Cargando…

Universal early warning signals of phase transitions in climate systems

The potential for complex systems to exhibit tipping points in which an equilibrium state undergoes a sudden and often irreversible shift is well established, but prediction of these events using standard forecast modelling techniques is quite difficult. This has led to the development of an alterna...

Descripción completa

Detalles Bibliográficos
Autores principales: Dylewsky, Daniel, Lenton, Timothy M., Scheffer, Marten, Bury, Thomas M., Fletcher, Christopher G., Anand, Madhur, Bauch, Chris T.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Royal Society 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10072946/
https://www.ncbi.nlm.nih.gov/pubmed/37015262
http://dx.doi.org/10.1098/rsif.2022.0562
_version_ 1785019486433181696
author Dylewsky, Daniel
Lenton, Timothy M.
Scheffer, Marten
Bury, Thomas M.
Fletcher, Christopher G.
Anand, Madhur
Bauch, Chris T.
author_facet Dylewsky, Daniel
Lenton, Timothy M.
Scheffer, Marten
Bury, Thomas M.
Fletcher, Christopher G.
Anand, Madhur
Bauch, Chris T.
author_sort Dylewsky, Daniel
collection PubMed
description The potential for complex systems to exhibit tipping points in which an equilibrium state undergoes a sudden and often irreversible shift is well established, but prediction of these events using standard forecast modelling techniques is quite difficult. This has led to the development of an alternative suite of methods that seek to identify signatures of critical phenomena in data, which are expected to occur in advance of many classes of dynamical bifurcation. Crucially, the manifestations of these critical phenomena are generic across a variety of systems, meaning that data-intensive deep learning methods can be trained on (abundant) synthetic data and plausibly prove effective when transferred to (more limited) empirical datasets. This paper provides a proof of concept for this approach as applied to lattice phase transitions: a deep neural network trained exclusively on two-dimensional Ising model phase transitions is tested on a number of real and simulated climate systems with considerable success. Its accuracy frequently surpasses that of conventional statistical indicators, with performance shown to be consistently improved by the inclusion of spatial indicators. Tools such as this may offer valuable insight into climate tipping events, as remote sensing measurements provide increasingly abundant data on complex geospatially resolved Earth systems.
format Online
Article
Text
id pubmed-10072946
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher The Royal Society
record_format MEDLINE/PubMed
spelling pubmed-100729462023-04-05 Universal early warning signals of phase transitions in climate systems Dylewsky, Daniel Lenton, Timothy M. Scheffer, Marten Bury, Thomas M. Fletcher, Christopher G. Anand, Madhur Bauch, Chris T. J R Soc Interface Life Sciences–Mathematics interface The potential for complex systems to exhibit tipping points in which an equilibrium state undergoes a sudden and often irreversible shift is well established, but prediction of these events using standard forecast modelling techniques is quite difficult. This has led to the development of an alternative suite of methods that seek to identify signatures of critical phenomena in data, which are expected to occur in advance of many classes of dynamical bifurcation. Crucially, the manifestations of these critical phenomena are generic across a variety of systems, meaning that data-intensive deep learning methods can be trained on (abundant) synthetic data and plausibly prove effective when transferred to (more limited) empirical datasets. This paper provides a proof of concept for this approach as applied to lattice phase transitions: a deep neural network trained exclusively on two-dimensional Ising model phase transitions is tested on a number of real and simulated climate systems with considerable success. Its accuracy frequently surpasses that of conventional statistical indicators, with performance shown to be consistently improved by the inclusion of spatial indicators. Tools such as this may offer valuable insight into climate tipping events, as remote sensing measurements provide increasingly abundant data on complex geospatially resolved Earth systems. The Royal Society 2023-04-05 /pmc/articles/PMC10072946/ /pubmed/37015262 http://dx.doi.org/10.1098/rsif.2022.0562 Text en © 2023 The Authors. https://creativecommons.org/licenses/by/4.0/Published by the Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/ (https://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, provided the original author and source are credited.
spellingShingle Life Sciences–Mathematics interface
Dylewsky, Daniel
Lenton, Timothy M.
Scheffer, Marten
Bury, Thomas M.
Fletcher, Christopher G.
Anand, Madhur
Bauch, Chris T.
Universal early warning signals of phase transitions in climate systems
title Universal early warning signals of phase transitions in climate systems
title_full Universal early warning signals of phase transitions in climate systems
title_fullStr Universal early warning signals of phase transitions in climate systems
title_full_unstemmed Universal early warning signals of phase transitions in climate systems
title_short Universal early warning signals of phase transitions in climate systems
title_sort universal early warning signals of phase transitions in climate systems
topic Life Sciences–Mathematics interface
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10072946/
https://www.ncbi.nlm.nih.gov/pubmed/37015262
http://dx.doi.org/10.1098/rsif.2022.0562
work_keys_str_mv AT dylewskydaniel universalearlywarningsignalsofphasetransitionsinclimatesystems
AT lentontimothym universalearlywarningsignalsofphasetransitionsinclimatesystems
AT scheffermarten universalearlywarningsignalsofphasetransitionsinclimatesystems
AT burythomasm universalearlywarningsignalsofphasetransitionsinclimatesystems
AT fletcherchristopherg universalearlywarningsignalsofphasetransitionsinclimatesystems
AT anandmadhur universalearlywarningsignalsofphasetransitionsinclimatesystems
AT bauchchrist universalearlywarningsignalsofphasetransitionsinclimatesystems