Modeling of centennial Black Sea level changes as a basis for forecasting

Abstract

The statistical parameters that characterize sea-level regime along the NW Black Sea have been calculated and analyzed.Monthly and yearly averages of sea level position for the 1876-2002 observation period (Odessa station) were used to study fluctuations in sea level.We performed spectral (Fourier Transform) and frequency-temporal (Wavelet) analyses of time domains, which characterize the level changes in the study region.The cross- correlation analysis of sea-level regime and water balance was also included in the study.We compared the low-frequency elements of water-level regime with factors that characterize climate changes in the Northern Hemisphere in the 20th century. Wavelet analysis provides wide possibilities for investigating time-domain periodicity. The basis of this method was generated in the mid 1980s by Grossman and Morle as an alternative to Fourier transform to be used for temporal/spatial domains with expressed heterogeneity. Wavelet transform divides the analyzed process into its constituent waves and components of various scales, and also provides time-specific process information.