This thesis examines the application of paleolimnological techniques, pioneered and developed predominately in northern hemisphere freshwater environments, to sub-tropical east Australian estuaries. The aim of this research was to determine how successfully these techniques - in particular those related to coring, dating, diatom analysis and stable isotope analysis – can be either applied or adapted to the estuarine environment to reconstruct nutrient levels in the absence of monitoring data. This will allow pre-impact environmental conditions of estuaries to be determined, which in turn will aid future management decisions.
Fifty-two estuaries from a spatially large geographic area were initially sampled, to create a dataset that consisted of surface sediment diatoms, and associated water chemistry data from point water samples. The sampling strategy of extracting point samples from estuaries was the only feasible option due to budgetary, spatial and time constraints. Principal Components Analysis (PCA) identified that a relationship between point sampling water chemistry and pre-determined classifications of estuarine health does exist (OzCoasts ratings, and ANZECC trigger values). Hence, a point-water sampling strategy can have merit as sampling method in sub-tropical east Australian estuaries, particularly when assessing total nutrient levels.
The application of the statistical methodologies PCA, Canonical Correspondence Analysis (CCA), and variance partitioning, to the water chemistry and surface sediment diatom data determined that total phosphorus (TP) exhibited significant influence on diatom distribution in the fifty-two estuaries, although this signal may be confounded with total nitrogen. Two transfer functions techniques, Weighted Averaging (WA), and Weighted Averaging Partial Least Squares (WAPLS), were used to create inference models for TP on numerous groupings of estuaries. The group that had all estuaries removed that had below detection limit levels of bio-available nutrients (PO4, NH4, NO2, NOx) in their point water samples gave the strongest transfer function scores (r2jack = 0.69, RMSEP = 0.027 log10 mg TP l).
Five cores were extracted from depositional areas within three sub-tropical estuaries of differing environmental health; the Richmond River, NSW; the Burrum River, QLD; and Moreton Bay, QLD. These cores were put through a process of elimination involving loss-on-ignition testing and paleomagnetic profiling, to ensure that the cores of best integrity for paleolimnological analyses were selected. These cores were analysed using a combination of 210Pb and AMS 14C methods. Core chronologies were constructed for the riverine Richmond River and Burrum River. However, the open embayment Moreton Bay 210Pb results indicated background unsupported 210Pb levels, with 14C results also inconclusive, demonstrating sediment deposition from mixed sources.
Results from diatom and δ13C and δ15N stable isotope analyses indicated that the total phosphorus levels in the Richmond River have fluctuated since 1920, and are most likely a function of climate. In the Burrum River, diatom assemblages, and the upper and lower limits for δ13C and δ15N stable isotopes, have shown very little variation over the past 5000 years. These minimal changes to water quality over the time period indicate that the Burrum River has value as a suitable reference site for benchmarking impacted sub-tropical estuaries. In the Moreton Bay core, zero diatom preservation was observed, probably due to the salinity and temperatures characteristic of sub-tropical open embayment estuaries. Reconstructions using geochemical and isotopic signals were mixed, but did indicate that Moreton Bay has been receiving sediment loads from different catchment sources for the past 10 000 years.
This thesis has demonstrated that paleolimnological techniques can be applied and adapted successfully in sub-tropical riverine estuaries. The inferences for the Richmond River can be used by environmental managers to influence decision making relating to nutrient levels in the Richmond River. In the Burrum River, paleolimnological techniques were applied successfully to identify the site as having benchmarking value. This research has also demonstrated the difficulties that can be encountered when applying paleolimnological techniques to more dynamic open embayment estuaries, such as Moreton Bay. Future studies attempting to reconstruct environmental histories of estuaries using paleolimnological techniques should incorporate river-influenced locations rather than marine-dominated sites, to provide increased probabilities for diatom preservation and definitive geochemical signals.