Predicting food web responses to biomanipulation using Bayesian Belief Network: Assessment of accuracy and applicability using in-situ exclosure experiments

Influence of surface properties and antecedent environmental conditions on particulate-associated metals in surface runoff

Particulate-associated trace metals have been regarded as an important pollution source for urban surface runoff. Cd, Pb, Cu, Zn and total solids (TS) washed off two different surfaces (low-elevated facade and road surfaces) under two kinds of antecedent environmental conditions (dry and snow-melting) were determined in this study. Wet-vacuuming sweeping (WVS) and surface washing (SW) methods, representing the maximum pollution potential and common rainfall-induced wash-off condition respectively, were used to collect the particulate matters. The result shows that the wash-off concentrations of trace metals were found in the order of Cd (2.28 ± 2.08 μg/l) < Pb (435.85 ± 412.61 μg/l) < Cu (0.93 ± 0.61 mg/l) < Zn (2.52 ± 2.30 mg/l). The snow-melting process had a considerable influence on the wash-off concentrations of the trace metals on both road and facade surfaces. It reduced >38% and >79% of metals and TS concentrations in the facade surface and road surface runoff respectively. The wash-off concentrations of Cd, Cu, and Zn on the road surface 45-780% higher than those on the facade surfaces. The sensitivity analysis based on the Bayesian network indicates that the wash-off concentrations of metals were mainly dependent on the antecedent environmental conditions or the surface properties while the sampling methods had a minor influence. Therefore, to accurately model the pollutant migration in the surface runoff requires an improving method considering different surfaces and antecedent environment conditions.