Reconstruction of Global Ocean Surface pCO₂ and Air‐Sea CO₂ Flux: Based on Multigrained Cascade Forest Model

Abstract Quantifying the role of air‐sea CO 2 exchange is essential for accurately estimating the global carbon balance, which is dependent on the spatial and temporal resolution of ocean surface carbon dioxide partial pressure (). When dealing with the global ocean as a vast and complex system, most existing studies tend to partition the global ocean into small‐scale regions. To account for interactions of environmental variables across multiple regions, we used machine learning algorithms to holistically reconstruct a 20‐year global map at a high resolution of 4 × 4 km based on products from the Moderate Resolution Imaging Spectroradiometer, reanalysis data, and Surface Ocean CO 2 Atlas. Three machine learning methods were compared, with multigrained cascade forest (gcForest) demonstrating the highest accuracy in global reconstruction (r 2 of 0.92, root mean square error of 13.46, and mean absolute error of 7.34 μatm). The global has shown a steady increase at an average annual growth rate of 1.95 ± 0.05 μatm yr −1 , controlled mainly by sea surface temperature and chlorophyll concentration. This study covers an ocean area of approximately 335 × km 2 , encompassing over 95% of the annual average carbon sink area. During 20 years, the daily CO 2 flux decreased by 0.44 mmol m −2 d −1 , while the proportion of carbon sink area remained constant, indicating ocean's carbon uptake capacity per unit area has been increasing.