Climate model simulations are used to examine possible future climates by simulating plausible scenarios (e.g. increasing atmospheric CO2, tropical deforestation) into the future using representations of inputs (i.e. forcings), storage between components of the climate system and and transfers between components (see Figure 8.4 and Chapter 11). The periods of time shown in Figure 8.4 refer to:
1 Forcing times. The characteristic timespans over which natural and anthropogenic changes of input occur. In the case of the former, these can be periods of solar radiation cycles or the effect of volcanism and in the case of the latter the average time interval over which significant changes of such anthropogenic effects as increased atmospheric CO2 occur.
2 Storage times. For each compartment of the atmosphere and ocean subsystems these are the average times taken for an input of thermal energy to diffuse and mix within the compartment. For the earth subsystem, the average times are those required for inputs of water to move through each compartment.
Model simulations can be performed in several different ways. A common procedure is to analyse the model's sensitivity to a specified change in a single variable. This may involve changes in external forcing (increased/decreased solar radiation, atmospheric CO2 concentrations, or a volcanic dust layer), surface boundary conditions (orography, land surface albedo, continental ice sheets) or in the model physics (modifying the convective scheme or the treatment of biosphere exchanges). In these simulations, the model
Source: After Saltzman (1983).
is allowed to reach a new equilibrium and the result is compared with a control experiment. A second approach is to conduct a genuine climate change experiment where, for example, the climate is allowed to evolve as atmospheric trace gas concentrations are increased at a specified annual rate (a transient experiment).
A key issue in assessments of greenhouse gas-induced warming is the sensitivity of global climate to CO2 doubling which is projected to occur in the mid-twenty-first century extrapolating current trends. Atmospheric GCM simulations for equilibrium condition changes, with a simple ocean treatment, indicate an increase in global mean surface air temperature of 2.5 to 5°C, comparing 1 X CO2 and 2 X CO2 concentrations in the models. The range is in part the result of a dependence of the temperature change on the temperature level simulated for the base-state 1 X CO2, and in part arises from the variations in the strength of feedback mechanisms incorporated in the models, particularly atmospheric water vapour, clouds, snow cover and sea ice. Use of coupled atmosphere-ocean models, however, suggest only a 1-2°C surface warming for century-long transient or doubled CO2 experiments (see Chapter 13).
Was this article helpful?