Using the HMI/SDO vector magnetic field observations, we studied the relation of degree of magnetic nonpotentiality with the observed flare/CME in active regions. From a sample of 77 flare/CME cases, we found a general relation that degree of non-potentiality is positively correlated with the flare strength and the associated CME speeds. Since the magnetic flux in the flare-ribbon area is more related to the reconnection, we trace the strong gradient polarity inversion line (SGPIL), Schrijver's R value manually along the flare-ribbon extent. Manually detected SGPIL length and R values show higher correlation with the flare strength and CME speed than the automatically traced values without flare-ribbon information. It highlights the difficulty of predicting the flare strength and CME speed a priori from the pre-flare magnetograms used in flare prediction models. Although the total, potential magnetic energy proxies show weak positive correlation, the decrease in free energy exhibits higher correlation (0.56) with the flare strength and CME speed. Moreover, the eruptive flares have threshold of SGPIL length (31Mm), R value (1.6 × 10 19 Mx), free-energy decrease (2 × 10 31 erg) compared to confined ones. In 90% eruptive flares, the decay-index curve is steeper reaching n crit = 1.5 within 42Mm, whereas it is beyond 42Mm in > 70% confined flares. While indicating the improved statistics in the predictive capability of the AR eruptive behaviour with the flare-ribbon information, our study provides threshold magnetic properties for a flare to be eruptive.

The background field is assumed to play prime role in the erupting structures like prominences. In the flux rope models, the critical decay index (n c ) is a measure of the rate at which background field intensity decreases with height over the flux rope or erupting structure. In the real observations, the critical height of the background field is unknown, so a typical value of n c = 1.5 is adopted from the numerical studies. In this study, we determined the n c of 10 prominence eruptions (PEs). The prominence height in 3D is derived from two-perspective observations of Solar Dynamics Observatory and Solar TErrestrial RElations Observatory. Synoptic maps of photospheric radial magnetic field are used to construct the background field in the corona. During the eruption, the height-time curve of the sample events exhibits the slow and fast-rise phases and is fitted with the linear-cum-exponential model. From this model, the onset height of fast-rise motion is determined and is considered as the critical height for the onset of the torus-instability because the erupting structure is allowed to expand exponentially provided there is no strapping background field. Corresponding to the critical height, the n c values of our sample events are varied to be in the range of 0.8-1.3. Additionally, the kinematic analysis suggests that the acceleration of PEs associated with flares are significantly enhanced compared to flare-less PEs. We found that the flare magnetic reconnection is the dominant contributor than the torus-instability to the acceleration process during the fast-rise phase of flareassociated PEs in low corona (< 1.3R ).

Using observations from Solar Dynamics Observatory, we studied an interesting example of a sigmoid formation and eruption from small-scale flux canceling regions of active region (AR) 11942. Analysis of HMI and AIA observations infer that initially the AR is compact and bipolar in nature, evolved to sheared configuration consisting of inverse J-shaped loops hosting a filament channel over a couple of days. By tracking the photospheric magnetic features, shearing and converging motions are observed to play a prime role in the development of S-shaped loops and further flux cancellation leads to tethercutting reconnection of J-loops. This phase is co-temporal with the filament rise motion followed by sigmoid eruption at 21:32 UT on January 6. The flux rope rises in phases of slow (v avg = 26 km s −1 ) and fast (a avg = 55 ms −2 ) rise motion categorizing the CME as slow with an associated weak C1.0 class X-ray flare. The flare ribbon separation velocity peaks at around peak time of the flare at which maximum reconnection rate (2.14 Vcm −1 ) occurs. Further, the EUV light-curves of 131, 171Å have delayed peaks of 130 minutes compared to 94Å and is explained by differential emission measure. Our analysis suggests that the energy release is proceeded in a much long time duration, manifesting the onset of filament rise and eventual eruption driven by converging and canceling flux in the photosphere. Unlike strong eruption events, the observed slow CME and weak flare are indications of slow runway tether-cutting reconnection where most of the sheared arcade is relaxed during the extended post phase of the eruption.

Using the observations from the Solar Dynamics Observatory, we study an eruption of a hot-channel flux rope (FR) near the solar limb on 2015 February 9. The pre-eruptive structure is visible mainly in EUV 131 Å images, with two highly sheared loop structures. They undergo a slow rising motion and then reconnect to form an eruptive hot channel, as in the tether-cutting reconnection model. The J-shaped flare ribbons trace the footpoint of the FR that is identified as the hot channel. Initially, the hot channel is observed to rise slowly at 40 km s−1, followed by an exponential rise from 22:55 UT at a coronal height of 87 ± 2 Mm. Following the onset of the eruption at 23:00 UT, the flare reconnection then adds to the acceleration process of the coronal mass ejection (CME) within 3 R ⊙. Later on, the CME continues to accelerate at 8 m s−2 during its propagation period. Further, the eruption also launched type II radio bursts, which were followed by type III and type IVm radio bursts. The start and end times of the type IVm burst correspond to the CME’s core height of 1.5 and 6.1 R ⊙, respectively. Also, the spectral index is negative, suggesting that nonthermal electrons are trapped in the closed loop structure. Accompanied by this type IVm burst, this event is unique in the sense that the flare ribbons are very clearly observed together with the erupting hot channel, which strongly suggests that the hooked parts of the J-shaped flare ribbons outline the boundary of the erupting FR.

The abrupt and permanent changes of the photospheric magnetic field in the localized regions of active regions during solar flares, called magnetic imprints (MIs), have been observed for nearly the past three decades. The well-known coronal implosion model is assumed to explain such flare-associated changes but the complete physical understanding is still missing and debatable. In this study, we made a systematic analysis of flare-related changes of the photospheric magnetic field during 21 flares (14 eruptive and seven noneruptive) using the 135 s cadence vector magnetogram data obtained from the Helioseismic and Magnetic Imager. The MI regions for eruptive flares are found to be strongly localized, whereas the majority of noneruptive events (>70%) have scattered imprint regions. To quantify the strength of the MIs, we derived the integrated change of horizontal field and the total change of Lorentz force over an area. These quantities correlate well with the flare strength, irrespective of whether flares are eruptive or not, or have a short or long duration. Further, the free energy (FE), determined from virial theorem estimates, exhibits a statistically significant downward trend that starts around the flare time and is observed in the majority of flares. The change of FE during flares does not depend on eruptivity but has a strong positive correlation (≈0.8) with the Lorentz force change, indicating that part of the FE released would penetrate the photosphere. While these results strongly favor the idea of significant feedback from the corona on the photospheric magnetic field, the characteristics of MIs are quite indistinguishable from flares being eruptive or not.