Current sheets are believed to form in the wakes of erupting flux ropes and to enable the magnetic reconnection responsible for an associated flare. Multiwavelength observations of an eruption on 2017 September 10 show a long, linear feature widely taken as evidence of a current sheet viewed edge-on. The relation between the hightemperature, high-density plasma thus observed and any current sheet is not yet entirely clear. We estimate the magnetic field strength surrounding the sheet and conclude that approximately one-third of all flux in the active region was opened by the eruption. Subsequently decreasing field strength suggests that the open flux closed down over the next several hours through reconnection at a rate F 5 10 17 Mx s −1. We find in AIA observations evidence of downward-moving, dark structures analogous to either supra-arcade downflows, more typically observed above flare arcades viewed face-on, or supra-arcade downflowing loops, previously reported in flares viewed in this perspective. These features suggest that the plasma sheet is composed of the magnetic flux retracting after being reconnected high above the arcade. We model flux tube retraction following reconnection to show that this process can generate high densities and temperatures as observed in the plasma sheet. The retracting flux tubes reach their highest temperatures at the end of their retraction, well below the site of reconnection, consistent with previous analysis of AIA and EIS data showing a peak in the plasma temperature near the base of this particular sheet.

In accepted models, magnetic tension drives reconnected magnetic flux away from the reconnection site at the local Alfvén speed. Numerous observational signatures of these outflows have been identified in solar flares, notable among them being supra-arcade downflows (SADs), almost none move at the Alfvén speed as predicted by models. Well-studied examples of SADs or SAD loops found in the flare of 2017 September 10 (SOL2017-09-10T15:35:00) move at a quarter or less of the expected Alfvén speed. Among those reasons posited to explain such discrepancies is the possibility that reconnected flux experiences a drag force during its outflow. Drag has not been included in previous reconnection models. Here, we develop the first such model in order to test the possibility that drag can explain sub-alfveńic reconnection outflows. Our model uses thin flux tube dynamics, previously shown to match features of flare observations other than outflow speed, including for the 2017 September 10 flare. We supplement the dynamics with a drag force representing the tube’s interaction with surrounding plasma through the formation of a wake. The wake’s width appears as a parameter in the force. We perform simulations, varying the drag parameter and synthesizing EUV observations, to test whether a drag force can produce a reasonable fit to observed features of the September 10 flare. We find that that slower retraction increases the brightness of emission and lowers the temperature of the synthetic plasma sheet. With proper choice of parameters the drag enables the simulation to agree reasonably with the observations.

Flare reconnection is expected to occur at some point within a large-scale coronal current sheet. The structure of the magnetic field outside this sheet is almost certain to affect the flare, especially its energy release. Different models for reconnection have invoked different structures for the current sheet’s magnetic field and different locations for the reconnection electric field within it. Models invoking Petschek-type reconnection often use a uniform field. Others invoke a field bounded by two Y-points with a field strength maximum between them and propose this maximum as the site of the reconnection electric field. Still other models, such as the collapsing trap model, require that the field strength peak at or near the edge of the current sheet and propose that reconnection occurs above this peak. At present there is no agreement as to where reconnection might occur within a global current sheet. We study the post-reconnection dynamics under all these scenarios, seeking potentially observable differences between them. We find that reconnection occurring above the point of strongest field leads to the highest density and the highest emission measure of the hottest material. This scenario offers a possible explanation of superhot coronal sources seen in some flares.

Spectroscopic observations of flare ribbons typically show chromospheric evaporation flows, which are subsonic for their high temperatures. This contrasts with many numerical simulations where evaporation is typically supersonic. These simulations typically assume flow along a flux tube with a uniform cross-sectional area. A simple model of the magnetic canopy, however, includes many regions of low magnetic field strength, where flux tubes achieve local maxima in their cross-sectional area. These are analgous to a chamber in a flow tube. We find that one-third of all field lines in a model have some form of chamber through which evaporation flow must pass. Using a one-dimensional isothermal hydrodynamic code, we simulated supersonic flow through an assortment of chambers and found that a subset of solutions exhibit a stationary standing shock within the chamber. These shocked solutions have slower and denser upflows than a flow through a uniform tube would. We use our solution to construct synthetic spectral lines and find that the shocked solutions show higher emission and lower Doppler shifts. When these synthetic lines are combined into an ensemble representing a single canopy cell, the composite line appears slower, even subsonic, than expected due to the outsized contribution from shocked solutions.