ABSTRACT
The purpose of this study is to evaluate the efficiency of hybrid constructed wetlands (HCWs) in a rural mountainous area. The experiment was set up in small rural community named Tidili within the region of Marrakech, Morocco. The wastewater treatment plant was composed of three vertical flow constructed wetlands (VFCWs) working in parallel, followed by two parallel horizontal-subsurface flow constructed wetlands (HFCWs), with hydraulic loading rates of 0.5 and 0.75 m3/m2.d, respectively. The two units were planted with Phragmites australis at a density of 4 plants/m2. Wastewater samples were collected at the inlet of the storage tank and at the outlet of the whole system (VFCWs, HFCWs) stages. The main removal percentages of total suspended solids (TSS), biochemical oxygen demand measured in a 5-day test (BOD5), chemical oxygen demand (COD), total nitrogen, and total phosphorus were respectively 95%, 93%, 91%, 67%, and 62%. The system showed a very high capacity to remove total coliforms, fecal coliforms, and fecal streptococci (4.46, 4.31, and 4.10 Log units, respectively). Artificial neural networks (ANNs) were used to model the quality parameters (TSS, BOD5, COD) and total coliforms and fecal streptococci. Based on the obtained results, the ANN model could be considered as an efficient tool to predict the studied phytoremediation performances using HCWs.
Funding
This work was supported by the Pole of Competences on Water and Environment (PC2E); the National Centre for Studies and Research on Water and Energy (CNEREE), University of Cadi Ayyad and the Tidili Wastewater Reuse and Sewerage Project (PAREUZT) funded by the US-AID.