Development and testing of a model for Micro-Organism Prediction in Urban Stormwater (MOPUS)

Accurate estimation of the levels of microorganisms in urban stormwater is needed for stormwater harvesting and to ensure that our streams and bays are safe for recreational uses. The aim of this research was to develop and test a simple urban stormwater microorganism model (Micro-Organism Prediction in Urban Stormwater - MOPUS) which is spatially lumped and coupled to a rainfall runoff model. The microorganism model has surface and subsurface components to simulate build-up and wash-off of microorganisms from the impervious surfaces of the catchment and the stormwater pipes, respectively. The rainfall-runoff model simulates processes from both pervious and impervious surfaces. Both models are conceptual and represent important processes in a simplified manner, thereby limiting the number of calibration coefficients (five for each model) while maintaining accuracy. The coupled model has been tested using a large Escherichia coli (E. coli) dataset collected from four urban catchments in Melbourne, Australia. For each catchment, around 20 well sampled pollutographs were available. Reasonably good predictions were obtained at each site for both instantaneous flow rates (Nash Sutcliffe E between 0.62 and 0.89) and E. coli concentrations (E = 0.25-0.45). Event E. coli peaks (E = 0.42-0.75), E. coli loads (E = 0.48-0.86) and event mean E. coli concentrations (E = 0.56-0.76) were also well estimated. In general, it has been demonstrated that, with further development and testing, MOPUS appears capable of reliable predictions of E. coli discharges from urban stormwater systems, allowing its use as a planning tool for urban catchments. (C) 2011 Elsevier B.V. All rights reserved.