The observed multi-decadal SSR variations cannot be explained by changes in TSI, which are an order of magnitude smaller (Willson and Mordvinov, 2003). They therefore have to originate from alterations in the transparency of the atmosphere, which depends on the presence of clouds, aerosols and radiatively active gases (Kvalevag and Myhre, 2007; Kim and Ramanathan, 2008). Cloud cover changes (Section 2.5.7) effectively modulate SSR on an interannual basis, but their contribution to the longer-term SSR trends is ambiguous. Although cloud cover changes were found to explain the trends in some areas (e.g., Liley, 2009), this is not always the case, particularly in relatively polluted regions (Qian et al., 2006; Norris and Wild, 2007, 2009; Wild, 2009; Kudo et al., 2012). SSR dimming and brightening has also been observed under cloudless atmospheres at various locations, pointing to a prominent role of atmospheric aerosols (Wild et al., 2005; Qian et al., 2007; Ruckstuhl et al., 2008; Sanchez-Lorenzo et al., 2009; Wang et al., 2009b; Zerefos et al., 2009).

Aerosols can directly attenuate SSR by scattering and absorbing solar radiation, or indirectly, through their ability to act as cloud condensation nuclei, thereby changing cloud reflectivity and lifetime (Chapter 7). SSR dimming and brightening is often reconcilable with trends in anthropogenic emission histories and atmospheric aerosol loadings (Stern, 2006; Streets et al., 2006; Mishchenko et al., 2007; Ruckstuhl et al., 2008; Ohvril et al., 2009; Russak, 2009; Streets et al., 2009; Cermak et al., 2010; Wild, 2012). Recent trends in aerosol optical depth derived from satellites indicate a decline in Europe since 2000 (Section 2.2.3), in line with evidence from SSR observations. However, direct aerosol effects alone may not be able to account for the full extent of the observed SSR changes in remote regions with low pollution levels (Dutton and Bodhaine, 2001; Schwartz, 2005). Aerosol indirect effects have not yet been well quantified, but have the potential to amplify aerosol-induced SSR trends, particularly in relatively pristine environments, such as over oceans (Wild, 2012).

SSR trends are also qualitatively in line with observed multi-decadal surface warming trends (Chapter 10), with generally smaller warming rates during phases of declining SSR, and larger warming rates in phases of increasing SSR (Wild et al., 2007). This is seen more pronounced for the relatively polluted NH than the more pristine SH (Wild, 2012). For Europe, Vautard et al. (2009) found that a decline in the frequency of low-visibility conditions such as fog, mist and haze over the past 30 years and associated SSR increase may be responsible for 10 to 20% of Europe’s recent daytime warming, and 50% of Eastern European warming. Philipona (2012) noted that both warming and brightening are weaker in the European Alps compared to the surrounding lowlands with stronger aerosol declines since 1981.

Reanalyses and observationally based methods have been used to show that increased atmospheric moisture with warming (Willett et al., 2008; Section 2.5) enhances thermal radiative emission of the atmosphere to the surface, leading to reduced net thermal cooling of the surface (Prata, 2008; Allan, 2009; Philipona et al., 2009; Wang and Liang, 2009).

In summary, the evidence for widespread multi-decadal variations in solar radiation incident on land surfaces has been substantiated since AR4, with many of the observational records showing a decline from the 1950s to the 1980s (‘dimming’), and a partial recovery thereafter (‘brightening’). Confidence in these changes is high in regions with high station densities such as over Europe and parts of Asia. These likely changes are generally supported by observed changes in related, but more widely measured variables, such as sunshine duration, DTR and hydrological quantities, and are often in line with aerosol emission patterns. Over some remote land areas and over the oceans, confidence is low owing to the lack of direct observations, which hamper a truly global assessment. Satellite-derived SSR fluxes support the existence of brightening also over oceans, but are less consistent over land surface where direct aerosol effects become more important. There are also indications for increasing downward thermal and net radiation at terrestrial stations since the early 1990s with medium confidence.