Feature

The Muck Stops Here

The rain comes down and the water has to go somewhere -- but what are the effects of stormwater runoff on urban estuaries?

Don Morrisey

Stormwater runoff -- rainwater running off roofs, roads and other impermeable surfaces -- from urban areas has become an increasing headache for environmental managers and planners in recent years. In the past, its environmental impact was overshadowed by that of industrial discharges. As these have been increasingly brought under control, environmental effects of diffuse sources of contamination, such as stormwater, have come to the fore. Many of the effects of stormwater are cumulative, and thereby potentially compromise the sustainability of urban development.

During the process of development, there is an evolution of the effects of stormwater runoff on receiving environments such as the streams, rivers, estuaries and other coastal areas into which they eventually drain. An increased amount of sediment in the runoff is one of the first effects to emerge. During construction, soils are exposed to erosion by rain. The resultant loads of sediment often exceed the removal capacity of natural drainage systems such as streams and rivers. Sediments accumulate in the streams, changing their morphology and character, and affecting the animals and plants that live in them.

Urbanisation leads to an increase in impermeable surface area, as bush and pasture change to roads and roofs, causing high stream-flows to become more frequent. This increased flow leads to periodic resuspension of the sediment deposited in the streams and its gradual removal downstream over a period of several years. It also produces widening of stream channels to accommodate the increased volume of water, a process that may take two or three decades. Drainage systems in urban areas are often straightened and lined with erosion-resistant materials to mitigate the effects of increased volumes of runoff; this also has a dramatic effect on the ecology of the streams.

In addition to increased sediment loads, urbanisation causes other changes in water quality. As the water passes over roads and other surfaces, it picks up dust, litter and chemical contaminants, such as heavy metals and various organic compounds. Many of these contaminants are toxic to aquatic organisms and represent another mechanism by which runoff from urban areas may have an impact on receiving environments.

Among the heavy metals, the most significant are copper (derived from the brake-linings of vehicles) and zinc (from tyre wear). Despite the recent end to its use in petrol, it is likely that lead present in the environment will also continue to contaminate stormwater for some time. Organic contaminants include polycyclic aromatic hydrocarbons (PAH), and pesticides and herbicides used in parks and gardens. Many of these contaminants are bound to the surface of particles of sediment, rather than being in solution, and this has important implications for their fate and for their ecological effects.

Stormwater may also contain microbial contaminants (bacteria and viruses), derived from wild and domestic animals and from leaking septic tanks or malfunctions of sewage-treatment systems. Although there is no clear evidence of adverse effects on human health from microbial contamination by stormwater (either through direct contact or consumption of shellfish grown in receiving-waters), the potential for such effects exists and represents another cause for concern.

As urbanisation proceeds, sediments and associated contaminants are deposited by stormwater in streams and rivers. From there they make their way slowly down to the sea. Because most of New Zealand's larger towns are situated on the edges of harbours or estuaries, these represent the most important type of receiving environment.

Estuaries are generally depositional environments -- they contain areas where water currents are sufficiently weak that fine sediments are able to settle from suspension and accumulate to form mudflats and sandflats. When the catchments of estuaries are urbanised, sediments and associated contaminants derived from stormwater begin to accumulate in sheltered parts of the estuaries. Many of the contaminants in stormwater do not break down in the environment. As urbanised areas mature, the amount of sediment present in runoff decreases but the concentrations of contaminants increase, so the concentrations of contaminants in the receiving environment's sediments gradually increase.

Contaminant Model

Staff at the National Institute of Water and Atmospheric Research Ltd (NIWA), in association with the Auckland Regional Council, have recently developed a model to predict the rate of accumulation of stormwater contaminants in the sheltered estuaries of the Upper Waitemata Harbour near Auckland. By predicting patterns and rates of increase in the concentrations of contaminants in estuaries with urbanised catchments, the model will help environmental managers plan such developments and prioritise treatment measures.

The model uses published information on the volumes of runoff from land under different stages of urbanisation, and the concentrations of sediment and contaminants in that runoff. The resulting estimates of the cumulative inputs of these materials are combined with estimates of the volume of sediments already present in the estuary, and the concentrations of contaminants in them, to generate predictions of the concentrations of contaminants in the sediments of the estuary at increasing times after the start of catchment urbanisation. The model also assumes that 75% of the sediment entering the upper estuary will be deposited in a "settling zone" near the freshwater input and equivalent in area to about 4% of the area of the catchment. This assumption is derived from studies of the effectiveness of retention-ponds in removing suspended material from stormwater. The model produces a graph of the rate of increase of the concentrations of contaminants through time and can be modified to predict the consequences of various treatment options.

The model has been tested by making predictions of the concentrations of copper, lead and zinc in estuaries near Auckland whose catchments have already been urbanised. These estuaries were then sampled to determine the actual concentrations. The results of this study showed that the model works well, at least in the types of small, sheltered estuaries for which it was developed.

