The Influence Of Solar Activity On Climate And The Ice Cover

Solar activity (SA) includes a complex of physical phenomena that take place on the Sun that lead to variations in solar emissions (electromagnetic and corpuscular). Various indicators are used for quantitative characterization of SA. These include sunspot indices, solar cycle duration, changing diameter of the Sun, geomagnetic indices, solar wind indices, etc. The most widely used is the Wolf number, which expresses a relative number of solar spots and their groupings on the visible solar disc. Changes in the Wolf number over time have allowed detection of their cyclicity ("Schwabe-Wolf's law''; Vitinsky, 1973; Rapp, 2008). The average duration of this cycle is presently 11.1 years, but this has varied widely over the past centuries. The cycles are numbered in a system known as Zurich numbering (Vitinsky, 1973; Rapp, 2008). Other indicators, discussed below, are also applied in studies of SA that characterize various aspects of solar activity and its influence on geophysical phenomena. Cycles detected for various indicators include those lasting 22 years and 80-90 years (Vitinsky, 1973).

Scientists' opinions on the role of SA in climate change on Earth differ significantly, from complete denial that it has any role (Monin, 1969) to attributing full determination and control (Bashkirtsev and Mashnich, 2004; Yegorov, 2004). The first to investigate the relationship between solar activity and sea ice extent was Vize (1944b,c). Comparing the sea ice extent of the Barents Sea using annual Wolf numbers, he found that the correlation coefficients characterizing this relationship have quite high values during some periods; however, the sign of the relationship changes from one period to another.

Maksimov (1954, 1955, 1970) and his students undertook major studies of solar activity's influence on the sea ice extent of the Arctic Seas. As early as 1954-1955, he proposed a "component-harmonic method of calculation and forecast of sea ice extent in some regions based on a periodogram analysis of a series of annual observational data.'' In addition, to "solar-based" 11-year and century-long fluctuations, the method took into account a 6-year cycle generated by the pole tide and a 19-year cycle connected with a long-period lunar declination tide. Further development of this method resulted in a "component-genetic" method: instead of using the results of a periodogram analysis, some component relationships were used to characterize an imaginary periodic part of the forecasted characteristic, expressed both by Wolf numbers and by specially calculated coefficients (Maksimov, 1970). A relationship between the total sea ice extent of the Arctic Seas and solar activity (the Wolf numbers) was claimed by Kovalev (1967). He found that the sign of the relationship changes from one period to another, and explained it by the fact that the inverse relationship of sea ice extent and SA is invoked when the 11-year fluctuations of sea ice extent of helio-physical origin are in a phase opposite to fluctuations of shorter, 6-8-year (geophysical) periods.

In subsequent years, several studies were published in which the Wolf numbers were used as an indicator of SA to make forecasts of sea ice extent of some Arctic Seas and their regions (Chaplygin and Yanes, 1968; Santsevich, 1970; Karklin, 1977; Karklin and Teitelbaum, 1987). Kupetsky (1969, 1974, 1977) describes a method for superimposing even and odd cycles of SA to forecast sea ice extent. In addition to the Wolf numbers, an index of geomagnetic planetary perturbation M, proposed by 0l' (1969), was also used. However, the statistical significance of such methods was questioned by Kovalev and Spichkin (1977).

Lassen and Friis-Christensen (1991) also noted the similarity between long-term variations in the Greenland Sea ice-cover area and SA expressed by Wolf numbers, and they pointed out the ambiguous relationship between SA and sea ice extent west and east of Greenland. The same study reveals a close correlation (—0.95) between the surface air temperature in the Northern Hemisphere (for 1861-1989) and the duration of the SA cycle: warming exhibits shorter cycles (about 10 years) and cooling longer cycles (about 11.5 years). Very early in the twentieth century Mei-nardus and Shott had noted opposite sea ice extent conditions west and east of Greenland determined by atmospheric circulation regimes (Alekseev et al., 1998).

A study by Sleptsov-Shevlevich (1991) claimed an important role for a 22-year ("magnetic") SA cycle in interannual variations in sea ice extent of the Arctic Seas as well as other oceanographic, meteorological, and geophysical phenomena. This cycle can be correlated with the change in sign of the Sun's magnetic fields at the transition from one 11-year cycle of SA to another. This author concluded that a 2-year cycle of sea ice extent and a 6-7-year cycle of fluctuations in the location of the North Pole occur together within a 22-year cycle.

