Kinetics of domestic wastewater COD removal by subsurface flow constructed wetlands using different plant species in temperate period

Abstract

The objective of this work has been to study the kinetics of domestic wastewater chemical oxygen demand (COD) removal when using constructed wetlands (CWs) with different plant species and at different water depths. Kinetic rate constants were obtained by using several kinetic models, and also combined with different hydraulic models. Synthetic wastewater was fed to five identical pilot-scale CWs, planted with different species (CW1: unplanted; CW2: Phragmites australis; CW3: Lythrum salicaria; CW4: Cladium mariscus; CW5: Iris pseudacorus). Wastewater was treated under continuous operation during 5 months. Water samples were taken along intermediate points at the wetland, and also at three different depths (top, medium depth and bottom). The COD experimental data were fitted to different kinetic models previously and extensively reported in the literature: the K-C and K-C* equations, and also the ‘retardation’ model. Also, the effect of the hydraulics characteristics was considered. Apart from the ideal plug flow assumption, two different flow models were used when integrating the mass balance equations: the plug flow with dispersion model and a detention time gamma distribution (DTGD) model. The more developed plants (CW3 and CW5) were the ones that caused an increase in COD removal rate compared with the unplanted wetland. Differences in COD removal rates were observed at different depths in the unplanted wetland, and the higher rate constant values were obtained near the wetlands top. On the contrary, the higher plants development in CW3 and CW5 eliminated the influence of water depth. The retardation model offered the best mathematical fitting to the experimental data. By using non ideal flow models, an increase in the rate constant values was always obtained, especially in the wetlands whose hydraulic behaviour was very far from the ideal plug flow. The rate constants values obtained using the DTGD model were higher (25–54%), compared to the values obtained if ideal flow was considered. These results could aid the design of CW, particularly in temperate periods.