Athabasca River Water Quality Modelling 1990 Update

Resource Type
Authors
Gordon Macdonald
Alfred Rademacher
Resource Date:
1992
Page Length
45

The calibrated DOSTOC, NUSTOC and UNSTOC models from 1989 were used effectively to simulate measured 1990 Athabasca River water quality. Changes in effluent loadings at the two pulp mills and an increase in river flows in the lower Athabasca Basin accounted for some noticeable improvements in water quality from that reported in Noton and Shaw (1990). Dissolved oxygen simulations for 1990 incorporated ultimate BOD and kinetic rate information from the long-term BOD laboratory measurements. The laboratory results indicated that at Millar Western, ultimate BOD amounts were substantially reduced from 1989 measurements, and as well the BOD oxidation rate was significantly lower. A lower oxidation rate combined with a reduced BOD ultimate loading accounted for most of the improvements in observed river dissolved oxygen concentrations. Weldwood ultimate BOD levels were also reduced on the sampling date, however, there was not a significant change in the oxidation rate. Sediment oxygen demand measured in 1990 was not appreciably different from measurements in 1989 despite substantial reductions in BOD loadings. There was, however, insufficient information to modify the modelling approach and consequently SOD was assumed to change proportionately with BOD. Phosphorus was modelled as particulate, dissolved and as total (the sum of two forms). Changes in total phosphorus concentrations in the river were greater below. Millar Western than Weldwood. The total phosphorus levels measured and simulated for 1990 were lower than previously measured. The simulation of nitrogen included organic nitrogen, nitrate, ammonia and total nitrogen. An ammonia oxidation rate was derived using the loss of ammonia and gain in nitrate with river distance. Difficulty arose over the mass balance of ammonia and nitrate at White court where measured downstream concentrations were greater than the sum of the Millar Western effluent and Macleod River contributions. Total nitrogen levels simulated and measured in the Athabasca River compare with other measured levels (Noton and Shaw 1990). Colour was modelled as a conservative parameter. Measured and simulated colour levels in the upper Athabasca River were lower in the 1990 survey than 1989 due to reduced colour inputs from Weldwood. Suspended solids simulations included settling of material below the two mill discharges. Millar Western's influence on river suspended solids appears to be reduced from 1989, however the longitudinal river pattern is similar. Following the modelling approach used for the Peace and Wapiti/Smoky River systems, the WASP model was configured for the Athabasca River and used to simulate two representative organic compounds, 2,4,6 trichlorophenol (TCP) and dehydroabietic acid (DHA). The WASP simulations predict concentrations in the water column and the sediments, as dissolved or sediment sorbed phases. Results indicate differences between TCP and DHA in their behaviour in the receiving waters. Expected concentrations in the benthic sediments or water column are lower than conventional analytical techniques could detect.