Prominent Periodic Oscillatory Signal Detection of Sea Level Variability over China Seas from T/P Altimetry

Abstract

The multitaper frequency-domain singular value decomposition (MTM-SVD) approach is used to isolate and detect oscillatory spatiotemporal signals of sea level variability over China Seas and their surroundings from TOPEX/POSEIDON altimetric data for six complete years. Some distinct periodic oscillatory signals at the 99% confidence level for the statistical significance are detected by MTM-SVD. In the East China Sea and the Yellow Sea, the most prominent oscillatory signals are the annual cycle and the high frequency cycle with periods between 56 and 63 days. In the South China Sea, sea level variability is somewhat more complex. Various prominent periodic signals - annual cycle, quasi-biennial, greater than three years, and those corresponding to seasonal cycles - are detected. We also make an attempt to combine the wavelet analysis with a principal component analysis (Wavelet-PCA) for the description of episodic sea level variability in a spatial data set. The spatial and temporal characteristics of the dominant periodic signal detected by MTM-SVD over China Seas are presented by the Wavelet-PCA approach.

Keywords:

• China Seas Multitaper Spectrum Principal Component Analysis Sea Level Variability T/P Altimetry Wavelet