We are also able to examine the spectra in light of the intervening decades of internal wave and turbulence measurements, predominantly from vertical profilers. Here, similar spectra are related to directly measured turbulence dissipation rates. Patches of high temperature variance have aspect ratios on the order of 100:1 ( Rosenblum and Marmorino 1990). It was also recognized that the shape and amplitude at high wavenumbers were likely set by turbulence. The low wavenumbers in these spectra have usually been interpreted as internal waves, and they inspired early versions of the empirical spectrum of internal waves (hereinafter referred to as GM75 Garrett and Munk 1975). 1b, slightly red at low wavenumbers and blue at high wavenumbers. These spectra share many of the characteristics of the spectra sketched in Fig. There is a substantial literature describing horizontal temperature spectra in the ocean, much of it overlapping the scales discussed here ( McKean and Ewart 1974 Katz 1975 Katz and Briscoe 1979 Marmorino 1987 Dugan et al. Conversely, the signature of turbulence extends to lower wavenumbers in horizontal than in vertical spectra ( Fig. Slope spectra are thought to be dominated by internal waves at low wavenumbers but, since the vertical-to-horizontal aspect ratio of internal waves is small, their influence on horizontal slope spectra does not extend to as high wavenumbers as in vertical strain spectra ( Klymak and Moum 2007, hereinafter Part I). In this paper, isopycnal slope spectra 1 collected in the deep ocean near the Hawaiian Ridge are fit to a simple internal wave model and the Batchelor spectrum for turbulence ( Batchelor 1959). Much effort has gone into estimating the rate of turbulence dissipation in the ocean, either by direct measurement ( Osborn and Cox 1972), or by characterizing the internal wave field and assuming that the rate of energy lost from it is a predictable function of its strength ( Henyey et al. In the ocean’s interior, turbulence is usually thought to be the result of breaking small-scale internal waves. Turbulence in the ocean enhances mixing of tracers and momentum by creating small-scale variance that increases gradients across which molecular diffusion acts. The broad bandwidth of the turbulence subrange means that a fit of spectral amplitude to the Batchelor model yields reasonable estimates of ε, even when applied at scales of tens of meters that in vertical profiles would be obscured by other fine structure. Scales between 1 m are modeled by a linear combination of internal waves and turbulence while at larger scales internal waves dominate. The turbulence spectral subrange ( k x > 0.4 cpm) responds to the dissipation rate as predicted by the Batchelor model spectrum, both in amplitude and towed vertical coherence. A four-order-of-magnitude range in turbulence dissipation rates at this site reveals that isopycnal slope spectra ∝ ε 2/3 k 1/3 x. The turbulence subrange of isopycnal slope spectra extends to surprisingly large horizontal wavelengths (>100 m). The spectra were compared with turbulence dissipation rates ε that are estimated using shear probes. Isopycnal slope spectra were computed from thermistor data obtained using a microstructure platform towed through turbulence generated by internal tidal motions near the Hawaiian Ridge.
0 kommentar(er)
