Catherine Neelamraju

Can you predict the concentrations of one nutrient species using the known concentrations of other nutrient species?

The Queensland Government has been collecting nutrient concentrations in South-East Queensland (SEQ) and Great Barrier Reef (GBR) catchments for almost 2 decades, and in that time has amassed approximately 400,000 datapoints. With such large datasets, we are now in a position to develop statistical relationships between the concentrations of all the water quality parameters. Such relationships can be used to set up Quality Assurance/Quality Control rules for current and future sampling programs.

Furthermore, the logistics of sampling and analysing those samples is a major expense for the Queensland Government, and the Great Barrier Reef Catchment Loads Monitoring Program (GBRCLMP) and SEQ Catchment Loads Monitoring Program are always looking for avenues to reduce cost or provide justification for the reduction or simplification of sampling programs.

This project will suit a student with an interest in building statistical models of environmental datasets.

Can the annual dissipation of pesticides at the GBRCLMP sites be described mathematically?

The Queensland Government has been collecting pesticide concentrations in South-East Queensland (SEQ) and Great Barrier Reef (GBR) catchments since 2009.  Pesticide runoff at the beginning of the wet season generally results in dramatic increases in pesticide concentrations in receiving waters.  These concentrations subsequently reduce, presumably because the amount of pesticide available for run off reduces as the wet season progresses. This results in a pattern of pesticide dissipation that is repeated each year. The pattern is most evident in the high-risk catchments.

Each site potentially has its own dissipation pattern and raises the question: can you describe the site-specific dissipation of pesticides using a mathematical function, and is this pattern common to different pesticides?

This project will suit a student with an interest in building mathematical models of environmental datasets.

Fingerprinting water: Using real-time spectra to predict the concentrations of pesticides

This project will explore options to value add to real time data currently being collected by TriOS OPUS probes deployed for the Great Barrier Reef Catchment Loads Monitoring Program. The probes provide nitrate and total suspended solid (TSS) equivalent data. But there is potential to derive greater value from the probes. The potential to use the real time TSS data calculate sediment loads, rather than the traditional approach of using grab samples, would be investigated.

There is also potential to use the spectra generated by TriOS OPUS probes to predict concentrations of pesticides (e.g., atrazine, diuron, with nitrate, phosphate and imidacloprid previously analysed by another honours project in 2023, the thesis is available and this research can be built on or new approaches can be used) in rivers that discharge to the Great Barrier Reef. Establishing such relationships could enable real-time water quality monitoring of waterways for a greater number of water quality parameters and would enable quicker and cheaper methods of monitoring waterways within the Great Barrier Reef catchments.

Predicting the required sample coverage for a flood event

The Great Barrier Reef Catchment Loads Monitoring Program (GBRCLMP) monitors water quality during ambient conditions as well as during events. Events, or periods of increased flow, account for a significant portion of annual transport of contaminants.

The GBRCLMP directs significant resources to ensuring that events are well sampled. Calculation of an event load requires samples to be captured across the rising, peak and falling stages of the hydrograph. To ensure that a load can be calculated, more samples are collected than analysed. Once the event has passed, analysis of the hydrograph determines which samples will be analysed. The effort to oversample, review and then discard extra samples is considerable. Contaminant loads for an event are a function of catchment characteristics (size of the catchment, topography, geology and soil types and catchment landuse) as well as event specific characteristics (rainfall intensity and distribution, groundcover and land management practices).

This project aims to analyse monitoring data from the GBRCLMP to determine whether catchment characteristics are the primary driver of event characteristics and therefore could be predicted based on knowledge of site behaviour. The work can potentially reduce the costs associated with oversampling while ensuring the GBRCLMP continues to provide robust sampling for loads calculation.