These ideas were further developed in studies by Latukhov and Sleptsov-Shevlevich (1995) and Sleptsov-Shevlevich and Zakharov (1996) who focused on "100-year" variations in sea ice extent that correlate with the magnetic perturbation index Kp, which is related to SA: when Kp increases, the sea ice extent of the sub-Atlantic Arctic Seas east of Greenland decreases while it increases in East-Canadian waters, and vice versa. The first of these studies also analyzes the causes for error in a super-long-range forecast of sea ice extent in the sub-Atlantic Arctic Seas by Maksimov (1955) and proposes a new forecast based on the updated duration of the 100-year cycle of solar activity: the epochal maximum sea ice extent in this region is expected around 2023.

Close relationships between the total number of large anomalies (>1.2o-) in Arctic Seas sea ice extent (regardless of the sign of the anomalies) during 11-year SA cycles and the sum of the Wolf numbers in the corresponding cycles were revealed by Karklin and Kovalev (1994; Figure 5.2). Averaging of the Wolf numbers for each 11-year SA cycle, performed in this study as well as in some other studies, excludes an analysis of the dependence of ice conditions on SA fluctuations within the 11-year solar cycles. The relationship depicted in Figure 5.2 reflects the influence of longer SA variations.

Shirochkov and Makarova (1998) outlined a new approach, similar to that for other Earth climate characteristics, for studying the relationship between long-period changes in SA and sea ice extent in the Arctic Seas. The solar wind dynamic pressure (PJW), which depends on the density of particle fluxes from the Sun to the Earth and

Figure 5.2. Relationship between the number of large sea ice extent anomalies (N) in the Arctic Seas in AugustSeptember to the total value of Wolf numbers in 11-year solar cycles. Figures near the points indicate numbers of 11-year solar cycles using Zurich numbering.

Figure 5.2. Relationship between the number of large sea ice extent anomalies (N) in the Arctic Seas in AugustSeptember to the total value of Wolf numbers in 11-year solar cycles. Figures near the points indicate numbers of 11-year solar cycles using Zurich numbering.

their speeds, was used as an indicator of SA. The value of Psw is measured by special satellite instruments. An analysis of the relationship between the Psw index and the ice-cover area in the Greenland, Barents, and Kara seas, and the sea ice extent of East Canadian waters for April-July, suggested a strong inverse relationship for the Greenland and Barents Seas: increased solar wind pressure is accompanied by decreased sea ice extent (the correlation coefficients are —0.77 and —0.64, respectively), whereas for East Canadian waters (mainly Baffin Bay), a weak direct correlation (0.30) is noted. This appears to confirm the conclusions reached by comparing sea ice extent of the study areas with SA expressed by Wolf numbers, considering that there is a significant although unstable negative correlation between the SA indicators used: from —0.60 for the period 1964-1980 to —0.33 for the period 1980-1996 (Shirochkov and Makarova, 1998).

This far-from-complete review of studies concerning the relationship between sea ice extent and SA indicates the presence of possible little-studied natural mechanisms that might affect these putative relationships.

At present, there is a broadly held opinion that variations in solar activity are primarily expressed as changes in atmospheric pressure fields and atmospheric circulation that result in anomalies of other hydrometeorological elements (Karklin, 1973). Atmospheric pressure changes at sea level with periods of about 11 and 22 years have received the most attention.

Studies of air pressure field changes caused by solar activity (Maksimov, 1970; Sleptsov-Shevlevich et al., 1991) reveal standing waves in Earth's atmosphere, with periods corresponding to known cycles of solar activity. These include the 11-year sunspot cycle (its duration varies from 8 to 17 years), and a 22-year cycle (varying from 18-28 years), which is related to two things: 1) the discovery by Hale that the sign of solar spots' magnetic polarity changes from one 11-year cycle to another (Vitinsky, 1973), and 2) features of changes in magnetic perturbations that were revealed by Ol' (1969).

The first of these waves is characterized by a stable positive relationship between atmospheric pressure and solar activity (the Wolf numbers) in the high latitudes of the Northern Hemisphere and by a negative relationship in temperate latitudes. The position of the wave's nodal line changes from cycle to cycle, which results in an extensive zone of unstable solar-atmospheric relations encircling the high-latitude area of stable direct relations with the boundary in the North European Basin at approximately 70°N (Maksimov and Sleptsov-Shevlevich, 1963; Gasyukov and Smirnov, 1967; Karklin, 1978).

