We developed a new numerical code that is able to perform 2.5D simulations of a magnetohydrodynamic (MHD) wave propagation in the corona, and its interaction with a low density region, such as a coronal hole (CH). We show that the impact of the wave on the CH leads to different effects, such as reflection and transmission of the incoming wave, stationary features at the CH boundary, or formation of a density depletion. We present a comprehensive analysis of the morphology and kinematics of primary and secondary waves, i.e. we describe in detail the temporal evolution of density, magnetic field, plasma flow velocity, phase speed and position of the wave amplitude. Effects like reflection, refraction and transmisson of the wave strongly support the theory that large scale disturbances in the corona are fast MHD waves and build the major distinction to the competing pseudo-wave theory. The formation of stationary bright fronts was one of the main reasons for the development of pseudo-waves. Here we show that stationary bright fronts can be produced by the interactions of an MHD wave with a CH. We find secondary waves that are traversing through the CH and we show that one part of these traversing waves leaves the CH again, while another part is being reflected at the CH boundary inside the CH. We observe a density depletion that is moving in the opposite direction of the primary wave propagation. We show that the primary wave pushes the CH boundary to the right, caused by the wave front exerting dynamic pressure on the CH.

We report observations of an eruptive X2.8 flare on 2013 May 13, which shows two distinct episodes of energy release in the impulsive phase. The first episode is characterized by the eruption of a magnetic flux rope, similar to the energyrelease process in most standard eruptive flares. While the second episode, which is stronger than the first normal one and shows enhanced high-energy X-ray and even γ-ray emissions, is closely associated with magnetic reconnection of a largescale loop in the aftermath of the eruption. The reconnection inflow of the loop leg is observed in the Solar Dynamics Observatory (SDO)/Atmospheric Imaging Assembly (AIA) 304Å passband and accelerates towards the reconnection region to a speed as high as ∼130 km/s. Simultaneously the corresponding outflow jets are observed in the AIA hot passbands with a speed of ∼740 km/s and mean temperature of ∼14 MK. RHESSI observations show a strong burst of hard X-ray (HXR) and γ-ray emissions with hard electron spectra of δ ≈ 3, exhibiting a soft-hard-harder behavior. A distinct altitude decrease of the HXR loop-top source coincides with the inward swing of the loop leg observed in the AIA 304Å passband, which is suggested to be related to the coronal implosion. This fast inflow of magnetic flux contained in the loop leg greatly enhances the reconnection rate and results in very efficient particle acceleration in the secondstep reconnection, which also helps to achieve a second higher temperature peak up to T ≈ 30 MK.

We analyze a confined multiple-ribbon M2.1 flare (SOL2015-01-29T11:42) that originated from a fan-spine coronal magnetic field configuration, within active region NOAA 12268. The observed ribbons form in two steps. First, two primary ribbons form at the main flare site, followed by the formation of secondary ribbons at remote locations. We observe a number of plasma flows at extreme-ultraviolet temperatures during the early phase of the flare (as early as 15 min before the onset) propagating towards the formation site of the secondary ribbons. The secondary ribbon formation is co-temporal with the arrival of the pre-flare generated plasma flows. The primary ribbons are co-spatial with RHESSI hard X-ray sources, whereas no enhanced X-ray emission is detected at the secondary ribbons sites. The (E)UV emission, associated with the secondary ribbons, peaks ∼1 min after the last RHESSI hard X-ray enhancement. A nonlinear force-free model of the coronal magnetic field reveals that the secondary flare ribbons are not directly connected to the primary ribbons, but to regions nearby. Detailed analysis suggests that the secondary brightenings are produced due to dissipation of kinetic energy of the plasma flows (heating due to compression), and not due to non-thermal particles accelerated by magnetic reconnection, as is the case for the primary ribbons.

We report a detailed study of the pre-eruption activities that led to the occurrence of an M9.0 flare/CME event on 2012 October 20 in NOAA AR 11598. This includes the study of the preceding confined C2.4 flare that occurred on the same AR ∼25 minutes earlier. We observed that the M9.0 flare occurred as a consequence of two distinct triggering events well separated in time. The first triggering episode occurred as early as ∼20 minutes before the onset of the M9.0 flare, evidenced by the destabilization and rise of a pre-existing filament to a new position of equilibrium at a higher coronal altitude during the decay phase of the C2.4 flare. This brought the system to a magnetic configuration where the establishment of the second triggering event was favorable. The second triggering episode occurred ∼17 minutes later, during the early phase of the M9.0 flare, evidenced by the further rise of the filament and successful ejection. The second trigger is followed by a flare precursor phase, characterized by non-thermal emission and the sequential formation of a hot channel as shown by the SDO/AIA DEM (differential emission measure) maps, the RHESSI X-ray images and spectra. These observations are suggestive of magnetic reconnection and particle acceleration that can explain the precursor phase and can be directly related to the formation of the hot channel. We discuss on the triggering mechanisms, their implications during the early and precursor phases and highlight the importance of early activities and preceding small confined flares to understand the initiation of large eruptive flares.

We analyze a confined flare that developed a hot cusp-like structure high in the corona (H∼66 Mm). A growing cusp-shaped flare arcade is a typical feature in the standard model of eruptive flares, caused by magnetic reconnection at progressively larger coronal heights. In contrast, we observe a static hot cusp during a confined flare. Despite an initial vertical temperature distribution similar to that in eruptive flares, we observe a distinctly different evolution during the late (decay) phase, in the form of prolonged hot emission. The distinct cusp shape, rooted at locations of non-thermal precursor activity, was likely caused by a magnetic field arcade that kinked near the top. Our observations indicate that the prolonged heating was a result of slow local reconnection and an increased thermal pressure near the kinked apexes due to continuous plasma upflows.