Development of generalized empirical models for comparing effectiveness of wastewater nutrient removal technologies

Abstract

The effectiveness of nutrient removal approaches was quantified at four wastewater treatment facilities (WWTFs) using mechanistic models. Generalized empirical models were developed applying statistical methods on the predicted values characterizing nutrient removal as a function of influent wastewater quality. The empirical models provide a framework to estimate nutrient removal effectiveness and inform system-level decisions on technology adoption. When carbon limited, more sophisticated approaches like five-stage Bardenpho and nitrite shunt provide the most notable benefit in removal efficiency (67% ± 3.3% and 89% ± 2.8%, respectively for total nitrogen (TN)), but little benefit is estimated under non-carbon-limited conditions between traditional solutions like anaerobic, anoxic, oxic (A2O), and advanced process configurations like five-stage Bardenpho (82% ± 2.8% and 85% ± 3.3%, respectively for TN). Sidestream physical/chemical processes can provide improvement in removal efficiency particularly at carbon-limited WWTFs, but negligible benefit is estimated with adoption of sidestream biological processes.