This paper develops a planetary model to predict the occurrence of intermediate range periodicity in solar activity, in particular the ~155 day Rieger periodicity in flare activity. It is shown that periodicity at half integer multiples of the period of Mercury occurs consistently in indices of solar activity. For this reason the planetary model is based on the triggering of sunspot emergence when planetary tides peak at times of conjunction of Mercury with Venus, Earth and/or Jupiter. The periodicity of components in the planetary model match reasonably well the observed intermediate periodicity in sunspot and flare activity with the strongest model component occurring at 155 day period. A comparison of filtered versions of the model and the N.O.A.A. flare index at 155 day periodicity demonstrates the potential for both short term, (within a solar cycle), and longer term, (over several solar cycles), predictive capability of the model. However, this assessment finds that prediction of 155 day periodicity in flare activity is effective only when the north and south hemispheric components of flare activity are in-phase.

Rieger et al (1984) reported observations of a 154 day periodicity in flares during solar cycle 21. This paper discusses the observations in the light of a simple empirical planetary model of sunspot emergence. The planetary model predicts sunspot emergence when Mercury and Earth approach conjunction and Mercury approaches the Sun. We show that the reported times of flare activity are coherent with the planetary model. While the base period of the model is 170 days, the average model period, over the interval of flare recordings, is 157 days due to a 180 degree phase change in the planetary forcing near the middle of the record interval. We conclude that the periodicity at 154 days arises when the phase change in planetary forcing and the resulting progressive phase change in total sunspot area emergence and flare occurrence shifts the major peak in the flare spectrum from the planetary forcing period, 170 days, to 154 days.

This paper is concerned with intermediate range periodicity in the sunspot area spectrum. An empirical model of sunspot area emergence based on Mercury planet conjunctions was developed and the spectra of the model variation and the sunspot area variation compared. By including solar cycle amplitude modulation and the effect of solar magnetic field reversal the model was able to predict fine detail in the sunspot area spectrum. As Mercury planet conjunctions occur predictably it was possible to compare the time variation of band limited components of sunspot area with the corresponding component variations in the model. When the model component variation was stable corresponding components of sunspot area lagged the model variation by a few tens of days. When a 180 degree phase change occurred in the model variation the corresponding component of sunspot area followed the change over an interval of a few hundred days, first by decreasing to zero and then emerging in phase with the model variation. Where periodicities in sunspot area did not match periods of Fourier components of the model the autocorrelation function of the model variation provided an indirect numerical link to these periodicities. Included in these indirect periodicities were components at periods of about 138, 235, 355, 472, 530, 605, 833, 1190 and 2020 days. Direct periodicities in the model, those linked directly to Fourier components of sunspot area, included components at about 88, 116, 176, 200, 292 and 405 days. The model was developed without allowing for a variable threshold for sunspot emergence. However, it was found that with a variable threshold above which the model variation is effective in triggering sunspots, weak Fourier components at the indirect periods were generated. The model with threshold then encompasses all of the significant intermediate range periodicities in sunspot area.

In this paper we examine the possibility that intermediate range periodicities in sunspot area can be linked to Mercury or to Mercury-planet conjunction periods. We show that peaks at the 45 day Mercury-Jupiter conjunction period, the 116 day first harmonic period of the Mercury-Earth conjunction and the 289 day period of the sub harmonic Mercury-Venus/Mercury-Earth conjunctions are prominent in spectra of sunspot area. We observe two prominent peaks close to the 88 day Mercury periodicity and two prominent peaks close to the 176 day first sub harmonic of Mercury periodicity. To confirm that the peaks arise as sidebands we band pass filter the sunspot area record to isolate the 88 and 176 day components of sunspot area. The components occur in episodes of duration from 1.5 to 4 years, with successive episodes usually overlapping in time but, for significant intervals in the record, the episodes were discrete, i.e. not overlapping. The time variation of the components was compared with the time variation of the tidal effect of Mercury. We were able to show that whenever episodes were discrete the component of sunspot area during the episode was either in-phase or in anti-phase with the Mercury tidal effect. We interpret this as indicating a connection between planetary motion and sunspot emergence. When several discrete episodes of the components occurred during a solar cycle the spectrum of sunspot area during that solar cycle exhibited periodicities at sidebands to the 88 day or 176 day periods, the periodicity of the sidebands depending on the duration of the episodes. A simple model based on amplitude modulation of 88 day and 176 day period sinusoids was able to consistently reproduce the spectral peaks, or Rieger quasi-periodicities, observed in the spectra of sunspot area.

There have been numerous reports of quasiperiodicities in solar activity in the intermediate period range. However, no accepted explanation for the episodic occurrence of quasiperiodicities has emerged. This paper examines the possibility that the periodicities are associated with a Mercury Sun interaction of base period 88 days. To test this idea we band pass filter the 140 year long daily sunspot area data to obtain the 88 day period and 176 day sub harmonic period components of the data and compare the time variation of the components with the time variation of the orbital radius of Mercury, or more specifically with the time variation of the tidal effect of Mercury. We were able to show that, when successive episodes of the occurrence of the 88 day period component were discrete and not overlapping in time, the time variation of this component of sunspot area was either exactly in-phase or exactly in anti-phase with the time variation of tidal effect. A similar result was obtained for the 176 day period component. When several discrete episodes of the components occurred during a solar cycle the spectrum of the sunspot area data exhibited strong sidebands with periods dependent on the duration of the episodes. A simple model based on episode modulation and solar cycle modulation of 88 day and sub harmonic period sinusoids reproduced most of the spectral peaks observed in the intermediate range of sunspot area periodicity. This is compelling evidence of a link between the motion of Mercury and the periodic emergence of sunspots. It is proposed that the link involves magnetic surface waves with mode periods close to the sub harmonic periods associated with Mercury and the triggering of sunspot emergence by the waves.

