Impermeable urban surfaces such as roads, carparks and roofs contribute pollutants via untreated runoff to urban waterways, causing adverse effects on the aquatic ecosystem. Total suspended sediment, copper and zinc, in particular, have been identified as key pollutants of concern in urban runoff. If urban waterways are to be sufficiently protected from runoff pollutants throughout effective stormwater management, then an easy-to-implement means of identifying specific pollutant sources is needed to enable targeted stormwater management planning and decision-making.

An event-based pollutant model, Modelled Estimates of Discharges for Urban Stormwater Assessments (MEDUSA), was developed to address these needs. It uses rainfall characteristics (antecedent dry days, intensity, duration, rainfall pH) and surface characteristics (material type, usage, condition, age) to predict sediment and heavy metal loads generated from individual impermeable urban surfaces.

The model was calibrated and validated on subcatchments up to 250 ha (620 acres). A key application has been to an industrial catchment that contributes stormwater to a sensitive coastal lagoon ecosystem in Timaru, New Zealand, enabling spatial mapping of highest load generation. Providing load predictions at the individual surface scale allows engagement with property owners to improve their onsite stormwater management. Furthermore, MEDUSA can assess the pollutant reduction from various management scenarios, including the effect of new roof material policies, maintenance, and at-source compared to end-of-catchment treatment. The model is now being rolled out across Christchurch city, New Zealand (>260 km2 (100 square miles)). City wide applications benefit a wide range of stakeholders, and ultimately contribute to healthier urban waterway ecosystems. Although the current focus has been on TSS, Zn, and Cu loadings, the model will be extended to incorporate nutrients.