Solar Magnetic Data Analysis for the FLARECAST Project

The Flare Likelihood And Region Eruption foreCASTing (FLARECAST) project is an EU H2020-funded consortium project aiming to develop an advanced solar flare forecasting system by implementing state-of-the-art solar data analysis and flare prediction algorithms. The Solar Physics Group at Trinity College Dublin is in charge of the analysis of observational data to extract solar active region properties that serve as input for the prediction algorithms. The calculated active region properties correspond to a non-exhaustive list of parameters that have demonstrated a strong flare association, such as Schrijver's R-value, the Fourier power spectrum exponent, the effective connected magnetic field (Beff), the horizontal field decay index, and the weighted length of strong-gradient polarity inversion lines (WLSG). Parameters were calculated from Spaceweather HMI Active Region Patch (SHARP) magnetograms, a data product of the Helioseismic and Magnetic Imager (HMI) magnetograph on the Solar Dynamics Observatory (SDO). SHARPs provide photospheric vector-magnetic field (B) images in near-realtime. For this study, results from a statistical study performed on a robust subsample of the entire SHARP dataset will be presented. In the framework of the FLARECAST predictor component, this study focuses, for the first time, on differences between parameter values found when the radial magnetic field component, Br, is used instead of the line-of-sight component, Blos. The effect of active region longitudinal position is discussed, as well as the flare association of the properties.