Probabilistic assessments of the marine biogeochemical cycles of calcium carbonate, nitrous oxide, and oxygen

Battaglia, Gianna (2017). Probabilistic assessments of the marine biogeochemical cycles of calcium carbonate, nitrous oxide, and oxygen (Unpublished). (Dissertation, Universität Bern, Philosophisch–naturwissenschaftliche Fakultät, Physikalisches Institut, Abteilung für Klima– und Umweltphysik)

Text battaglia17phd.pdf - Other Restricted to registered users only Available under License BORIS Standard License. Download (48MB)

The current anthropogenic perturbation to the carbon cycle and resulting climate change is expected to have profound implications for biogeochemical cycling, ecosystem dynamics, and human livelihood. Quantification of biogeochemical cycles and their feedbacks to climate are usually hampered by sparse observational datasets, such that biogeochemical flux estimates usually come with large uncertainties. In this thesis, marine biogeochemical studies on calcium carbonate (CaCO3), nitrous oxide (N2O), and oxygen (O2) are presented relying on Bayesian, probabilistic ensemble approaches with the Bern3D Earth System Model of Intermediate Complexity (EMIC). Export and dissolution fluxes of CaCO3, and emissions of N2O including their uncertainties are estimated for modern ocean conditions. In addition, the long-term evolution of different Earth system variables are investigated with a focus on ocean deoxygenation. Given the long residence time of anthropogenic CO2 in the atmosphere, and long equilibration timescale of the ocean overturning circulation, such multi-millennial, ensemble perspectives are required when dealing with Earth system feedbacks and climate-related ocean hazards (Chapter 1).
The Bern3D EMIC used in this thesis is a cost-efficient, three-dimensional, dynamic ocean model coupled to a two-dimensional energy and moisture balance model of the atmosphere, and to a marine biogeochemistry module. It is designed for ensemble simulations recognizing parameter uncertainty, and for long-term, multi-millennial projections of Earth system variables. The probabilistic approach judges model versions with small deviations from available observations more probable than models with large deviations from the observations (Chapter 2).
CaCO3 cycling in the ocean is an important element of the global carbon cycle as it co-governs the distribution of carbon and alkalinity in the water column. Chapter 3, published in Biogeosciences, presents a probabilistic, quantitative assessment of the cycling of CaCO3 and constrains CaCO3 export fluxes out of the surface ocean, dissolution within the water column, and the flux to the ocean floor with observations. In this assessment, different observational data, the influence of ocean-sediment interactions, and ocean transport uncertainties are considered. Three main findings emerged. First, observation-constrained CaCO3 export is large in the Southern Ocean, the tropical Indo-Pacific, the northern Pacific and relatively small in the Atlantic. Second, dissolution within the 200 to 1500 m depth range is substantially lower than inferred in previous studies. These earlier estimates rest on a widely used method that neglects physical tracer transport. As such, their estimates are likely biased high. Third, parameters and mechanisms governing water column dissolution are hardly constrained by currently 4 available observational datasets. Consequently, simple saturation-independent dissolution rate parameterization may be applied in Earth system models to minimize computational costs.
N2O is an atmospheric greenhouse gas, and variations in its atmospheric concentration are informative of nitrogen and carbon cycle processes in terrestrial and marine systems. In recognition of recent observational evidence, a new parameterization for marine N2O pathways from nitrification and denitrification is developed in Chapter 4. A range of observational data is used in a Bayesian, probabilistic framework to constrain modern net N2O production based on a 1,000-member ensemble of the Bern3D model. To our knowledge, it is the first time that N2O fluxes from explicit denitrification and nitrification are included in a 3D ocean model. The observation-constrained results narrow the range in N2O emissions considerably compared to the most recent assessment by IPCC. Results are consistent with the global N2O budget and estimates of total denitrification. Probabilistic projections over the 21st century and the next 8,000 years reveal intricate, transient interactions between the marine carbon cycle, N2O, oxygen, and climate. For example, O2 minimum zones are projected to expand in volume, and the oceanic O2 inventory is projected to decrease by approximately a factor of two within the next 2,000 years, with surprisingly small impacts on marine N2O emissions and related N2O-climate feedbacks.
As an ongoing follow-up study, marine emissions of N2O have been investigated for the Last Glacial Maximum climate state and past climate changes. New high-resolution ice core reconstructions of atmospheric N2O and its isotopes are available for model evaluation. Understanding of atmospheric N2O variations offers an additional constraint on glacial-interglacial change.
The hazard of decreasing marine oxygen and changes in metabolic viability due to anthropogenic climate change are further analyzed in Chapter 5. The Bern3D simulations reveal that the long-term fate of oceanic oxygen is characterized by an initial decline followed by a recovery phase, with the peak decline occurring long after the end of the 21st century. For business as usual scenarios, the ocean oxygen content is projected to decrease by 40% over the next thousand years. This would likely have severe consequences for marine life. Global warming and oxygen loss are linked and meeting the warming target of the Paris climate agreement effectively limits related marine hazards. Marine hazards from warming and deoxygenation add to the list of long-term Earth system commitments including acidification and sea-level rise.
A large number of physical and biogeochemical tracers can be investigated in probabilistic applications with the Bern3D model on multi-millennial timescales. Simulations and reconstructions of past climate change can be used for process understanding and model evaluation, and to put anthropogenic climate change into perspective. Many international model intercomparison projects exist which should be used in future studies to evaluate model robustness (Chapter 6).

Interest & Impact

Downloads

2 since deposited on 22 Feb 2024
2 in the past 12 months

Citations

5 Citations in Web of Science ®
6 Citations in Scopus

Search

in Google Scholar™

Services

Actions (login required)

Edit item

Actions (login required)

Edit item