Optimizing ocean biogeochemical models: Harnessing satellite ocean color data spatially varying parameter estimation

Ocean biogeochemical (BGC) models help us study the global carbon cycle. These models have many parameters, but the best values for these parameters are not precisely known, which means the models might not be perfectly accurate. Usually, these parameters are kept constant in time and space in the models, but in reality, they can change in time and location based on the physical and biogeochemical context. In this study, we used satellite ocean color data to estimate the best values of some of the parameters of an ocean BGC model- the Regulated Ecosystem Model 2 (REcoM2). We allowed the parameters in the model to change over time and to vary with the location to make the model more realistic. To do this, we applied the ensemble data assimilation technique, which objectively searches for the best values for the uncertain parameters by optimally matching simulated output and observational data. We found that using the estimated values of the parameters made the REcoM2 model much better in terms of being more in line with observations, reducing errors by 52%. We are now more confident about using this model to understand ocean biogeochemistry and its role in the global carbon cycle.