Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4391318 | Ecological Engineering | 2006 | 13 Pages |
To protect receiving waters from stormwater pollution, stormwater managers need to be able to predict the performance of proposed stormwater treatment measures, under variable operating conditions. This paper describes the development of a model, capable of predicting the performance of stormwater wetlands, ponds, vegetated swales, sediment basins and biofilters, with a single algorithm. The model describes two principal processes: (a) water quality behaviour and (b) hydrodynamic behaviour.Water quality is described by a first-order kinetic decay model (named the “k–C*” model, after its two parameters, the decay rate, k, and equilibrium concentration, C*). However, since pollutant removal depends on flow behaviour, the continuously stirred tank reactor (CSTR) concept is used to account for the hydrodynamics within a treatment device. Where the device has a high degree of turbulence or short-circuiting (such as in a sediment basin), the k–C* model is applied through a small number of CSTRs in series, whereas a well-designed wetland with even flow distribution is modelled by a high number of CSTRs. The unified model has been successfully tested on a series of treatment measures—a wetland, pond, swale, grass filter, gravel filter, and large lake. Necessary research to address limitations and assumptions of the model is described.