Mechanisms of internal wave generation, decay and energy transfer in high resolution ocean models

Abstract

Internal waves play a major role in diapycnal ocean mixing, and in turn have a broad impact many ocean processes from general circulation to local nutrient availability. However, the nature of internal waves makes them difficult to observe, and for the most part we must rely on sparse in-situ observations to study the way they propagate, dissipate and impact ocean dynamics. Numerical models, then, play an important role in our understanding of internal waves. Much research has been done employing idealised, high resolution models to understand the intricacies of internal wave mechanics. However, due to their spanning large horizontal scales, with wavelengths varying from hundreds of metres to hundreds of kilometres, internal waves are seldom represented directly in general circulations ocean models, and their effects must be parameterised. These parameterisations schemes are able to emulate the expected energy inputs and outputs of the wave field, but are limited in their ability to represent the spatial variability, as well as the diversity of mechanisms at play in the generation, propagation, momentum or energy transfer, and mixing caused by the waves.

In my thesis, I undertook two studies on internal waves in high resolution ocean models. First, a novel internal wave-generation mechanism that is not accounted for by any modern internal wave parameterisation scheme. This study employed a high resolution idealised model of a storm passing over the top of a submarine ridge, emulating a field campaign in which anomalous bottom-generated internal waves were generated. We found that given a strong enough storm near to a significant bathymetric feature, bottom generated internal waves are readily generated which could explain some of the unaccounted for bottom mixing found in other studies. Secondly, I study the generation, propagation and dissipation of the Tasman Sea internal wave beam in a suite of high resolution regional models. Given that internal waves typically cascade their energy towards smaller scales before they are able to dissipate or break, we hypothesised that at different model resolutions, the energy pathway the internal wave beam would differ as more or less of the internal wave spectrum was resolved under different horizontal model resolutions. This study helps us to understand how well models represent internal wave processes when they are high enough resolution that parameterisation schemes must be turned off, but they are still only able to represent a portion of the internal wave spectrum.

Finally, I designed and developed a software package for setting up and running regional ocean models with Modular Ocean Model version 6. This package addresses that major barrier to entry of new oceanographers or climate scientists who don't have a strong computational background, or the time to fiddle for months with a model configuration before it will run. This package automates the regridding, metadata, and file encoding of all of the required inputs, as well as setting up the parameter files to work with the users' chosen experiments. This package has gained traction already within the Australian communities, and is started being picked up overseas. It was published in JOSS in August of 2024.