Analyzing Koopman approaches to physics-informed machine learning for long-term sea-surface temperature forecasting (2020-09-15T00:00:00.000000Z)

TL;DR

It is shown that the Koopman approach consistently outperforms baselines, and the utility of the additional assumptions and methods in this sea-surface temperature domain is discussed.

Abstract

Accurately predicting sea-surface temperature weeks to months into the future is an important step toward long term weather forecasting. Standard atmosphere-ocean coupled numerical models provide accurate sea-surface forecasts on the scale of a few days to a few weeks, but many important weather systems require greater foresight. In this paper we propose machine-learning approaches sea-surface temperature forecasting that are accurate on the scale of dozens of weeks. Our approach is based in Koopman operator theory, a useful tool for dynamical systems modelling. With this approach, we predict sea surface temperature in the Gulf of Mexico up to 180 days into the future based on a present image of thermal conditions and three years of historical training data. We evaluate the combination of a basic Koopman method with a convolutional autoencoder, and a newly proposed "consistent Koopman" method, in various permutations. We show that the Koopman approach consistently outperforms baselines, and we discuss the utility of our additional assumptions and methods in this sea-surface temperature domain.

Authors

Julian Rice

1 papers

Wenwei Xu

2 papers

Andrew August

1 papers

TL;DR

Abstract

Authors

References33 items

Integrating Physics-Based Modeling with Machine Learning: A Survey

Forecasting Sequential Data using Consistent Koopman Autoencoders

A spatiotemporal deep learning model for sea surface temperature field prediction using time-series satellite data

Physics-informed Autoencoders for Lyapunov-stable Fluid Flow Prediction

Informed Machine Learning - Towards a Taxonomy of Explicit Integration of Knowledge into Machine Learning

Shallow Learning for Fluid Flow Reconstruction with Limited Sensors and Limited Data

Deep learning and process understanding for data-driven Earth system science

A CFCC-LSTM Model for Sea Surface Temperature Prediction

Linearly-Recurrent Autoencoder Networks for Learning Dynamics

Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equations

Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations

Deep learning for physical processes: incorporating prior scientific knowledge

Learning Koopman Invariant Subspaces for Dynamic Mode Decomposition

Prediction of Sea Surface Temperature Using Long Short-Term Memory

Prediction of Sea Surface Temperature by Combining Numerical and Neural Techniques

Artificial neural network method for solving the Navier–Stokes equations

Applied Koopmanism.

Understanding the difficulty of training deep feedforward neural networks

High-resolution ensemble forecasting for the Gulf of Mexico eddies and fronts

Daily High-Resolution-Blended Analyses for Sea Surface Temperature

An exercise in forecasting loop current and eddy frontal positions in the Gulf of Mexico

Gulf of Mexico Hypoxia, A.K.A. “The Dead Zone”

Neural network modeling for near wall turbulent flow

Effects of a Warm Oceanic Feature on Hurricane Opal

Forecasting the equatorial Pacific sea surface temperatures by neural network models

Hamiltonian Systems and Transformation in Hilbert Space.

Equations

Temperature prediction

Introduction to Koopman operator theory of dynamical systems

Deep Learning

Noaa real-time ocean forecasting system rtofs-(atlantic): National weather ser-vice

Comparisons between gulf of mexico sea surface temperature anomalies and southern us severe thunder-storm frequency in the cool season

THE GENERALIZATION OF ‘STUDENT'S’ PROBLEM WHEN SEVERAL DIFFERENT POPULATION VARLANCES ARE INVOLVED

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names