Influence of GEOTRACES data distribution and misfit function choice on objective parameter retrieval in a marine zinc cycle model

[摘要] Biogeochemical model behaviour for micronutrients istypically hard to constrain because of the sparsity of observational data,the difficulty of determining parameters in situ, and uncertainties inobservations and models. Here, we assess the influence of data distribution,model uncertainty, and the misfit function on objective parameter optimisation ina model of the oceanic cycle of zinc (Zn), an essential micronutrient formarine phytoplankton with a long whole-ocean residence time. We aim toinvestigate whether observational constraints are sufficient forreconstruction of biogeochemical model behaviour, given that the Zn datacoverage provided by the GEOTRACES Intermediate Data Product 2017 is sparse.Furthermore, we aim to assess how optimisation results are affected by thechoice of the misfit function and by confounding factors such as analyticaluncertainty in the data or biases in the model related to either seasonalvariability or the larger-scale circulation. The model framework appliedherein combines a marine Zn cycling model with a state-of-the-art estimationof distribution algorithm (Covariance Matrix Adaption Evolution Strategy,CMA-ES) to optimise the model towards synthetic data in an ensemble of 26optimisations. Provided with a target field that can be perfectly reproducedby the model, optimisation retrieves parameter values perfectly regardlessof data coverage. As differences between the model and the system underlyingthe target field increase, the choice of the misfit function can greatly impactoptimisation results, while limitation of data coverage is in most cases ofsubordinate significance. In cases where optimisation to full or limiteddata coverage produces relatively distinct model behaviours, we find thatapplying a misfit metric that compensates for differences in data coveragebetween ocean basins considerably improves agreement between optimisationresults obtained with the two data situations.

[学科分类] 大气科学