Spatial and temporal variability in the Holocene sea-level record of the South Australian coastline

• Published in: Sedimentary geology 2002-06, Vol.150 (1), p.153-169
• Main Authors: Belperio, A.P., Harvey, N., Bourman, R.P.
• Format: Article
• Language: English
• Subjects: Australia; Coast; Holocene; Marine; Radiocarbon dating; Sea level
• ISSN: 0037-0738; 1879-0968
• DOI: 10.1016/S0037-0738(01)00273-1

Studies of past sea levels based on radiocarbon-dated field samples suffer variously from compilation of information from disparate locations and the imprecise nature of the dated indicators. Geographic segregation and systematic elimination of poor samples from time–depth data sets leads to improved interpretation of Holocene sea-level history. An example is presented from Southern Australia, where this is achieved through selection of higher-resolution palaeoenvironmental indicators, separation of transgressive from regressive populations, and geographic regionalisation of data. In Southern Australia, fossil sea-level indicators preserved in prograding coastal plain settings include seagrass, sandflat, mangrove, samphire and chenier ridge organo–sedimentary remains. These provide sea-level time–depth data points with a variety of elevational and dating errors. Preserved organic remains at the transition from Posidonia seagrass to intertidal sandflat environment, and from sandflat to Avicennia mangrove environment provide the most precise local data for tracing sea-level change. Time–depth plots of 233 dated sea-level indicators from South Australia generate a broad sea-level envelope tracing the Southern Australian transgression from 10,000 to ca. 6000 radiocarbon years BP, followed by a more or less consistent level to the present. Finer details of sea-level behaviour are only apparent after systematic selection, separation and regionalisation of the data. The data indicate a very rapid sea-level rise in the early Holocene, at about 16 mm/year, reaching present levels at 6400 years BP. This was followed by regionally variable regression and emergence of the land of 1–3 m, a process that continues to the present. The systematic increase from 1 to 3 m in the elevation of the 6400-year BP highstand with distance away from the shelf margin is consistent with a hydro-isostatic origin for the emergence as predicted by geophysical models.