Numerical modelling of climate: Scientific basis and uncertainties in climate change projections

This seminar provides an overview of the research effort undertaken in recent decades in the field of numerical modelling of climate change. First, we briefly describe the characteristics of global circulation models that reproduce the dynamics of climate and that allow to simulate future developments based on the forcing induced by human activities (change in the composition of the atmosphere by CO2 , etc.). These forcings are known as future emission scenarios and represent the main source of uncertainty for climate change modelling, hence, it is common to consider a set of scenarios that try to cover the spectrum of possible future situations, more or less optimistic for any study of climate change. Moreover, the influence of other sources of uncertainty is also analyzed, like the pitfalls of the models themselves (for example, the parameterization of unresolved physical processes dynamics). This uncertainty makes it necessary to address the problem of climate change from a probabilistic point of view, using new techniques based on ensemble prediction to quantify uncertainty. This feature, coupled with the enormous computational cost needed to solve numerically these models, makes the spatial resolution of the data quite coarse (hundreds of kilometers). Therefore, in recent years have developed different strategies for regional climate change projections, which provide more detail on areas for impact studies.

The seminar has an informative and informal, open discussion of all aspects that may arise. An informative article on the subject can be downloaded http://www.meteo.unican.es/reviewCambioClima.