Brian Harker and Alexei Pevtsov
National Solar Observatory, 950 N. Cherry Ave, Tucson, AZ, 85719 USA
Solar eruptions, produced by the release of stored magnetic energy in regions of complex magnetic field topology, are at the forefront of modern space weather studies. The behavior of these complex magnetic fields before, during, and after an eruption is a long-studied, yet still hotly-debated topic in the solar physics community. One phenomenon which has been brought back into the spotlight in recent years is that of transient magnetic field reversals: patches of magnetic field which seem to “flip” their polarity from positive to negative (or vice versa) in response to the initiation of an energetic solar flare.
In a recent paper1 published by the Astrophysical Journal, Harker and Pevtsov produce a detailed case study of one such transient magnetic reversal, observed in NOAA 11429 during the M7.9 flare on 13 March 2012. This transient represents the most limb-adjacent ever studied in detail, and as such it presents a unique perspective into the origin and behavior of these phenomena.
The authors utilize line-of-sight (LOS) magnetograms and Stokes profile information obtained by the Helioseismic and Magnetic Imager (HMI) instrument aboard the Solar Dynamics Observatory (SDO). SDO/HMI provides fantastic views of the photospheric magnetic field at very high spatial resolution, and on very fast time-scales. Using line-of-sight magnetograms and Dopplergrams, the authors observed a small kernel of transient polarity-reversed magnetic field near the umbra/penumbra boundary of NOAA 11429 on 13 March 2012 (see Figure 1), which was co-spatial with a white-light flare kernel seen in both line-core and continuum intensity images. Analysis of 135s cadence full Stokes vector data confirmed this reversal, very clearly showing Stokes V lobes reversing their order during the course of the flare. Figure 2 presents a more detailed temporal evolution of the transient in a pixel mosaic, showing LOS field strength observed by SDO/HMI as well as derived from a Milne-Eddington inversion.
Figure 1 |Temporal evolution of LOS flux in NOAA 11429, as measured by SDO/HMI. Note the appearance of a transient patch of oppositely-signed magnetic flux (black) emerging into the positive polarity center-side penumbra (white), shown in the red circle. The arrow points along the solar north direction, in the image plane.
An interesting result from this data was the lack of any signatures of line-core emission in the Stokes I profiles during the flare, which is traditionally seen in such magnetic transient regions. The authors then conclude that flare-emission cannot produce the observed polarity reversal seen in the magnetic transient. In the absence of any mechanism which could produce the observed reversal as a result of changes in the observed spectral line, the authors procede with several forward-modelling experiments, seeking to explain the reversal as a by-product of instrumental and/or algorithmic anomalies in the SDO/HMI observation and/or data reduction protocols. By simulating the integration time of a typical SDO/HMI line-of-sight observation, Harker and Pevtsov were able to rule out the presence of either gradual or impulsive changes in the plasma velocity during the observation window as the source of the magnetic transient. Furthermore, an investigation in the sinc-interpolation scheme used by SDO/HMI to reconstruct the Stokes profiles from individual filtergrams taken at slightly different times revealed no major “contamination” which could cause such a polarity reversal.
Figure 2 |Pixel mosaic showing the LOS flux in a 9 x 11 pixel region where the MT was observed, as a function of time. This subfield is centered on the image feature bounded by the red circle in Figure 1. The vertical, dotted lines represent flare onset time and peak X-ray flux time, respectively, and the horizontal, dashed line is at 0 G. The flaring region hosts a kernel of magnetic flux (panels outlined in red) which completely reverses sign over the course of approximately 6 minutes around the time of maximum rate-of-change of the X-ray flux. Surrounding pixels which show an impulsive decrease in flux (but no sign reversal) are outlined in blue, and x-y coordinates of the pixels are shown relative to the lower-left panel. The LOS flux in this region derived from a Milne-Eddington inversion are shown simultaneously in orange and light blue, for the MT and surrounding pixels, respectively.
Based on these observations and experiments, Harker and Pevtsov conclude that the observed magnetic transient is not an artifact created by the particulars of the SDO/HMI observation process, but instead represent a real change in the vector magnetic field; viewing geometry and projection effects on this very limb-adjacent transient suggest that it manifested in the horizontal component of the magnetic field (relative to the solar surface). This can be the case with an impulsive rotation of the field vector in response to the flare.
 Harker, B.J. and Pevtsov, A.A., 2013, ApJ, 778, 175