Rachel Howe
School of Physics and Astronomy, University of Birmingham, Birmingham B15 2TT, UK

The most obvious marker of the solar cycle is the pattern of the belts where sunspots appear migrating from mid-latitudes towards the equator. Accompanying this sunspot ‘butterfly diagram’ is a pattern of migrating belts of faster- and slower-than-average rotation, known as the torsional oscillation^[1]. This pattern can be detected in surface Doppler measurements and, through helioseismology, below the surface as well – it has been shown to penetrate through most of the bulk of the convection zone. The belts of faster rotation appear at mid-latitudes a year or two before the new-cycle activity appears. With Cycle 24 approaching minimum, we would expect to start seeing the flow band associated with the new cycle appearing at mid-latitudes. Figure 1 shows the flows at 0.99 R_⊙ over the whole period of helioseismic observations with SOHO/MDI and SDO/HMI, as inferred from global rotation inversions with a temporal mean subtracted at each latitude. The black contours show the 5 Gs level of the magnetic field.

Figure 1| Rotation-rate residuals at 0.99R_⊙ from MDI (1996-2010) and HMI (2011-present).

In this representation, the new flow belt can be seen – it has reached about 35° latitude in the most recent data^[2]– but it is not very distinct. Also, the strong poleward-migrating branch that was seen from around 1999-2006 during Solar Cycle 23 appears to be absent in Cycle 24.

Figure 2| Rotation-rate residuals at 0.99R_⊙ in Cycle 23 (lef) and Cycle 24 (right) with the mean subtracted being over the first eight years of each cycle. The black contours show the 5-Gs level of the magnetic field strength.

However, if we compare the cycles by removing a mean tailor-made for each cycle, we see that the residuals for each cycle look more similar (Figure 2). The mean here was calculated over the first 8 years of each cycle – 1996.5-2004.5 and 2009-2017. The main difference between the two means is that the rotation at higher latitudes is slower – by about 4 nHz, or a bit more than 1% at 60° latitude – in Cycle 24 than it was in Cycle 23. This is believed to be related to the weaker polar fields in Cycle 24^[3]. The Cycle 24 poleward branch is now visible (though it is weak and noisy compared to the Cycle 23 one) and we can see the beginning of the poleward-moving belt of slower-than-average rotation around 50°-60° latitude in the most recent data.

Figure 3| Rotation rate residuals as a function of time at selected locations, as inferred from two different types of inversions of MDI and HMI data and from data from the Global Oscillation Network Group (GONG).

The acceleration at 30°-35° and the deceleration at 60° can be seen when we plot the residuals as a function of time at selected locations, as in Figure 3. The timing of the new-cycle flow can give us some hints about when the new-cycle activity will appear: historically it usually takes about two years for the flow belt to move from 35 to 25 degrees, and when it reaches 25 degrees the new-cycle activity becomes widespread^[4]. So we would not expect to see much Cycle 25 activity before the middle of 2019 at the earliest.

References

[1] Howard, R., & LaBonte, B. J. 1980, ApJL, 239, L33
[2] Howe, R., Hill, F., Komm, R., et al. 2018, ApJL, 862, L5
[3] Howe, R., Christensen-Dalsgaard, J., Hill, F., et al. 2013, ApJL, 767, L20
[4] Howe, R., Hill, F., Komm, R., et al. 2011, in Journal of Physics Conference Series, 271, 012074