|
Advanced treatment for nitrogen and phosphorus removal is required at many municipal wastewater treatment plants
to meet increasingly stringent discharge standards. The proven approach nitrogen removal typically has involved
biological nitrification/denitrification, while phosphorus removal has been accomplished by some combination of
physical-chemical and biological processes. While there is a wide array of available processes for nutrient removal,
there is little in the literature that indicates the statistical reliability of the processes in meeting specific
discharge standards. This paper will present full-scale performance results and reliability data for nutrient removal
obtained from treatment plants located around the United States. Reliability will be statistically assessed by
calculating the coefficient of variation (COV) for annual or seasonal effluent data. Higher values of the COV
indicate a larger standard deviation relative to the mean, so that a reliability target of plus or minus one standard
deviation about the mean will be relatively large, meaning the process is less reliable.
For nitrogen removal, a limit of 5 mg/L nitrogen can largely be obtained via one stage nitrification/denitrification,
while a limit of 3 mg/L nitrogen will require two anoxic stages to reliably meet the goal. For phosphorus removal,
chemical addition at two application points is needed to reliably obtain concentrations under 0.5 mg/L, while to meet
a limit of 0.1 mg/L, a filtration system is typically required for reliable results. Data showing the COV for effluent
nutrients at a number of facilities around the United States will be presented, with the conclusion being what processes
have the most promise for economically meeting specified targets.
|
