Constraining bubble dynamics and mixing with dissolved gases: Implications for productivity measurements by oxygen mass balance

We used a dynamic mixed layer model to determine carbon export by the oxygen mass balance method from a time series of O-2/Ar, N-2/Ar and Ne measurements collected at station ALOHA near Hawaii from July 2000 to June 2001. The inert gas measurements constrain the flux of oxygen into the mixed layer from small, collapsing bubbles (injection) to be greater than or equal to the flux from larger bubbles (exchange), with mean estimates of the ratio in the range of 1-2. We also show that monthly observations of temperature and inert gases cannot constrain the rate of diapycnal mixing at this location, because of uncertainties in air-sea heat flux estimates and bubble dynamics. Organic carbon export from the mixed layer calculated from our dataset was 1.1 +/- 0.5 mol C m(-2) yr(-1), with most of the error deriving from uncertainties in the parameterization of diffusive gas exchange with wind speed. Our estimates of carbon export from the zone beneath the mixed layer but still in the euphotic zone ranged from 0 to 0.6 mol C m(-2) yr(-1) as the rate of background diapycnal mixing was increased from 0. 1 to 1.0 cm(2) s(-1). We conclude that the oxygen mass balance method has errors of about a factor of two in areas similar to the subtropical North Pacific, with the main uncertainties deriving from mixing rates and the parameterization of diffusive gas exchange.