Technical note: Evaluation of three machine learning models for surface ocean CO<sub>2</sub> mapping

Technical note: Evaluation of three machine learning models for surface ocean CO<sub>2</sub> mapping

Reconstructing surface ocean CO<sub>2</sub> from scarce measurements plays an important role in estimating oceanic CO<sub>2</sub> uptake. There are varying degrees of differences among the 14 models included in the Surface Ocean CO<sub>2</sub> Mapping (SOCOM) inte...

Full description

Bibliographic Details
Main Authors: J. Zeng, T. Matsunaga, N. Saigusa, T. Shirai, S.-I. Nakaoka, Z.-H. Tan
Format: Article
Language:English
Published: Copernicus Publications 2017-04-01
Series:Ocean Science
Online Access:http://www.ocean-sci.net/13/303/2017/os-13-303-2017.pdf
id doaj-1182a206b5bb415ab225619bb3dbc911
record_format Article
spelling doaj-1182a206b5bb415ab225619bb3dbc9112020-11-25T01:49:11ZengCopernicus PublicationsOcean Science1812-07841812-07922017-04-0113230331310.5194/os-13-303-2017Technical note: Evaluation of three machine learning models for surface ocean CO<sub>2</sub> mappingJ. Zeng0T. Matsunaga1N. Saigusa2T. Shirai3S.-I. Nakaoka4Z.-H. Tan5Centre for Global Environmental Research, National Institute for Environmental Studies, Tsukuba, Ibaraki, JapanCentre for Global Environmental Research, National Institute for Environmental Studies, Tsukuba, Ibaraki, JapanCentre for Global Environmental Research, National Institute for Environmental Studies, Tsukuba, Ibaraki, JapanCentre for Global Environmental Research, National Institute for Environmental Studies, Tsukuba, Ibaraki, JapanCentre for Global Environmental Research, National Institute for Environmental Studies, Tsukuba, Ibaraki, JapanInstitute of Tropical Agriculture and Forestry, Hainan University, Haikou, Hainan, ChinaReconstructing surface ocean CO<sub>2</sub> from scarce measurements plays an important role in estimating oceanic CO<sub>2</sub> uptake. There are varying degrees of differences among the 14 models included in the Surface Ocean CO<sub>2</sub> Mapping (SOCOM) inter-comparison initiative, in which five models used neural networks. This investigation evaluates two neural networks used in SOCOM, self-organizing maps and feedforward neural networks, and introduces a machine learning model called a support vector machine for ocean CO<sub>2</sub> mapping. The technique note provides a practical guide to selecting the models.http://www.ocean-sci.net/13/303/2017/os-13-303-2017.pdf
collection DOAJ
language English
format Article
sources DOAJ
author J. Zeng
T. Matsunaga
N. Saigusa
T. Shirai
S.-I. Nakaoka
Z.-H. Tan
spellingShingle J. Zeng
T. Matsunaga
N. Saigusa
T. Shirai
S.-I. Nakaoka
Z.-H. Tan
Technical note: Evaluation of three machine learning models for surface ocean CO<sub>2</sub> mapping
Ocean Science
author_facet J. Zeng
T. Matsunaga
N. Saigusa
T. Shirai
S.-I. Nakaoka
Z.-H. Tan
author_sort J. Zeng
title Technical note: Evaluation of three machine learning models for surface ocean CO<sub>2</sub> mapping
title_short Technical note: Evaluation of three machine learning models for surface ocean CO<sub>2</sub> mapping
title_full Technical note: Evaluation of three machine learning models for surface ocean CO<sub>2</sub> mapping
title_fullStr Technical note: Evaluation of three machine learning models for surface ocean CO<sub>2</sub> mapping
title_full_unstemmed Technical note: Evaluation of three machine learning models for surface ocean CO<sub>2</sub> mapping
title_sort technical note: evaluation of three machine learning models for surface ocean co<sub>2</sub> mapping
publisher Copernicus Publications
series Ocean Science
issn 1812-0784
1812-0792
publishDate 2017-04-01
description Reconstructing surface ocean CO<sub>2</sub> from scarce measurements plays an important role in estimating oceanic CO<sub>2</sub> uptake. There are varying degrees of differences among the 14 models included in the Surface Ocean CO<sub>2</sub> Mapping (SOCOM) inter-comparison initiative, in which five models used neural networks. This investigation evaluates two neural networks used in SOCOM, self-organizing maps and feedforward neural networks, and introduces a machine learning model called a support vector machine for ocean CO<sub>2</sub> mapping. The technique note provides a practical guide to selecting the models.
url http://www.ocean-sci.net/13/303/2017/os-13-303-2017.pdf
work_keys_str_mv AT jzeng technicalnoteevaluationofthreemachinelearningmodelsforsurfaceoceancosub2submapping
AT tmatsunaga technicalnoteevaluationofthreemachinelearningmodelsforsurfaceoceancosub2submapping
AT nsaigusa technicalnoteevaluationofthreemachinelearningmodelsforsurfaceoceancosub2submapping
AT tshirai technicalnoteevaluationofthreemachinelearningmodelsforsurfaceoceancosub2submapping
AT sinakaoka technicalnoteevaluationofthreemachinelearningmodelsforsurfaceoceancosub2submapping
AT zhtan technicalnoteevaluationofthreemachinelearningmodelsforsurfaceoceancosub2submapping
_version_ 1725008232965996544

Similar Items