Search

We analyze 152 large confined flares (GOES class $\geq$M1.0 and $\leq$$45^{\circ}$ from disk center) during 2010$-$2019, and classify them into two types according to the criterion taken from the work of Li et al. (2019). "Type I" flares are characterized by slipping motions of flare loops and ribbons and a stable filament underlying the flare loops. "Type II" flares are associated with the failed eruptions of the filaments, which can be explained by the classical 2D flare model. A total of 59 flares are "Type I" flares (about 40\%) and 93 events are "Type II" flares (about 60\%). There are significant differences in distributions of the total unsigned magnetic flux ($\Phi$$_\mathrm{AR}$) of active regions (ARs) producing the two types of confined flares, with "Type I" confined flares from ARs with a larger $\Phi$$_{AR}$ than "Type II". We calculate the mean shear angle $\Psi$$_\mathrm{HFED}$ within the core of an AR prior to the flare onset, and find that it is slightly smaller for "Type I" flares than that for "Type II" events. The relative non-potentiality parameter $\Psi$$_\mathrm{HFED}$/$\Phi$$_\mathrm{AR}$ has the best performance in distinguishing the two types of flares. About 73\% of "Type I" confined flares have $\Psi$$_\mathrm{HFED}$/$\Phi$$_\mathrm{AR}$$<$1.0$\times$$10^{-21}$ degree Mx$^{-1}$, and about 66\% of "Type II" confined events have $\Psi$$_\mathrm{HFED}$/$\Phi$$_\mathrm{AR}$$\geq$1.0$\times$$10^{-21}$ degree Mx$^{-1}$. We suggest that "Type I" confined flares cannot be explained by the standard flare model in 2D/3D, and the occurrence of multiple slipping magnetic reconnections within the complex magnetic systems probably leads to the observed flare., Comment: 11 pages, 8 figures