However, in-situ sensors are increasingly being used to monitor river water quality at fine time scales. The advantage to using these continuous in-situ sensors is that data can be analysed in real-time to detect water-quality issues that may be detrimental to ecosystem and public health (i.e. global anomalies). The limitation is that in-situ sensors are prone to drift due to biofouling (i.e. local anomalies) and thus require regular maintenance.
In this project, you will build on existing anomaly-detection research being undertaken at ARC Centre of Excellence in Mathematical and Statistical Frontiers (ACEMS), implement automated algorithms for detecting and classifying sensor anomalies and, if possible, automate data corrections and/or metadata quality coding. In addition, relationships between low-cost sensor data (e.g. temperature, conductivity, turbidity, nitrate etc.) and sediment and nutrient concentrations will be explored, with the future objective of using surrogate measurements to produce real-time nutrient and sediment load estimates.
Positions available: 2
Appointment: 8 months, starting immediately
Expertise in statistics, mathematics, machine learning or data science
R programming skills, or similar
Ability to work in a team and to collaborate effectively in-person and remotely
Good written and verbal communication skills
Desire to work on real-world problems
Please contact Erin Peterson (Erin.Peterson@qut.edu.au) with questions about the positions.
The positions will be located in Brisbane, starting immediately. They will be open until suitable candidates are identified.