The parameterization of air-sea gas fluxes has long been an area of study in oceanography. Standard parameterizations used in most climate models have a single diffusive term that scales as a function of wind speed. However, relatively insoluble gases are impacted by breaking waves that inject bubbles into the ocean. Oxygen, argon, and nitrogen are all substantially impacted by bubble injection and many air-sea flux parameterizations include bubble components in an attempt to deal with this issue.
The air-sea flux of oxygen is the largest term in the upper ocean oxygen balance. In order to identify the flux parameterization that best captures the bubble injection of oxygen we tested a range of air-sea flux parameterizations, in part by using my upper ocean gas model and comparing the results to mooring-derived nitrogen concentrations (Emerson and Bushinsky, 2016). We found that the Liang et al. (2013) parameterization best reproduced both mooring nitrogen data and water column noble gas measurements. Subsequently, we tuned the bubble component of the Liang et al. (2013) parameterization against wintertime periods when high wind speeds drive strong bubble injection (Yang, Emerson, and Bushinsky, 2017).
In the summer of 2017 I supervised Rebecca Barber, a Princeton Environmental Institute summer intern who analyzed output from an ocean model that included the Liang et al. (2013) parametrization. She compared this output to the standard gas flux parameterization that did not include bubbles and found a significant improvement in modeled noble gases.