What Is the Suitable Sampling Frequency for Water Quality Monitoring in Full-Scale Constructed Wetlands Treating Tail Water?

Abstract

Three years of hourly COD and NH4+-N measurements for two full-scale integrated constructed wetlands (CWs) treating secondary effluents from sewage treatment plants (STPs) were used to quantify the proper sampling frequency (SF). The modified coefficient of variation (CVm) and average variation rate (VRa) were calculated to monitor the dynamics and annual average performance, respectively. It was found that (1) under CVm 5%, VRa 5%, and VRm 5%, the sampling intervals (SI) of COD can be set as 1.19 h, 526.5 h, and 110.1 h, respectively, and the SI of NH4+-N should be 4.51 h, 66.3 h, and 26.8 h, respectively; (2) under CVm 10%, VRa 10%, and VRm 10%, the monitoring intervals of COD can be set as 11.92 h, 1401.7 h, and 233.5 h, respectively, and the monitoring intervals of NH4+-N should be 30.73 h, 139.3 h, and 50.5 h, respectively. Therefore, to meet the need of monitoring the dynamic changes in data, hourly and 4 h SIs were recommended for COD and NH4+-N evaluation, respectively, when it is necessary to consider the operation and maintenance costs at the same time, 11 h and 30 h SIs were proper for COD and NH4+-N evaluation, respectively. The methods proposed in this study could provide reference to improve the management and evaluation level of full-scale CWs.