Climate Change 2001:
Working Group I: The Scientific Basis
Other reports in this collection

9.1.2 New Types of Model Experiments since 1995

The progression of experiments including additional forcings has continued and new experiments with additional greenhouse gases (such as ozone, CFCs, etc., as well as CO2) will be assessed in this chapter.

In contrast to the two global coupled climate models in the 1990 Assessment, the Coupled Model Intercomparison Project (CMIP) (Meehl et al., 2000a) includes output from about twenty AOGCMs worldwide, with roughly half of them using flux adjustment. Nineteen of them have been used to perform idealised 1%/yr CO2-increase climate change experiments suitable for direct intercomparison and these are analysed here. Roughly half that number have also been used in more detailed scenario experiments with time evolutions of forcings including at least CO2 and sulphate aerosols for 20th and 21st century climate. Since there are some differences in the climate changes simulated by various models even if the same forcing scenario is used, the models are compared to assess the uncertainties in the responses. The comparison of 20th century climate simulations with observations (see Chapter 8) has given us more confidence in the abilities of the models to simulate possible future climate changes in the 21st century and reduced the uncertainty in the model projections (see Chapter 14). The newer model integrations without flux adjustment give us indications of how far we have come in removing biases in the model components. The results from CMIP confirm what was noted in the SAR in that the basic patterns of climate system response to external forcing are relatively robust in models with and without flux adjustment (Gregory and Mitchell, 1997; Fanning and Weaver, 1997; Meehl et al., 2000a). This also gives us more confidence in the results from the models still using flux adjustment.

The IPCC data distribution centre (DDC) has collected results from a number of transient scenario experiments. They start at an early time of industrialisation and most have been run with and without the inclusion of the direct effect of sulphate aerosols. Note that most models do not use other forcings described in Chapter 6 such as soot, the indirect effect of sulphate aerosols, or land-use changes. Forcing estimates for the direct effect of sulphate aerosols and other trace gases included in the DDC models are given in Chapter 6. Several models also include effects of tropospheric and stratospheric ozone changes.

Additionally, multi-member ensemble integrations have been run with single models with the same forcing. So-called "stabilisation" experiments have also been run with the atmospheric greenhouse gas concentrations increasing by 1%/yr or following an IPCC scenario, until CO2-doubling, tripling or quadrupling. The greenhouse gas concentration is then kept fixed and the model integrations continue for several hundred years in order to study the commitment to climate change. The 1%/yr rate of increase for future climate, although larger than actual CO2 increase observed to date, is meant to account for the radiative effects of CO2 and other trace gases in the future and is often referred to as "equivalent CO2" (see discussion in Section 9.2.1). This rate of increase in radiative forcing is often used in model intercomparison studies to assess general features of model response to such forcing.

Continues on next page



Other reports in this collection