Sediment Transport and Metals Modeling in an Urban Stream - The Don River, Toronto
Mansoor, Sannan Zahid
MetadataShow full item record
The Don River watershed has been subjected to rapid urbanization over the last few decades. As a result, vast area of built-up land has shifted the watershed’s hydrologic cycle towards lower infiltration and higher runoff rates. Such a drastic hydrologic change has resulted in frequent flooding, channel widening and erosion, and poor water quality in the region. Metals sourced from roads, landfills, industrial effluents, and wastewater treatment plant are a particularly damaging component to the system and need to be quantified and addressed. A research study was conducted by (Louie, 2014) to quantify the trace metals distribution in the Don River system and study the spatial and temporal trends of copper, lead, and zinc concentrations. It recognized the limitations in quantifying such information on a watershed scale. Efforts have been made to restore the natural water cycle of the watershed by the local authorities such as the Toronto and Region Conservation Authority (TRCA). Regional Watershed Monitoring Program (RWMP) was launched by TRCA in 2002 to monitor the surface water quality in the region. Moreover, Wet Weather Flow Management Guidelines (WWFMG) (City of Toronto, 2006) is a document currently used to design stormwater management solutions and restoration plans to control the surface water quantity and quality in the region. Challenges related to quantification of sediments and associated metals flushing through the system can be addressed through implementing appropriate modeling tools. Hydrologic models are commonly used, but they lack the capability to model instream processes that are important in case of metals. Metals can bind to the sediments and can remain in the system for years creating ‘hot spots’ of deposition with possibly elevated local levels of other pollutants. Incorporating the simulation of instream processes can enable understanding of temporal and spatial distribution of sediments and metals in detail, which is required for advanced infrastructure planning and informed decision making to restore the river network where possible and mitigate the damage where it is not. The research aims to advance this understanding through the help of a 1-dimensional (1D) numerical model of the lower Don River extending from Taylor Creek South to the mouth of the river at Keating Channel. Total length of this reach is 9.81Km and it is confluent with two primary tributaries of the Don River, the East Don and the West Don. The metals which are focused in this study are copper, lead, and zinc as they are primarily sourced from urban centers. Hydrologic model and a hydraulic model are used in this thesis. A program is developed as a secondary objective of this thesis to link the urban hydrologic model of the river to the hydraulic model to efficiently set up the latter for detailed modeling of instream processes. Two commercially available modeling packages are linked in this thesis. The first model is an urban watershed modeling tool called PCSWMM. TRCA has developed a hydrologic model of the entire Don River watershed using this program. Their calibrated model currently simulates the hydrology for a time span of 40 days from June 20 to July 30, 2008. The model provided by the TRCA is extended to a longer period in this thesis, and the modules for sediment and metals buildup and wash-off are activated and parameterized to simulate input loads to the channel. A second model called the Environmental Fluid Dynamics Code (EFDC) is used for advanced hydrodynamic, sediment transport, and metals fate and transport modeling of the lower Don River. The EFDC model is necessary because PCSWMM does not have the capability to simulate instream physical processes related to sediment and metals transport. Examples of processes that can be simulated in EFDC that are not possible in PCSWMM include erosion, deposition, and resuspension of sediments along with diffusion and sorption of metals to sediments. PCSWMM cannot simulate sediment bed dynamics and its pollutant composition. It only has the capability to estimate pollutant loads from subcatchments using buildup and wash-off models and land use information. It routs these loads through the hydraulic network using a completely mixed or plug flow assumption. Therefore, a dedicated model that can simulate the governing physical processes in an integrated manner is required. EFDC Explorer is used to develop a representative 1D hydrodynamic, sediment transport, and metals transport model in a coupled approach. EFDC Explorer is the commercially available user interface for pre and post processing of the EFDC model. The existing PCSWMM model of the Don River was upgraded and verified to provide pollutant loads from subcatchments spanning the time period of interest from May to August 2010. The linking of the PCSWMM and EFDC model is achieved through development of a program written in MATLAB® R2014b. This program, called the SWMM to EFDC Model Setup tool (STEMS), creates the grid and boundary condition files in a format compatible with EFDC and reports other information for efficient setup of the EFDC model. It can be applied to any river network modelled in PCSWMM for further analysis in EFDC. The comparison between the results of EFDC and PCSWMM model showed that the EFDC model better predicted measured suspended sediment and metals loads in comparison to the PCSWMM model alone. The hydraulic results of the two models were similar and showed high correlation. This suggested high sensitivity of EFDC hydraulic results to the boundary conditions provided by PCSWMM. However, the sediment and metal results were clearly different for the two models. The superior performance of the EFDC model further highlighted the importance of instream physical processes in sediment and contaminant transport rather than adopting simplifying assumptions. The relation of suspended sediment and total metal concentrations with river discharge suggested good agreement with the observed data set at the Todmorden monitoring station provided by TRCA and Environment Canada. Baseflow levels suggested that metals are deposited during low flow periods along with sediments and this material is resuspended during high flow events. Moreover, resulting sediment bed metal concentrations at the mouth of the river also agreed with the suggested trend provided by TRCA for the dredged sediment in the Keating Channel. These results verified that the model is representative of the actual conditions. It can be used as a predictive tool to estimate the total metal loads flushed from the river associated with the deposited sediments.