Can stormwater harvesting restore pre-development flows in urban catchments in South East Queensland?

Increases in the impervious area due to urbanisation have been shown to have negative impacts on the physical and ecological condition of streams, primarily through increased volume and frequency of runoff. The harvesting and detention of runoff has a potential to decrease this impact. This paper describes the effects of urbanisation on catchment flow and of stormwater harvesting on reducing those adverse impacts on a stream in South East Queensland (SEQ), Australia. A largely undeveloped catchment located southeast of Brisbane city was calibrated and validated using the Stormwater Management Model (SWMM). This model was used to investigate the effect of a range of future increases in urbanisation (represented by impervious area) on stream hydrology as well as the potential of stormwater harvesting to return the catchments to predevelopment flow conditions. Stormwater harvesting was modelled according to flow frequency measures specified in current SEQ development guidelines. These guidelines stipulate the capture of the first 10 mm of runoff from impervious areas of 0-40% and the first 15 mm from impervious areas of 40% or greater for urban developments. We found that increases in the impervious area resulted in increases in the mean, frequency and duration of high flows, and an increase in the mean rate of rise and fall for storm events in the catchment. However, the predevelopment (non-urbanised) flow distribution was very flashy in comparison with all urbanised scenarios; i.e. it had the quickest response to rainfall indicated by a high rate of rise to and fall from peak flow volume, followed by a return to zero flow conditions. Capturing the runoff according to the development guidelines resulted in a reduction in flow towards the flow distribution of a lower impervious area, however this was insufficient to meet predevelopment conditions. This suggests a stronger influence of impervious areas in this catchment on the volume of runoff than flow frequency measures are able to ameliorate.