Detection of diurnal and semidiurnal tidal signatures from continuous GPS daily vertical residual time series by frequency mixing method

The diurnal and semidiurnal tidal displacements in the continuous GPS (CGPS) daily coordinate time series cannot be perfectly removed by ocean tide models, especially in coastal regions. The residual tidal displacements will propagate to the daily time series as alias signals with longer periods ranging from about 2 weeks to 1 year, which can be theoretically calculated. However, the alias signals with long periods (e.g., more than half a year) are difficult to detect through alias harmonic analysis when the length of the time series is insufficient. To extract all the diurnal and semidiurnal tidal displacements from the daily coordinate time series, we propose the frequency mixing method, that is, converting the signals from a high frequency to a low frequency, before the harmonic analysis. We demonstrate the feasibility of this method using simulated data and analyze the spectra of real daily vertical residual time series from eight globally distributed CGPS stations. The solar-related ocean tidal constituents (K1, P1, K2, and S2) are clearly detected from the time series by this method, verifying that the FES2004 model on solar-related ocean tidal displacement should be improved.