Functional clustering for Italian climate zones identification

摘要: This work presents a functional clustering procedure applied to meteorological time series. Our proposal combines series interpolation with smoothing penalized B-spline and the partitioning around medoids algorithm. final goal is obtain homogeneous climate zones of Italy. We compare this approach standard methods based on combination principal component analysis Cluster Analysis (CA) we discuss it in relation other approaches Fourier CA. show that simpler than from methodological interpretability point view. Indeed, becomes natural find clear connection between mathematical results physical variability mechanisms. how choice basis expansion (splines, Fourier) affects propose some comments their use. The for classification formed by monthly values temperature precipitation recorded during period 1971–2000 over 95 94 Italian monitoring stations, respectively. An assessment climatic patterns presented prove consistency comparison obtained different used judge data approach.