Abstract
Fjords have been recognised as pivotal marginal marine settings for long-term carbon sequestration. The temperate fjords of Fiordland, New Zealand may display some of the largest rates of carbon burial due to large quantities of terrestrially derived organic matter transported into the fjords. This terrestrial organic matter combined with reduced fjord circulation leads to high rates of carbon burial within fjord basins. However, the current understanding of carbon burial is poorly understood. Some research has suggested that increased dissolved oxygen (DO) within the bottom water layer of the water column leads to increased organic matter degradation, therefore decreasing carbon stored within the sediment, however, this is not well constrained.
Sedimentary metal-based reduction oxidation (redox) proxies are commonly used in variable redox marine environments to provide a method of reconstructing past climate-driven changes in bottom water DO. However, they have rarely been applied to fjord settings, including Fiordland. This thesis applies a suite of metal-based redox proxies, derived from micro x-ray fluorescence (μXRF) and quadrupole inductively coupled plasma mass spectrometry (Q-ICP-MS), to 71 surface sediments from Fiordland’s fjords, which display a variety of redox conditions. This study will establish which proxies are able to reliably reconstruct the known bottom water DO content at each location, the analytical method/s available for quantification and recommend proxies for future investigations of redox conditions within Fiordland. Fiordland is an ideal location to ground truth and apply metal- based redox proxies to seafloor sediments because the basins and sub-basins of the fjords display a variety of different bottom water redox conditions and there is minimal anthropogenic release of metals into the region.
Additionally, the construction of two tailrace tunnels from the Manapouri Power Station (MPS) now discharges an additional 500 m³ s⁻¹ of freshwater into Deep Cove, Doubtful Sound. In the second phase of this study, a late Holocene sediment core from the depocenter of the Deep Cove basin was obtained and proxies assessed as reliable were applied downcore to determine if the additional freshwater discharge arising from tailrace construction has impacted the bottom water redox regime of Deep Cove and in turn its potential for carbon accumulation and storage.
A calibration was performed between Q-ICP-MS and μXRF derived datasets to first establish which elemental proxies can be reliably quantified using the μXRF technique. The proxies of Ca/Ti, Ba/Ti, Fe/Ti and Fe/Mn generated well-correlated datasets using both analytical techniques. Of these, the Ca/Ti and Ba/Ti worked particularly well. However, the Fe/Ti and Fe/Mn redox proxies were unable to reliably demonstrate that sedimentary iron enrichments stem solely from redox regime variations in the fjord settings of Fiordland.
An expanded suite of well-tested metal-based redox proxies acquired solely from Q-ICP-MS were applied to the same 71 surface sediments. Of the investigated proxies, Cd/Mo - Co x Mn(%), Uᴇꜰ, Moᴇꜰ, Uᴇꜰ/Moᴇꜰ, U/Th and V/Cr, were all deemed to reliably reconstruct both the subtle and large-scale modern redox conditions experienced within the fjord basins of Fiordland. Opposingly, proxies such as U/Al, Mo/Al, Fe/Al, Ni/Co and V/(V+Ni) were unable to either correctly determine the modern redox conditions or show subtle redox variations within Fiordland.
Reliable Q-ICP-MS-derived redox proxies were then applied to a sediment core from Deep Cove, spanning the late Holocene. The downcore data generally showed that the redox conditions pre-construction were highly variable. During the MPS construction, the changes in the bottom water redox regime induced by the additional freshwater discharge were within the extent recorded for natural climate-related redox variations. 40 years after the construction of the MPS first tailrace tunnel, equilibration of the bottom water redox regime in Deep Cove occurred although the upper 1 to 4 cm show a shift to weaker suboxic conditions. However, the long-term impact of the additional freshwater discharged on the redox conditions in Deep Cove still may not be known for some time.
Future studies should consider investigating cores from more marginal locations within Deep Cove as they may be more vulnerable to changing redox conditions and therefore may be more impacted by the additional freshwater from the Manapouri power station tailrace tunnels. Additionally, acquiring ²¹⁰Pb and ¹⁴C chronology for this core would precisely identify the start of the MPS construction therefore increasing the comparison between the pre- and post-construction phases.