303, USA), Timothy S. Newman (Department of Computer Science, University of Alabama in Huntsville), Luke Farris (Department of Computer Science, University of Alabama in Huntsville), Nikolai Pogorelov (Center for Space Plasma and Aeronomic Research, University of Alabama in Huntsville; Department of Space Science, University of Alabama in Huntsville), Prajun Trital (Department of Computer Science, University of Alabama in Huntsville)

We investigate whether incorporating the time evolution of magnetic properties of active regions improves solar flare forecasting beyond models that use only single-time magnetic snapshots. Using the SDO/HMI SHARP data set, we compare baseline classifiers trained on 20 single-time SHARP-based parameters with augmented classifiers that additionally include temporal-difference features (“deltas”), defined as Delta = SHARP(T_lead) – SHARP(T_older), constructed over a range of pre-flare intervals. We evaluate three representative machine-learning models: a random forest classifier (RFC), quadratic discriminant analysis (QDA), and a support vector machine (SVM), for binary prediction of major (M/X) versus non-major (B/C and no-flare) events using leave-one-out cross-validation. Forecast skills are assessed with the True Skill Statistic (TSS) and Heidke Skill Score (HSS2). We find that adding temporal-difference features improves performance for all three models, with the most consistent gains occurring when the endpoint of the delta construction is 23-24 hr before flare onset. The improvement is stronger and more robust for RFC and QDA, and more variable for SVM. TSS and HSS2 show consistent behavior, indicating that the benefit is not metric-dependent. These results show that a compact representation of SHARP-parameter evolution provides useful predictive information beyond single-time-point features, and they identify an approximately 1-day preflare timescale as particularly informative for major-flare forecasting.