An Early Prediction of the Amplitude of Solar Cycle 25

A “Solar Dynamo” (SODA) Index prediction of the amplitude of Solar Cycle 25 is described. The SODA Index combines values of the solar polar magnetic field and the solar spectral irradiance at 10.7 cm to create a precursor of future solar activity. The result is an envelope of solar activity that minimizes the 11-year period of the sunspot cycle. We show that the variation in time of the SODA Index is similar to several wavelet transforms of the solar spectral irradiance at 10.7 cm. Polar field predictions for Solar Cycles 21?–?24 are used to show the success of the polar field precursor in previous sunspot cycles. Using the present value of the SODA index, we estimate that the next cycle’s smoothed peak activity will be about \(140 \pm30\) solar flux units for the 10.7 cm radio flux and a Version 2 sunspot number of \(135 \pm25\). This suggests that Solar Cycle 25 will be comparable to Solar Cycle 24. The estimated peak is expected to occur near \(2025.2 \pm1.5\) year. Because the current approach uses data prior to solar minimum, these estimates may improve as the upcoming solar minimum draws closer.     

Temporal and Periodic Variations of Sunspot Counts in Flaring and Non-Flaring Active Regions

We analyzed temporal and periodic variations of sunspot counts (SSCs) in flaring (C-, M-, or X-class flares), and non-flaring active regions (ARs) for nearly three solar cycles (1986 through 2016). Our main findings are as follows: i) temporal variations of monthly means of the daily total SSCs in flaring and non-flaring ARs behave differently during a solar cycle and the behavior varies from one cycle to another; during Solar Cycle 23 temporal SSC profiles of non-flaring ARs are wider than those of flaring ARs, while they are almost the same during Solar Cycle 22 and the current Cycle 24. The SSC profiles show a multi-peak structure and the second peak of flaring ARs dominates the current Cycle 24, while the difference between peaks is less pronounced during Solar Cycles 22 and 23. The first and second SSC peaks of non-flaring ARs have comparable magnitude in the current solar cycle, while the first peak is nearly absent in the case of the flaring ARs of the same cycle. ii) Periodic variations observed in the SSCs profiles of flaring and non-flaring ARs derived from the multi-taper method (MTM) spectrum and wavelet scalograms are quite different as well, and they vary from one solar cycle to another. The largest detected period in flaring ARs is \(113\pm 1.6~\mbox{days}\) while we detected much longer periodicities (\(327\pm 13\), \(312 \pm 11\), and \(256\pm 8~\mbox{days}\)) in the non-flaring AR profiles. No meaningful periodicities were detected in the MTM spectrum of flaring ARs exceeding \(55\pm 0.7~\mbox{days}\) during Solar Cycles 22 and 24, while a \(113\pm 1.3~\mbox{days}\) period was detected in flaring ARs of Solar Cycle 23. For the non-flaring ARs the largest detected period was only \(31\pm 0.2~\mbox{days}\) for Cycle 22 and \(72\pm 1.3~\mbox{days}\) for the current Cycle 24, while the largest measured period was \(327\pm 13~\mbox{days}\) during Solar Cycle 23.     

Long-Term Variations of Solar Differential Rotation and Sunspot Activity: Revisited

Long-term variations of solar differential rotation and sunspot activity are investigated through re-analyzing the data on parameters of the differential-rotation law obtained by Makarov, Tlatov, and Callebaut (Solar Phys. 170, 373, 1997), Javaraiah, Bertello, and Ulrich (Astrophys. J. 626, 579, 2005a; Solar Phys. 232, 25, 2005b), and Javaraiah et al. (Solar Phys. 257, 61, 2009). Our results indicate that the solar-surface-rotation rate at the Equator (indicated by the A-parameter of the standard solar-rotation law) shows a secular decrease since Cycle 12 onwards, given by about 1?–?1.5×10^?3 (deg?day^?1?year^?1). The B-parameter of the standard differential-rotation law seems to also show a secular decrease since Cycle 12 onwards, but of weak statistical significance. The rotation rate averaged over latitudes 0^°?–?40^° does not show a secular trend of statistical significance. Moreover, the average sunspot area shows a secular increase of statistical significance since Cycle 12 onwards, while a negative correlation is found between the level of sunspot activity (indicated by the average sunspot area) and the solar equatorial rotation on long-term scales.     

