Since ancient times it has been known that the sun is the main source of energy driving the Earth's climate system (e.g. Gilliland, 1989). Changes in the solar constant were predicted by Milankovitch, who calculated the gravitational effect of the planets on the orbital parameters of the Earth. Until recently, it has been generally believed that the energy output from the sun is fairly stable. Since 1978, however, high-precision measurements of solar irradiance by satellite-based radiometers have shown that the solar constant has changed in phase with solar activity by 1-1.5 per mil. These measurements and observations raise three fundamental questions:
(1) How large can solar variability be on longer time scales?
(2) What is the sensitivity of the climate system to changes in solar forcing?
(3) To what extent are present and past climate changes caused by solar forcing?
Concerning question 1, observations of 20 solar type stars, exhibiting clear activity variations, indicate that changes of the solar energy output of about 1 per cent cannot be excluded, especially when considering longer time scales. The second question is difficult to answer, because the relative changes in the solar constant are significantly larger for shorter wavelengths. In addition, several processes within the atmosphere directly or indirectly affect the radiation balance and may amplify the solar effect. To the third question, there is growing evidence that periods with special climatic conditions have coincided with time intervals of extreme solar activity. Quiet sun periods, like the Maunder minimum, have tended to be colder and synchronous with the expansion of Earth's glaciers. Cosmogenic radionuclides (10Be and 14C) make it possible to extend the record of solar activity from a few centuries to several millennia, and to investigate solar and natural climate variations during recent millennia.
Variations in solar radiation have been regarded as a significant contributor to climate change. Sunspots, dark areas on the sun's surface, are indicators of changes in solar activity. Sunspot observations over the last two centuries indicate an 11 yr periodicity, with a longer but less pronounced 78 yr periodicity. Records of past solar changes include measurements of the cosmogenic isotopes of 14C and 10Be. 14C production can be obtained by comparing radiocarbon dates from tree rings with the calendar ages (Stuiver et al.,
1991), while 10Be production can be measured in ice cores (Beer et al., 1992). Dansgaard and Oeschger (1989) found that the close correspondence between the two records may be attributed to solar output variations. Spectral analysis of the 14C record suggests periodicities of 11 and 22 years (Hale cycles), an 88-year cycle (Geisberg cycle), and 200 and 2500-year cycles (Rind and Overpeck, 1993). From studies of ice cores (Dansgaard et al., 1984), ocean cores (Pestieux et al., 1987), tree rings (Sonett and Finney, 1990), varved sediments (Anderson,
1992) and lake-level variations (Magny, 1993), there seems to be some empirical support for a relationship between solar output and climate change. The Maunder minimum, the most recent episode of reduced sunspot activity, occurred during the coldest part of the Little Ice Age. During the Maunder minimum the reduction of solar insolation was in the order of 0.25 per cent, equivalent to a global temperature decline of around 0.5°C. Empirical evidence suggests, however, a temperature decline during the Little Ice Age in the range of 0.5-1.5°C (Rind and Overpeck, 1993). This suggests that other factors must be invoked to explain the Little Ice Age cooling, such as the North Atlantic thermohaline circulation (Stuiver and Brazunias, 1993).
Irradiance varies with sunspot number, but their direct climatic effect is minor over an 11-year sunspot cycle. Over the last 1000 years there have been periods when sunspot numbers were near zero: the Wolf (ad 1280-1350), Sporer (ad 1416-1534) and Maunder (ad 1654-1714) minima. Wigley (1988) related 14C anomalies to glacier fluctuations throughout most of the Holocene. Wigley and Kelly (1990) found a statistically significant correlation between the global glacier advances of Rothlisberger (1986) and 14 C concentration during the Holocene. They interpreted the results, however, only as a strong indication since many uncertainties were involved in the analyses. Karlen and Kuyi-lenstierna (1996) compared Holocene climate changes in Scandinavia with changes in solar irradiation. For most of the last 9000 years they found a fairly good correspondence between cold events and <514C anomalies. The general Holocene cooling trend was suggested to be a combined result of land uplift after deglaciation and orbitally forced irradiation changes. Kelly and Wigley (1990) found that the influence of the enhanced greenhouse effect on global mean temperature dominated over the direct influence of solar variability. Prior to the period of direct sunspot observations, the differences between radiocarbon years and calendar years, as measured in tree rings, reflect anomalies in atmospheric 14C concentration (e.g. Stuiver and Brazunias, 1993; Kromer and Becker, 1993).
Variations in solar radiation caused by changes in the Earth's orbital parameters (mainly precession) had significant effects on Holocene global climate (COHMAP Members, 1988). In the early Holocene this involved a greater input of solar radiation to the top of the atmosphere during northern hemisphere summers and southern hemisphere winters.
Maximum summer solar radiation at high latitudes of the northern hemisphere occurred at about 11,000 bp (Berger, 1978), giving summer solstice radiation 7-8 per cent greater than at present at 60°N. During the Holocene, summer insolation gradually reduced to present values. Prior to approximately 4500 bp, maximum insolation at 60°N occurred during mid-summer (June-July). Between 4500 and 4000 bp, however, maximum insolation at 60°N changed to late summer/autumn (August-September) (Berger, 1978).
A high-resolution oxygen isotope record (mainly reflecting temperature variations) of the GISP2 ice core from Summit Greenland was analysed for solar influences (Stuiver et al., 1997). The atmospheric 14 C record was used as a proxy of solar change and compared with the oxygen isotope signal obtained from centimetre-scale isotope measurements from the period subsequent to ad 818. The analysis suggested a solar component to the forcing of Greenland climate during this millennium. The climatic response of the cold interval associated with the Maunder sunspot minimum, the Medieval warm period and the Little Ice Age temperature decline seem to have been related to solar climate forcing. For the rest of the Holocene, the oxygen isotope record shows more frequent fluctuations than the 14C record. Ocean-atmospheric circulation forcing of climate may therefore have dominated over other forcing mechanisms during this interval.
Was this article helpful?