Selecting appropriate sample locations in cell-network treatment wetlands Huaguo Wang and James W. Jawitz (Soil and Water Science Department, University of Florida)

Pollutant removal effectiveness of a treatment wetland is generally estimated from the relative difference between inlet and outlet concentrations. However, this estimation becomes complicated when the wetland is composed of multiple hydraulically connected cells. In such cell-network structures, the wetland or portion of the wetland that is of interest may have multiple inlets or outlets. Varying water parcel residence times may exist along different water paths, resulting in different levels of pollutant treatment effectiveness. The lumped residence time distribution (RTD) of the wetland will not likely represent the RTD of water parcels traveling between each inlet and outlet. Thus, if water samples are not collected at all inlets or outlets, , the estimation of pollutant removal effectiveness for the wetland may be inaccurate. Here, a method is introduced to specify such potential estimation errors by incorporating hydraulic performance models such as tanks-in-series into the k-C* pollutant removal model. The method was applied to data from the Orlando Easterly Wetland. Errors resulting from applying RTDs measured at one sample location to another location were on the order of 10-50%.