Toward the Physics-Based Prediction of Solar Flares Using the Triggered Feedback Instability Model

Abstract:

Solar flares are believed to be the explosive liberation of magnetic energy in the solar corona and may cause space weather disturbances. However, the physical condition for their onset is not yet well understood, and thus our predictability of when, where, and how large events will occur is not sufficiently reliable yet. Which kind of instability determines the critical condition for the onset of solar eruptions is a key question of this problem, because magnetohydrodynamic (MHD) instabilities must play an important role in driving solar eruptions. Although the torus instability (TI) and the kink-mode instability are candidates of the key driver of solar eruptions, the recent observation suggested that those instabilities could not explain the onset condition of solar flares (Jing et al. 2018).

On the other hand, Ishiguro and Kusano (2017) proposed that a new instability called double-arc instability (DAI) may work as the initial driver of solar flares and it can trigger the onset of solar eruptions. Though the DAI is one of the hoop-force driven instabilities like the TI, the critical condition of the DAI, given by a new parameter kappa, is different from the TI. The DAI can explain the triggered feedback-instability model which is proposed by Kusano et al. (2012) based on the ensemble simulation study.

In this paper, based on the triggered feedback-instability model, we have developed a flare prediction model which can evaluate the stability for the DAI and the capability of energy liberation within each active region using the nonlinear force-free field extrapolation. We analyzed the evolution of Active Region NOAA 11158 by applying this model to the Space-weather HMI Active Region Patches (SHARP). The results are well consistent with the flare activities of the active region, and they clearly show that the active region became less stable several hours before the onset of the X2.2 flare which occurred at 01:57 UT on Feb. 15, 2011. This model can also predict the location where a flare is most likely to occur and how much magnetic free energy can be released if a flare occurs at some location. The predicted results of location and the estimation of flare releasing energy are also consistent with the observations. Finally, we discuss the applicability of this model for the operational forecast of solar flares.