Table 5.1 indicates the significance of solar activity in the variability of atmospheric pressure fields over the Arctic Ocean using the average values of large-scale wind-field vorticity in years of increased and decreased SA expressed by Wolf numbers. The characteristics that indicate the intensity of cyclonic (anti-cyclonic) circulation were calculated for two regions of the Arctic Ocean: the North European (J{) and Arctic (/2) basins, averaged for the winter months (October-March). For these calculations, equations expressing the Laplacian of atmospheric pressure distribution similar to Equation 4.11 were applied. To calculate the /2 value, a triangle replaced the squares used for calculating /0 and / (see Figure 4.22).

Table 5.1. Wolf numbers (£ W), cyclonicity indices in the western (/1) and eastern (J2) regions of the Arctic Ocean, and total sea ice extent (thousand km2) in the western (L1) and eastern (L2) regions during cycles of increased and decreased solar activity

Period, years

£ w













































































Table 5.1 provides Wolf number sums for 5-year time intervals, characterizing the periods of increased and decreased SA (cycles 17-22). It also shows the values of J1 and J2 averaged for October-March, expressing the intensity of cyclonic (anti-cyclonic) circulation in the indicated regions (Figure 4.22), and the sea ice extent values averaged for the same 5-year periods for the western (L1) and eastern (L2) regions. Based on these data, corresponding average values were calculated to characterize the periods of increased and decreased SA during 17-22-year cycles (Table 5.2).

As Table 5.2 shows, at an almost triple (on average) change in the Wolf numbers during the 11-year cycle—from a 5-year period of increased SA (501) to a 5-year period of decreased SA (179)—the intensity of the Icelandic cyclonic circulation slightly increases on average (from 64 to 82 units), and the intensity of the anti-cyclonic circulation in the Arctic High drops significantly (on average from —61 to —35 units). That is, the cyclonicity increases in both regions. These results match the character of the 11-year fluctuations in the baric field discussed above: with an increasing Wolf number, the atmospheric pressure in high latitudes increases, and

Table 5.2. Wolf number (£ W) averages for periods of increased and decreased solar activity, cyclonicity indices for the western (/1) and eastern (J2) regions of the Arctic Ocean, and total sea ice extent (thousand km2) in the western (L1) and eastern (L2) Eurasian Arctic regions


£ w





Increased solar activity






Decreased solar activity












with a decreasing Wolf number, it decreases (Karklin, 1978). Increased cyclonicity is accompanied by decreased sea ice extent in the western region and increased sea ice extent in the eastern region. Increases in anticyclonicity have the opposite effect on the ice cover area in both regions. It is important to note that the influence of solar activity on the baric field is more pronounced in the Arctic Basin than in the North European Basin. This pattern can be probably be explained by the phenomenon noted above that a nodal line dividing the areas of the standing solar-determined baric wave with a different sign of corresponding anomalies often passes across the North-European Basin.

In analyzing the influence of the 11-year SA cycle on atmospheric circulation, it is important to remember that satellite data available since 1978 show that the difference between maximum and minimum solar radiation in the 11-year cycle is only 2W/m2 (0.15 % of the average solar constant value) (Bashkirtsev and Mashnich, 2004; Rapp, 2008). Therefore, explaining the manifestation of this cycle in the Earth's atmosphere requires accounting for the corresponding variations in UV radiation, particle fluxes, galactic rays, and the presence of trigger mechanisms that could cause energetically insignificant variations in incoming solar radiation to result in significant weather changes in the Arctic.

The character of the spatial distribution of the relationship between atmospheric pressure and solar activity in a 22-year wave is similar to an 11-year wave both in the location of the loops at high and temperate latitudes and in the location of the nodal line (zones of unstable relations). During even numbered 11-year cycles (using Zurich numbering), atmospheric pressure decreases in the near-pole region and increases at temperate latitudes. On the contrary, during odd numbered cycles, the atmospheric pressure increases at high latitudes and decreases at temperate latitudes. The nodal line of this wave passes between 55°N and 60°N (Maksimov and Sleptsov-Shevlevich, 1971; Ol' and Sleptsov-Shevlevich, 1972; Karklin, 1973). Similar fluctuations in atmospheric pressure were also observed in the Southern Hemisphere (Maksimov and Sleptsov-Shevlevich, 1963).