We present evidence that the allowed periods of equatorially trapped Rossby wave modes on the Sun coincide closely with the 88 day period and 176 day sub harmonic period of Mercury and evidence of Rossby waves on the Sun at the same periods. To test the hypothesis that Rossby waves trigger the emergence of sunspots we use band pass filtering to obtain the 88 day period and 176 day period components of hemispherical sunspot area and compare the variations to the tidal height variation on the surface of the Sun due to Mercury. We find that the two components of hemispherical sunspot area occur in several episodes or activations of duration 2 to 6 years during each solar cycle. When the activations are discrete the variation of the 88 day and 176 day period components are phase coherent with the tidal height variation and a 180 degree phase change is evident between successive activations. We use this result to demonstrate that Rieger type quasi-periodicities in sunspot activity are, in most reported cases, periodicities associated with sidebands of the 88 day and 176 day period variation. We find that, in variables that arise from activity in both solar hemispheres, phase coherence with the Mercury tidal effect occurs at infrequent intervals but, when it does occur, it is possible to observe phase coherence of variables such as sunspot number, solar flares, solar wind speed and cosmic ray flux with the Mercury tidal effect. Sensitivity of a range of variables to the tidal effect is calculated for such intervals in solar cycle 23.

Recent work indicates that 27 day variations in cosmic ray flux during 2007 2009 are phase locked to 27 day variations in cloud and surface temperature at Shetland. Here we extend the study to other regions including Central England, US and Australia and to several other annual intervals that exhibit strong 27 day variation in cosmic ray flux and sunspot area. Band pass filtering was used to obtain 27 day components of daily maximum temperature in each region and 27 day components of cloud variation were determined, in Australia only, from solar exposure records. When cosmic ray flux is the dominant influence phase locked variations in surface temperature occur in each of the regions with, however, in phase or anti phase variation in different regions. Similar phase locking of 27 day variation in surface temperature to sunspot area variation occurs when sunspot activity is the dominant influence with indications that changes from in phase to anti phase variation are linked to flipping of sunspot activity from one active longitude to another. The 27 day component of cloud in Australia was phase locked to the 27 day component of temperature in Central England in two of the intervals studied, 1997 and 2005, indicating the global nature of the connection between solar disturbance and the lower atmosphere. It was observed that very large swings in 27 day temperature components are often correlated with 27 day variation in solar activity.

We show that there is a previously unreported quad-annual variation in the ~13.5 day component of the aa index of geomagnetic activity. We derive a model based on the planetary tidal effect at the solar surface due to Mercury and Jupiter that, when combined with an equinoctial response of the magnetosphere, predicts the times of occurrence of predominantly quad-annual variation or predominantly semi-annual variation in the ~13.5 day component of the aa index. In support of the model we show that, during years when the quad-annual variation in the ~13.5 day component of aa index is predominant there is a large component at the 88 day periodicity of Mercury in the ~13.5 day component of solar wind speed. As further support for the model we establish that significant peaks in the aa index spectrum are due to an 88 day modulation of 27 day period solar activity. The model also predicts the occurrence of planetary tidal effect maximum in anti-phase with solar cycle maximums around 1970 and we show this is consistent with prior observations of higher solar emissions during the 1975 solar cycle minimum than in the following solar cycle maximum. This effect offers a possible explanation of the reduced solar cycle maximums around 1970 and during the Dalton Minimum.

We develop a basic model of the time variation of geomagnetic activity and show that the model predicts, with decreasing levels of exactitude, the time variation of the ~27 day period components of geomagnetic aa index, cloud cover and surface temperature during several years near solar cycle minima. We interpret this as indicating that there is a connection between the ~27 day variation of geomagnetic activity and the ~27 day variations of cloud cover and surface temperature with the decreasing correlations between model variation and aa index, cloud cover and surface temperature variation due to delays and phase shifts between the three variables some of which are obvious, such as the 180 degree seasonal phase shift between cloud cover and surface temperature, and others less certain. We find that, while the components of cloud cover and surface temperature influenced by geomagnetic activity amount to, on average, about 20% of the overall variations, the influence may be several times higher during the semi-annual maxima in geomagnetic activity that occur around the equinoxes.

We report the first observation of a 27 day period component in daily maximum temperature recorded at widely spaced locations in Australia. The 27 day component, extracted by band pass filtering, is correlated with the variation of daily solar radio flux during years close to solar minimum. We demonstrate that the correlation is related to the emergence of regions of solar activity on the Sun separated, temporally, from the emergence of other active regions. In this situation, which occurs only near solar minimum, the observed 27 day variation of temperature can be in phase or out of phase with the 27 day variation of solar activity. During solar maximum correlation of temperature and solar activity is much less defined. The amplitude of the 27 day temperature response to solar activity is large, at times as high as 6 degrees C, and much larger than the well documented temperature response to the 11 year cycle of solar activity. We demonstrate that the 27 day temperature response is localised to the Australian continent and increases towards the centre of the continent. Several aspects of the 27 day temperature response are consistent with a 27 day period solar input to the Earth continents generating large scale planetary waves.