Optimum Estimation of Sea‐Air Carbon Dioxide Fluxes Considering Diurnal Variations

A significant issue in estimating sea‐air CO 2 flux (FCO 2 ) in marginal seas is the reliance on single‐survey data, ignoring significant diurnal variations and causing large uncertainties. Here, we analyzed 12 days of observations from March to September 2012 near Xiaomai Island in the Southern Yellow Sea to investigate the factors influencing FCO 2 and identified wind speed (WS 10m ) and surface seawater CO 2 partial pressure ( p CO 2,sw ) as the primary sources of uncertainty. Based on these, we used the monthly average WS 10m instead of real‐time WS 10m to calculate CO 2 flux, significantly reducing the standard deviation (σ) of FCO 2 from 2.58 to 0.53 mmol·C·m −2 ·d −1 . In addition, the pronounced diurnal variation in p CO 2,sw introduces considerable uncertainty in CO 2 flux observations. Meanwhile, we found that photosynthetically active radiation (PAR) is a major factor controlling the diurnal variation of p CO 2,sw . Therefore, we developed a PAR‐based correction method for p CO 2,sw , which further lowered the σ of FCO 2 to 0.32 mmol·C·m −2 ·d −1 . These improvements provide a new method to reduce uncertainties in FCO 2 estimates derived from single‐survey data in marginal seas.