Assessment of the realisations of EURO-CORDEX projections in the simulation of rainfall regimes in Spain

Climate model projections are usually utilised to assess how the climate will change or develop in the future. Climate models are simplified representations of the Earth's climate system, that allow us to know the possible evolution of climate in the future (Garijo and Mediero 2018). Climate models are forced with the expected evolution of a set of variables, such as greenhouse gas emissions, to assess how climate change will affect climatic variables, such as precipitation. Global Climate Models (GCM) usually have a gross spatial resolution. Thus, regionalization methods are used to obtain results at a higher spatial resolution. In Spain, there are two main data sources of climate change projections under the Fifth Assessment Report (AR5) of the IPCC: AEMET ('Agencia Estatal de Meteorología', in Spanish) and CORDEX. AEMET generated climatic projections in a set of existing rain-gauging stations throughout Spain by statistical regionalisation. The CORDEX project offers several realisations of Regional Climate Models (RCM), producing climatic data over a mesh focused on a given area of the planet. For example, EURO-CORDEX focuses on Europe (Jacob et al. 2014). Despite several studies were carried out in recent years assessing how climate models simulate the current climate (Frei et al. 2003; Jacob et al. 2007; Mascaro et al. 2018), few studies were focused on Spain. A comparison between AEMET projections and observations at rain-gauging stations was conducted by Garijo et al. (2018), finding that such projections characterise better the mean climate behaviour than the extremes. However, no studies examine how EURO-CORDEX model simulations fit observed data over the Iberian Peninsula, and, therefore, how proper simulations of EURO-CORDEX climate models are for consequent studies. In this manuscript, EURO-CORDEX projections are investigated, to assess how precipitation simulated by climate models fit the actual behaviour of observed precipitation time series. Such assessment is based on a set of statistics that represent the spatial and temporal distribution of rainfall.