The impact of mesoscale flows on the identification, energy partitioning and propagation of near-inertial internal waves

Abstract

Near-inertial waves (NIWs) contain a significant fraction of the ocean internal wave energy and can propagate long distances from their source before breaking. The dissipation of those waves contributes to the lightening of abyssal water masses and is thus important for sustaining the overturning circulation. Despite the crucial contribution of near-inertial waves to the ocean energy budget, a complete understanding of the physical processes that determine the waves generation, propagation and destruction is still lacking.

Past studies have highlighted the influence of background flows, in the form of mesoscale eddies and jets, on NIWs. Mesoscale flows not only affect the generation mechanism of the waves but also refract the waves once generated, thereby imprinting on the near-inertial energy distribution. The deep propagation of NIWs in anticyclonic eddies has been observed in numerical simulations while the ability of eddies and jets to amplify the waves has been demonstrated in idealised configurations. However, the magnitude of wave amplification by eddies in more realistic conditions is yet to be quantified and uncertainties remain regarding how background flows affect the partitioning of near-inertial energy across different wavelengths. Lastly, the impact of turbulent background flows is often neglected when identifying NIWs; this exclusion consequently leads to an inaccurate estimate of their energy content. In this thesis, we employ a realistic eddy-resolving numerical simulation of the North Pacific to address those research gaps in relation to wind-generated NIWs.

We first study the impacts of background mean flows on NIWs as a function of the waves horizontal wavelength. Two distinct cases are analysed: the propagation and energetics of large-scale NIWs in negative relative vorticity flows and the behaviour of small-scale NIWs in high shear. The large-scale waves rapidly propagate to depth in negative vorticity areas, while their small-scale counterparts are confined to shallower depths in high shear regions. Additionally, the results show that the mean-to-wave energy exchanges provide a considerable source of energy, similar in magnitude to wind-work, to NIWs.

Secondly, we study the inherent intermittency and localised variations of NIWs . Traditional Fourier analysis falls short in resolving the temporal evolution of the waves frequency and in detecting the localised variations of NIWs of different wavelengths. Wavelet analysis is utilised to overcome these limitations and to characterise the near-inertial wave field across various temporal and spatial scales. The propagating behaviour, whether dispersive or trapped, of waves with a particular wavenumber is found to be depend on the scales of the dominant background flow. Furthermore, the dominant wavelengths of NIWs being generated are found to be dependent on the scales of the prevailing eddies.

Lastly, the thesis examines how a varying background flow velocity can alter the NIWs properties. Background relative vorticity modifies the lower bound of the frequency bandwidth over which the waves can exist, while Doppler shifts alter the wave intrinsic frequency. Both effects complicate the identification of the waves and the estimation of their energy content. We employ different types of adaptive frequency filters to isolate the effects of relative vorticity and Doppler shift on the apparent near-inertial wave energy. Spectral filters neglecting background vorticity effects result in an underestimation of near-inertial energy by 40% in anticyclonic structures and an overestimation by 100% in cyclonic flows.

The findings presented in this thesis advance our understanding of the influence of realistic turbulent background flows on the energetics, partitioning and propagation of near-inertial waves, and provide a new analysis tool for quantifying these effects in future studies. Show more