A probabilistic view on modelling weather regimes
Abstract The statistical modelling of weather regimes encompasses the definition of a framework involving a dimensionality reduction step to reduce the highly sparse feature space of weather anomaly maps, and unsupervised learning techniques to correctly categorize regimes. Across the literature...
Published in: | International Journal of Climatology |
---|---|
Main Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Wiley
2022
|
Subjects: | |
Online Access: | http://dx.doi.org/10.1002/joc.7942 https://onlinelibrary.wiley.com/doi/pdf/10.1002/joc.7942 https://onlinelibrary.wiley.com/doi/full-xml/10.1002/joc.7942 https://rmets.onlinelibrary.wiley.com/doi/pdf/10.1002/joc.7942 |
id |
crwiley:10.1002/joc.7942 |
---|---|
record_format |
openpolar |
spelling |
crwiley:10.1002/joc.7942 2024-06-02T08:11:24+00:00 A probabilistic view on modelling weather regimes Baldo, Alessandro Locatelli, Robin 2022 http://dx.doi.org/10.1002/joc.7942 https://onlinelibrary.wiley.com/doi/pdf/10.1002/joc.7942 https://onlinelibrary.wiley.com/doi/full-xml/10.1002/joc.7942 https://rmets.onlinelibrary.wiley.com/doi/pdf/10.1002/joc.7942 en eng Wiley http://onlinelibrary.wiley.com/termsAndConditions#vor International Journal of Climatology volume 43, issue 4, page 1710-1730 ISSN 0899-8418 1097-0088 journal-article 2022 crwiley https://doi.org/10.1002/joc.7942 2024-05-03T11:49:02Z Abstract The statistical modelling of weather regimes encompasses the definition of a framework involving a dimensionality reduction step to reduce the highly sparse feature space of weather anomaly maps, and unsupervised learning techniques to correctly categorize regimes. Across the literature's methodology, the two stages stick to a recurrent scheme: Empirical Orthogonal Functions or Principal Component Analysis are used to assess the former; a standard K‐Means clustering algorithm maps each datapoint to the closest‐matching regime. However, such a combination has to cope with an overall reduction in the modelling accuracy. In our study, we re‐think both the two steps in favour of a more dynamic methodology which we apply to the last 42‐years winters' geopotential anomaly maps in the North Atlantic‐European zone. The dimensionality reduction is tackled by means of Variational Autoencoders, leading to better compressed feature spaces, enhancing the datapoints separability. Finally, we employ two probabilistic clustering methods based on Gaussian and Dirichlet Process mixture models, enabling a more faithful recognition of weather regimes, and allowing to reproduce their dynamics and transitions. Article in Journal/Newspaper North Atlantic Wiley Online Library International Journal of Climatology 43 4 1710 1730 |
institution |
Open Polar |
collection |
Wiley Online Library |
op_collection_id |
crwiley |
language |
English |
description |
Abstract The statistical modelling of weather regimes encompasses the definition of a framework involving a dimensionality reduction step to reduce the highly sparse feature space of weather anomaly maps, and unsupervised learning techniques to correctly categorize regimes. Across the literature's methodology, the two stages stick to a recurrent scheme: Empirical Orthogonal Functions or Principal Component Analysis are used to assess the former; a standard K‐Means clustering algorithm maps each datapoint to the closest‐matching regime. However, such a combination has to cope with an overall reduction in the modelling accuracy. In our study, we re‐think both the two steps in favour of a more dynamic methodology which we apply to the last 42‐years winters' geopotential anomaly maps in the North Atlantic‐European zone. The dimensionality reduction is tackled by means of Variational Autoencoders, leading to better compressed feature spaces, enhancing the datapoints separability. Finally, we employ two probabilistic clustering methods based on Gaussian and Dirichlet Process mixture models, enabling a more faithful recognition of weather regimes, and allowing to reproduce their dynamics and transitions. |
format |
Article in Journal/Newspaper |
author |
Baldo, Alessandro Locatelli, Robin |
spellingShingle |
Baldo, Alessandro Locatelli, Robin A probabilistic view on modelling weather regimes |
author_facet |
Baldo, Alessandro Locatelli, Robin |
author_sort |
Baldo, Alessandro |
title |
A probabilistic view on modelling weather regimes |
title_short |
A probabilistic view on modelling weather regimes |
title_full |
A probabilistic view on modelling weather regimes |
title_fullStr |
A probabilistic view on modelling weather regimes |
title_full_unstemmed |
A probabilistic view on modelling weather regimes |
title_sort |
probabilistic view on modelling weather regimes |
publisher |
Wiley |
publishDate |
2022 |
url |
http://dx.doi.org/10.1002/joc.7942 https://onlinelibrary.wiley.com/doi/pdf/10.1002/joc.7942 https://onlinelibrary.wiley.com/doi/full-xml/10.1002/joc.7942 https://rmets.onlinelibrary.wiley.com/doi/pdf/10.1002/joc.7942 |
genre |
North Atlantic |
genre_facet |
North Atlantic |
op_source |
International Journal of Climatology volume 43, issue 4, page 1710-1730 ISSN 0899-8418 1097-0088 |
op_rights |
http://onlinelibrary.wiley.com/termsAndConditions#vor |
op_doi |
https://doi.org/10.1002/joc.7942 |
container_title |
International Journal of Climatology |
container_volume |
43 |
container_issue |
4 |
container_start_page |
1710 |
op_container_end_page |
1730 |
_version_ |
1800757538601304064 |