Search

Abstract: The debate on the enhanced greenhouse effect continues, confusing the climate change impact analysis and the decision makers. Hence, this paper is addressed to the extreme regional-scale evaluation of global climate model (GCM). Given that assessment of models is crucial to consider future scenarios in climate change impact analyses, the exploratory work is essential, as it identifies the strengths and weaknesses of GCM to regional-scale differences in model performance, providing insights on regional-scale climate interactions and processes. That is why this manuscript attempts to compare the spatial pattern of a longer temporal aggregation period (1960/61–1989/90) for extreme temperatures simulated by a particular GCM (HadCM3) to that of NCEP Reanalyses, which are considered as “truth”. Climatic risk analyses and forecasting, concern among other topics, assess an interpretation in intensity and frequency of recurrence of long duration extreme values (heat/cold waves). One study assesses the impact of extreme events analysing the temporal dependence of extreme events. The present study leads to the conclusion that the HadCM3 Simulations do not reproduce reasonably the NCEP Reanalyses over the Modelling the Impact of Climate Extremes (MICE—EU Project) domain, even despite the fact that the climatology of extremes has demonstrated very similar spatial patterns. Consequently, all diagnostics of extremes led support to the fitted Generalised Extreme Value (GEV) model for extrapolation, suggesting that the HadCM3 is not adequate for forecasting, simulating future scenarios of extremes based on the analysis of risk or probability maps for spatial uncertainties. The greatest difference between the two models is in terms of the precision of estimation. [Copyright &y& Elsevier]