Modeling highway runoff pollutant levels using a data driven model

Pollutants accumulated on road pavement during dry periods are washed off the surface with runoff water during rainfall events, presenting a potentially hazardous non-point source of pollution. Estimation of pollutant loads in these runoff waters is required for developing mitigation and management strategies, yet the numerous factors involved and their complex interconnected influences make straightforward assessment almost impossible. Data driven models (DDMs) have lately been used in water and environmental research and have shown very good prediction ability. The proposed methodology of a coupled MT-GA model provides an effective, accurate and easily calibrated predictive model for EMC of highway runoff pollutants. The models were trained and verified using a comprehensive data set of runoff events monitored in various highways in California, USA. EMCs of Cr, Pb, Zn, TOC and TSS were modeled, using different combinations of explanatory variables. The models' prediction ability in terms of correlation between predicted and actual values of both training and verification data was mostly higher than previously reported values. Pb_Total was modeled with an outcome of R² of 0.95 on training data and 0.43 on verification data. The developed model for TOC achieved R² values of 0.91 and 0.49 on training and verification data respectively.