Numerical modelling of oil-sands tailings dam breach runout and overland flow.

Tailings dams, used for containing the residue of mining processes, are very important elements of the Alberta oil-sands industry in Canada. Potential breach of any of these dams can have catastrophic impact on the environment, economy and human health and safety. Therefore, understanding the after-breach processes is a crucial step in hazard analysis and response planning. This paper studies the potential consequence of a hypothetical oil-sands tailings dam breach by performing numerical simulations of the runout and non-Newtonian overland flow of tailings, including the resulting flooding condition and subsequent spill to nearby water bodies. A non-Newtonian dam-breach model with a visco-plastic rheological relationship is used for this purpose. The model is first validated using the 2014 Mount Polley tailings dam breach in British Columbia, before its application to investigate the flooding volume, extent, and downstream hydrograph of a hypothetical breach from a selected oil-sands tailings dam. The validation results show that the model is able to reproduce the flooding extent and water level variation (due to breach wave) at a downstream lake. The oil-sands tailings spill simulation study demonstrated the importance of considering the non-Newtonian behaviour of tailings materials as the non-Newtonian approach resulted in twice as long flood travel time and slightly less spill volume to the downstream river (i.e. Lower Athabasca River) as that of a Newtonian fluid (i.e. water). The results are also found to be highly sensitive to the rheological parameters of the tailings materials such as their viscosity and yield stress that need to be determined through proper calibration.

Normalized dynamic behavior of combined sewer overflow discharges for source water characterization and management.

As one of the major sources of surface water quality impairments, Combined Sewer Overflows (CSOs) are of concern when receiving waters are used for drinking water supplies. Given the large number and variability in CSO discharges and loads, there is a need for a general methodology for estimating discharges for environmental planning and source water protection. Detailed data on CSO flowrates, contaminant concentrations including Total Suspended Solids (TSS), Escherichia coli (E. coli), caffeine (CAF) and acetaminophen (ACE) were used to develop a simple loading model that was then verified using discharge and concentration data from other CSO and stormwater events in the literature. The variability of the parameters within each event was analyzed by normalizing flowrate, concentration and event duration to their respective peak values. The normalized flowrate data indicate that the second decile of the discharge periods was associated with peak flowrates. The dynamic behavior of CSO flowrates can be characterized by a linearly increasing trend and then a logarithmically decreasing trend in terms of normalized values. The samples captured during the first decile of the events were illustrated to be a better representation of peak concentrations of all four contaminants. By analyzing the discharge period in three sections (i.e. 1st decile, 2nd decile and remainder), a semi-probabilistic CSO loading model is proposed for the entire discharge period taking into account the variability of the phenomena. Findings can help water managers and utilities to characterize their source waters for better planning and to more efficiently design sampling campaigns for capturing peak concentrations at drinking water treatment plants.

Microbial risk associated with CSOs upstream of drinking water sources in a transboundary river using hydrodynamic and water quality modeling.

Urban source water protection planning requires the characterization of sources of contamination upstream of drinking water intakes. Elevated pathogen concentrations following Combined Sewer Overflows (CSOs) represent a threat to human health. Quantifying peak pathogen concentrations at the intakes of drinking water plants is a challenge due to the variability of CSO occurrences and uncertainties with regards to the fate and transport mechanisms from discharge points to source water supplies. Here, a two-dimensional deterministic hydrodynamic and water quality model is used to study the fluvial contaminant transport and the impacts of the upstream CSO discharges on the downstream concentrations of Escherichia coli in the raw water supply of two drinking water plants, located on a large river. CSO dynamic loading characteristics were considered for a variety of discharges. As a result of limited Cryptosporidium data, a probability distribution of the ratio of E. coli to Cryptosporidium based on historical data was used to estimate microbial risk from simulated CSO-induced E. coli concentrations. During optimal operational performance of the plants, the daily risk target was met (based on the mean concentration during the peak) for 80% to 90% of CSO events. For suboptimal performance of the plants, these values dropped to 40% to 55%. Mean annual microbial risk following CSO discharge events was more dependent on treatment performance rather than the number of CSO occurrences. The effect of CSO-associated short term risk on the mean annual risk is largely dependent on the treatment performance as well as representativeness of the baseline condition at the intakes, demonstrating the need for assessment of treatment efficacy. The results of this study will enable water utilities and managers with a tool to investigate the potential alternatives in reducing the microbial risk associated with CSOs.

