Xudong Sun, Todd Hoeksema, Yang Liu, & Junwei Zhao
W. W. Hansen Experimental Physics Laboratory, Stanford University, Stanford, CA 94305
The Sun’s polar magnetic field changes sign during the maximum phase of each cycle, manifesting the large-scale poloidal field reversal. The apparent cause is the flux from decaying active regions (ARs), which migrates poleward owing to meridional flow and diffusion. The long-term net flux has the same polarity of the trailing sunspots and cancels the old-cycle polar field.
We utilized HMI line-of-sight observations to characterize this reversal process for Cycle 24. We calculated the mean radial field in various latitudinal bins for each magnetogram assuming all field vectors are radial, and performed an area-weighted running average to estimate the full-Sun mean. The high cadence data enable deep averaging (over 3,000 magnetograms for each measurement), which significantly lowers the statistical uncertainty.
Fig 1 | Magnetic field reversal. (a) SIDC hemispheric sunspot number. (b) Mean polar field above 60°. Vertical dotted lines indicate the reversal times. (c) Magnetic butterfly diagram constructed by zonally averaging magnetograms. Contours show polarity inversion; their intersections with horizontal dotted lines indicate reversals at ±60° (star symbols). Multiple reversals took place in the north. N1-N6 and S1-S5 mark individual flux surges.
The magnetic activity, characterized by the sunspot number (SSN), is low and north-south asymmetric (Fig. 1a). The maximum hemispheric SSN is about 60% of Cycle 23; the north peaked almost two years earlier than the south. The polar fields are asymmetric too (Fig. 1b). The north and south reversed in November 2012 and March 2014 respectively, about 16 months apart. The asymmetry is clearly related to the asymmetric poleward flux “surges” that contain remnant AR flux (Fig. 1c). Individual surges may have either polarity; we have demonstrated that the sign is dictated by the changing tilt angle of the ARs.
We found a curious anti-correlation between the mean field of these surges and the near-surface meridional flow velocity at mid latitudes as inferred from time-distance helioseismology (Fig. 2a). That is, the poleward flow is generally slower when the surge field has the trailing sunspot polarity, and faster otherwise. We have shown (see  for details) that this interesting sign dependence can be explained in the surface flux transport (SFT) framework, if we evoke an observed, field-dependent converging flow towards the ARs (Fig. 2b) as well as Joy’s law. Incorporating such observation-based, two-dimensional meridional flow profile may improve the SFT modeling of cycle magnitude.
Fig 2 |Meridional flow and magnetic field. (a) Scatter plot of the mean field of the surges (Br) and the residual meridional velocity from helioseismology (δuy). A four-year mean profile is subtracted to obtain δuy; northward flow is positive. Both Br and δuy are sampled in a 10° latitudinal band just poleward of the activity belt. Each symbol represents a Carrington-rotation (CR) mean. Yellow and cyan represents north and south, respectively. (b) Inflow pattern around AR 11106 after removing the mean flow. The maximum flow speed is ~30 m s-1. The background shows magnetic field saturated at ±200 G.
Data from the Wilcox Solar Observatory (http://wso.stanford.edu/gifs/Polar.gif) suggest that the current reversal is slower than the last three cycles. The rebuild of the new cycle field has also been slow. The north showed multiple changes of sign near 60° (Fig. 1c); the northern polar field remains close to zero even two years after the reversal (Fig. 1b). Because the maximum polar field is a good indicator of the magnitude of the next cycle, Cycle 25 may be a very weak one if the trend continues.
 Sun, X., Hoeksema, J. T., Liu, Y., & Zhao, J. 2015, ApJ, 798, 114 (http://arxiv.org/abs/1410.8867)
 Zhao, J., Kosovichev, A. G., & Bogart, R. S. 2014, ApJL, 789, L7
 Wang, Y.-M., Nash, A.G., & Sheeley, Jr., N. R. 1989, Science, 245, 712
 Gizon, L., Duvall, Jr., T. L., & Larsen, R. M. 2001, IAU Symposium, 203, 189
 Svalgaard, L., Cliver, E. W., & Kamide, Y. 2005, GRL, 32, 1104