Wavelet analysis of δ18O time series of monthly precipitation

A method based on wavelet transform is introduced to analyze δ¹⁸O time series in precipitation. Multi-Resolution Analysis (MRA) and periodicity analysis were conducted with Dmey wavelet function on the monthly average values of δ¹⁸O in precipitation from1961 to 2005 observed by the Vienna station, which provided a data-based approximation on the evolution of monthly oxygen-18 composition in precipitation in the Vienna station. It was found that the δ¹⁸O dynamics in precipitation was relatively stable and had a decrease trend on the whole during 1961 ~ 1982. Periodicity analysis showed that the enrichment- depletion periodicity of oxygen-18 composition at the scale of 14.5years (174months) and 7.5 years (88months) was 3 and 5 times, respectively. Periodically regular pattern of wavelet transform coefficient (W_f (α, b)) at the scale of 4.2 years was consistent with that at the scale of 7.5 years. The average fluctuation ranges of wavelet transform coefficients at the scale of 4.2 years and 7.5 years were bigger than that at the scale of 14.5years, which indicated that the time series of δ¹⁸O in precipitation at the Vienna station had prominent periods of 4~7.5years.