Correlation and spectral analysis of daily solar radio flux

Correlation and spectral analysis of solar radio flux density and sunspot number near the maximum of the sunspot cycle has indicated the existence of

1. long period amplitude modulation of the slowly varying component (SVC) of radio emission
2. coronal storage over a period of the order of three solar rotations
3. fast decay (one solar rotation period or less) of gyromagnetic emissions from radio sources
4. shift in location of chromospheric sources compared to those of either the upper corona or the photosphere.

     

The Solar Cycle 23?�C?24 Minimum. A Benchmark in Solar Variability and Effects in the Heliosphere

Given the numerous ground-based and space-based experiments producing the database for the Cycle 23??C?24 Minimum epoch from September 2008 to May 2009, we have an extraordinary opportunity to understand its effects throughout the heliosphere. We use solar radiative output in this period to obtain minimum values for three measures of the Sun??s radiative output: the total solar irradiance, the Mg ii index, and the 10.7 cm solar radio flux. The derived values are included in the research summaries as a means to exchange ideas and data for this long minimum in solar activity.     

Seismology of a Sunspot Atmosphere

Two competing theories of sunspot oscillations are discussed. It is pointed out that the normal mode (eigenoscillations) theory is in contradiction with a number of observations. The reasons for this are discussed. The revised filter theory of the three-minute sunspot oscillations is outlined. It is shown that the reason for the occurrence of the multipassband filter for the slow waves is the interference that appears from the multilayer structure of the sunspot atmosphere. In contrast with Zhugzhda and Locans (Sov. Astron. Lett. 7, 25?–?27, 1981) it is shown that along with the Fabry?–?Perot chromospheric passband the cutoff frequency passband and a number of the high-frequency passbands occur. The effect of the nonlinearity of the sunspot oscillations in the upper chromosphere and the transition region is taken into account. The spectra of the distinct empirical models of the sunspot atmosphere are explored. An example of the interpretation of the sunspot oscillations based on the revised filter theory is presented. Only the filter theory can explain the complicated behavior of the oscillations across the sunspot. The observations provide evidence of the nonuniformity of the sunspot atmosphere.     

A Comparison of the 10.7-cm Radio Flux Values and the International Sunspot Numbers for Solar Activity Cycles 19, 20, and 21

A nonlinear analysis of the daily 10.7-cm radio flux values for each of Solar Cycles 19, 20, and 21 is used to determine if the results match those of the International Sunspot Numbers for each of these cycles. Fractals and chaos are described and a brief review of utilizing fractals and chaos is given. The origin of the 10.7-cm radio flux is discussed and a short review of recent work discussing its measurement and its relation to the international sunspot number and other proxies for solar activity cycles given. The parameters used to describe chaos for the 10.7-cm radio flux are discussed. The length of the data sets for either statistical analysis or nonlinear analysis of the 10.7-cm radio flux values is considered. These results indicate that the 10.7-cm radio flux values appear to be stochastic for Cycle 19 and chaotic for Cycles 20 and 21. The International Sunspot Numbers show similar behavior for these three cycles. A day-by-day comparison of the dimensionless 10.7-cm radio flux values and the dimensionless International Sunspot Numbers differences shows a linear trend. The results remain consistent in that the 10.7-cm radio flux values indicate, as did the International Sunspot Numbers, that there is a transition from stochastic behavior for Cycle 19 to chaotic behavior in Cycles 20 and 21. The day-by-day comparison of the 10.7-cm radio flux values and the International Sunspot Numbers emphasizes that the 10.7?cm radio flux values are responding to the magnetic field associated with the sunspots.     

Phase Relationship Between Sunspot Number, Flare Index and Solar Radio Flux

To understand better the variation of solar activity indicators originated at different layers of the solar atmosphere with respect to sunspot cycles, we carried out a study of phase relationship between sunspot number, flare index and solar radio flux at 2800 MHz from January 1966 to May 2008 by using cross-correlation analysis. The main results are as follows: (1) The flare index and sunspot number have synchronous phase for cycles 21 and 22 in the northern hemisphere and for cycle 20 in the southern hemisphere. (2) The flare index has a noticeable time lead with respect to sunspot number for cycles 20 and 23 in the northern hemisphere and for cycles 22 and 23 in the southern hemisphere. (3) For the entire Sun, the flare index has a noticeable time lead for cycles 20 and 23, a time lag for cycle 21, and no time lag or time lead for cycle 22 with respect to sunspot number. (4) The solar radio flux has a time lag for cycles 22 and 23 and no time lag or time lead for cycles 20 and 21 with respect to sunspot number. (5) For the four cycles, the sunspot number and flare index in the northern hemisphere are all leading to the ones in the southern hemisphere. These results may be instructive to the physical processes of flare energy storage and dissipation.     

