Recent trends of extreme temperature indices for the Iberian Peninsula

Abstract

Climate change and extreme climate events have a significant impact on societies and ecosystems. As a result, climate change projections, especially related with extreme temperature events, have gained increasing importance due to their impacts on the well-being of the population and ecosystems. However, most studies in the field are based on coarse global climate models (GCMs). In this study, we perform a high resolution downscaling simulation to evaluate recent trends of extreme temperature indices. The model used was Weather Research and Forecast (WRF) forced by MPI-ESM-LR, which has been shown to be one of the more robust models to simulate European climate. The domain used in the simulations includes the Iberian Peninsula and the simulation covers the 1986–2005 period (i.e. recent past). In order to study extreme temperature events, trends were computed using the Theil-Sen method for a set of temperature indexes defined by the Expert Team on Climate Change Detection and Indices (ETCCDI). For this, daily values of minimum and maximum temperatures were used. The trends of the indexes were computed for annual and seasonal values and the Mann-Kendall Trend test was used to evaluate their statistical significance. In order to validate the results, a second simulation, in which WRF was forced by ERA-Interim, was performed. The results suggest an increase in the number of warm days and warm nights, especially during summer and negative trends for cold nights and cold days for the summer and spring. For the winter, contrary to the expected, the results suggest an increase in cold days and cold nights (warming hiatus). This behavior is supported by the WRF simulation forced by ERA-Interim for the autumn days, pointing to an extension of the warming hiatus phenomenon to the remaining seasons. These results should be used with caution since the period used to calculate the trends may not be long enough for this purpose. However, the general sign of trends are similar for both simulations despite some differences in their magnitudes.