Seasonality Effects on Nonlinear Properties of Hydrometeorological Records

摘要: Climatic time series, in general, and hydrological particular, exhibit pronounced annual periodicity. This periodicity its corresponding harmonics affect the nonlinear properties of relevant series (i.e. long-term volatility correlations width multifractal spectrum) thus have to be filtered out before studying fractal properties. We compare several filtering techniques find that order eliminate effects on it is necessary filter seasonal standard deviation addition mean, with conservation linear two-point . name proposed technique “phase substitution”, because employs Fourier phases series. The obtained results still indicate nonlinearity river data, strength being weaker than under previously used techniques.