Modelling features of the biological pump and its impact on marine oxygen distribution
The marine biological pump not just impacts the uptake of atmospheric CO2 but also contributes to the regulation of ocean dissolved oxygen concentrations. The degree of ocean oxygenation has varied strongly throughout earth’s history. After several periods of oxygen depletion, the ocean currently exhibits relatively high oxygen concentrations. However, in the past 50 years, a decrease in oxygen concentrations of 2% in the global ocean has been observed and it is expected that the oxygen concentration will decrease even further with global change conditions, reducing the habitat volume of hypoxia-sensitive pelagic species. Although the interplay between supply of oxygen by ventilation and its consumption by biogeochemical processes is generally known, it is still unclear to which degree both processes influence the global marine oxygen distribution even under today’s climate conditions. Thus, this thesis focuses on features of the biological pump that might impact the marine oxygen distribution. Moreover, a comprehensive understanding of processes that influence the oxygen distribution is important to be able to estimate potential changes under future global change scenarios. Global models are an important tool to get a deeper insight into determinative processes for the marine oxygen distribution. In this thesis, three approaches regarding the biological pump are tested to advance the understanding of processes that determine the oxygen distribution under current climate conditions, which, in turn, potentially enable understanding of the expansion of oxygen minimum zones (OMZs) under future global change conditions: In the second chapter of this thesis, I test two competing feedbacks, which impact future oxygen concentrations, in the University of Victoria Earth System Climate Model (UVic ESCM) of intermediate complexity. This study shows, that the warming-induced phosphorus-oxygen feedback at the sediment-water interface and the resulting potential increase of released phosphorus does not constitute a major feedback in our model. It thus seems that other processes control the strength of future deoxygenation. In the third chapter of this thesis, a global biogeochemical ocean model is coupled to a particle aggregation model, which, using an appropriate parameterisation, improves the vertical and lateral representation of OMZs compared to the original model without aggregation. As there are still uncertainties in the parameterisation of the particle aggregation, a model calibration against an observed particle dataset seems necessary. In the fourth chapter two new processes influencing particle dynamics, namely particle breakup (disaggregation of large particles into smaller ones) and mesozooplankton migration are included in the biogeochemical model, which is optimised against observed particles, dissolved inorganic tracers and the overlap between modelled and observed OMZs. This study further improves the representation of OMZs. However, it also shows that the model is not able to represent shallow and deep particles realistically at the same time, which indicates that important processes that enhance particle export flux are still unknown and thus not considered in the model parameterisation.