Photospheric and coronal magnetic fields in six magnetographs

Abstract

Aims. We study the long-term evolution of photospheric and coronal magnetic fields and the heliospheric current sheet (HCS), especially its north-south asymmetry. Special attention is paid to the reliability of the six data sets used in this study and to the consistency of the results based on these data sets. Methods. We use synoptic maps constructed from Wilcox Solar Observatory (WSO), Mount Wilson Observatory (MWO), Kitt Peak (KP), SOLIS, SOHO/MDI, and SDO/HMI measurements of the photospheric field and the potential field source surface (PFSS) model. Results. The six data sets depict a fairly similar long-term evolution of magnetic fields and the heliospheric current sheet, including polarity reversals and hemispheric asymmetry. However, there are time intervals of several years long, when first KP measurements in the 1970s and 1980s, and later WSO measurements in the 1990s and early 2000s, significantly deviate from the other simultaneous data sets, reflecting likely errors at these times. All of the six magnetographs agree on the southward shift of the heliospheric current sheet (the so-called bashful ballerina phenomenon) in the declining to minimum phase of the solar cycle during a few years of the five included cycles. We show that during solar cycles 20-22, the southward shift of the HCS is mainly due to the axial quadrupole term, reflecting the stronger magnetic field intensity at the southern pole during these times. During cycle 23 the asymmetry is less persistent and mainly due to higher harmonics than the quadrupole term. Currently, in the early declining phase of cycle 24, the HCS is also shifted southward and is mainly due to the axial quadrupole as for most earlier cycles. This further emphasizes the special character of the global solar field during cycle 23.

Comparing Coronal and Heliospheric Magnetic Fields Over Several Solar Cycles

“…Figure 12 shows that the tilt of the HCS decreased very rapidly and quite early in the declining phase of cycle 22. We also note that the HCS was southward shifted in 1992-1995 (Virtanen & Mursula 2016), in agreement with decreasing match since 1992. However, the peak in 1994 is most likely related to the activation of the southern polar coronal holes at this time, as described by Mursula & Virtanen (2012).…”

“…Surprisingly, the match percentage decreases dramatically from 1984 to 1985. Interestingly, this is exactly the time when the HCS starts being systematically shifted southward for about 3 years, marking the so-called bashful ballerina period (Mursula & Hiltula 2003;Zhao et al 2005;Virtanen & Mursula 2016). (We will study the effect of the HCS shift in more detail in a separate paper; J. S. Koskela et al 2016, in preparation.)…”

“…Another uncertainty is caused by the different coverage of solar polar regions by the observatories and the different methods in filling them. In this paper the synoptic maps have been preprocessed in the same way as in Virtanen & Mursula (2016).…”

“…There is a very good correspondence in polarity match between the different data sets, except in 1998 and 2000-2002 when the WSO polarity match is significantly lower than MDI polarity match. Interestingly, a recent study (Virtanen & Mursula 2016) found that WSO synoptic maps have problems (partly unknown earlier) exactly (and only) at these times. The good agreement between the four instruments at other times can be easily understood by a recent observation that the PFSS solutions based on the lowest harmonic coefficients are very similar in different data sets (I. Virtanen & K. Mursula 2016, in preparation).…”

“…SOLIS also yields slightly larger p-values than MDI and HMI during the overlapping time. The effect of saturation during field observation is strongest in WSO, which causes the intensity level of WSO synoptic maps to be lower than for the other magnetograms (Riley et al 2014;Virtanen & Mursula 2016). Saturation may occur if the magnetic field of the active region is very high; hence the Zeeman shifted spectral line will be outside of the spectrometer slit and cannot be detected.…”