| dc.description.abstract | The pH of the Arctic Ocean is decreasing three to four times faster than other ocean basins. But the carbonate chemistry data from which this has been determined is spatially sparse and seasonally biased and so our knowledge on how the region is responding more widely to ocean acidification is incomplete. This same ocean hosts commercially important fish stocks, but its short food web means that small changes in species abundance can have dramatic effects across the complete ecosystem. Therefore, to study ecosystem health and to monitor future food security, the changing carbonate chemistry conditions of the Arctic Ocean need to be monitored. This will enable assessment of the impact that ocean acidification is likely having on these food webs and fish stocks, although this requires novel observational methods to observe Arctic wide carbonate chemistry. This thesis identifies the scientific priorities for addressing these needs, and then demonstrates how satellite and in situ observations in synergy can be used to begin addressing them.
Chapter 1 gives a brief overview of ocean acidification in the Arctic Ocean and methods to monitor the changing carbonate chemistry. Based upon the published literature, Chapter 2 (published as Green et al., 2021) provides a metadata analysis and resulting scientific roadmap which identifies three inter-linked research priorities: 1) The need to establish organisms and ecosystem physiochemical baselines by increasing the coverage of Arctic physicochemical observations; 2) The need to understand the variability of these conditions and the stressors likely being experienced by Arctic ecosystems; 3) The need to map life histories and fish stocks against these observed stressors. The remaining chapters of this thesis then begin to address these research priorities for one fish stock, Atlantic Cod, and one stressor, the impact of the changing carbonate system.
In Chapter 3 (now submitted as Green et al., submitted) published Arctic empirical total alkalinity and dissolved inorganic carbon algorithms, using inputs from different observational-data from satellites and in situ re-analyses, were implemented and evaluated against a large in situ measurement dataset. Despite these approaches being able to accurately describe variations in surface total alkalinity and dissolved inorganic carbon, when used to calculate the full carbonate system, they lead to propagated uncertainties in key derived parameters (e.g., pH) that make them unusable for assessing impacts on wild fish stocks. In Chapter 4, optimal re-training of these algorithms improves the combined uncertainty of two Arctic Ocean total alkalinity and three dissolved inorganic carbon algorithms, enabling pH and aragonite saturation state to be calculated with accuracies of 0.07 and 0.2, respectively, making these approaches suitable for fish stock impact assessments. In Chapter 5 this approach enabled the generation of a 27 year time series of the full carbonate chemistry for the Arctic Ocean Atlantic influenced Seas. This time series identifies that surface conditions of the Atlantic influenced Seas are already reaching pH minima (pH <7.7) and pCO2 maxima (pCO2> 1100) shown to cause negative impacts on Atlantic Cod in laboratory studies. The regions identified here that already experience low pH and high pCO2 could be the focus of natural analogue studies. These findings also suggest that laboratory studies need to expand the range in pH values being used to test Arctic Ocean species response to current and future ocean acidification. | en_GB |
