Simulation of constructed wetlands treating combined sewer overflow using HYDRUS/CW2D

• Published in: Ecological engineering 2016-02, Vol.87, p.340-347
• Main Authors: Pálfy, T.G., Molle, P., Langergraber, G., Troesch, S., Gourdon, R., Meyer, D.
• Format: Article
• Language: English
• Subjects: Biokinetics; Combined sewer overflow; Constructed wetland; CW2D; Environment and Society; Environmental Sciences; Process-based model
• ISSN: 0925-8574; 1872-6992
• DOI: 10.1016/j.ecoleng.2015.11.048

•We simulated combined sewer overflow constructed wetlands (CSO CWs) in HYDRUS/CW2D.•Previous research showed strong limitations simulating CSO CWs in HYDRUS/CW2D.•We calibrated biokinetic and other parameters using laboratory column experiments.•We report quazi-stable biomass and a good fit for single loads and a load series.•Former limitations were overcome and validation can continue with full scale. Constructed Wetland 2D (CW2D) is a biokinetic model describing microbial dynamics and transformation and degradation processes in subsurface flow constructed wetlands (CWs). The implementation of CW2D in HYDRUS (©PC Progress s.r.o.) was verified for application on CWs treating combined sewer overflow (CSO CWs). CSO CWs mitigate pollutant and hydraulic shock on receiving waters. Their loadings are stochastic in terms of periodicity, volume and quality. Their storage basin and outflow limitation causes cycles of saturated (intra-event) and unloaded (inter-event) states. The need for verification is due to this stochasticity. Key parameters to overcome the limitations identified by earlier studies were (1) biokinetic parameters, (2) fractionation of COD between readily and slowly biodegradable and inert forms and (3) adsorption of inert COD. With the new settings inoculation runs yielded stable biomass and domain conditions. These were successfully used as initial conditions for calibration and validation. Laboratory column experiments formed the basis of comparison, including single loads and a load series. The goodness of fit was quantified by an updated method. Good fit was reached to COD and NH4-N. Fitting to NO3-N was not a target; still, dynamics are discussed.