Differences in atmospheric circulation during odd and even SA cycles are convincingly reflected in average NAO index values (Gudkovich et al., 2004). The average twentieth century NAO anomaly was predominantly positive in the even cycles and negative in the odd cycles. These results suggest a high probability of intensified zonal transports in the atmosphere of temperate latitudes during even SA cycles and their weakening during odd cycles, which is consistent with the behavior of the 22-year wave.

According to available estimates (Karklin, 1978), the contribution of solar-induced fluctuations to the variability of atmospheric pressure is 10 to 30% for both the 11- and 22-year waves, which influences the general circulation of the atmosphere, especially the intensity of west-to-east air flow at temperate and high latitudes. This is confirmed by investigations of the relationship between SA and recurrence of the main atmospheric circulation forms (Girs, 1960) and the number of elementary synoptic processes during a year (Dmitriyev, 1994). Some studies found that the location of atmospheric action centers and their development changes with the 11-year SA rhythm (Abramov, 1967; Karklin, 1975).

The relationship between SA and Arctic baric fields influences a complex of factors that determine the long-term variability of the ice state in the Arctic Seas. One of these factors is ice export from the Arctic Basin to the Greenland Sea. A monthly comparison of ice exported through Fram Strait from 1946 to 1999 was carried out using the method developed by Gudkovich and Nikolayeva (1963) and averaged for each of the last five 11-year cycles of SA, using the Wolf-number average for each cycle. It indicated that the ice export increases with an increase in the Wolf number average during odd SA cycles and decreases during even cycles (Figure 5.3). This pattern helps to explain the dependence of the ice cover state of the Arctic Seas on SA.

S, 1000km2/month 651

1QOodd W, cycle

Figure 5.3.

Dependence of monthly ice export area (S) from the Arctic Basin to the Greenland Sea on the Wolf number average (Wc) for a cycle during the odd (19, 21) and even (18, 20, 22) cycles of solar activity.

Bashkirtsev and Mashnich (2004) provide evidence of relationships between SA and various hydrometeorological indicators in different regions. Zherebtsov and Kovalenko (2001) note a close correlation (0.97) between averaged 11-year solar cycle Wolf numbers and surface air temperatures in the Baikal area.

Studies by Reid (2000) and Makarov and Tlatov (2000) conclude that variations in surface air temperature over the oceans are similar to Wolf number variations. Multiyear fluctuations of global surface air temperature (GSAT) exhibit periodicity similar to solar activity: the Schwabe cycle (11 years), the Hale cycle (22 years), and the Fritz cycle (about 60 years) which are also evident in the Sun's large-scale magnetic field and in the aurora borealis (Bashkirtsev and Mashnich, 2004). Diagrams shown by Bashkirtsev and Mashnich (2004), based on the SA data obtained by Nagovitsyn et al. (Nagovitsyn et al., 2004; Nagovitsyn, 2007), illustrate the similarity of smoothed variations of SA and GSAT for 1100-2000 and extrapolated to 2300. SA variations for 1611-2005 in a form of yearly sunspot numbers are reproduced in Figure 5.4. Smoothed curves in this figure suggest cycles lasting about 200 years, which can explain the intra-secular trends considered above. The figure also reflects Earth's coldest period in the last millennium, coinciding with the known Maunder


Figure 5.4. Observed and simulated yearly sunspot areas (Greenwich general system) for 1611-2005 (Nagovitsyn et al., 2004; Nagovitsyn, 2007). (1) Approximation by a polynomial to the sixth power. (2) FFT-filter by 11 points (years).

, number sea











Figure 5.4. Observed and simulated yearly sunspot areas (Greenwich general system) for 1611-2005 (Nagovitsyn et al., 2004; Nagovitsyn, 2007). (1) Approximation by a polynomial to the sixth power. (2) FFT-filter by 11 points (years).

minimum in SA variations (1645-1715). The same article by Bashkirtsev and Mashnich (2004, p. 136) also provides an important scheme for interpreting the results set forth in Section 5.4: "the largest GSAT occurs during Hale and Fritz cycle synchronism: 1880, 1940,2000 and 2060.'' This provides additional evidence for a 60-year cycle in Earth's climate fluctuations.

