Population Socioeconomics Predicted Using Wastewater

Abstract

Municipal wastewater typically contains many drugs and anthropogenic chemicals or biomarkers. The occurrence of these chemicals in wastewater is linked to the socioeconomic characteristics of the contributing population. Based on these relationships, we propose, execute and evaluate a novel model for predicting population socioeconomics. Specifically, we used biomarkers in wastewater to predict 37 socioeconomic characteristics of populations during the Australian Census. The resultant model was further tested on nine other populations separate from the training data set. Prediction performance in the test populations (defined as accuracy ± SD) fit within 75% and 125% for many features such as catchment median age, and specific measures of educational attainment (e.g., high school completion) and employment (e.g., managerial employment). Considering the relative ease, low cost and high frequency at which wastewater samples can be collected and analyzed, wastewater analysis could be used as a complementary technique for assessing population socioeconomics.