Is torus instability crucial for solar coronal mass ejections?

Coronal mass ejections (CMEs) are the most violent eruptions in the solar atmosphere. What are the progenitors? How are the progenitors triggered to erupt? How are CMEs accelerated and how do they propagate in the interplanetary space? All these issues still remain elusive. In particular, a lot of efforts have been taken to the triggering mechanisms. From both theoretical and observational points of view, a CME progenitor can be triggered from a metastable state to eruption through either a resistive process, e.g., magnetic reconnection, or ideal magnetohydrodynamic (MHD) instabilities, e.g., kink and torus instabilities. About thirty years ago, it was once thought that CMEs can be due to an ideal MHD process, which opens up the initially closed magnetic loops, and the ensuing magnetic reconnection leads to a solar flare, which is dispensible for the CMEs. Considering the Aly-Sturrock constraint and the intimate correlation between CMEs and flares, it was then realized that magnetic reconnection may play a crucial role later. However, in the past fourteen years, there was an increasing trend to emphasize torus instability in CMEs, not only for their triggering, but also for the final eruption. On one hand, we believe that torus instability, as well as kink instability and magnetic reconnection, is one of the possible trigger mechanisms for CMEs; on the other hand, we are suspicious of some papers on how the torus instability was identified. Our main arguments are summarized as follows: (1) Torus instability is only one of the trigger mechanisms. Other mechanisms may work as well, and sometimes several mechanisms may be at work simultaneously. Therefore, it is not surprising at all that a CME erupts while the criterion of torus instability is not satisfied. (2) Torus instability can serve as a trigger mechanism, but presumably not a driven mechanism. Whether a CME can erupt or not depends on both external conditions, e.g., the background magnetic field, and internal conditions, e.g., the nonpotentiality of the core magnetic field in the source region and how fast the reconnection proceeds. Based on the Aly-Sturrock constraint, a line-tied flux rope can never erupt without reconnection or mass drainage even when the background magnetic field satisfies the criterion of torus instability everywhere from low corona all the way to the interplanetary space. When some papers claimed the key role of torus instability, magnetic reconnection was happening as well, which makes their conclusions unconvincing. So, it will be critical to distinguish resistive processes from ideal MHD instabilities in future studies. (3) When studying the torus instability, it is better to calculate the 3-dimensional distribution of the decay index of the background magnetic field. Checking the decay index along one direction is not sufficient. (4) When studying the torus instability, it might be meaningless to examine the decay index only at the apex of a flux rope. A local loss of equilibrium never means the catastrophe of a whole structure. We need to take in account the distribution of the decay index along the major part of the flux rope in the source region. © 2019, Science Press. All right reserved.