Subsurface flow constructed wetlands performance evaluation, modeling, and statistical analysis.

摘要

Constructed wetlands have been demonstrated successfully in treating wastewater for removal of organic matter, nitrogen, and phosphorus through various physical, chemical and biological mechanisms. However, wetland performance is difficult to predict due to the complex reaction mechanisms between different wetland components (e.g. vegetation, substrate, and hydrology) involved in the treatment. Currently, design of a subsurface flow constructed wetland follows a first-order, plug flow empirical model which is reported to be over-simplified. A subsurface flow constructed wetland system in southeastern Nebraska serving a small community was monitored during its first five years of operation. Its performance was analyzed and compared with the prediction of the empirical model. A sequential nitrogen transformation model was developed to identify the major nitrogen removal mechanisms and a nitrogen mass balance was performed, based on the kinetic rates calculated from the sequential model. In addition, a factor analysis was applied to determine the major controlling factors of the wetland performance and a multiple regression model was used to predict the wetland performance.; The empirical design model, adopting an exponential reduction relationship, in general over estimated the removal efficiency. The simplicity of the empirical model does not allow the model to accommodate the complex multiple reactions, for example, sequential nitrogen reactions (organic nitrogen → ammonium, ammonium → nitrate, nitrate → nitrogen gas). The sequential nitrogen transformation model successfully calibrated the different rates of nitrogen reactions and model predictions overall reflected the average performance. The nitrogen mass balance identified the sources and the sinks for nitrogen in the wetland. Nitrification-denitrification and nitrogen uptake by the vegetation were found to be major nitrogen removal mechanisms. Finally, the statistical analysis identified that the climate effect was the most important factor for the performance. The multiple regression prediction models reproduce the effluent TCBOD₅, TKN and TP concentrations satisfactorily. The sand filter which follow the wetland, found to increase the water oxygen level and improve the COD removal as well as nitrification, can serve as a polishing facility to the wetland.