Temperature influence on nitrogen removal in a hybrid constructed wetland system in Northern Italy

•We studied the efficiency of a full-scale hybrid constructed wetland system (HCW).•We investigated the seasonal influence in the removal performance of the system.•We observed the presence of a temperature breakpoint that caused variation in nutrient removal.•We observed higher TN and NO3-N percentage abatement at temperatures above 14.2 °C.

The objective of this research was to investigate the efficiency and seasonal performance of a full-scale hybrid constructed wetland system (HCW) in reducing total nitrogen (TN), ammonia nitrogen (NH4-N) and nitrate nitrogen (NO3-N). HCW with a total area of about 130 m2 and hydraulic load of 2 m3/day was composed of three subsurface flow vertical systems (VF), working in parallel and one horizontal (HF) connected in series. The system was loaded daily with synthetic wastewater having an average concentration of TN of 250 mg/L (about 125 mg/L of NH4-N and 125 mg/L of NO3-N). Water samples were collected and analyzed from May to July 2011 and from January 2012 to July 2012. Variations were observed in nutrient removal performance related to temperature.During the whole monitoring period median reduction efficiency (RE) in the HCW was TN 95%, NH4-N 95% and NO3-N 93%, although three sub-periods characterized by different performances have been observed. During the first period (from May to July 2011) the RE was positive for the three nitrogen forms considered, whereas from January to the end of March 2012 the RE was lower, particularly for TN and NO3-N. From April 2012, when the temperature rose above 14.8 °C, there was an increase in the performance that reached the 2011 values.Internal production of NO3-N was observed, mainly in the VF systems between January and March 2012. The median removals of mass pollutants per m2 of HCW per day were TN 3.1 g/m2/d, NH4-N 1.5 g/m2/d, NO3-N 1.5 g/m2/d. Segmented regression analysis identified a breakpoint at 14.2 °C for wastewater temperature that caused variations in TN and NO3-N concentration reduction performances. According to this approach the abatement was always positively correlated with temperature, but different regression slopes were obtained below and above the breakpoint. In particular, with lower temperature the abatement of NO3-N and TN increased by 1.7 and 2.0% per °C of temperature increase; with temperature higher than 14.2 °C the increase in abatement due to increased temperature was sharper, especially for NO3-N.