Cyclic and Long-Term Variation of Sunspot Magnetic Fields

Measurements from the Mount Wilson Observatory (MWO) were used to study the long-term variations of sunspot field strengths from 1920 to 1958. Following a modified approach similar to that presented in Pevtsov et al. (Astrophys. J. Lett. 742, L36, 2011), we selected the sunspot with the strongest measured field strength for each observing week and computed monthly averages of these weekly maximum field strengths. The data show the solar cycle variation of the peak field strengths with an amplitude of about 500?–?700 gauss (G), but no statistically significant long-term trends. Next, we used the sunspot observations from the Royal Greenwich Observatory (RGO) to establish a relationship between the sunspot areas and the sunspot field strengths for cycles 15?–?19. This relationship was used to create a proxy of the peak magnetic field strength based on sunspot areas from the RGO and the USAF/NOAA network for the period from 1874 to early 2012. Over this interval, the magnetic field proxy shows a clear solar cycle variation with an amplitude of 500?–?700 G and a weaker long-term trend. From 1874 to around 1920, the mean value of magnetic field proxy increases by about 300?–?350 G, and, following a broad maximum in 1920?–?1960, it decreases by about 300 G. Using the proxy for the magnetic field strength as the reference, we scaled the MWO field measurements to the measurements of the magnetic fields in Pevtsov et al. (2011) to construct a combined data set of maximum sunspot field strengths extending from 1920 to early 2012. This combined data set shows strong solar cycle variations and no significant long-term trend (the linear fit to the data yields a slope of ??0.2±0.8 G?year^?1). On the other hand, the peak sunspot field strengths observed at the minimum of the solar cycle show a gradual decline over the last three minima (corresponding to cycles 21?–?23) with a mean downward trend of ≈?15 G?year^?1.     

Modelling solar irradiance variations with an area dependent photometric sunspot index

The He 1083 nm line equivalent width and the 10.7 cm radio flux are employed to model the total solar irradiance corrected for sunspot deficit. A new area dependent photometric sunspot index (APSI) based on sunspot photometry by Steinegger et al. (1990) is used to correct the irradiance data for sunspot deficits. Two periods of time are investigated: firstly, the 1980–1989 period between the maxima of solar cycles 21 and 22; this period is covered by ACRIM I irradiance data. Secondly, the 1978–92 period which includes both maxima; here, the revised Nimbus-7 ERB data are used.For both He 1083 nm and 10.7 cm radio flux irradiance models as well as ACRIM I and ERB irradiance data, the APSI yields an improved fit compared to the one obtained with the standard Photometric Sunspot Index (PSI) which uses a constant bolometric spot contrast. With APSI, the standard deviation calculated from daily values is 0.461 Wm^–2 for the period 1980–89 modelling ACRIM I vs. He 1083 nm, as compared to 0.478 when PSI is used, and to 0.531 for the uncorrected ACRIM series. A similar improvement is obtained for the same period modelling ERB vs. He 1083 nm, while there is almost no improvement for the long period.As a general result the models provide a good fit with the spot-deficit.-corrected irradiance only during the period between the maxima. If both maxima are included (period 1978–92) the He 1083 nm and 10.7 cm radio flux models show appreciably larger discrepancies to the irradiances corrected for PSI or APSI.     

