Turbulence Estimates from a Free Drifting Platform at a River Inlet

At river inlets, waves breaking from winds, depth, and/or currents inject turbulence into the ocean surface, driving mean currents, and potentially enhancing plume mixing. However measurements of turbulence near the ocean surface are complicated by the waves themselves. Fixed-frame measurements of turbulent velocities are contaminated by wave orbitals, making the calculation of turbulent kinetic energy (TKE) dissipation rate difficult, and limiting their distance from the free surface. Recent measurements have therefore focused on floating, free-drifting platforms to measure near-surface turbulence. These platforms enable measurements within the wave crest; but they suffer from contamination due to buoy motion. Therefore spatial processing methods (i.e., structure functions) have been used to estimate dissipation rates rather than the more commonly used time-domain (spectral) methods. 

We use high-resolution velocity data collected from Doppler profilers mounted on free drifting "SWIFT" buoys deployed at the Columbia River Mouth in 2013 to evaluate and compare three turbulence processing methods: spatial structure function, vertical wavenumber spectra, and frequency spectra using the inertial-advective subrange. The buoy translational and rotational motions are first removed from the velocity data with a method similar to that which is used for ship-based wind speed and wind stress measurements. We find that correcting for these motions is essential for calculation of TKE dissipation rate, especially when using the spectral methods. Existing vertical scalings for dissipation rates are successful at describing the general shape of profiles, however dissipation rates decay faster than expected. These scalings can be used to prescribe location and intensity of wave breaking turbulence in coastal environments.