Return to Lab
Ms.

 

Name: Wai Sum Chan (Grace)

PhD Candidate

Email: wai.s.chan@adelaide.edu.au

Project Title: Spatial and Temporal Variability of Phytoplankton Community in Drinking Water Reservoirs

Supervisor:
A/Prof Friedrich Recknagel, School of Earth and Environmental Sciences, The University of Adelaide
A/Prof Justin Brookes, School of Earth and Environmental Sciences, The University of Adelaide
Dr David Lewis, School of Chemical Engineering, The University of Adelaide

Collaborations:
School of Earth and Environmental Sciences, The University of Adelaide
School of Chemical Engineering, The University of Adelaide
CRC for Water Quality and Treatment
Shinshu Univerisity (Japan)
Institute of Hydrobiology, Chinese Academy of Sciences


Project Summary: The occurrence of cyanobacterial blooms and their toxins in freshwater bodies is a global environmental concern posing potential health threats to humans and animals as well as aesthetic problems. Serious Microcystis blooms followed by high microcystin concentration often occur in summer in Lake Suwa (Japan) and Torrens Lake (South Australia). Using wide ranges of physical, chemical and biological variables monitored in both shallow eutrophic lakes, the project aims to investigate the following:

(1) Ordination and clustering long-term time-series data of Lake Suwa (12 years) and the short-term time-series data of Torrens Lake (3 years) individually by non-supervised Artificial Neural Network (ANN). This method ordinates and clusters the causal relationships of complex, non-linear limnological parameters onto two-dimensional maps with regards to the seasonality and habitat conditions.

(2) Forecasting the timing and the magnitude of outbreaks of Microcystis populations and microcystin concentrations for 7-days-ahead in Lake Suwa by means of recurrent supervised ANN.

(3) Discovery of explanatory rule sets for forecasting timing and magnitudes of outbreaks of Microcystis populations and microcystin concentrations for 7-days-ahead in Lake Suwa by Evolutionary Algorithms (EA).

(4) Testing validity of models developed in (2) and (3) for forecasting and understanding outbreaks of Microcystis populations and microcystin concentrations in Torrens Lake.

(5) Testing validity of models developed according to (2) and (3) but trained by a combined data set of Lake Suwa and Torrens Lake for forecasting and understanding outbreaks of Microcystis populations and microcystin concentrations in Torrens Lake.

Outcomes of the project aim to better understand and forecast ecological processes and climate conditions responsible for temporary contamination of freshwater lakes by microcystin.

Publications:
Chan WS, Recknagel F, Cao H, Park H-P (2007) Elucidation and short-term forecasting of microcystin concentrations in Lake Suwa (Japan) by means of artificial neural networks and evolutionary algorithms. Water Research 41, 2247-2255. Welk A, Recknagel F, Cao H, Chan WS, Talib A (2008) Rule-based agents for forecasting algal population dynamics in freshwater lakes discovered by hybrid evolutionary algorithms. Ecological Informatics in press.

Research Interest:
Phytoplankton, cyanobacteria, toxins, freshwater, internal wave, Artificial Neural Networks, Evolutionary Algorithms