Microbial Water Quality Modelling of Recreational Beaches in Windsor Essex Region
Standing
Graduate (PhD)
Type of Proposal
Oral Presentation
Faculty
Faculty of Engineering
Faculty Sponsor
Rajesh Seth
Proposal
High microbial contamination levels deemed to be unsafe for beach goers resulting in the closure of beaches are routinely observed in Lake St. Clair. Major tributaries to Lake St. Clair are dominant flow and nutrient contributors to the lake and are expected to bring microbial contamination to the lake - a specific area of interest is Sandpoint Beach. In order to better track the sources of microbial contamination, find out the level contribution from major tributaries and to have the capability of to predict the level of microbial contamination at the beaches in real time, a model with the capability to predict the unsteady hydrodynamics in 3D space and time is necessary. In this study, a 3D, high-resolution model of Lake St. Clair was set up over a four-month period (June-September) of 2010 using the 3-Dimensional coupled Hydrodynamic-Aquatic Ecosystem Model, (AEM3D) framework. Escherichia coli (E. coli) fate and transport were modelled assuming a first-order decay rate. The performance of the model in producing water temperature and microbial concentration is evaluated through comparison against one month measured (from previous study) data and satellite images. High agreement between simulated water temperature and Great Lakes Surface Environmental Analysis (GLSEA) images was observed with Root Mean Square Error (RMSE) from 0.79 to 1.55 °C, and Normalized Root Mean Square Error (NRMSE) from 0.32 to 0.63 °C. The vertical distribution of temperature revealed little to no stratification in the water column. Both the model predicted profile and the temperature values were in good agreement with the observed buoy data with RMSE and Mean Absolute Error (MAE) values of 0.8 and 0.6 respectively. Simulated E. coli concentrations showed that maximum predicted fecal concentrations from the combined input of the major tributaries to be less than 100 CFU/100 mL for most of the lake and less than 10 CFU/100 mL at Sandpoint Beach. These results indicate that the major tributaries are unlikely responsible for the observed E. coli exceedances at Sandpoint Beach.
Grand Challenges
Healthy Great Lakes
Microbial Water Quality Modelling of Recreational Beaches in Windsor Essex Region
High microbial contamination levels deemed to be unsafe for beach goers resulting in the closure of beaches are routinely observed in Lake St. Clair. Major tributaries to Lake St. Clair are dominant flow and nutrient contributors to the lake and are expected to bring microbial contamination to the lake - a specific area of interest is Sandpoint Beach. In order to better track the sources of microbial contamination, find out the level contribution from major tributaries and to have the capability of to predict the level of microbial contamination at the beaches in real time, a model with the capability to predict the unsteady hydrodynamics in 3D space and time is necessary. In this study, a 3D, high-resolution model of Lake St. Clair was set up over a four-month period (June-September) of 2010 using the 3-Dimensional coupled Hydrodynamic-Aquatic Ecosystem Model, (AEM3D) framework. Escherichia coli (E. coli) fate and transport were modelled assuming a first-order decay rate. The performance of the model in producing water temperature and microbial concentration is evaluated through comparison against one month measured (from previous study) data and satellite images. High agreement between simulated water temperature and Great Lakes Surface Environmental Analysis (GLSEA) images was observed with Root Mean Square Error (RMSE) from 0.79 to 1.55 °C, and Normalized Root Mean Square Error (NRMSE) from 0.32 to 0.63 °C. The vertical distribution of temperature revealed little to no stratification in the water column. Both the model predicted profile and the temperature values were in good agreement with the observed buoy data with RMSE and Mean Absolute Error (MAE) values of 0.8 and 0.6 respectively. Simulated E. coli concentrations showed that maximum predicted fecal concentrations from the combined input of the major tributaries to be less than 100 CFU/100 mL for most of the lake and less than 10 CFU/100 mL at Sandpoint Beach. These results indicate that the major tributaries are unlikely responsible for the observed E. coli exceedances at Sandpoint Beach.