Near-inertial mixing: Modulation of shear, strain and microstructure at low latitude

We report direct, quantitative measurements of mixing associated with three cycles of a single, energetic, downward-propagating near-inertial wave in the Banda Sea at 6.5 degreesS, 128 degreesE during October 1998. The wave dominates the shear, containing 70% of the total variance. Simultaneous depth/time series of shear, strain, Froude number (Fr), and microstructure allow direct computation of their coherence and phase from 50-120 m, for 14 days. In this depth range, 72% of diapycnal diffusivity (68% of dissipation) occurs in three distinct pulses, spaced at the inertial period of 4.4 days. These are collocated with maxima of transverse shear, strain and Fr. Inertial-band log diapycnal diffusivity, log(10) Kp, is coherent at the 95% confidence level with both components of shear and Froude number. In this data set, strain is more important than shear in modulating Fr. Owing to the low latitude, the inertial frequency (f(o)=1/4.4 cycles per day) is much smaller than the diurnal and tidal frequencies. Consequently, near-inertial motions may be studied separately from tides and other motions via time-domain filtering. Semiempirical WKB plane-wave solutions with observed frequency omega (o) = 1.02f(o) and vertical scale 100 m explain 66% and 42% of inertial-band shear and strain variance, respectively. On the basis of the observed phase relationship between shear and strain, the wave is propagating equatorward, toward 295 degrees true. Ratios of shear to strain and of parallel to transverse shear suggest that the wave's intrinsic frequency omega (1) approximate to1.18f(eff). This indicates that background vorticity has lowered the effective Coriolis frequency, f(eff) = f(o) + zeta /2, relative to its planetary value, f(o) [Kunze, 1985]. Ray tracing suggests that the wave was generated near 6.9 degreesS, 130.6 degreesE, similar to 20 days prior to the cruise, coincident with the end of high winds associated with the SE monsoon. A slab mixed layer model [Pollard and Millard, 1970], forced with National Center for Environmental Prediction (NCEP) model surface winds, confirms that fluxes from the wind to the ocean at this time were sufficient to generate the wave. A very simple model shows that mixing by monsoon-generated inertial waves may add an important and strongly time-dependent aspect to some regions' energy budgets.