Uncertainty Quantification for Inferring Marine Biogeochemical Cycle

Abstract:

In this talk, I give a hierarchical-statistical-modelling framework for uncertainty quantification for inferring marine biogeochemical cycles. Uncertainties appear in the (noisy and incomplete) data, the (imperfect) scientific models, and the (unknown) parameters that drive the models. Inference on the scientific process is obtained through the posterior distribution. Our approach is applied to an NPZD scientific model and a dataset from Ocean Station Papa in the northeastern Pacific Ocean.