When considering the relationship of variations in the baric field and SA both lasting more than a century, an increase in the Wolf number corresponds to intensified cyclonic activity over the Arctic Basin. For example, Gudkovich et al. (2005) found that the average rate of increase in the Wolf number in the twentieth century was about 50 units/100 years. In the opinion of Gribbin and Lamb (1978), the assumption that such long-term relationships exist strongly suggests the importance of investigating the processes occurring in the atmosphere of our planet.

The discussion above suggests that large-scale changes in atmospheric circulation can be caused by SA at some time scales. However, the physical mechanisms for interaction between solar processes and Earth's troposphere are not yet resolved. Studies devoted to this issue include Maksimov (1970a), Mustel (1974), Vitinsky et al. (1976), German and Goldberg (1981), Krymsky (1994), Kondratyev and Nikolsky (1995), as well as many others. Proposed mechanisms range from the possible influence of SA on gravity and the solar constant to hypotheses regarding the impact of chemical, condensation, and electrical processes induced by anomalies of wave or particle solar energy fluxes. After satellite and radiosonde observations showed that solar wind causes heating and expansion of the upper layer of Earth's atmosphere, questions arose regarding the mechanisms for energy transfer from the stratosphere and magnetosphere to the troposphere.

Based on the law of conservation of momentum in the atmosphere, Sytinsky (1987) concluded that heating of the upper layers of the atmosphere by a solar particle flux causes the atmosphere's moment of inertia to increase, resulting in a decrease in the angular rotation speed of the atmosphere relative to the Earth, and the atmospheric pressure distribution gradually adapts to this change. A decrease in particle flux, whose energy is largely absorbed near the poles, has the opposite effect. These effects are also influenced by the direction of interplanetary magnetic fields, which change at the transitions between odd and even solar cycles. According to Bothner and Schwenn (1998), the magnetic fields of solar emissions during odd cycles are primarily directed opposite to Earth's magnetic field, which allows solar particles to enter Earth's atmosphere and leads to an increase in GSAT.

Theoretical studies by Krymsky (1994) concluded that the solar wind transfers not only mass, energy, and impulse to the magnetosphere, but also the moment of impulse, which is then transferred to the atmosphere and to the Earth as a result of turbulent viscosity. In this way, solar activity causes variations in the super-rotation of the atmosphere, in zonal circulation intensity, in atmospheric pressure distribution, and even in Earth's angular speed. The angular speed also directly depends on the rotation moment and corresponding torsional oscillations excited by the solar wind in the electromagnetically connected mantle and core of the Earth. Krymsky (1994) categorically rejects the opinion of some scientists regarding the transfer of the moment of impulse to the atmosphere from the Earth, because this would require "Earth, at insignificant angular speed changes, to impart to the atmosphere a rotation with a speed one million times greater'' (p. 42). The proposed theory purports to explain, at least qualitatively, many phenomena in the atmosphere and solid Earth that are influenced by SA: cyclic changes in atmospheric circulation, angular speed of Earth's rotation, nutation of its axis, etc.

A different approach is used by Sleptsov-Shevlevich (1991) who assumes that changes in the angular speed of Earth's rotation due to solar wind lead to corresponding gravity changes as a sum of gravity and centrifugal force, causing cyclic "deformations of all Earth's shells,'' i.e., redistribution of the water and air masses, climate changes, etc. A decrease in the Earth's angular speed leads to displacement of air masses toward high latitudes, and its increase displaces the air masses toward mid latitudes, which influences the processes of cyclogenesis, prevailing trajectories of cyclones, etc. In our opinion, one of many weak points in this hypothesis is the fact that the changes in the atmosphere can occur only after Earth's angular speed variations are transferred to its air shell. As shown above, real mechanisms for this are absent.

