ABSTRACT
Constructed wetlands (CWs) are natural-friendly method to eliminate the pollutants stormwater pollutants. For this study, a pilot-scale treatment system consisting of a sedimentation tank (ST) and five recirculated fill-and-drain wetlands (namely CW-1, CW-2, CW-3, CW-4 and CW-5) were constructed to treat the first-flush from the impervious road. From bottom to top, the main substrates of CWs were selected as 0 cm woodchip + 60 cm pumice for CW-1, 15 cm woodchip + 45 cm pumice for CW-2, 30 cm woodchip + 30 cm pumice CW-3, 45 cm woodchip + 15 cm pumice CW-4 and 60 cm woodchip + 0 cm pumice for CW-5. During the operational period, the reduction efficiencies of TSS, total COD, total nitrogen and total phosphorus were 89–100%, 52–100%, 31–87% and 72–100% by CW-1, 92–100%, –27% to 78%, 8–85%, 49–94% by CW-2, 89–100%, –97% to 77%, –29% to 86%, –46% to 94% by CW-3, 89–100%, –115 to 69%, –21% to 99%, 35–94% by CW-4 and 59–100%, –342%to 88%, –20% to 88%, –77% to 99% by CW-5, respectively. Overall, the reduction efficiency, except for nitrogen, declined as the amount of woodchip increased. As excepted, the employment of woodchip improved denitrification via providing with carbon sources, resulting in low NO3–N but unfavourable COD concentrations in the effluent. And the combination use of woodchip and pumice as CW-3 or CW-4 is suggested, whereas the specific amount of woodchip and pumice should be determined based on the regional environmental goal.
GRAPHICAL ABSTRACT
![](/cms/asset/96d38cb6-783e-45b4-837a-11fe1e94720c/tent_a_1543356_uf0001_oc.jpg)
Acknowledgement
The authors are grateful for the support.
Disclosure statement
No potential conflict of interest was reported by the authors.