Reduce, Reuse, Recycle: Curating re-calibrated radiocarbon proxies refine Holocene sea-level curve for crucial far-field region

Mr Hayden Golding1, Dr Zoe Thomas2, Dr Kirti Lal1, A/Prof Kerrylee Rogers1, Prof Chris Turney2, Dr Amy Dougherty1,2,3

1University of Wollongong, Wollongong, Australia, 2University of New South Wales, Sydney, Australia, 3GeoInsights, Wollongong, Australia

The impacts of global mean sea-level rise (GMSLR) are expected to be spatially and temporally complex, highlighting the need for well-informed assessments of future change for coastal communities and cities around the world. Accurate projections of sea-level rise depend on models that consider past climatic, tectonic, and isostatic responses to the transfer of water between the cryosphere and ocean. Records of relative sea-level from tectonically stable, far-field regions have the potential to improve understanding of past meltwater contributions to GMSLR, informing on models. However, these key regions remain under-represented within the global atlas of postglacial sea-levels (e.g., Australia). Despite decades of research from southeast Australia, considerable uncertainty surrounds whether Holocene sea-level stabilised at present levels, fell gradually from a prolonged highstand or experienced multiple small-scale oscillations. This study aims to update the mid- to late-Holocene sea-level curve for southeast Australia by incorporating recently published data and re-analysing previously synthesised sea-level proxies using protocols from the global sea-level atlas. The resulting set of radiocarbon dates were calibrated to sideral years against the Southern Hemisphere calibration (SHCal20) dataset using OxCAL 4.4. Simply re-calibrating and re-analysing the same proxies used to construct the previous sea-level curve, resulted in noticeable shifts in the location and error of each individual indicator. This combined with a twofold increase in the total number of data points has expanded the lower limit of the pre-existing range (+0.5 m) down to contemporary sea level. This new reconstruction primarily aligns with the previous envelope that extends up to a maximum highstand of ~2 m between ~8 to 2 ka followed by a gradual fall to present. Merging disparate and imperfect datasets diffuses the ability to accurately interpret sea-level change. To refine this sea-level reconstruction, each proxy used is objectively critiqued and data with high uncertainties eliminated to produce a curve of higher confidence with a more precise envelope. This is of particular importance as much of the available data originate from localised, geomorphic studies and not specifically gathered to be used for sea-level reconstruction. With 75% of the points removed the curve is reduced to two main proxies, mangrove deposits that capture sea-level rise and tubeworm encrustations that document sea-level fall. Together these proxies exhibit a highstand of ~1.8 m between 7 to 4.5 ka, indicating prolonged meltwater contributions in the mid- to late- Holocene. This is incongruous with current regional models and given that these individual pieces of evidence are scattered along >500 km of coastline, it appears unlikely that they are anomalous. The preservation of these optimal proxies is sparse, highlighting the value of a single, widely distributed proxy that records both rising and falling sea-level. This would aid the construction of curves and data-driven models in other far-field regions to capture subtle variations in relative sea-level highstands.


My name is Hayden Golding, I am currently a 2nd year PhD candidate at the University of Wollongong and have a broad interest in environmental science and geomorphology. My research involves studying past coastal behaviour at various timescales (decades to millennia) to understand potential future impacts of anthropogenic climate change.

  • 00


  • 00


  • 00


  • 00



Nov 08 - 19 2021