Identification and modelling the HRT distribution in subsurface constructed wetland.

Abstract

This study focused on the identification of the hydrodynamics of a horizontal subsurface constructed wetland (HSSF-CW) located in Beijing wildlife rescue and rehabilitation center, Beijing. The effects of plant growth of iris tectorum on the hydrodynamic behaviours were studied and the distribution of the hydraulic residence time was simulated by several mathematical models in order to understand the fluctuations and mixing processes of pollutants in the HSSF-CW. Treatment performance of the HSSF-CW was evaluated by comparing the area-based removal rates of different pollutants. According to the results, water depth has a negative effect on the plant growth and a larger hydraulic loading rate is not conducive to the growth of wetland plants. Modelling the probability density of the residence time distribution indicated that the shorter hydraulic residence time of 10.16 hours compared with a theoretical hydraulic residence time of 12.81 hours was responsible for the lower removal efficiency of pollutants (T-P: 0.17 ± 0.04 g m(-2) day(-1), T-N: 1.10 ± 0.05 g m(-2) day(-1), PO(4)-P: 0.08 ± 0.04 g m(-2) day(-1), NH(4)-N: 0.19 ± 0.02 g m(-2) day(-1), NO(3)-N: 0.52 ± 0.03 g m(-2) day(-1), Chl_a: 18.26 ± 0.09 g m(-2) day(-1)). The results of a superposition simulation of residence time distribution indicated that the asymmetric double sigmoidal (asym2sig) model is competent at providing a reasonable match between the measured and the predicted values to some extent. Based on the good fit of the experimental datasets by the asym2sig probability density function, the mathematical expectation approximated to the actual hydraulic residence time (10.16 hours) of the HSSF-CW.