Articles | Volume 13, issue 2
Ocean Sci., 13, 303–313, 2017
https://doi.org/10.5194/os-13-303-2017
Ocean Sci., 13, 303–313, 2017
https://doi.org/10.5194/os-13-303-2017

Technical note 19 Apr 2017

Technical note | 19 Apr 2017

Technical note: Evaluation of three machine learning models for surface ocean CO2 mapping

Jiye Zeng et al.

Related authors

Harmonization of global surface ocean pCO2 mapped products and their flux calculations; an improved estimate of the ocean carbon sink
Amanda R. Fay, Luke Gregor, Peter Landschützer, Galen A. McKinley, Nicolas Gruber, Marion Gehlen, Yosuke Iida, Goulven G. Laruelle, Christian Rödenbeck, and Jiye Zeng
Earth Syst. Sci. Data Discuss., https://doi.org/10.5194/essd-2021-16,https://doi.org/10.5194/essd-2021-16, 2021
Preprint under review for ESSD
Short summary
Data-based estimates of the ocean carbon sink variability – first results of the Surface Ocean pCO2 Mapping intercomparison (SOCOM)
C. Rödenbeck, D. C. E. Bakker, N. Gruber, Y. Iida, A. R. Jacobson, S. Jones, P. Landschützer, N. Metzl, S. Nakaoka, A. Olsen, G.-H. Park, P. Peylin, K. B. Rodgers, T. P. Sasse, U. Schuster, J. D. Shutler, V. Valsala, R. Wanninkhof, and J. Zeng
Biogeosciences, 12, 7251–7278, https://doi.org/10.5194/bg-12-7251-2015,https://doi.org/10.5194/bg-12-7251-2015, 2015
Short summary

Related subject area

Depth range: Surface | Approach: Data Assimilation | Geographical range: All Geographic Regions | Phenomena: Air-Sea Fluxes
The OceanFlux Greenhouse Gases methodology for deriving a sea surface climatology of CO2 fugacity in support of air–sea gas flux studies
L. M. Goddijn-Murphy, D. K. Woolf, P. E. Land, J. D. Shutler, and C. Donlon
Ocean Sci., 11, 519–541, https://doi.org/10.5194/os-11-519-2015,https://doi.org/10.5194/os-11-519-2015, 2015
Short summary
Optimal adjustment of the atmospheric forcing parameters of ocean models using sea surface temperature data assimilation
M. Meinvielle, J.-M. Brankart, P. Brasseur, B. Barnier, R. Dussin, and J. Verron
Ocean Sci., 9, 867–883, https://doi.org/10.5194/os-9-867-2013,https://doi.org/10.5194/os-9-867-2013, 2013
Global surface-ocean pCO2 and sea–air CO2 flux variability from an observation-driven ocean mixed-layer scheme
C. Rödenbeck, R. F. Keeling, D. C. E. Bakker, N. Metzl, A. Olsen, C. Sabine, and M. Heimann
Ocean Sci., 9, 193–216, https://doi.org/10.5194/os-9-193-2013,https://doi.org/10.5194/os-9-193-2013, 2013

Cited articles

Abe, T., Kanaya, S., Kinouchi, M., Ichiba, Y., Kozuki, T., and Ikemura, T.: A Novel Bioinformatic Strategy for Unveiling Hidden Genome Signatures of Eukaryotes: Self-Organizing Map of Oligonucleotide Frequency, Genom. Inform., 13, 12–20, 2002.
Basak, D., Pal, S., and Patranabis, D. C.: Support vector regression, Neu. Inf. Pro.-Letters and Reviews, 11, 203–224, 2007.
Chierici, M., Fransson, A., and Nojiri, Y.: Biogeochemical processes as drivers of surface fCO2 in contrasting provinces in the subarctic North Pacific Ocean, Global Biogeochem. Cy., 20, GB1009, https://doi.org/10.1029/2004GB002356, 2006.
Download
Short summary
Three machine learning models were investigated for the reconstruction of global surface ocean CO2 concentration. They include self-organizing maps (SOMs), feedforward neural networks (FNNs), and support vector machines (SVMs). 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 problems, it is sensitive to data scaling and its discrete interpolation may not be good for some applications.