Machine Learning analysis of E. coli data from the Yarra River

Project scope

• This project will explore the value of perspicuous Machine Learning models to analyse long-term E. coli monitoring data from the Yarra River.
• Will use high-quality observations from the Yarra Watch program and associated hydrometeorological data.
• Investigating (i) the key drivers of extreme E. coli events and (ii) the potential to predict E. coli concentrations from rainfall and streamflow.
• Performed as a desktop study based at the ANU School of Engineering.
• The work will serve as a proof-of-concept study.
• Suitable for a full-year final-year student project for ENGN4350, ENGN4712 or similar School of Engineering or School of Computing research-based courses.

Project description

This project will adapt an existing machine-learning based dissolved oxygen (DO) modelling framework to analyse long-term E. coli monitoring data from the Yarra River collected through the Yarra Watch program. Using E. coli observations from four monitoring sites together with rainfall and streamflow data, the study will investigate the key drivers of extreme E. coli events and assess the potential to predict E. coli concentrations from hydroclimatic variables. This project will provide great opportunity to collaborate industrial stakeholder at Melbourne Water, with flexibility in scope to focus on process understanding, predictive performance, or both, depending on stakeholder priorities. The project is envisaged to provide insights into both process understanding and predictive capability to support future water quality management applications.

Information for applicants

The ideal candidate will:

• have a strong interest in environmental or water quality modelling, and
• be motivated to work with real-world monitoring data to address applied water management problems.

Essential skills and background

• A passion in environmental or catchment modelling.
• Coursework in hydrology, environmental engineering, environmental science, data science, or a related discipline.
• Programming skills (e.g. R or Python) for data analysis.
• Familiarity with statistical or data-driven modelling concepts.
• Ability to work independently and communicate results clearly.
• Good academic standing (e.g. solid GPA).
• Strong organisational skills and attention to detail Desirable (nice-to-have).
• Experience with timeseries analysis or environmental datasets.
• Knowledge of water quality processes (e.g. microbial indicators, dissolved oxygen, nutrients).
• Exposure to machine learning, GAMs, or regression models.
• Interest in applied research and collaboration with industry partners (e.g. water utilities).

For more background on the Yarra Watch program, visit https://www.epa.vic.gov.au/check-air-and-water-quality

Takeaways

The student will: 

• gain hands-on experience working in laboratory environment
• contribute to experiment design and test planning
• learn accelerate UV ageing techniques
• perform lifetime measurements
• analyse and interpret experimental data.

How to apply

If you are interested, please send a copy of your CV (resume) and academic transcript to the project supervisor via email.