|
Scientists study global warming using computer-driven models of the climate. These models are based on the application of the principles of the physics of fluid dynamics, heat transfer and other processes. Simplifications are necessary, since the power of computers is limited. The models enable calculation of predictions of the impact of future increases in greenhouse gas concentrations on the climate. Starting with identical scenarios for changes in concentration, different models nevertheless give different forecasts of temperature : the sensitivity of the climate to increases in greenhouse gases is very variable. The models adopted by the IPCC predict warming of 1 to 6 °C between 1990 and 2100 (see picture 1). Models have also been developed to study the causes of recent modifications in climate by comparing the modifications observed to those predicted by the models under different assumptions about human and natural causes. Current models reproduce quite precisely the global temperature variations observed during the last century, but they do not simulate all aspects of the climate. They suggest that, at least since 1975, global warming has been mainly caused by greenhouse gas emissions. One of the principal problems with current models is the representation of clouds, which are a lot smaller in size than the grid of the model. Other problems are linked to indirect effects and solar variability. Depending on their complexity, the models take account of the atmosphere, the oceans, carbon cycles, sulphur and the chemistry of the atmosphere (including the impact of the ozone layer). The latest models under development, which will enable more reliable regional predictions to be obtained, are coupled atmosphere-ocean models.
To find out more >>
nternational Collaborative Model : Offer the power of your computers for the running of climate models: www.climateprediction.net
|