Abstract:The South China Sea (SCS) is the largest marginal sea of the North Pacific Ocean, where intensive field observations, including mappings of the sea surface partial pressure of CO 2 ( p CO 2 ), have been conducted over the last 2 decades. It is one of the most studied marginal seas in terms of carbon cycling and could thus be a model system for marginal sea carbon research. However, the cruise-based sea surface p CO 2 datasets are still temporally and spatially sparse. Using a machine-learning-based method facilitated by empirical orthogonal function (EOF) analysis, this study provides a reconstructed dataset of the monthly sea surface p CO 2 in the SCS with a reasonably high spatial resolution (0.05 ∘ × 0.05 ∘ ) and temporal coverage between 2003 and 2020. The data input to our model includes remote-sensing-derived sea surface salinity, sea surface temperature, and chlorophyll, the spatial pattern of p CO 2 constrained by EOF, atmospheric p CO 2 , and time labels (month). We validated our reconstruction with three independent testing datasets that are not involved in the model training. Among them, Test 1 includes 10 % of our in situ data, Test 2 contains four independent in situ datasets corresponding to the four seasons, and Test 3 is an in situ monthly dataset available from 2003–2019 at the South East Asia Time-series Study (SEATs) station located in the northern basin of the SCS. Our Test 1 validation demonstrated that the reconstructed p CO 2 field successfully simulated the spatial and temporal patterns of sea surface p CO 2 observations. The root mean square error (RMSE) between our reconstructed data and in situ data in Test 1 averaged ∼10 μatm , which is much smaller (by ∼50 %) than that between the remote-sensing-derived data and in situ data. Test 2 verified the accuracy of our retrieval algorithm in months lacking observations, showing a relatively small bias (RMSE of ∼8 μatm ). Test 3 evaluated the accuracy of the reconstructed long-term trend, showing that, at the SEATs station, the difference between the reconstructed p CO 2 and in situ data ranged from −10 to 4 μatm ( −2.5 % to 1 %). In addition to the typical machine learning performance metrics, we assessed the uncertainty resulting from reconstruction bias and its feature sensitivity. These validations and uncertainty analyses strongly suggest that our reconstruction effectively captures the main spatial and temporal features of sea surface p CO 2 distributions in the SCS. Using the reconstructed dataset, we show the long-term trends of sea surface p CO 2 in five subregions of the SCS with differing physicobiogeochemical characteristics. We show that mesoscale processes such as the Pearl River plume and China coastal currents significantly impact sea surface p CO 2 in the SCS during different seasons. While the SCS is overall a weak source of atmospheric CO 2 , the northern SCS acts as a sink, showing a trend of increasing strength over the past 2 decades. The data used in this article are available at https://doi.org/10.57760/sciencedb.02050 (Wang and Dai, 2022).

Unraveling the Cenozoic carbon cycle by reconstructing carbonate compensation depth (CCD)

Annual Resolution Records of Sea-Level Change Since 1850 CE Reconstructed from Coral Δ18o from the South China Sea

Elevated Anthropogenic CO2 Invasion and Stimulated Carbonate Dissolution in the South China Sea Basin

Reconciling the Apparent Discrepancy Between Cenozoic Deep‐Sea Temperatures From Proxies and From Benthic Oxygen Isotope Deconvolution

Exploring Cyclic Changes of the Ocean Carbon Reservoir

Validating the Deep Time Carbonate Carbon Isotope Records: Effect of Benthic Flux on Seafloor Carbonate

Deep carbon cycles constrained by a large-scale mantle Mg isotope anomaly in eastern China

Evolution of the Cenozoic Carbon Cycle: the Roles of Tectonics and CO2 Fertilization

Seasonal and Interannual Variability of Coccolithophore Flux in the Northern South China Sea

Late Quaternary Carbon Cycling Responses to Environmental Change Revealed by Multi-Proxy Analyses of a Sediment Core from an Upland Lake in Southwest China

Toward a Cenozoic history of atmospheric CO 2

Deep Water [CO32−] and Circulation in the South China Sea over the Last Glacial Cycle

Evolution of paleo-climate and seawater pH from the late Permian to postindustrial periods recorded by boron isotopes and B/Ca in biogenic carbonates

Spatial reconstruction of long-term (2003–2020) sea surface pCO2 in the South China Sea using a machine-learning-based regression method aided by empirical orthogonal function analysis

Long-term variations in terrestrial carbon cycles and atmospheric CO 2 levels: Exploring impacts on global ecosystem and climate in the aftermath of end-cretaceous mass extinction

High-resolution paired carbon isotopic records from the Meishucun section in South China: Implications for carbon cycling and environmental changes during the Ediacaran-Cambrian transition

Sustained Deep Pacific Carbon Storage after the Mid‐Pleistocene Transition Linked to Enhanced Southern Ocean Stratification

Aridity-driven Decoupling of Δ13c Between Pedogenic Carbonate and Soil Organic Matter

Precession-driven Changes in Air-Sea CO2 Exchange by East Asian Summer Monsoon in the Western Tropical Pacific Since MIS 6

Longer-term terrestrial responses in the aftermath of the end-Cretaceous mass extinction

A model of Phanerozoic cycles of carbon and calcium in the global ocean: Evaluation and constraints on ocean chemistry and input fluxes