Limits of solar flare forecasting models and new deep learning approach*

Reliable forecasting models are necessary to mitigate the risks posed by solar flares to human technology. This study introduces a novel deep learning forecasting approach while emphasizing the need for performance evaluation methods tailored to better highlight current models' limitations. In particular, we show that models reaching state-of-the-art performance with traditional metrics have similar explanatory power to no-skill persistence models and notably struggle to forecast change in activity significantly better than random guesses. We also discuss shortcomings in traditional evaluation metrics like the True Skill Statistic (TSS), which we show to be mathematically dependent on the class balance for specific models. We introduce patch-distributed CNNs, which allow us to perform full-disk forecasts while providing event probabilities in solar subregions and position predictions. This new framework offers similar information to active region (AR)-based forecasting models while bypassing the problem of unrecorded and misattributed flares that are detrimental to machine learning training. As a result, the model also operates independently of prior feature extraction and AR detection, thus offering promising operational utility with minimal external dependencies. Finally, a method is proposed for constructing balanced and independent cross-validation folds for full-disk models. Models combining Solar Dynamic Observatory (SDO)/Atmospheric Imaging Assembly EUV images as inputs show improved performance compared to employing SDO/HMI photospheric magnetograms, with a TSS of 0.74 for the C+ model and 0.62 for the M+ model.