Predicting Solar Cycle 24 Using a Geomagnetic Precursor Pair

We describe using Ap and F_10.7 as a geomagnetic-precursor pair to predict the amplitude of Solar Cycle 24. The precursor is created by using F_10.7 to remove the direct solar-activity component of Ap. Four peaks are seen in the precursor function during the decline of Solar Cycle 23. A recurrence index that is generated by a local correlation of Ap is then used to determine which peak is the correct precursor. The earliest peak is the most prominent but coincides with high levels of non-recurrent solar activity associated with the intense solar activity of October and November 2003. The second and third peaks coincide with some recurrent activity on the Sun and show that a weak cycle precursor closely following a period of strong solar activity may be difficult to resolve. A fourth peak, which appears in early 2008 and has recurrent activity similar to precursors of earlier solar cycles, appears to be the “true” precursor peak for Solar Cycle 24 and predicts the smallest amplitude for Solar Cycle 24. To determine the timing of peak activity it is noted that the average time between the precursor peak and the following maximum is ≈?6.4 years. Hence, Solar Cycle 24 would peak during 2014. Several effects contribute to the smaller prediction when compared with other geomagnetic-precursor predictions. During Solar Cycle 23 the correlation between sunspot number and F_10.7 shows that F_10.7 is higher than the equivalent sunspot number over most of the cycle, implying that the sunspot number underestimates the solar-activity component described by F_10.7. During 2003 the correlation between aa and Ap shows that aa is 10 % higher than the value predicted from Ap, leading to an overestimate of the aa precursor for that year. However, the most important difference is the lack of recurrent activity in the first three peaks and the presence of significant recurrent activity in the fourth. While the prediction is for an amplitude of Solar Cycle 24 of 65±20 in smoothed sunspot number, a below-average amplitude for Solar Cycle 24, with maximum at 2014.5±0.5, we conclude that Solar Cycle 24 will be no stronger than average and could be much weaker than average.     

Correlations Between CME Parameters and Sunspot Activity

Smoothed monthly mean coronal mass ejection (CME) parameters (speed, acceleration, central position angle, angular width, mass, and kinetic energy) for Cycle 23 are cross-analyzed, showing that there is a high correlation between most of them. The CME acceleration (a) is highly correlated with the reciprocal of its mass (M), with a correlation coefficient r=0.899. The force (Ma) to drive a CME is found to be well anti-correlated with the sunspot number (R _z), r=?0.750. The relationships between CME parameters and R _z can be well described by an integral response model with a decay time scale of about 11 months. The correlation coefficients of CME parameters with the reconstructed series based on this model (\(\overline{r}_{\mathrm{f1}}=0.886\)) are higher than the linear correlation coefficients of the parameters with R _z (\(\overline{r}_{\mathrm{0}}=0.830\)). If a double decay integral response model is used (with two decay time scales of about 6 and 60 months), the correlations between CME parameters and R _z improve (\(\overline{r}_{\mathrm{f2}}=0.906\)). The time delays between CME parameters with respect to R _z are also well predicted by this model (19/22=86%); the average time delays are 19 months for the reconstructed and 22 months for the original time series. The model implies that CMEs are related to the accumulation of solar magnetic energy. These relationships can help in understanding the mechanisms at work during the solar cycle.     

A Relationship Between the Solar Rotation and Activity Analysed by Tracing Sunspot Groups

The sunspot position published in the data bases of the Greenwich Photoheliographic Results (GPR), the US Air Force Solar Optical Observing Network and National Oceanic and Atmospheric Administration (USAF/NOAA), and of the Debrecen Photoheliographic Data (DPD) in the period 1874 to 2016 were used to calculate yearly values of the solar differential-rotation parameters \(A\) and \(B\). These differential-rotation parameters were compared with the solar-activity level. We found that the Sun rotates more differentially at the minimum than at the maximum of activity during the epoch 1977?–?2016. An inverse correlation between equatorial rotation and solar activity was found using the recently revised sunspot number. The secular decrease of the equatorial rotation rate that accompanies the increase in activity stopped in the last part of the twentieth century. It was noted that when a significant peak in equatorial rotation velocity is observed during activity minimum, the next maximum is weaker than the previous one.     

Cycling Changes in the Amplitudes of the 27-Day Variation of the Galactic Cosmic Ray Intensity

We study quasi-periodical changes in the amplitudes of the 27-day variation of the galactic cosmic ray (GCR) intensity, and the parameters of solar wind and solar activity. We have recently found quasi-periodicity of three to four Carrington rotation periods (3?–?4 CRP) in the amplitudes of the 27-day variation of the GCR intensity (Gil and Alania in J. Atmos. Solar-Terr. Phys. 73, 294, 2011). A similar recurrence is recognized in parameters of solar activity (sunspot number, solar radio flux) and solar wind (components of the interplanetary magnetic field, solar wind velocity). We believe that the 3?–?4 CRP periodicity, among other periodicities, observed in the amplitudes of the 27-day variation of the GCR intensity is caused by a specific cycling structure of the Sun’s magnetic field, which may originate from the turbulent nature of the solar dynamo.     

