Mesoscale Eddy Detection Based on Hilbert Marginal Spectrum

Hilbert marginal spectrum is an adaptive signal processing method, which is suitable for processing nonlinear and non-stationary signal and has less energy leakage. Using the Hilbert marginal spectrum of sea level anomaly (SLA) and seawater temperature, this paper successfully detects the mesoscale eddy. Variational mode decomposition (VMD) based on reverse weighted-permutation entropy (RWPE) is used to obtain the marginal spectrum of the non-stationary time series. Simulations verify that the performance of the Hilbert marginal spectrum is better than that of the Fourier spectrum when dealing with non-stationary signals. The marginal spectrum of the SLA and that of the seawater temperature at different depths reveal the periodic and vertical characteristics of the mesoscale eddy, which exhibits a good consistency with the actual phenomenon.