Estimation of effluent quality parameters from an activated sludge system using quantitative image analysis

The efficiency of an activated sludge system is generally evaluated by determining several key parameters related to organic matter removal, nitrification and/or denitrification processes. Off-line methods for the determination of these parameters are commonly labor, time consuming, and environmentally harmful. In contrast, quantitative image analysis (QIA) has been recognized as a prompt method for assessing activated sludge contents and structure. In the present study an activated sludge system was operated under different experimental conditions leading to a variety of operational data. Key parameters such as chemical oxygen demand (COD) and ammonium (N-NH4+), and nitrate (N-NO3−) concentrations, throughout the experimental periods, were measured by classical analytical techniques. QIA was further used for the microbial community characterization. Partial least squares (PLS) models were used to correlate QIA information and the aforementioned key parameters. It was found that the use of the morphological and physiological data allowed predicting, at some extent, the effluent COD, N-NH4+, and N-NO3− concentrations based on chemometric techniques.