Shirochkov and Makarova (1998) conclude that changes in solar wind pressure significantly affect the thermal regime of the middle atmosphere and even the state of the tropopause in Earth's polar regions. When solar wind pressure causes heating of the lower stratosphere, the tropopause "thins," which has some specific climatic consequences. These authors suggest a version of global electric circulation, which they have improved, as a possible mechanism for this relationship. An important element of this mechanism is a "giant spherical capacitor'' with a magnetopause as its external plate and Earth's surface as its internal plate. The energy accumulated by this capacitor increases with a decrease in the distance between the plates, which is subject to the influence of solar wind pressure. The local character of the influence of this parameter on hydrometeorological processes can be explained by the differences in electrical conductivity of the underlying surface of the water (for example, in the Greenland Sea and Davis Strait).

The latter assumption is, in our opinion, the weakest point in this interesting hypothesis. It appears that the known phenomenon of the "tropopause funnel'' (a significant decrease in the tropopause above the deep cyclone of low mobility; Khromov and Mamontova, 1974) should be examined as a possible physical mechanism for the relationship between the state of the tropopause height and processes in the troposphere. It is possible that the changes in the tropopause structure caused by solar wind pressure variations lead to significant changes in cyclogenesis in some specific regions, for example, at atmosphere action centers. If the increased solar wind pressure is accompanied by deepening of the Icelandic Low and its decrease with depression, then the ambiguity of the relationship of Psw to sea ice extent in the Greenland and Barents seas on the one hand, and the ice cover area in Davis Strait, on the other hand, becomes clear. Deepening of the depression contributes to increased heat advection to the former regions and cold advection to the latter, and vice versa. The facts confirming this hypothesis are presented below.

To reveal the relationship between solar wind pressure and the Icelandic Low, the anomalies of atmospheric pressure at sea level for October-March, averaged by five points in square 1 of Figure 4.22 (1965-1995), were compared with the anomalies Psw for the same period. In 70% of the cases, the anomalies in the two categories had opposite signs. This means that an increase in solar wind is typically accompanied by a decrease in atmospheric pressure in the area of the Icelandic Low, and, on the contrary, weakening solar wind results in an increase in atmospheric pressure. This supports the hypothesis that the solar wind impacts the baric field. There are grounds to suppose that the Arctic Oscillation (AO) phenomenon considered in section 4.2 reflects the influence of SA on Earth's baric field: the distribution of atmospheric pressure anomalies in the AO phases shown in Figure 4.8 closely matches the changes in atmospheric pressure fields at high and temperate latitudes during corresponding SA cycles (Karklin, 1973, 1978). The fact that both phenomena include cycles lasting about 10 and 20 years also supports an AO/SA relationship.

The 10-year cycle, the physical mechanism that causes corresponding changes in atmospheric pressure fields, is probably connected with the Wolf cycle of solar activity, which allows intrusion of charged solar wind particles into the upper layers of Earth's atmosphere at high latitudes. This would appear to cause an experimentally determined effect (Sytinsky, 1987) of atmospheric heating and expansion in the zone of particle intrusion, the corresponding appearance of horizontal atmospheric pressure gradients, and redistribution of air masses, which can trigger convection processes in the troposphere.

The "20-year" cycle of baric field oscillations over the Arctic Ocean and the sea ice extent of its seas, which are components of the AO, can also be related to the known 22-year fluctuation in solar activity (Hale's law) evident in solar magnetic field sign changes. Unfortunately, there are no convincing hypotheses on the possible mechanisms for such relationships. In some studies (Jose, 1965; Vasilieva, 1997; Vasilieva et al., 2002), processes occurring in the interior of the sun that cause magnetic field changes, fluctuations in the solar diameter, and other phenomena are explained by solar dissipative processes being determined by the distance between the center of the sun and the mass center of the solar system. The period of these fluctuations is close to the period of synodic revolution of Jupiter and Saturn (19.86 years).

In Raspopov et al. (2004), 20-25-year cyclic climate changes were revealed on the basis of a dendrochronological analysis of extensive data available on the northern forests of Arctic Eurasia for the period from 1458 to 1975. Their analysis of fluctuations in solar activity, based on measurements of radionuclide (14C) concentrations in annual tree-trunk rings, identified the cyclic climate changes and solar activity under consideration here. Raspopov et al. (2004) conclude that as a result of nonlinear impact, the influence of solar activity and related cosmic ray fluxes can increase significantly (three- to fivefold) the inherent internal oscillations of the atmosphere-ocean-continent system. This may express the nature of double-ten-year global climate fluctuations. Potential mechanisms for such fluctuations are considered below.

Was this article helpful?

0 0

Post a comment