Analyzing Koopman approaches to physics-informed machine learning for long-term sea-surface temperature forecasting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rice, Julian, Xu, Wenwei, August, Andrew
Format: Preprint
Published: 2020
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918060309348352
author Rice, Julian
Xu, Wenwei
August, Andrew
author_facet Rice, Julian
Xu, Wenwei
August, Andrew
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2010_00399
institution arXiv
publishDate 2020
record_format arxiv
spellingShingle Analyzing Koopman approaches to physics-informed machine learning for long-term sea-surface temperature forecasting
Rice, Julian
Xu, Wenwei
August, Andrew
Geophysics
Machine Learning
Atmospheric and Oceanic Physics
Computational Physics
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.
title Analyzing Koopman approaches to physics-informed machine learning for long-term sea-surface temperature forecasting
topic Geophysics
Machine Learning
Atmospheric and Oceanic Physics
Computational Physics
url https://arxiv.org/abs/2010.00399