Statistical Modeling of Batch Versus Continuous Feeding Strategies for Pollutant Removal by Tropical Subsurface Flow Constructed Wetlands

In this study, six mesocosms, including three subsurface flow constructed wetlands (SSF CWs) and three sand filters (without plants) were set up at the campus of Nanyang Technological University, Singapore. The objective of this study was to compare the removal efficiencies of chemical oxygen demand (COD), nitrogen (N) and total phosphorus (TP) under batch and continuous operational modes. Three factors, namely, with/without the presence of vegetation, operational modes (batch and continuous) and hydraulic residence time were investigated by two statistical models including factorial design and its rationale analysis and a quadratic polynomial regression model with ANOVA, to find the relationships between the contaminant removal efficiencies and the affecting factors, as well as to determine the dominant variables and how each of them interact for each parameter.