Update of the Spain02 gridded observational dataset for EURO-CORDEX evaluation: assessing the effect of the interpolation methodology

Abstract

Observational gridded products are commonly used to evaluate the performance of regional climate models. To this aim, gridded datasets should be comparable to the output of these models and, thus, should represent grid-cell area-averaged values and, whenever possible, they should be defined on the same spatial domains as the models, in order to avoid re-gridding or re-projection. In this study, we present an update of the Spain02 gridded observational dataset for daily precipitation and mean temperature building on the grids defined for the EURO-CORDEX initiative. In order to assess and intercompare different interpolation approaches, we analysed (1) two standard methodologies (ordinary kriging and thin plate splines), (2) three horizontal resolutions: 0.11°, 0.22° and 0.44° (matching the rotated EURO-CORDEX and ENSEMBLES grids), (3) two different approaches to guarantee either area-averaged or point representativity of the resulting grid values and (4) including/excluding orography as a covariable in the interpolation procedure. Besides introducing the new gridded datasets, in this work we also present some preliminary results on the sensitivity of temperature and precipitation (both mean and extreme regimes) to all these factors.