Development of hight resolution gridded datasets of monthly temperature since 1916 for Spain

Abstract

This article describes the methodology used in the Spanish State Meteorological Agency (AEMET) for obtaining gridded datasets of monthly minimum, maximum and mean temperature with 1 × 1 km spatial resolution for Spain, covering the period 1916-2018. These datasets have been created for climate analysis and monitoring, and will be updated periodically to extend the time coverage. The data used to produce the grids have undergone a quality control process in order to remove or correct erroneous data. The spatial interpolation method consists on a multiple linear regression with ordinary kriging of the regression residuals, using elevation, easting, northing and distance to the coast as independent variables in the regression. The performance of the interpolation method and the accuracy of the grids are evaluated using a cross-validation approach to estimate the errors. Some examples of derived products are shown, as well as a temperature analysis over the 1916-2018 period in Spain based on the gridded datasets.