## General Design Procedures

All constructed wetland systems can be considered to be attached-growth biological reactors, and their performance can be estimated with first-order plug-flow kinetics for BOD and nitrogen removal. Design models are presented in this chapter for removal of BOD, TSS, ammonia nitrogen, nitrate, total nitrogen, and phosphorus, for both FWS and SSF wetlands. In some cases, an alternative model from other sources is also presented for comparison purposes because a universal consensus does not exist on the "best" design approach. The basic relationship for plug-flow reactors is given by Equation 6.8:

where

Ce = Effluent constituent concentration (mg/L). C0 = Influent constituent concentration (mg/L). Kt = Temperature-dependent, first-order reaction rate constant (d-1). t = Hydraulic residence time (d).

The hydraulic residence time in the wetland can be calculated with Equation 6.9:

where

L = Length of the wetland cell (ft; m). W = Width of the wetland cell (ft; m). y = Depth of water in the wetland cell (ft; m).

n = Porosity, or the space available for water to flow through the wetland. Vegetation and litter occupy some space in the FWS wetland, and the media, roots, and other solids do the same in the SSF case. Porosity is a percent (expressed as a decimal). Q = The average flow through the wetland (ft3/d; m3/d):

It is necessary to determine the average flow with Equation 6.10 to compensate for water losses or gains via seepage or precipitation as the wastewater flows through the wetland. A conservative design might assume no seepage and adopt reasonable estimates for evapotranspiration losses and rainfall gains from local records for each month of concern. This requires a preliminary assumption regarding the surface area of the wetland so the volume of water lost or added can be calculated. It is usually reasonable for a preliminary design estimate to assume that Qout equals Qin.

It is then possible to determine the surface area of the wetland by combining Equation 6.8 and Equation 6.9:

KTyn where As is the surface area of wetland (ft2; m2). The value used for KT in Equation 6.1 or Equation 6.4 depends on the pollutant that must be removed and on the temperature; these aspects are presented in later sections of this chapter.

Because the biological reactions involved in treatment are temperature dependent it is necessary, for a proper design, to estimate the water temperature in the wetland. The performance and basic feasibility of FWS wetlands in very cold climates are also influenced by ice formation on the system. In the extreme case, a relatively shallow wetland might freeze to the bottom and effective treatment would cease. This chapter contains calculation procedures for estimating water temperatures in the wetland and for estimating the thickness of ice that will form.

The hydraulic design of the wetland is just as important as the models that determine pollutant removal because those models are based on the critical plug-flow assumption, with uniform flow across the wetland cross-section and minimal short-circuiting. Many of the early designs of both SSF and FWS wetlands did not give sufficient consideration to the hydraulic requirements, and the result was often unexpected flow conditions including short-circuiting and adverse impacts on expected performance. These problems can be avoided using the hydraulic design procedures in this chapter.

A valid design requires consideration of hydraulics and the thermal aspects, as well as removal kinetics. The procedure is usually iterative in that it is necessary to assume a water depth and temperature to solve the kinetic equations. These will predict the wetland area required to remove the pollutant of concern. The pollutant requiring the largest area for removal is the limiting design parameter (LDP), and it controls the size of the wetland. When the wetland area is known, the thermal equations can be used to determine the theoretical water temperature in the wetland. If the original assumed water temperature and this calculated temperature do not agree, further iterations of the calculations are required until the two temperature values converge. The last step is to use the appropriate hydraulic calculations to determine the final aspect ratio (length-to-width) and flow velocity in the wetland. If these final values differ significantly from those assumed for the thermal calculations, further iterations may be necessary.

0 0