FIGURE 14.55 Reconstructed total solar irradiance from 1610 to 1995 using an 11-year solar cycle plus a longer term component of variability (adapted from Lean et al., 1995a).
Age" from 1450 to f850. The long-term component has been scaled to agree with the estimate of an overall increase in total irradiance from the Maunder Minimum to the present of 0.24% (Lean et al., 1992, 1995b).
Lean et al. (1995a) have used this reconstruction to estimate how much of the increase in surface temperatures in the Northern Hemisphere can be explained due to solar variability. They concluded that about half of the observed Northern Hemisphere surface temperature increase of 0.55°C since 1860 is due to solar variability but that it only accounts for about a third of the 0.36°C increase since f970. Similar conclusions have been reached by a number of researchers (e.g., see Kelly and Wigley, 1992; Schlesinger and Ra-mankutty, 1992; Scuderi, 1993; Crowley and Kim, 1996; Solanki and Fligge, 1998; Cliver et al., 1998; and Wilson, 1998). However, Fröhlich and Lean (1998) have reexamined the solar irradiance record since 1978; they concluded that the irradiances in 1986 and 1996 were similar and that changes in the solar flux during this period could not have contributed significantly to the observed changes in global mean surface temperature.
In short, it is clear that variations in solar output have played a major role in determining the earth's climate in the past, and understanding and quantifying this variability are critical for understanding anthropogenic influences on global climate. The observed temperature increases over approximately the past three decades are larger than expected from solar variability and have been interpreted by many researchers in this field to be the first signs of anthropogenic perturbations on climate.
As discussed in Section C.la, major volcanic eruptions have been observed to alter the earth's climate through injection of large amounts of S02 into the stratosphere. There it is oxidized to sulfate particles that scatter incoming solar radiation, leading to cooling at the earth's surface. These particles also absorb long-wavelength terrestrial infrared radiation, warming the stratosphere (Fig. 14.30). While this absorption of infrared increases the downward emission of infrared from the stratosphere into the troposphere, i.e., causes a positive radiative forcing, the effect is much smaller than the direct scattering of solar radiation. As a result, the major overall net effect of volcanic eruptions is cooling (Robock and Mao, 1995).
However, it should be noted that the effect is somewhat geographically and temporally variable. For example, Robock and Mao (1995) have examined climate records since about 1850 and correlated them to volcanic eruptions both before and after removal of the effects of the El Nino-Southern Oscillation (ENSO) signal. While the effect of volcanic eruptions on the global mean surface temperature is cooling, there are circumstances where the effect is not only smaller than the mean, but warming was observed. For example, in the first winter following a number of different volcanic eruptions, the Northern Hemisphere and Eurasia on average warmed, in contrast to northern Africa and southwestern Asia, which cooled. Robock and Mao propose that the warming is due to changes in the winter circulation pattern, associated with an enhanced polar vortex, which lowers the extent of normal winter cooling.
Figure 14.56 demonstrates the overall cooling effect of volcanic eruptions in the Northern Hemisphere over the past six centuries, reconstructed using the effects of temperature on tree ring densities (Briffa et al., 1998). The relationship between the average summer monthly mean land and marine temperatures in the Northern Hemisphere and tree ring density was determined for the period from 1881 to 1960. This was then applied to measured tree ring densities to obtain the temperature anomalies for the entire 600 years compared to the f881-1960 period. Some of the major volcanic eruptions are also marked on the diagram, clearly demonstrating their association with significant cooling. Similar conclusions have been reached using ice core data (e.g., White et al., 1997; Clausen et al., 1997; Taylor et al., 1997; Zielinski et al., 1997).
Volcanic eruptions provide an opportunity for testing not only our current understanding of the direct effects of aerosol particles due to backscattering but also the sensitivity of the climate system to such perturbations. Thus, after the initial short-term effects on temperature, the coupled atmosphere-ocean-land system responds on a longer time scale through a complex set of feedback mechanisms. As discussed by Lindzen and Giannitsis (1998), the effects of multiple volcanic eruptions should provide a better test of our understanding of such feedbacks than is provided by a single eruption.
Oceans have an enormous effect on climate through many different mechanisms that are beyond the scope of this book. Globally, oceans absorb heat and greenhouse gases such as C02 from the atmosphere (Fig. 14.11), both moderating such changes (e.g., Schneider et al., 1997; Bush and Philander, 1998) and providing a time lag in the response to atmospheric perturbations (e.g., Wigley, 1995). Other phenomena such as the El Nino-Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO) clearly also have substantial
Was this article helpful?