Reconstruction of global surface ocean p CO 2 using region-specific predictors based on a stepwise FFNN regression algorithm

[摘要] Various machine learning methods were attempted in the global mapping of surface ocean partial pressure of CO 2 ( p CO 2 )to reduce the uncertainty of the global ocean CO 2 sink estimate due to undersampling of p CO 2 . In previous research, thepredictors of p CO 2 were usually selected empirically based on theoretic drivers of surface ocean p CO 2 , and thesame combination of predictors was applied in all areas except where there was a lack of coverage. However, the differences between the drivers of surface ocean p CO 2 in different regions were not considered. In this work, we combined the stepwise regression algorithm and a feed-forwardneural network (FFNN) to select predictors of p CO 2 based on the mean absolute error in each of the 11 biogeochemical provincesdefined by the self-organizing map (SOM) method. Based on the predictors selected, a monthly global 1 ∘   ×  1 ∘ surface ocean p CO 2 product from January 1992 to August 2019 was constructed. Validation of different combinations of predictors based on theSurface Ocean CO 2 Atlas (SOCAT) dataset version 2020 and independent observations from time series stations was carried out. The prediction of p CO 2 basedon region-specific predictors selected by the stepwise FFNN algorithm was more precise than that based on predictors from previousresearch. Applying the FFNN size-improving algorithm in each province decreased the mean absolute error (MAE) of the global estimate to11.32  µatm and the root mean square error (RMSE) to 17.99  µatm . The script file of the stepwise FFNN algorithm and p CO 2 product are distributed through the Institute of Oceanology of the Chinese Academy of Sciences Marine Science Data Center(IOCAS, https://doi.org/10.12157/iocas.2021.0022 , Zhong, 2021.

[学科分类] 大气科学