Urbanisation and stormwater management in South East Queensland – Synthesis and recommendations

The ecological health of waterways is generally known to be impacted by the hydrologic and water quality changes which occur as a consequence of urbanisation. The aims of the research reported here were: to develop detailed characterisation of the hydrological, water quality and ecological impacts of urbanisation in SEQ across a range of catchments; to tease apart the likely causes of ecological impacts; and, having done so, to make a set of recommendations about how urbanisation might be managed differently to help avoid waterway ecological degradation. SEQ has a sub-tropical climate and the existing literature on the impacts of urbanisation has been developed mostly focussed on temperate climate conditions. This report provides a synthesis of the range of research results generated by the project and a set of management and research recommendations developed in critical response to the results. Twelve catchments in the Brisbane and Gold Coast areas of SEQ were gauged hydrologically for three years to yield a sufficient quantity and quality of flow and rainfall data to develop reliable catchment models using the Stormwater Management Model (SWMM) platform. In addition, the total impervious area (TIA) for those catchments was determined from aerial photographs. SWMM models were successfully developed for eight of those catchments using a generic algorithm automatic parameterisation approach. At the same time as flow data was gathered, water quality data on pH, temperature, conductivity and turbidity was gathered for each catchment using Sonde instrumentation to allow the impacts of water quality change on ecological health to be assessed. These models were used to assess how hydrology changes with urbanisation intensity and pattern. First of all, a set of baseline simulations were run using long time-series hourly rainfall data for the catchments to investigate how increasing TIA impacts on catchment hydrology. Next, three predevelopment (no urbanisation) catchment models were used to simulate, using long time-series hourly rainfall, the impacts of increasing levels of urbanisation as characterised by per cent TIA (%TIA). From the modelling results, urbanisation is clearly associated with changes in hydrology, but the changes are complex. Whilst there are some generalities (increases in high flow condition duration, increases in mean flow and 90th percentile flow, increases in the frequency and rate of runoff event rise), the hydrological impact of urbanisation depends on catchment characteristics, including size, slope, time of concentration (ToC), sub-catchment sizes and distributions, and on the pattern of urbanisation itself across sub-catchments and the catchment as a whole. The same urbanisation pattern can exert a qualitatively different impact hydrologically, depending on the composition of the whole catchment in terms of sub-catchments. Maximum hourly flows appear not to be impacted by urbanisation, but 90th percentile hourly flows and mean hourly flows are impacted, both increasing with urbanisation. The number of runoff events increases with urbanisation and the size of the rise and fall in flow with each event also increases with urbanisation. The proportion of time spent under high flow conditions tends to increase with urbanisation for any given catchment, but not necessarily so – there can be some catchment specific decreases in high flow spell duration under urbanisation, depending on sub-catchment characteristics. The mean of high flow spells may increase, but not necessarily so. The proportion of time spent under low flow conditions tends to decrease with urbanisation, probably as a consequence of the streams studied being ephemeral in their pre-development state rather than strongly base flow supported and perennial. To understand how urbanisation affects the ecology of urban streams and waterways, a conceptual model was developed to articulate the range of potential mechanisms, and these mechanisms were then investigated through a mixture of means, by way of: statistically analysing the relationships between urban land use (particularly imperviousness) and Ecosystem Health Monitoring Program (EHMP) score and indicators; characterising macroinvertebrate assemblages present in three selected case study sites (one reference and two urban) and how these assemblages vary between summer and winter seasons or high and low flow conditions; and relating the assemblage data to hydrological and water quality variables in the sites concerned. The research reported here clearly indicates that there are negative aquatic ecological impacts associated with urbanisation in SEQ. In particular, the EHMP analysis demonstrates that urbanisation (as a lumped land use category) is associated with decreases in macroinvertebrate richness, and increase in the proportion of alien fish species observed. TIA, either lumped or weighted to mimic the effect of directly connected impervious area (DCIA), was not observed to exert a strong impact on any ecological variables. The ecological results tend to indicate that the hydrological changes following urbanisation are not significant degrading factors in themselves, rather, the water quality variables, particularly temperature range, are more likely to be important. The association of lumped urban land use with ecological impact and the simultaneous lack of ecological impact associated with IA (TIA or proxied DCIA) raise the question as to whether the process of urbanisation, i.e. the process of construction, is the primary source of ecologically degrading waterway impact in SEQ, rather than the on-going impact of impervious area runoff flows. Whilst urban and pre-development streams had similar levels of macroinvertebrate species richness and diversity, and similar distributions of habitat availability (riffle and pool proportions), there were significant differences over time (seasonally) within each stream type and between each stream type in relation to species composition. Pool species composition in both urban and pre-development streams was found to be stable over time, i.e. not affected by higher summer or lower winter flows. Conversely, riffle species composition in the urban stream was found to vary significantly over time, with lower diversity in the lower flow winter months, suggesting the importance of water quality changes rather than flow changes as a driver of assemblage change. As with hydrological impact, the mechanisms of ecological change from urbanisation are complicated and based partly on catchment specific features, e.g., the winter flow supporting upstream wetlands in Stable Swamp Creek and the ecologically locally devastating iron floc problems at Blunder Creek. Finally, the evaluation of the Qld frequent flow management objectives (FFMOs) as an ecologically oriented flow management policy instrument designed to avoid the ecological impacts associated with urbanisation, suggests that they will bring catchment hydrographs back towards their pre-development profile, but are insufficiently strong. The FFMOs will have an effect which is partly dependent on catchment characteristics, the distribution and sizes of sub-catchments and the spatial pattern of urbanisation.