Applications of Deep Learning to Ocean Data Inference and Subgrid Parameterization

By T. Bolton, L. Zanna. Journal of Advances in Modeling Earth Systems, 2018

353 citations recorded by the source corpus.

Oceanographic observations are limited by sampling rates, while ocean models are limited by finite resolution and high viscosity and diffusion coefficients. Therefore, both data from observations and ocean models lack information at small and fast scales. Methods are needed to either extract information, extrapolate, or upscale existing oceanographic data sets, to account for or represent unresolved physical processes. Here we use machine learning to leverage observations and model data by predicting unresolved turbulent processes and subsurface flow fields. As a proof of concept, we train convolutional neural networks on degraded data from a high‐resolution quasi‐geostrophic ocean model. We demonstrate that convolutional neural networks successfully replicate the spatiotemporal variability of the subgrid eddy momentum forcing, are capable of generalizing to a range of dynamical behaviors, and can be forced to respect global momentum conservation. The training data of our convolutional neural networks can be subsampled to 10–20% of the original size without a significant decrease in accuracy. We also show that the subsurface flow field can be predicted using only information at the surface (e.g., using only satellite altimetry data). Our results indicate that data‐driven approaches can be exploited to predict both subgrid and large‐scale processes, while respecting physical principles, even when data are limited to a particular region or external forcing. Our in‐depth study presents evidence for the successful design of ocean eddy parameterizations for implementation in coarse‐resolution climate models.

Subfields: Atmospheric Science, Global and Planetary Change, Oceanography

View Paper: https://www.semanticscholar.org/paper/cd750a59358efcc61f0e5112b71e876cc0cc0204

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atmospheric Science

In Global and Planetary Change

In Oceanography

Citations in the corpus

Integrating Physics-Based Modeling with Machine Learning: A Survey (2020, neutral): “Parameterization [198] [157] [203] [135] [169] [95] [43] [90] [186] [212] [29] [33] [189] [292] [23] [24] [285] [21] [23] [294]”
Challenges and Prospects in Ocean Circulation Models (2019, background): “• Increasing use of artificial intelligence, neural networks, and deep learning as part of the development and improvement of parameterizations–particularly in their numerical optimization for efficiency (Schneider et al., 2017; Gentine et al., 2018; Bolton and Zanna, 2019).”
Physics-informed machine learning: case studies for weather and climate modelling (2021, background): “…part of complex multi-scale processes, which can be used to augment or replace existing parameterizations in weather and climate models [36,46,47,91–95 PIML also offers novel ways of emulating a chain of coupled processes, for example, the hydrological cycle, or even entire weather and climate…”
Enforcing Analytic Constraints in Neural Networks Emulating Physical Systems. (2019, background): “Despite the need for physically informed NNs for complex physical systems [2–5], enforcing hard constraints [6] has been limited to physical systems governed by specific equations, such as advection equations [7–9], Reynolds-averaged Navier-Stokes equations [10,11], boundary conditions of idealized ”
Resolving and Parameterising the Ocean Mesoscale in Earth System Models (2020, background): “deep learning) to design ocean eddy parameterisation for implementation in the coarser ocean models [65, 66] (see the ‘Advances in Parameterising the Mesoscale for Future Earth System Models’ section for further discussion).”