Effects of Hysteresis Between Maximum CME Speed Index and Typical Solar Activity Indicators During Cycle 23

Using the smoothed time series of maximum CME speed index for solar cycle 23, it is found that this index, analyzed jointly with six other solar activity indicators, shows a hysteresis phenomenon. The total solar irradiance, coronal index, solar radio flux (10.7?cm), Mg?ii core-to-wing ratio, sunspot area, and H?? flare index follow different paths for the ascending and the descending phases of solar cycle?23, while a saturation effect exists at the maximum phase of the cycle. However, the separations between the paths are not the same for the different solar activity indicators used: the H?? flare index and total solar irradiance depict broad loops, while the Mg?ii core-to-wing ratio and sunspot area depict narrow hysteresis loops. The lag times of these indices with respect to the maximum CME speed index are discussed, confirming that the hysteresis represents a clue in the search for physical processes responsible for changing solar emission.     

Reconstruction of Wolf Sunspot Numbers on the Basis of Spectral Characteristics and Estimates of Associated Radio Flux and Solar Wind Parameters for the Last Millennium

A reconstruction of sunspot numbers for the last 1000 years was obtained using a sum of sine waves derived from spectral analysis of the time series of sunspot number R _z for the period 1700–1999. The time series was decomposed in frequency levels using the wavelet transform, and an iterative regression model (ARIST) was used to identify the amplitude and phase of the main periodicities. The 1000-year reconstructed sunspot number reproduces well the great maximums and minimums in solar activity, identified in cosmonuclides variation records, and, specifically, the epochs of the Oort, Wolf, Spörer, Maunder, and Dalton Minimums as well the Medieval and Modern Maximums. The average sunspot number activity in each anomalous period was used in linear equations to obtain estimates of the solar radio flux F _10.7, solar wind velocity, and the southward component of the interplanetary magnetic field.     

Identification of Gleissberg Cycles and a Rising Trend in a 315-Year-Long Series of Sunspot Numbers

We show in this short note that the method of singular spectrum analysis (SSA) is able to clearly extract a strong, clean, and clear component from the longest available sunspot (International Sunspot Number, ISN) time series (1700?–?2015) that cannot be an artifact of the method and that can be safely identified as the Gleissberg cycle. This is not a small component, as it accounts for 13% of the total variance of the total original signal. Almost three and a half clear Gleissberg cycles are identified in the sunspot number series. Four extended solar minima (XSM) are determined by SSA, the latest around 2000 (Cycle 23/24 minimum). Several authors have argued in favor of a double-peaked structure for the Gleissberg cycle, with one peak between 55 and 59 years and another between 88 and 97 years. We find no evidence of the former: solar activity contains an important component that has undergone clear oscillations of \(\approx90\) years over the past three centuries, with some small but systematic longer-term evolution of “instantaneous” period and amplitude. Half of the variance of solar activity on these time scales can be satisfactorily reproduced as the sum of a monotonous multi-secular increase, a \(\approx90\)-year Gleissberg cycle, and a double-peaked (\(\approx10.0\) and 11.0 years) Schwabe cycle (the sum amounts to 46% of the total variance of the signal). The Gleissberg-cycle component definitely needs to be addressed when attempting to build dynamo models of solar activity. The first SSA component offers evidence of an increasing long-term trend in sunspot numbers, which is compatible with the existence of the modern grand maximum.     

Long-Term Modulation of Cosmic-Ray Intensity and Correlation Analysis Using Solar and Heliospheric Parameters

