Probabilistic neural networks for predicting energy dissipation rates in geophysical turbulent flows

TL;DR

This paper introduces a probabilistic neural network model trained on high-resolution ocean-like turbulence simulations to accurately predict energy dissipation rates, outperforming traditional models and capturing complex physics.

Contribution

The novel probabilistic neural network effectively models turbulent energy dissipation in stratified flows, incorporating physics from density gradients, and improves upon existing theoretical methods.

Findings

PNN outperforms baseline models in predicting dissipation rates.

The model captures the tails of the distribution more accurately.

Sensitivity analysis links improvements to physics from density gradients.

Abstract

Motivated by oceanographic observational datasets, we propose a probabilistic neural network (PNN) model for calculating turbulent energy dissipation rates from vertical columns of velocity and density gradients in density stratified turbulent flows. We train and test the model on high-resolution simulations of decaying turbulence designed to emulate geophysical conditions similar to those found in the ocean. The PNN model outperforms a baseline theoretical model widely used to compute dissipation rates from oceanographic observations of vertical shear, being more robust in capturing the tails of the output distributions at multiple different time points during turbulent decay. A differential sensitivity analysis indicates that this improvement may be attributed to the ability of the network to capture additional underlying physics introduced by density gradients in the flow.