On The Origin, Structure, And Evolution Of Coronal Mass Ejections

Coronal mass ejections (CMEs) are large-scale eruptive events in the solar system and have severe impacts on the solar-terrestrial environment. Understanding the origin, structure, and evolution of CMEs is significantly important to predict space weather. In this thesis, we first study the origin and evolution of CME-related flux rope and its role in re-flaring of the post-flare loop systems (Chapter3). Second, we analyze the initia-tion mechanisms of three CMEs in detail, as well as the role of background magnetic field in determining whether a CME is a successful eruption (Chapter4). We fur-ther determine the temperature and density of different components of CMEs through differential emission measure (DEM) technology (Chapter5). Third, we analyze the impulsive acceleration process of three CMEs and study the post-impulsive-phase ac-celeration in the propagation phase (Chapter6). Finally, we disclose the physical nature of the global-scale EUV disturbances (EUV wave) driven by the early expansion of a CME (Chapter7). The main results are as follows.From the observations of the2011March8CME, we find that the magnetic flux rope, a volumetric current channel running over the polarity inversion line, exists prior to the CME onset. It initially appears as a twisted and writhed channel structure with a temperature as high as10MK, and then transforms toward a semi-circular shape dur-ing a slow-rise phase, which is followed by a fast acceleration and onset of a flare. By analyzing a limb eruption that occurred on2010November3, we further find that the flux rope can be enhanced during the impulsive acceleration phase. In the meantime, it rapidly moves outward and stretches the overlying magnetic field upward, forming a CME. Therefore, the CME consists of multi-temperature structures:the hot flux rope and the cool leading front (LF). In the wake of the CME eruption, magnetic reconnec-tion forms a set of new loop systems, named post-flare loops. We find that the post-flare loops can expand to form a second CME. Using nonlinear force-free field (NLFFF) ex-trapolation based on ground-based vector magnetic field data, we confirm that the rapid emergence and twisting of a flux rope system underneath the post-flare loops results in such a re-eruption.CMEs usually experience a slow rise phase before the impulsive acceleration phase. For the CME that occurred on2008April26, its host active region appeared as a simple bipolar magnetic field, and the whole magnetic field structure can be characterized by a sigmoidal core field constrained by an overlying bipolar arcade field. Moreover, some pre-eruption EUV brightenings occurred within only the core field contrary to the remote brightenings as expected from the breakout model. Therefore, we believe that the initial tether-cutting reconnection, as indicated by the EUV brightenings, causes the slow rise of the sigmoidal magnetic structure. For other two CMEs that took place on2011March7and8, respectively, magnetic reconnection is also likely the cause of the slow rise phase. However, the trigger of the impulsive acceleration phase is attributed to torus instability of the flux rope. In this respect, we further examine the distinct properties of confined flares (without CMEs) and eruptive flares (with CMEs). Through NLFFF extrapolation, we find that the decay index of the transverse magnetic field in the low corona (-10Mm) have a larger value for eruptive flares than that for confined ones. In addition, the strength of the transverse magnetic field over the eruptive flare sites is weaker than that over the confined ones. These results demonstrate that the strength and the decay index of background magnetic field may determine whether a flare is an eruptive or confined one.Through DEM analysis, we study the temperature and density properties of dif-ferent components of CMEs, which include the hot channel in the core region,(the magnetic flux rope), the bright loop-like LF, and the coronal dimming in the wake of the CME. We find that the flux rope has the highest average temperature (>8MK) and density (-1.0×109cm-3), resulting in an enhanced emission measure (EM) over a broad temperature range (3-20MK). On the other hand, the CME LF has a rela-tively cool temperature (-2MK) and a narrow temperature range (1-3MK), similar to the pre-eruption coronal temperature; the difference is that the density in the LF is in-creased by～2%-33%compared with the pre-eruption coronal value. For coronal dim-mings, the temperature varies in a broader range (1-4MK), but the density decreases by～35%-40%. These observational results show that:(1) the CME core regions are significantly heated, presumably through magnetic reconnection;(2) the CME LFs are a consequence of compression of the ambient plasma by the expansion of the CME core region; and (3) the dimmings are largely caused by the plasma rarefaction associated with the eruption.Through detailed kinematic analysis, we find that, during the impulsive acceleration phase of CMEs, the time profile of the CME acceleration in the inner corona is con- sistent with the time profile of the reconnection electric field and the RHESSI15-25keV hard X-ray flux curve of the associated flare. It implies that CMEs and associated flares are two distinct aspects of the same physical process, the key of which is mag-netic reconnection taking place underneath the flux rope. We also examine in detail the post-impulsive-phase acceleration of CMEs in the propagation phase, immediately following the main impulsive acceleration phase. After a statistical study, we find that the CMEs associated with flares of a long-decay time tend to have a positive post-impulsive-phase acceleration, even though some of them have already obtained a high speed at the end of the impulsive acceleration. On the other hand, the CMEs associated with flares of a short-decay time tend to have a significant deceleration. In the scattering plot of all events, there is a weak correlation between the CME post-impulsive-phase acceleration and the flare decay time. Some CMEs, mostly slow or weak ones associ-ated with flares of a short-decay time, do not follow such a trend, owing to the relatively stronger solar wind dragging force on these events.Finally, we address the nature of EUV waves through direct observations of the formation of a diffuse wave driven by the expansion of a CME and its subsequent separation from the CME front. Following the eruption onset, the CME exhibits a strong lateral expansion. During this phase, the expansion speed of the CME bubble increases from100km s-1to450km s-1in only6min. An important finding is that a diffuse wave front starts to separate from the front of the expanding bubble shortly after the lateral expansion slows down. A type-Ⅱ burst is also formed nearly at this time. After the separation, two distinct fronts propagate with different kinematic properties. The diffuse front travels across the entire solar disk; while the sharp front rises up, forming the CME ejecta with the diffuse front ahead of it. These observations suggest that the previously termed EUV wave is a composite phenomenon and driven by the CME expansion. While the CME expansion is accelerating, the wave front is cospatial with the CME front; thus the two fronts are indiscernible. In the end of the acceleration phase, the wave moves away from the CME front with a gradually increasing distance between them.In summary, we have investigated various aspects of the CME origin, structure, and evolution, as well as the nature of the associated EUV waves. The results are helpful in improving our understanding of the physics mechanisms of solar eruptions and in predicting space weather.