Lihua Cui (Professor in Environmental Science and Engineering)

Constructed wetlands (CWs), which is consisted of wetland plants, wetland substrate, micro-organisms and other constituent elements, are effective systems to treat wastewater through the complex combination of physical, chemical and biological processes. In a mature constructed wetland system, a large number of micro-organisms and substrates, plants and other organisms interconnected to form an independent ecosystem. When wastewater flows into the wetlands, suspended solids (SS) can be intercepted by substrates and plant roots, while dissolved organic pollutants can be removed by biofilm through adsorption, assimilation and dissimilation.

Water flow direction (a) and schematic diagram (b) of hybrid vertical down-flow constructed wetlands systems.

Recently, We designed a new constructed wetland system named two-step hybrid vertical down-flow constructed wetland systems. The results showed that the hydraulic loading and step-feeding ratio were two crucial factors determining the removal efficiency of most pollutants, while substrate types only significantly affected the removal of COD and NH4+-N. Generally, the lower the hydraulic loading, the better removal efficiency of all contaminants, except for TN. By contrast, the increase of step-feeding ratio would slightly reduce the removal rate of ammonium and TP but obviously promoted the TN removal. Therefore, the optimal operation of this CWs could be achieved with low hydraulic loading combined with 50% of step-feeding ratio when TN removal is the priority, whereas medium or low hydraulic loading without step-feeding would be suitable when TN removal is not taken into consideration. The obtained results can provide us with a guideline for design and optimization of hybrid vertical flow constructed wetland systems to improve the pollutants removal from domestic sewage.

Significance (P value) in general linear model analysis of variation between three factors and five indexes：**Correlation is significant at the 0.01 level，* Correlation is significant at the 0.05 level.