In the SAR (Santer et al., 1996c), pattern-based detection studies took into account changes in well-mixed greenhouse gases (often represented by an equivalent increase in CO2), the direct effect of sulphate aerosols (usually represented by a seasonally constant change in surface albedo) and the influence of changes in strato-spheric ozone. Recent studies have also included the effect of increases in tropospheric ozone and a representation of the indirect effect of sulphate aerosols on cloud albedo. Many models now include the individual greenhouse gases (as opposed to a CO2 equivalent) and include an interactive sulphur cycle and an explicit treatment of scattering by aerosols (as opposed to using prescribed changes in surface albedo). Note that representation of the sulphur cycle in climate models is not as detailed as in the offline sulphur cycle models reported in Chapter 5. Detection and attribution studies to date have not taken into account other forcing agents discussed in Chapter 6, including biogenic aerosols, black carbon, mineral dust and changes in land use. Estimates of the spatial and temporal variation of these factors have not been available long enough to have been included in model simulations suitable for detection studies. In general, the neglected forcings are estimated to be small globally and there may be a large degree of cancellation in their global mean effect (see Chapter 6, Figure 6.8). It is less clear that the individual forcings will cancel regionally. As discussed in Section 12.4, this will add further uncertainty in the attribution of the response to individual forcing agents, although we believe it is unlikely to affect our conclusions about the effects of increases in well-mixed greenhouse gases on very large spatial scales.
Global mean anthropogenic forcing
The largest and most certain change in radiative forcing since the pre-industrial period is an increase of about 2.3 Wm-2 due to an increase in well-mixed greenhouse gases (Chapter 6, Figure 6.8 and Table 6.1). Radiative forcing here is taken to be the net downward radiative flux at the tropopause (see Chapter 6). Smaller, less certain contributions have come from increases in tropospheric ozone (about 0.3 Wm-2), the direct effect of increases in sulphate aerosols (about -0.4 Wm-2) and decreases in stratospheric ozone (about -0.2 Wm-2). There is a very uncertain and possibly large negative contribution from the indirect effects of aerosols. Other factors such as that due to increases in fossil fuel organic carbon, aviation, changes in land use and mineral dust are very poorly known and not yet incorporated into simulations used in formal detection studies. Their contribution is generally believed to be small relative to well-mixed greenhouse gases, though they could be of importance on regional scales.
In order to assess temperature changes over the last two decades, Hansen et al. (1997b) estimated the net radiative forcing due to changes in greenhouse gases (including ozone), solar variations and stratospheric aerosols from 1979 to 1995 from the best available measurements of the forcing agents. The negative forcing due to volcanoes and decreases in stratospheric ozone compensated for a substantial fraction of the increase in greenhouse gas forcing in this period (see Chapter 6, Table 6.13).
Patterns of anthropogenic forcing
Many of the new detection studies take into account the spatial variation of climate response, which will depend to some extent on the pattern of forcing (see also Section 12.2.3). The patterns of forcing vary considerably (see Chapter 6, Figure 6.7). The magnitude of the overall forcing due to increases in well-mixed greenhouse gases varies from almost 3 Wm-2 in the sub-tropics to about 1 Wm-2 around the poles. The warming due to increases in tropospheric ozone is mainly in the tropics and northern sub-tropics. Decreases in stratospheric ozone observed over the last couple of decades have produced negative forcing of up to about 0.5 Wm-2 around Antarctica. The direct effect of sulphate aerosols predominates in the Northern Hemisphere industrial regions where the negative forcing may exceed 2 Wm-2 locally.
Temporal variations in forcing
Some of the new detection studies take into account the temporal as well as spatial variations in climate response (see Section 18.104.22.168). Hence the temporal variation of forcing is also important. The forcing due to well-mixed greenhouse gases (and tropospheric ozone) has increased slowly in the first half of the century, and much more rapidly in recent decades (Chapter 6, Figure 6.8). Contributions from other factors are smaller and more uncertain. Sulphur emissions increased steadily until World War I, then levelled off, and increased more rapidly in the 1950s, though not as fast as greenhouse gas emissions. This is reflected in estimates of the direct radiative effect of increases in sulphate aerosols. Given the almost monotonic increase in greenhouse gas forcing in recent decades, this means the ratio of sulphate to greenhouse gas forcing has probably been decreasing since about 1960 (see Chapter 6, Figure 6.8). This should be borne in mind when considering studies that attempt to detect a response to sulphate aerosols. The decreases in stratospheric ozone have been confined to the last two to three decades.
Uncertainties in aerosol forcing
Some recent studies have incorporated the indirect effect of increases in tropospheric aerosols. This is very poorly understood (see Chapter 6), but contributes a negative forcing which could be negligible or exceed 2 Wm-2. The upper limit would imply very little change in net global mean anthropogenic forcing over the last century although there would still be a quite strong spatial pattern of heating and cooling which may be incompatible with recent observed changes (see, for example, Mitchell et al., 1995a). A negligible indirect sulphate effect would imply a large increase in anthropogenic forcing in the last few decades. There is also a large range in the inter-hemispheric asymmetry in the different estimates of forcing (see Chapter 6, Table 6.4). Given this high level of uncertainty, studies using simulations including estimates of indirect sulphate forcing should be regarded as preliminary.
Well-mixed greenhouse gases make the largest and best-known contribution to changes in radiative forcing over the last century or so. There remains a large uncertainty in the magnitude and patterns of other factors, particularly those associated with the indirect effects of sulphate aerosol.
Other reports in this collection