Deep learning of mixing by two 'atoms' of stratified turbulence

Abstract

Current global ocean models rely on ad-hoc parameterizations of diapycnal mixing, in which the efficiency of mixing is globally assumed to be fixed at $20\%$, despite increasing evidence that this assumption is questionable. As an ansatz for small-scale ocean turbulence, we may focus on stratified shear flows susceptible to either Kelvin-Helmholtz (KHI) or Holmboe wave (HWI) instability. Recently, an unprecedented volume of data has been generated through direct numerical simulation (DNS) of these flows. In this paper, we describe the application of deep learning methods to the discovery of a generic parameterization of diapycnal mixing using the available DNS dataset. We furthermore demonstrate that the proposed model is far more universal compared to recently published parameterizations. We show that a neural network appropriately trained on KHI- and HWI-induced turbulence is capable of predicting mixing efficiency associated with unseen regions of the parameter space well beyond the range of the training data. Strikingly, the high-level patterns learned based on the KHI and weakly stratified HWI are `transferable' to predict HWI-induced mixing efficiency under much more strongly stratified conditions, suggesting that through the application of appropriate networks, significant universal abstractions of density stratified turbulent mixing have been recognized.

arXiv comment12 pages, 5 figures and 1 table. Revised version submitted to JFM Rapids

References in corpus (2)

• Earth System Modeling 2.0: A Blueprint for Models That Learn From Observations and Targeted High-Resolution Simulations
• Self-organized criticality of turbulence in strongly stratified mixing layers

Cited by in corpus (15)

• Super-resolution reconstruction of turbulent flows with machine learning
• Nonlinear mode decomposition with convolutional neural networks for fluid dynamics
• Analog forecasting of extreme-causing weather patterns using deep learning
• Data-driven prediction of a multi-scale Lorenz 96 chaotic system using deep learning methods: Reservoir computing, ANN, and RNN-LSTM
• Convolutional neural networks for fluid flow analysis: toward effective metamodeling and low-dimensionalization
• Stable a posteriori LES of 2D turbulence using convolutional neural networks: Backscattering analysis and generalization to higher Re via transfer learning
• Experimental velocity data estimation for imperfect particle images using machine learning
• Generalization techniques of neural networks for fluid flow estimation
• Data-driven super-parameterization using deep learning: Experimentation with multi-scale Lorenz 96 systems and transfer-learning
• Assessments of epistemic uncertainty using Gaussian stochastic weight averaging for fluid-flow regression
• Interpreting neural network models of residual scalar flux
• XLB: A differentiable massively parallel lattice Boltzmann library in Python
• Experimental properties of continuously-forced, shear-driven, stratified turbulence. Part 1. Mean flows, self-organisation, turbulent fractions
• Dimension Reduced Turbulent Flow Data From Deep Vector Quantizers
• Model order reduction with neural networks: Application to laminar and turbulent flows