A set of parameters that characterize the complexity and energy potential of solar active-regions is fed through several Machine Learning and conventional statistics algorithms to forecast solar flares.
A deep-learning method, Convolutional Neural Network, is developed to use the HMI’s line-of-sight magnetic field to forecast solar flares.
AR12192, the largest active region in Solar Cycle 24, produced 6 X-class flares, but none of them were associated with a CME. However, a much weaker flare, of M4.0-class, was associated with a CME. Magnetic field and morphological changes are analyzed during these flares to understand why this is the case.
44 strong flares are analyzed, and a few factors are identified to determine whether a flare will be eruptive or confined.
Magnetic field changes associated with solar flares, observed by the SDO/HMI, are surveyed, and permanent changes of magnetic field are found in the majority of flare events. Properties of the magnetic field changes are further investigated.
Statistical studies find that white-light flares from the Sun and from solar-type stars have similar energy-duration relations, but the stellar flares have shorter duration. Cooling effect and stronger magnetic field in the stellar corona are proposed to explain this difference.
An X9.3 flare excited strong yet unusual sunquakes.
We attempt to forecast M- and X-class solar flares using a support vector machine algorithm and 4 years of HMI vector magnetic field data. With the true skill statistic as the preferred metric to estimate the algorithm performances, we obtain good predictive abilities. This may be partly due to fine-tuning the algorithm for this purpose, and to a choice of 13 SHARP parameters obtained from vector magnetograms.
Reports of white-light ejecta above the limb of the Sun and imaged without the aid of a true coronagraph are exceedingly rare. Here we report the successful use of HMI to observe flare effects in the corona by the use of differencing as a substitute for an actual occulter.