Modelling the potential effects of Oil-Sands tailings pond breach on the water and sediment quality of the Lower Athabasca River.

Within the Oil-Sands industry in Alberta, Canada, tailings ponds are used as water recycling and tailings storage facilities (TSF) for mining activities. However, there could be possible circumstances under which a sudden breach of an embankment confining one of the TSFs may occur. Such a tailings pond breach would result in a sudden release of a huge volume of Oil Sands process-affected water (OSPW) and sediment slurry containing substantial amount of chemical constituents that would follow the downstream drainage paths and subsequently enter into the Lower Athabasca River (LAR). This study investigates the implications of OS tailings release on the water and sediment quality of the LAR by simulating the fate of sediment and associated chemicals corresponding to a hypothetical breach and release scenarios from a select set of tailings ponds using a two-dimensional hydrodynamic and constituent transport model. After predicting the total volume, time evolution and concentration of sediment and associated chemicals (metals, polycyclic aromatic hydrocarbons (PAHs) and naphthenic acids (NAs)) reaching the LAR, the transport and deposition of these materials within the study reach is simulated. The results show that, depending on tailings release locations, between 40 and 70% of the sediment and associated chemicals get deposited onto the river bed of the 160 km study reach while the rest leaves the study domain during the first three days following the release event. These sediment/chemicals deposited during the initial spill may also have long-term effects on the water quality and aquatic ecosystem of the river and the downstream delta. However, care has to be taken in interpreting the results as further analysis has shown that the outcomes of such model simulations are very sensitive to the various underlying assumptions as well as the values assigned to some model parameters representing the physical properties of the tailings material.

Two-dimensional numerical modelling of sediment and chemical constituent transport within the lower reaches of the Athabasca River.

Flows and transport of sediment and associated chemical constituents within the lower reaches of the Athabasca River between Fort McMurray and Embarrass Airport are investigated using a two-dimensional (2D) numerical model called Environmental Fluid Dynamics Code (EFDC). The river reach is characterized by complex geometry, including vegetated islands, alternating sand bars and an unpredictable thalweg. The models were setup and validated using available observed data in the region before using them to estimate the levels of cohesive sediment and a select set of chemical constituents, consisting of polycyclic aromatic hydrocarbons (PAHs) and metals, within the river system. Different flow scenarios were considered, and the results show that a large proportion of the cohesive sediment that gets deposited within the study domain originates from the main stem upstream inflow boundary, although Ells River may also contribute substantially during peak flow events. The floodplain, back channels and islands in the river system are found to be the major areas of concern for deposition of sediment and associated chemical constituents. Adsorbed chemical constituents also tend to be greater in the main channel water column, which has higher levels of total suspended sediments, compared to in the flood plain. Moreover, the levels of chemical constituents leaving the river system are found to depend very much on the corresponding river bed concentration levels, resulting in higher outflows with increases in their concentration in the bed sediment.

An integrated numerical framework for water quality modelling in cold-region rivers: A case of the lower Athabasca River.

There is a great deal of interest to determine the state and variations of water quality parameters in the lower Athabasca River (LAR) ecosystem, northern Alberta, Canada, due to industrial developments in the region. As a cold region river, the annual cycle of ice cover formation and breakup play a key role in water quality transformation and transportation processes. An integrated deterministic numerical modelling framework is developed and applied for long-term and detailed simulation of the state and variation (spatial and temporal) of major water quality constituents both in open-water and ice covered conditions in the lower Athabasca River (LAR). The framework is based on the a 1D and a 2D hydrodynamic and water quality models externally coupled with the 1D river ice process models to account for the cold season effects. The models are calibrated/validated using available measured data and applied for simulation of dissolved oxygen (DO) and nutrients (i.e., nitrogen and phosphorus). The results show the effect of winter ice cover on reducing the DO concentration, and a fluctuating temporal trend for DO and nutrients during summer periods with substantial differences in concentration between the main channel and flood plains. This numerical frame work can be the basis for future water quality scenario-based studies in the LAR.