An Alternative to Static Climatologies: Robust Estimation of Open Ocean CO2 Variables and Nutrient Concentrations From T, S, and O2 Data Using Bayesian Neural Networks

Henry Bittig (1) , Tobias Steinhoff (2) , Hervé Claustre (3) , Björn Fiedler (2) , Nancy Williams , Raphaëlle Sauzède , Arne Körtzinger (4) , Jean-Pierre Gattuso (3)

1 IOW - Leibniz Institute for Baltic Sea Research Warnemünde
2 GEOMAR - Helmholtz Centre for Ocean Research [Kiel]
3 LOV - Laboratoire d'océanographie de Villefranche
4 CAU - Christian-Albrechts-Universität zu Kiel = Christian-Albrechts University of Kiel = Université Christian-Albrechts de Kiel

which shows a significant, high latitude-intensified increase between +0.1 and +0.4 units per decade. This shows the utility that such transfer functions with realistic uncertainty estimates provide to ocean biogeochemistry and global climate change research. In addition, CANYON-B provides robust and accurate estimates of nitrate, phosphate, and silicate. Matlab and R code are available at https://github.com/HCBScienceProducts/.

Mots clés: carbon cycle; GLODAP; marine carbonate system; surface pCO 2 climatology; Revelle buffer factor increase; machine learning; nutrient estimation

Domaines: Océanographie; Interfaces continentales, environnement; Océan, Atmosphère; Environnement et Société; Milieux et Changements globaux

Henry Bittig, Tobias Steinhoff, Hervé Claustre, Björn Fiedler, Nancy Williams, et al.. An Alternative to Static Climatologies: Robust Estimation of Open Ocean CO2 Variables and Nutrient Concentrations From T, S, and O2 Data Using Bayesian Neural Networks. Frontiers in Marine Science, 2018, 5, ⟨10.3389/fmars.2018.00328⟩. ⟨hal-02345747⟩