Flare forecasting based on sunspot-groups characteristics

Our comprehension of solar flares is still lacking in many aspects and the possibility of observing active regions during the first phases of flare occurrence is limited by our capability of doing accurate flare forecasting. In order to give a contribution to this aspect, we focused our attention on the characteristics that must be fulfilled by sunspot-groups in order to be flare-productive. We addressed this problem using a statistical approach: first, we analyzed sunspot-groups parameters (i.e., Zürich class, magnetic configuration, area, morphology of the penumbra) and evolution; then, we performed a flare forecasting campaign, based on the results obtained in the first phase and on real-time observations. The results obtained by comparing the flare forecasting probability with the number of flares that have actually occurred are quite encouraging; we plan to improve this procedure by including a bigger statistical sampling.