The results presented in this chapter are based on simulations made with global climate models and apply to spacial scales of hundreds of kilometres and larger. Chapter 10 presents results for regional models which operate on smaller spatial scales. Climate change simulations are assessed for the period 1990 to 2100 and are based on a range of scenarios for projected changes in greenhouse gas concentrations and sulphate aerosol loadings (direct effect). A few Atmosphere-Ocean General Circulation Model (AOGCM) simulations include the effects of ozone and/or indirect effects of aerosols (see Table 9.1 for details). Most integrations¹ do not include the less dominant or less well understood forcings such as land-use changes, mineral dust, black carbon, etc. (see Chapter 6). No AOGCM simulations include estimates of future changes in solar forcing or in volcanic aerosol concentrations.

There are many more AOGCM projections of future climate available than was the case for the IPCC Second Assessment Report (IPCC, 1996) (hereafter SAR). We concentrate on the IS92a and draft SRES A2 and B2 scenarios. Some indication of uncertainty in the projections can be obtained by comparing the responses among models. The range and ensemble standard deviation are used as a measure of uncertainty in modelled response. The simulations are a combination of a forced climate change component together with internally generated natural variability. A number of modelling groups have produced ensembles of simulations where the projected forcing is the same but where variations in initial conditions result in different evolutions of the natural variability. Averaging these integrations preserves the forced climate change signal while averaging out the natural variability noise, and so gives a better estimate of the models� projected climate change.

For the AOGCM experiments, the mean change and the range in global average surface air temperature (SAT) for the 1961 to 1990 average to the mid-21st century (2021 to 2050) for IS92a is +1.3°C with a range from +0.8 to +1.7°C for greenhouse gas plus sulphates (GS) as opposed to +1.6°C with a range from +1.0 to +2.1°C for greenhouse gas only (G). For SRES A2 the mean is +1.1°C with a range from +0.5 to +1.4°C, and for B2, the mean is +1.2°C with a range from +0.5 to +1.7°C.
For the end of the 21st century (2071 to 2100), for the draft SRES marker scenario A2, the global average SAT change from AOGCMs compared with 1961 to 1990 is +3.0°C and the range is +1.3 to +4.5°C, and for B2 the mean SAT change is +2.2°C and the range is +0.9 to +3.4°C.

AOGCMs can only be integrated for a limited number of scenarios due to computational expense. Therefore, a simple climate model is used here for the projections of climate change for the next century. The simple model is tuned to simulate the response found in several of the AOGCMs used here. The forcings for the simple model are based on the radiative forcing estimates from Chapter 6, and are slightly different to the forcings used by the AOGCMs. The indirect aerosol forcing is scaled assuming a value of -0.8 Wm^-2 for 1990. Using the IS92 scenarios, the SAR gives a range for the global mean temperature change for 2100, relative to 1990, of +1 to +3.5°C. The estimated range for the six final illustrative SRES scenarios using updated methods is +1.4 to +5.6°C. The range for the full set of SRES scenarios is +1.4 to +5.8°C.

These estimates are larger than in the SAR, partly as a result of increases in the radiative forcing, especially the reduced estimated effects of sulphate aerosols in the second half of the 21st century. By construction, the new range of temperature responses given above includes the climate model response uncertainty and the uncertainty of the various future scenarios, but not the uncertainty associated with the radiative forcings, particularly aerosol. Note the AOGCM ranges above are 30-year averages for a period ending at the year 2100 compared to the average for the period 1961 to 1990, while the results for the simple model are for temperature changes at the year 2100 compared with the year 1990.

A traditional measure of climate response is equilibrium climate sensitivity derived from 2xCO₂ experiments with mixed-layer models, i.e., Atmospheric General Circulation Models (AGCMs) coupled to non-dynamic slab oceans, run to equilibrium. It has been cited historically to provide a calibration for models used in climate change experiments. The mean and standard deviation of this quantity from seventeen mixed-layer models used in the SAR are +3.8 and +0.8°C, respectively. The same quantities from fifteen models in active use are +3.5 and +0.9°C, not significantly different from the values in the SAR. These quantities are model dependent, and the previous estimated range for this quantity, widely cited as +1.5 to +4.5°C, still encompasses the more recent model sensitivity estimates.

A more relevant measure of transient climate change is the transient climate response (TCR). It is defined as the globally averaged surface air temperature change for AOGCMs at the time of CO₂ doubling in 1%/yr CO₂ increase experiments. The TCR combines elements of model sensitivity and factors that affect response (e.g., ocean heat uptake). It provides a useful measure for understanding climate system response and allows direct comparison of global coupled models. The range of TCR for current AOGCMs is +1.1 to +3.1°C with an average of 1.8°C. The 1%/yr CO₂ increase represents the changes in radiative forcing due to all greenhouse gases, hence this is a higher rate than is projected for CO₂ alone. This increase of radiative forcing lies on the high side of the SRES scenarios (note also that CO₂ doubles around mid-21st century in most of the scenarios). However these experiments are valuable for promoting the understanding of differences in the model responses.

Continues on next page

IPCC Homepage