Abstract

This study synthesizes spatial and temporal variations in surface seawater pCO2 (partial pressure of CO2) and associated air-sea CO2 fluxes in the largest marginal sea of the North Pacific, the South China Sea (SCS), based on a large dataset collected from 47 surveys during 2000–2018. We categorized the SCS into five domains featuring different physical and biogeochemical characteristics to better understand the seasonality of SCS pCO2 dynamics and constrain the CO2 fluxes. The five domains are (A) the northern SCS shelf, (B) the northern SCS slope, (C) the SCS basin, (D) West of the Luzon Strait, and (E) the western SCS. We found a large spatial variability in sea surface pCO2 in the SCS, except during winter when values remained in a narrow range of 300 to 360 μatm. In general, seasonal variability was evident in surface water pCO2 values from the northern SCS (Domains A, B and D), with lower values during the cold seasons and higher values during the warm seasons, except in the Pearl River plume (150–650 μatm) and the area off northwest Luzon where winter upwelling occurred (370–470 μatm). In the SCS basin and the western SCS (Domains C and E), pCO2 in surface waters was generally higher than in the atmosphere (380–420 μatm). We also revealed large intra-seasonal variations in the northern SCS during monsoonal transitions in both spring and fall. In spring, pCO2 increased with temperature in the northern SCS, which was a CO2 sink in March but became a CO2 source in May with April as a transitional month. Fall is also a transitional season for the northern SCS, where it changes from a CO2 source back to a CO2 sink. The area-weighted CO2 fluxes across the entire SCS were −1.1 ± 2.2 mmol m−2 d−1 in winter, 0.9 ± 0.9 mmol m−2 d−1 in spring, 2.5 ± 1.4 mmol m−2 d−1 in summer and 1.9 ± 1.1 mmol m−2 d−1 in fall. Nevertheless, on an annual basis, the average CO2 flux from the SCS was 1.2 ± 1.7 mmol m−2 d−1. Enhanced carbon sink on the northern SCS shelf was observed in winter. The annual average CO2 flux was significantly lower than the previous estimate, which can largely be attributed to the addition of new datasets in the previously under-sampled seasons and regions.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call