Journal cover Journal topic
Ocean Science An interactive open-access journal of the European Geosciences Union

Journal metrics

  • IF value: 2.985 IF 2.985
  • IF 5-year<br/> value: 2.790 IF 5-year
    2.790
  • CiteScore<br/> value: 2.70 CiteScore
    2.70
  • SNIP value: 1.191 SNIP 1.191
  • SJR value: 1.885 SJR 1.885
  • IPP value: 2.589 IPP 2.589
  • h5-index value: 25 h5-index 25
doi:10.5194/os-2016-73
© Author(s) 2016. This work is distributed
under the Creative Commons Attribution 3.0 License.
Technical note
25 Oct 2016
Review status
A revision of this discussion paper was accepted for the journal Ocean Science (OS) and is expected to appear here in due course.
Technical note: Evaluation of three machine learning models for surface ocean CO2 mapping
Jiye Zeng1, Tsuneo Matsunaga1, Nobuko Saigusa1, Tomoko Shirai1, Shin-ichiro Nakaoka1, and Zheng-Hong Tan2 1National Institute for Environmental Studies, Tsukuba, Japan
2Department of Environmental Science, Hainan University, China
Abstract. Reconstructing surface ocean CO2 from scarce measurements plays an important role in estimating oceanic CO2 uptake. There are varying degrees of differences among the 14 models included in the Surface Ocean CO2 Mapping (SOCOM) inter-comparison initiative, in which five models used neural networks. This investigation evaluates two neural networks used in SOCOM, self-organization map and feedforward neural network, and introduces a machine learning model called support vector machine for ocean CO2 mapping. The technique note provides a practical guide to selecting the models.

Citation: Zeng, J., Matsunaga, T., Saigusa, N., Shirai, T., Nakaoka, S.-I., and Tan, Z.-H.: Technical note: Evaluation of three machine learning models for surface ocean CO2 mapping, Ocean Sci. Discuss., doi:10.5194/os-2016-73, in review, 2016.
Jiye Zeng et al.
Jiye Zeng et al.

Viewed

Total article views: 178 (including HTML, PDF, and XML)

HTML PDF XML Total BibTeX EndNote
135 27 16 178 4 14

Views and downloads (calculated since 25 Oct 2016)

Cumulative views and downloads (calculated since 25 Oct 2016)

Viewed (geographical distribution)

Total article views: 178 (including HTML, PDF, and XML)

Thereof 177 with geography defined and 1 with unknown origin.

Country # Views %
  • 1

Saved

Discussed

Latest update: 27 Mar 2017
Publications Copernicus
Download
Short summary
Three machine learning models were investigated for the reconstruction of global surface ocean CO2 concentration. They include self-organization map (SOM), feed forward neural network (FNN), and support vector machine (SVM). Our results show that the SVM performs the best, the FNN the second, and the SOM the worst. While the SOM does not have over fitting problem, it is sensitive to data scaling and its discrete interpolation may not be good for some applications.
Three machine learning models were investigated for the reconstruction of global surface ocean...
Share