A Global Evaluation of Model Parameterizations for Carbon Export and Remineralization

The transport of biogenic particles from the surface ocean and their remineralization at depth constitutes the “biological pump” that moves carbon to the deep ocean. Export and remineralization are complex processes, but their parameterization in ocean models tends to be based on simple functions in which particles decay over fixed length scales, with or without association to mineral ballast. Parameterizations lack greater mechanistic realism in part because observations of particulate organic carbon flux using moored sediment traps and 234-Thorium are difficult, and thus sparse in space and time. In this study, we ask, “Does the available sediment trap data indicate which common parameterizations are most realistic?“ We use a one-dimensional version of the MITgcm that captures climatological productivity observed in open ocean provinces to compare observed export to several carbon flux parameterizations in use, including the “Martin curve”, exponential decay, and “ballast model.” Statistical tests are used to evaluate the model goodness of fit to the observations.