Addington Brook is a stormwater-affected urban stream in western Christchurch that is a priority catchment for stormwater management improvements, due to adverse impacts seen on the aquatic ecosystem of the Brook and downstream receiving environment. In order to better understand where and what contaminants are being generated in stormwater runoff from the catchment’s impermeable (roof, road and carpark) surfaces, an event-based contaminant load model, Modelled Estimates of Discharges for Urban Stormwater Assessment (MEDUSA), was applied to the Addington catchment to estimate the amount of total suspended solids (TSS), total zinc (Zn) and total copper (Cu) being generated by these surfaces (Charters 2016). The modelling identified large carparks and highly trafficked roads as primary contributors of TSS, galvanised roofs as primary contributors of Zn and highly trafficked roads and industrial carparks as primary contributors of Cu. In this current study, stormwater management scenarios that target primary contributors for each contaminant were simulated with MEDUSA, as follows: • S1 – Replacement of all zinc-based roofs with new Coloursteel® • S2 – Replacement of all zinc-based roofs with inert material • S3 – Replacement of only old galvanised roof with new Coloursteel® • S4 – Treatment of old galvanised roof runoff with on-site treatment • S5 – Treatment of a representative large commercial site with large-scale on-site treatment • S6 – Treatment of road runoff from highly trafficked roads • S7 – Treatment of the highest load generating carparks • S8 – Effect of removing copper generation from road via implementation of Cu-free brake pads The predicted annual loads for these scenarios have been compared against the ‘baseline’ annual load for current catchment conditions. For TSS management, the most significant reduction is achieved by targeting minor and major arterial roads and treating their runoff in a system such as a raingarden (12% TSS reduction in catchment). For zinc, targeting zinc-based roofs (i.e. Scenarios 1 and 2) is predicted to achieve the greatest reduction from baseline conditions (Zn reductions of 52-70%). Targeting only old galvanised roofs (Scenario 3; i.e. targeting a limited surface area) is predicted to achieve 15% Zn reduction. For TCu management, the removal of copper contributions onto roads and carparks via implementation of Cu-free brake pads is predicted to have a significant reduction on catchment copper loads (78%). There are several recommendations for future research to address limitations observed in this current study: Alternative treatment solutions are required that can provide greater zinc removal efficiencies than current treatment systems. The effectiveness of scenarios that are based on treatment (as opposed to source reduction) is limited by the level of zinc reduction that can be achieved in such systems. Roof runoff in particular has a high proportion of dissolved metals, which are poorly removed in typical sediment-removal focused treatment systems. There is a knowledge gap to be addressed in quantifying the proportional contributions from non-brake pad sources to better predict the effects of the Cu-free brake pad scenario. As the uptake of Cu-free brake pads will be phased in over a period of time, there is opportunity to monitor the change this brings to untreated runoff quality. This can then be compared to the data being collected in other international programmes. Street sweeping remains a valuable management method for reducing sediment and heavy metal pollution in road runoff, however a lack of good data on the effectiveness of street sweeping meant such a scenario was not modelled in this study. Further research is needed to quantify the amount of pollutants removed per frequency of sweeping, how widely this varies and what influences Christchurch’s local climate has on the removal rates, in order to be able to make effective decisions on sweeping frequency. Percent reductions in at-source pollutant loads predicted by this modelling may not result in an immediate equivalent reduction being observed in the receiving waterway. Resuspension of settled sediment within the stormwater network and within the Brook may result in higher TSS and metal loads instream. Pollutant transformations within the stormwater network and in stream are not well understood and further research is needed in both settings to better link the at-source runoff quality with the receiving waterway quality (and resultant impacts on the aquatic ecosystem).