IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group I: The Physical Science Basis Solar Forcing

The direct measurement of solar irradiance by satellite began less than 30 years ago, and over this period only very small changes are apparent (0.1% between the peak and trough of recent sunspot cycles, which equates to only about 0.2 W m–2 change in radiative forcing; Fröhlich and Lean (2004); see Section 2.7). Earlier extensions of irradiance change used in most model simulations are estimated by assuming a direct correlation with evidence of changing sunspot numbers and cosmogenic isotope production as recorded in ice cores (10Be) and tree rings (14C) (Lean et al., 1995; Crowley, 2000).

There is general agreement in the evolution of the different proxy records of solar activity such as cosmogenic isotopes, sunspot numbers or aurora observations, and the annually resolved records clearly depict the well-known 11-year solar cycle (Muscheler et al., 2006). For example, palaeoclimatic 10Be and 14C values are higher during times of low or absent sunspot numbers. During these periods, their production is high as the shielding of the Earth’s atmosphere from cosmic rays provided by the Sun’s open magnetic field is weak (Beer et al., 1998). However, the relationship between the isotopic records indicative of the Sun’s open magnetic field, sunspot numbers and the Sun’s closed magnetic field or energy output are not fully understood (Wang and Sheeley, 2003).

The cosmogenic isotope records have been linearly scaled to estimate solar energy output (Bard et al., 2000) in many climate simulations. More recent studies utilise physics-based models to estimate solar activity from the production rate of cosmogenic isotopes taking into account nonlinearities between isotope production and the Sun’s open magnetic flux and variations in the geomagnetic field (Solanki et al., 2004; Muscheler et al., 2005). Following this approach, Solanki et al. (2004) suggested that the current level of solar activity has been without precedent over the last 8 kyr. This is contradicted by a more recent analysis linking the isotope proxy records to instrumental data that identifies, for the last millennium, three periods (around AD 1785, 1600 and 1140) when solar activity was as high, or higher, than in the satellite era (Muscheler et al., 2006).

The magnitude of the long-term trend in solar irradiance remains uncertain. A reassessment of the stellar data (Hall and Lockwood, 2004) has been unable to confirm or refute the analysis by Baliunas and Jastrow (1990) that implied significant long-term solar irradiance changes, and also underpinned some of the earlier reconstructions (see Section 2.7). Several new studies (Lean et al., 2002; Foster, 2004; Foukal et al., 2004; Y.M. Wang et al., 2005) suggest that long-term irradiance changes were notably less than in earlier reconstructions (Hoyt and Schatten, 1993; Lean et al., 1995; Lockwood and Stamper, 1999; Bard et al., 2000; Fligge and Solanki, 2000; Lean, 2000) that were employed in a number of TAR climate change simulations and in many of the simulations shown in Figure 6.13d.

In the previous reconstructions, the 17th-century ‘Maunder Minimum’ total irradiance was 0.15 to 0.65% (irradiance change about 2.0 to 8.7 W m–2; radiative forcing about 0.36 to 1.55 W m–2) below the present-day mean (Figure 6.13b). Most of the recent studies (with the exception of Solanki and Krivova, 2003) calculate a reduction of only around 0.1% (irradiance change of the order of –1 W m–2, radiative forcing of –0.2 W m–2; section 2.7). Following these results, the magnitude of the radiative forcing used in Chapter 9 for the Maunder Minimum period is relatively small (–0.2 W m–2 relative to today).