Based on the monthly sunspot numbers (SSNs), the solar-flare index (SFI), grouped solar flares (GSFs), the tilt angle of heliospheric current sheet (HCS), and cosmic-ray intensity (CRI) for Solar Cycles 21?–?24, a detailed correlation study has been performed using the cycle-wise average correlation (with and without time lag) method as well as by the “running cross-correlation” method. It is found that the slope of regression lines between SSN and SFI, as well as between SSN and GSF, is continuously decreasing from Solar Cycle 21 to 24. The length of regression lines has significantly decreased during Cycles 23 and 24 in comparison to Cycles 21 and 22. The cross-correlation coefficient (without time lag) between SSN–CRI, SFI–CRI, and GSF–CRI has been found to be almost the same during Cycles 21 and 22, while during Cycles 23 and 24 it is significantly higher between SSN–CRI and HCS–CRI than for SFI–CRI and GSF–CRI. Considering time lags of 1 to 20 months, the maximum correlation coefficient (negative) amongst all of the sets of solar parameters is observed with almost the same time lags during Cycles 21?–?23, whereas exceptional behaviour of the time lag has been observed during Cycle 24, as the correlation coefficient attains its maximum value with two time lags (four and ten months) in the case of the SSN–CRI relationship. A remarkably large time lag (22 months) between HCS and CRI has been observed during the odd-numbered Cycle 21, whereas during another odd cycle, Cycle 23, the lag is small (nine months) in comparison to that for other solar/flare parameters (13?–?15 months). On the other hand, the time lag between SSN–CRI and HCS–CRI has been found to be almost the same during even-numbered Solar Cycles 22 and 24. A similar analysis has been performed between SFI and CRI, and it is found that the correlation coefficient is maximum at zero time lag during the present solar cycle. The GSFs have shown better maximum correlation with CRI as compared to SFI during Cycles 21 to 23, indicating that GSF could also be used as a significant solar parameter to study the cosmic-ray modulation. Furthermore, the running cross-correlation coefficient between SSN–CRI and HCS–CRI, as well as between solar-flare activity parameters (SFI and GSF) and CRI is observed to be strong during the ascending and descending phases of solar cycles. The level of cosmic-ray modulation during the period of investigation shows the appropriateness of different parameters in different cycles, and even during the different phases of a particular solar cycle. We have also studied the galactic cosmic-ray modulation in relation to combined solar and heliospheric parameters using the empirical model suggested by Paouris et al. (Solar Phys.280, 255, 2012). The proposed model for the calculation of the modulated cosmic-ray intensity obtained from the combination of solar and heliospheric parameter gives a very satisfactory value of standard deviation as well as \(R^{2}\) (the coefficient of determination) for Solar Cycles 21?–?24.     

Variation of solar irradiance and mode frequencies during Maunder minimum

Using the sunspot numbers reported during the Maunder minimum and the empirical relations between the mode frequencies and solar activity indices, the variations in the total solar irradiance and 10.7 cm radio flux for the period 1645 to 1715 is estimated. We find that the total solar irradiance and radio flux during the Maunder minimum decreased by 0.19% and 52% respectively, as compared to the values for solar cycle 22. This revised version was published online in July 2006 with corrections to the Cover Date.     

Short-Term Variations in the Equatorial Rotation Rate of Sunspot Groups

We have detected several periodicities in the solar equatorial rotation rate of sunspot groups in the catalog Greenwich Photoheliographic Results (GPR) during the period 1931?–?1976, the Solar Optical Observing Network (SOON) during the period 1977?–?2014, and the Debrecen Photoheliographic Data (DPD) during the period 1974?–?2014. We have compared the results from the fast Fourier transform (FFT), the maximum entropy method (MEM), and the Morlet wavelet power-spectra of the equatorial rotation rates determined from SOON and DPD sunspot-group data during the period 1986?–?2007 with those of the Mount Wilson Doppler-velocity data during the same period determined by Javaraiah et al. (Solar Phys. 257, 61, 2009). We have also compared the power-spectra computed from the DPD and the combined GPR and SOON sunspot-group data during the period 1974?–?2014 to those from the GPR sunspot-group data during the period 1931?–?1973. Our results suggest a ～?250-day period in the equatorial rotation rate determined from both the Mt. Wilson Doppler-velocity data and the sunspot-group data during 1986?–?2007. However, a wavelet analysis reveals that this periodicity appears mostly around 1991 in the velocity data, while it is present in most of the solar cycles covered by the sunspot-group data, mainly near the minimum epochs of the solar cycles. We also found the signature of a period of ～?1.4 years in the velocity data during 1990?–?1995, and in the equatorial rotation rate of sunspot groups mostly around the year 1956. The equatorial rotation rate of sunspot groups reveals a strong ～?1.6-year periodicity around 1933 and 1955, a weaker one around 1976, and a strong ～?1.8-year periodicity around 1943. Our analysis also suggests periodicities of ～?5 years, ～?7 years, and ～?17 years, as well as some other short-term periodicities. However, short-term periodicities are mostly present at the time of solar minima. Hence, short-term periodicities cannot be confirmed because of the larger uncertainty in the data.