In the stratified ocean interior, diapycnal turbulent mixing is mostly caused by breaking internal waves. This type of mixing is a key driver for the upwelling branch of the meridional overturning circulation. However, mixing due to internal waves is still poorly constrained in ocean models, due to a lack of understanding of: first, how much a certain internal wave field is able to mix; second, the regional distribution of internal wave energy itself.

Here, we tackle the second knowledge gap. Applying a machine-learning clustering technique to the internal wave energy level distribution from Argo data, we extract a small number of coherent clusters that characterize different types of oceanic internal wave fields. The resulting clusters can be explained in terms of the main regional controls of internal waves: the stratification configuration and the forcing mechanisms, among which are internal tides, wind-driven near-inertial waves, and eddy-wave interactions.

We propose a simple phenomenological internal wave generation model able to predict the mean global patterns of internal wave energy in different depth layers.

__________

Giovanni Dematteis’ research focuses primarily on nonlinear wave dynamics applied to oceanic internal waves. Giovanni studied Physics at Università di Torino, and obtained a PhD in Applied Mathematics from Politecnico di Torino in 2019. Since then, he has been a postdoc at Rensselaer Polytechnic Institute and Woods Hole Oceanographic Institution, USA, and then back at Università di Torino. There, he will become Associate Professor in the summer of 2026 after recently obtaining a European Research Council Starting Grant on mixing due to oceanic internal waves.