Estimating steroid oestrogen inputs into activated sludge treatment works and observations on their removal from the effluent

A method to predict steroid oestrogen inputs into sewage works is described and tested against available data. For oestradiol (E2), 68% of the predictions were within 50% of the actual measured value, and for oestrone (E1), 52% of the estimations were within 50% of the measured value. Predictions for ethinyloestradiol (EE2), which are particularly sensitive to assumptions on the number of people taking the oral contraceptive, were less accurate. Five Italian and three Dutch activated sludge treatment works (STW) were sampled on two to three occasions for E2, E1, EE2 and E3 (E3 only in Italy) in both influent and effluent waters. High concentrations of E3 were found in the inffluent, as predicted, with a mean of 57 ng/l and a mean of 10 ng/l in the effluent. Using the currently available data collected from composite samples, an average of 88% of E2 and 74% of E1 would appear to be removed by the activated sludge process. (C) 2000 Elsevier Science B.V. All rights reserved.