Nitrogen removal from domestic effluent using subsurface flow constructed wetlands: influence of depth, hydraulic residence time and pre-nitrification

This paper describes two studies into the BOD and TN removal performance of horizontal subsurface flow wetlands (reed beds) in subtropical Australia. The aim of the first study was to determine the influence of HRT and vertical position on BOD and TN concentration and removal performance in a 0.5m deep reed bed (System 1) by taking samples from three levels (or layers) in the water column at five points along the length of the bed. The aim of the second study was to investigate the TN removal performance of a treatment train consisting of a vertical flow intermittently dosed sand filter preceding a reed bed (System 2). Both systems were dosed with primary settled municipal wastewater (BOD 194 mg L-1; TN 49 mg L-1). System 1 achieved a TN load removal of 58% under a HLR of 22 mm day-1 (HRT 10.5 days), producing effluent BOD concentrations consistently less than 8 mg L-1. There was no significant difference in BOD attenuation rate between the three layers. While there were differences in both the nitrification and denitrification rates between the three layers, the TN concentration was found to decline steadily in all layers up to an HRT of 8.7 days. System 2 reduced TN influent load by 33%, less than half of which was removed by the reed bed. The lack of substantial TN removal within this reed bed was attributed to the low concentrations of BOD and consequent lack of dissolved organic carbon to drive the denitrification process.