Abstract
Oceanic turbulence very near the water surface controls heat, momentum, energy, and mass transfer at the air–sea interface. This study examines the use of a rising temperature microstructure profiler for determination of the rate of turbulent energy dissipation ε beneath a water surface dominated by wind-driven surface gravity waves. Under short-fetch wind waves there is sufficient turbulent energy to generate an inertial-convective subrange. Thus, as well as the conventional Batchelor spectrum fitting approach, ε can be estimated using the temperature spectrum at lower wavenumbers. The inertial-convective subrange-derived turbulent energy dissipation rate εi compares well with the Batchelor spectrum method εb for higher dissipation rates. Sensor limitations mean that these estimates will be a lower bound. The highest dissipation rate estimates were in the uppermost data bin, indicating the importance of resolving right to the surface. Velocimeter results show that the surface waves modulate the rise velocity of the profiler, with variations reaching 75% of the deep water profiler speed. This increases the uncertainty in the transformation of spectra from frequency to wavenumber space using Taylor's frozen-field hypothesis. The use of an inertial convective structure-function-derived energy dissipation rate avoids this transformation. However, the structure function results are not encouraging as they provide a very poor estimate due to the sensitivity of the calculations and the low signal-to-noise ratio in the data. Repeated profiling illustrates the variability of turbulence intensity near the surface and also enables a reliable estimate of the background temperature gradient to be derived. This provides an improved estimate of vertical diffusion of heat.
Corresponding author address: Craig Stevens, National Institute for Water and Atmospheric Research, Greta Point, P.O. Box 14-901, Kilbirnie, Wellington, New Zealand. Email: c.stevens@niwa.cri.nz
