Statistical methods for the analysis of climatic phenomena

Statistical Climatology investigates the application of statistics to atmospheric and climate science. This Thesis is intended to contribute some developments in this field. The Thesis is composed of two distinct applied problems: the first presents a functional clustering procedure applied to meteorological time series to obtain homogeneous climate zones; the second builds a hierarchical Bayesian model aiming at the prediction of 15- and 30-minutes cumulated precipitation at unknown locations and time using information on lightnings in the same area.