Modelling biofilm nitrogen transformations in constructed wetland mesocosms with fluctuating water levels

A mathematical model has been developed that attempts to reproduce patterns of nitrogen removal observed in experiments investigating constructed wetl and treatment of ammonium-rich wastewaters under a range of frequencies of water level fluctuation. The experiments were carried out using batch-fed g ravel-filled mesocosms, with and without plants, subjected to fluctuating o xygen input through a drain-and-refill regime. Experimental data showed tha t removal of ammoniacal-nitrogen (NH4-N) and chemical oxygen demand (COD) i ncreased markedly with fluctuation frequency. Plants also tended to enhance the removal of NH4-N and COD. For the highest fluctuation frequency (16 cy cles per day, plants absent), accumulation of oxidised nitrogen (NOx-N) was observed to continue even when the wastewater NH4-N had disappeared from s olution. A process-based numerical model was developed to elucidate the str ength of competing nitrogen transformation processes, which were postulated to be strongly influenced by biofilms and adsorption/desorption associated with gravel surfaces and organic matter, particularly when the mesocosm wa s empty and liquid on the biofilms was exposed to the atmosphere. A combina tion of thin-biofilm theory, the microbiological kinetics in the IAWQ activ ated sludge model No. 2, reversible sorption kinetics and mixing equations was used to demonstrate that very rapid initial decreases in NH4-N were lik ely to be caused by adsorption onto the gravel and that during the latter p art of the batch periods nitrification was likely to be controlled by the r ate of desorption of this NH4-N. Nitrification could therefore continue whe n NH4-N was almost absent from the bulk water. At moderate-to-high fluctuat ion frequencies (16 cycles per day) the presence of plants enhanced NH4-N r emoval and NOx-N accumulation, through a combination of direct uptake of NH 4-N and increased root-zone reaeration. (C) 2000 Elsevier Science B.V. All rights reserved.