Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning
In-situ observations of subsurface ocean temperatures are, in many regions, inconsistently distributed in time and space. These spatio-temporal inconsistencies in the observational network lead to difficulties in utilizing those observations effectively for ocean model evaluation or understanding la...
Published in: | Ocean Modelling |
---|---|
Main Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
2022
|
Subjects: | |
Online Access: | http://hdl.handle.net/10852/101078 https://doi.org/10.1016/j.ocemod.2022.102092 |
id |
ftoslouniv:oai:www.duo.uio.no:10852/101078 |
---|---|
record_format |
openpolar |
spelling |
ftoslouniv:oai:www.duo.uio.no:10852/101078 2023-05-15T14:26:16+02:00 Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning ENEngelskEnglishCharacterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning Thomas, Erin E. Müller, Malte 2022-08-15T11:15:53Z http://hdl.handle.net/10852/101078 https://doi.org/10.1016/j.ocemod.2022.102092 EN eng NFR/301450 NFR/276730 Thomas, Erin E. Müller, Malte . Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning. Ocean Modelling. 2022, 177 http://hdl.handle.net/10852/101078 2042947 info:ofi/fmt:kev:mtx:ctx&ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.jtitle=Ocean Modelling&rft.volume=177&rft.spage=&rft.date=2022 Ocean Modelling 177 13 https://doi.org/10.1016/j.ocemod.2022.102092 Attribution 4.0 International https://creativecommons.org/licenses/by/4.0/ 1463-5003 Journal article Tidsskriftartikkel Peer reviewed PublishedVersion 2022 ftoslouniv https://doi.org/10.1016/j.ocemod.2022.102092 2023-03-15T23:36:43Z In-situ observations of subsurface ocean temperatures are, in many regions, inconsistently distributed in time and space. These spatio-temporal inconsistencies in the observational network lead to difficulties in utilizing those observations effectively for ocean model evaluation or understanding larger-scale ocean characteristics. Model accuracy of subsurface ocean characteristics is especially important within regions that contain complex ocean structures. One such region is the European Arctic which not only contains several types of water masses with unique characteristics, but also wintertime sea ice coverage and complex bathymetry. This study presents an unsupervised neural networking technique that can be used in combination with traditional ocean model evaluation techniques to provide additional information on the accuracy of modeled vertical ocean temperature profiles. Self-organizing maps is an unsupervised machine learning technique that we apply to approximately twenty thousand Argo and CTD temperature profiles from 2012 to 2020 in the European Arctic to categorize the observed vertical ocean temperature structures in the top 150 m. The observed ocean profile categories, or neurons, defined by the self-organizing map show strong spatial and temporal dependencies. We then use the neuron weights, or the learned temperature profile structure of each neuron, to validate the spatial and temporal variability of modeled vertical temperature structures. This analysis gives us new insights about the model’s capabilities to reproduce specific vertical structures of the top-most ocean layer within different regions and seasons. Mapping modeled ocean temperature profiles onto the neuron-space of the observationally-defined self organized map highlights the potential of this method to advance our understanding of model deficiencies in that region. Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning Article in Journal/Newspaper Arctic Arctic Sea ice Universitet i Oslo: Digitale utgivelser ved UiO (DUO) Arctic Ocean Modelling 177 102092 |
institution |
Open Polar |
collection |
Universitet i Oslo: Digitale utgivelser ved UiO (DUO) |
op_collection_id |
ftoslouniv |
language |
English |
description |
In-situ observations of subsurface ocean temperatures are, in many regions, inconsistently distributed in time and space. These spatio-temporal inconsistencies in the observational network lead to difficulties in utilizing those observations effectively for ocean model evaluation or understanding larger-scale ocean characteristics. Model accuracy of subsurface ocean characteristics is especially important within regions that contain complex ocean structures. One such region is the European Arctic which not only contains several types of water masses with unique characteristics, but also wintertime sea ice coverage and complex bathymetry. This study presents an unsupervised neural networking technique that can be used in combination with traditional ocean model evaluation techniques to provide additional information on the accuracy of modeled vertical ocean temperature profiles. Self-organizing maps is an unsupervised machine learning technique that we apply to approximately twenty thousand Argo and CTD temperature profiles from 2012 to 2020 in the European Arctic to categorize the observed vertical ocean temperature structures in the top 150 m. The observed ocean profile categories, or neurons, defined by the self-organizing map show strong spatial and temporal dependencies. We then use the neuron weights, or the learned temperature profile structure of each neuron, to validate the spatial and temporal variability of modeled vertical temperature structures. This analysis gives us new insights about the model’s capabilities to reproduce specific vertical structures of the top-most ocean layer within different regions and seasons. Mapping modeled ocean temperature profiles onto the neuron-space of the observationally-defined self organized map highlights the potential of this method to advance our understanding of model deficiencies in that region. Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning |
format |
Article in Journal/Newspaper |
author |
Thomas, Erin E. Müller, Malte |
spellingShingle |
Thomas, Erin E. Müller, Malte Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning |
author_facet |
Thomas, Erin E. Müller, Malte |
author_sort |
Thomas, Erin E. |
title |
Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning |
title_short |
Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning |
title_full |
Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning |
title_fullStr |
Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning |
title_full_unstemmed |
Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning |
title_sort |
characterizing vertical upper ocean temperature structures in the european arctic through unsupervised machine learning |
publishDate |
2022 |
url |
http://hdl.handle.net/10852/101078 https://doi.org/10.1016/j.ocemod.2022.102092 |
geographic |
Arctic |
geographic_facet |
Arctic |
genre |
Arctic Arctic Sea ice |
genre_facet |
Arctic Arctic Sea ice |
op_source |
1463-5003 |
op_relation |
NFR/301450 NFR/276730 Thomas, Erin E. Müller, Malte . Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning. Ocean Modelling. 2022, 177 http://hdl.handle.net/10852/101078 2042947 info:ofi/fmt:kev:mtx:ctx&ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.jtitle=Ocean Modelling&rft.volume=177&rft.spage=&rft.date=2022 Ocean Modelling 177 13 https://doi.org/10.1016/j.ocemod.2022.102092 |
op_rights |
Attribution 4.0 International https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.1016/j.ocemod.2022.102092 |
container_title |
Ocean Modelling |
container_volume |
177 |
container_start_page |
102092 |
_version_ |
1766298733759168512 |