Carparks make up a large portion of impervious urban areas. Stormwater runoff from carparks can, therefore, be a significant source of pollutants to receiving urban waterways, particularly of total suspended solids (TSS), heavy metals (dissolved and particulate) and indicator bacteria (total coliform and E. coli). Traffic volume and patterns, vehicle type, and level of vehicle maintenance have been identified as primary factors that contribute to carpark pollutants in urban stormwater runoff. Materials deposited on carpark surfaces from atmospheric deposition can also be an indirect source of pollutants in carparks. The wash-off of these pollutants is dependent on various rainfall characteristics such as rainfall intensity, rainfall depth, antecedent dry days and rainfall duration. However, there is still a lack of quantitative information on how climatic conditions (particularly low-intensity rainfall climate) and carpark land use influence pollutant loads and the corresponding stormwater quality characteristics during first flush and steady-state conditions. There is also limited information on the characterization of pollutants from different carparks based on particle size distribution (PSD). Therefore, the aim of this research is to understand the dynamic nature of pollutant loadings and PSDs from urban carparks under low-intensity rainfall conditions. The long-term goal is to contribute to the development of guidelines for the selection of land use based stormwater treatment systems for individual carparks.

Carpark runoff samples were collected during 21 storm events (rainfall intensity varied from 0.39-6.87 mm/h with an average of 2.3 mm/h) from three urban carparks under different land uses (university, hospital and industrial) in Christchurch, New Zealand. Samples were collected over 14 months using grab and automatic samplers. Samples from first flush (FF) and steady-state (SS) were analyzed for TSS, heavy metals (dissolved and particulate), particle size and E. coli.

During first flush, the industrial carpark runoff quality had significantly higher pollutant concentrations. The positive correlations between each heavy metal (Zn, Cu, Pb) at the university and the industrial carparks indicated that in each of these sites heavy metals originated from a similar source. The higher TSS and ratios of different heavy metals at the industrial carpark suggested relatively higher wear and tear from larger commercial vehicles. Rainfall characteristics had a weak positive correlation with pollutant concentrations at all the carparks, but vehicular activities are likely to be the dominant factor in the observed range of contaminant yield values.

During steady-state flow conditions, TSS and heavy metals yield at the industrial carpark were statistically higher than other two carparks studied during period 1 (first 40 mins of rainfall). There was a subsequent reduction in mean pollutant yields from each period except for the university carpark. This shows that the increased rain duration resulted in a decreased supply of pollutants in each period from each carpark. A linear relationship between total and dissolved metal yields was found at the university and the hospital carpark. The data for all the carparks monitored did not show any seasonal effects on total pollutants yields. Larger rain events (>5 mm) had relatively greater pollutants yields than smaller rain events and average pollutant yields were found to be a relative maximum at 3-6 antecedent dry days.

Particle size during FF and SS showed a similar distribution pattern but had a slightly different median (D50) with an average of 80 and 110 µm respectively. Fine particles (<67 µm) represented 32-40% of the total particles from the carparks.

Indicator bacteria concentrations were statistically higher at the industrial carpark and also exceeded the recommended recreational guidelines established by the NZ Ministry for Environment.

The results of this study indicate that given the range of pollutant types and quantities coming off different carparks, the selection of treatment systems has to be targeted to the individual carpark land use and climatic characteristics. Furthermore, a combination of treatment systems may be needed to achieve optimal removal of pollutants of the types and quantities observed. Various nonstructural management strategies should also be considered to reduce the build-up of pollutants in carparks. These include frequent cleaning and maintaining downpipes and gutters, proper disposal of pet waste and litter, use of vegetation and grass to cover and stabilize exposed soil to prevent sediment wash-off, sweeping and vacuuming of carpark surfaces and so on.

Overall, the knowledge gained through this research contributes to a greater understanding of urban carpark runoff quality and appropriate selection of treatment systems and management of carparks, which will thus result in improvement of the water quality of urban waterways.