The analysis of the characteristics of weather and climate extremes at regional and local scale is crucial to adequately plan and prevent the socio-economic impacts arising in several fields of human activity. Different empirical extreme value indicators have been developed and applied. A set of them based on extreme percentiles and / or observed gusts calculated for each weather station (see, for example, the generic indicators working group CLIVAR ETCCDI http://cccma.seos.uvic.ca/ETCCDI) and other statistics estimated from extreme distributions fit to tail or tails of the data (return periods, etc.). However, most studies conducted to date have a global scale or they are limited to small geographic
regions and therefore are not appropriate to characterize the extremes in a national scale.
This project aims to make a systematic study of extreme events for precipitation and temperature on Spain, using the two methodologies for characterization of extreme events mentioned above, and considering both observations in local points and high resolution interpolated grids (in particular the Spain02 grid) that has exhibited a good performance for the characterization of extreme events. The main objective is to establish an appropriate set of high resolution observational datasets and to define a set of 20-30 indicators and an appropriate methodology for calculating return periods of 50 and 100 years for each of the seasons. By this way, one could characterize observed extreme regime of
precipitation and temperature in Spain at high resolution.
In addition to the observations, global simulations (4 GCMs) and regional (10 RCMs) provided by the European project ENSEMBLES (2004-2009) are also considered. Moreover, those simulations given by Spanish national strategic actions of generation of regional climate change scenarios in Spain with dynamic techniques (ESCENA) and statistics (esTcena) of the Ministry of Environment and Rural and Marine Affairs (2008-2011) will be also considered. These data allow examination of the ability of models to reproduce the observed extreme events and trends in projecting future scenarios of climate change. These data sets provide a great value to the project, since they may carry out various sensitivity studies of the results.
Finally, it is intended to further develop a more theoretical line on the characterization of extremes associated with spatial-temporal structures in practice the problem is usually addressed in an univariate way). This line was begun in an earlier project with excellent results and now the methodology will be applied to analyze whether the spatial structures associated with extreme events can be used to characterize or even improve the prediction of these phenomena.