CoAuthors

Maarten Buijsman (University of Southern Mississippi, United States)

We investigate the seasonal variability of internal tides in the global ocean using a six-year simulation of the Hybrid Coordinate Ocean Model (HYCOM). The non-stationarity of internal tides is caused by the time varying stratification, mesoscale variability, large-scale shifts in amphidromic points, and changes in ice cover. For example, the annual cycle in background flow and stratification modulates the internal tide amplitudes, causing the semidiurnal MA2 and MB2 frequencies. We use six years of steric sea surface height (SSSH) from a realistically forced global HYCOM simulation with a horizontal resolution of 8 km. This data set allows us to explore the global variability in both stationary and non-stationary tides. A least-squares harmonic analysis is used to extract SSSH amplitudes and phases for the M2, K1, O1, and S2 constituents for two-week/month-long time series segments. The amplitude and phase are used to construct the total internal tide for the bi-weekly/monthly time series for each tidal constituent. To compute the stationary internal tide time series, we extract the amplitude and phase of the principal tidal frequency and its annual modulates (e.g., M2, MA2, and MB2) over the entire six-year-long time series. Next, we subtract the stationary time series from the total bi-weekly/monthly time series to obtain the non-stationary bi-weekly/monthly time series for the tidal constituents. We find that the total variance for M2 internal tides for bi-weekly and monthly time series is on average 0.61 cm2 and 0.56 cm2, respectively, for seafloor depths greater than 250 m and a mean amplitude larger than the 50th percentile. Of this total variance, the fraction for bi-weekly/monthly time series due to stationary M2 tides, seasonal stationary tides, and non-stationary tides is 47%/51%, 4%/5%, and 49%/45%, respectively. We observe that the non-stationary fraction is higher for bi-weekly time series when compared to monthly time series. To validate our HYCOM simulation with observations, we also compare seasonal variance in stationary tides with altimetry data. This comparison shows a good agreement. We plan to repeat the same analysis for K1, O1, and S2 constituents. In addition, we also fit annual and semiannual seasonal cycles to the time series of monthly M2 harmonic constants to understand the seasonal variations in total internal tide amplitude. Internal tides generated in the coastal regions, such as Georges Bank, Arabian Sea, and Amazon Shelf, show high seasonal variability.