Magara, T., Shibata, K., and Yokoyama, T., "Evolution of Eruptive Flares. I. Plasmoid Dynamics in Eruptive Flares," ApJ 487, 437 (1997).
Click on the image above for a larger .gif version.
This cartoon represents a high level of refinement in the traditions of this school. It is important here because it clearly shows the "dimple" pushed into the loops envisioned to receive the impact of the outflow jet's ram pressure. This was originally found via less-elaborate simulations in the 1980s, as I recall, but I cannot remember where. Of course this feature of the simulations fails to predict the observations e.g. of the unusual Masuda flare, for which the feature associated with the termination of the reconnection outflow lies above the loops, rather than buried in them. Another point is the "long-lived strong soft X-ray source" called out in the right panel. This is of course an artifact of the 2D simulation. In 3D the overpressure implied by strong soft X-ray emission would result in immediate flows along the loops to get pressure equilibrium. Thus the high-density region should rapidly disappear. The cartoon also does not seem reasonable in terms of loop shapes: the "gradual phase" loops are shown as being less dipolar than the "impulsive phase" loops, which is inconsistent with the expectation of dipolarization. Finally, the hard X-ray footpoints are incorrectly placed. The excitation of the outer ribbon edges was always (and probably still is) the best evidence for the large-scale reconnection picture, so it is a shame to get this bit wrong!
10 March 2008