A multifractal-based climate analysis

This work consists of a multifractal analysis of surface air temperature in Western and Eastern Europe. The wavelet leaders method enables us to exhibit the monofractal nature of these signals. Then, we define a climate classification relying on the Zygmund-Hölder spaces and show that it matches the Köppen classification. By doing so, we evidence the existing relation between the regularity of the temperature data (obtained through functional spaces) and the type of climate. Finally, we give a climatic interpretation of these results.