Can pCO2 observation help constrain predator-prey interactions in a biological model?

The Scotian Shelf, located in the northwest North Atlantic at the junction of the subpolar and subtropical gyres, is characterized by a large, shelf-wide spring bloom initiating in late March when temperature is at its coldest. The spring bloom is well represented in a 3D biogeochemical model, that was optimized using Chl and NO₃ observations, and validated against Chl measurements from satellite, bottle samples and glider transects. Observations of surface pCO₂ during the spring bloom show a large decrease of 100-200 μatm across most of the shelf within less than two weeks. However, the model does not capture the observed drawdown of pCO₂ during the spring bloom well, indicating that it is missing an important dynamical aspect. Can the medium-complexity regional model reproduce the pCO₂ observations in addition to the observed Chl and NO₃? We address this question by optimizing region-specific biological parameters in a 1-D version of the 3-D model. The results suggest that by incorporating the pCO₂ observations into the optimization, we can more accurately constrain the dynamics between phytoplankton and zooplankton. Our study underscores the value of multiple observation types in constraining the processes underlying spring bloom dynamics, sometimes in unexpected ways.