Treatment Options

Various options are available for treating stormwater before it reaches estuaries. These include simple, relatively cheap measures such as constructing grassed or other vegetated channels (swales). These trap sediments and contaminants, reduce the rate of water movement and may enhance the infiltration of the water into the ground. This contrasts with the traditional practice of getting the water to the receiving environment as quickly as possible via straightened, concrete channels. More elaborate systems involve large treatment ponds, designed to detain the water so that suspended sediments and associated contaminants can settle out and later be removed from the pond.

None of these options is completely effective in removing contaminants and, at best, they can only reduce the rate of increase in the concentrations of contaminants in receiving environments. Given the high cost of installation of treatment-devices, a "best practical option" approach to managing stormwater has generally been adopted.

In situations where potential urban development threatens a highly valued, pristine environment, the only certain way to avoid environmental impacts from stormwater is to prevent development. This approach has been adopted by the Auckland Regional Council in the form of the "metropolitan urban limits", put forward in the Council's Regional Policy Statement. These limits represent planned, geographical restrictions on the extent of Auckland's urban sprawl. Implicit in this policy is the idea that future increases in the city's population will be contained largely within the existing urban area. Receiving environments for stormwater in this area have already been impacted and the policy thus seeks to avoid despoiling more and more areas as the population expands.

Concerns about the effects of stormwater on estuarine receiving environments relate not so much to the build-up of contaminants themselves, but to the possible ecological consequences of this build-up. As our studies of Auckland estuaries have shown, the predicted build-up does occur. There is also abundant evidence to show that, when present in high enough concentrations, many of these contaminants are toxic to marine and estuarine animals and plants, at least in laboratory tests. Stormwater produces relatively small concentrations of contaminants compared with, for example, industrial sources. Consequently, dramatic toxic effects, such as widespread mortality, are not generally seen. Work done by NIWA has, however, demonstrated other effects that, although more subtle, are potentially of similar ecological importance.

Ecological Effects

In the laboratory, these effects have included the reduced ability of juvenile bivalves (such as the wedge shell, Macomona liliana) to burrow into contaminated sediments. Juvenile wedge shells also preferred to settle onto uncontaminated rather than contaminated sediments where possible. In these experiments, the sediments tested were collected along a gradient of contamination in an estuary where stormwater was the principal source of contaminants. Although less dramatic than the mortality seen at larger concentrations, such effects may be just as important in changing the composition of communities of animals in the field.

Determining the broader, ecological significance of stormwater, however, requires that effects on individual animals, observable either in the laboratory or in the field, also produce ecologically important effects in the field. There are two complimentary approaches to identifying such effects. The first is to look for correlations between patterns of distribution of contaminants and those of organisms. Such evidence might take the form, for example, of reduced diversity of animals in sediments containing larger concentrations of stormwater-derived contaminants. The second approach is to conduct manipulative experiments in the field, comparing, say, changes in the biological communities in areas to which contaminants have been added to those in uncontaminated areas. These approaches are complimentary in that correlative studies can be conducted over large spatial scales but do not provide direct evidence of cause and effect between the nature of the biological communities and the concentrations of contaminants. Experimental studies, on the other hand, are more restricted in scale for logistical reasons but have the advantage of providing direct evidence of cause and effect.

Demonstration of ecological impacts in the field is generally difficult because populations and communities of organisms often show considerable spatial and temporal variability. This tends to mask impacts of particular human activities, especially those whose effects are likely to be subtle, such as stormwater contamination. There is some evidence of decreased diversity of animal communities in parts of the Manukau Harbour (near Auckland) where concentrations of contaminants are relatively large. In this particular case, however, as in many others where there are multiple sources of contaminants, it is very difficult to separate effects of stormwater-contamination from other sources.

The case is more clear-cut in small, urban estuaries where stormwater is the only significant source of contamination. Even in these environments, however, the presence of natural gradients in factors such as salinity and type of sediment may impose so strong a signal on the composition of communities of organisms that effects of contamination are obscured.

In a continuing study of the ecological effects of stormwater on urban estuaries, NIWA is taking a "weight of evidence" approach. The first line of approach incorporates correlative studies of the distributions of animals and contaminants in the sediments of several urban and rural estuaries near Auckland. This has provide evidence of ecological effects in terms of differences in communities of animals between estuaries with urban and rural catchments, and correlations of patterns of distribution of faunal communities with those of concentrations of contaminants.

The second line of approach is the development of standard laboratory toxicity tests of sediments contaminated by stormwater. These tests use native, marine or estuarine species and sublethal endpoints. The methodology of these tests, and of equivalent toxicity tests carried out in the field, is presently being refined.

The third line of approach, and the most robust in terms of demonstrating cause and effect, is the use of experimental field-studies. A large-scale, experimental study in the Manukau Harbour has provided evidence for such effects. The experiment involved contamination of an area of sandflat with the pesticide chlordane at concentrations similar to those found in other parts of the harbour. Application of the pesticide had no apparent effect on adult bivalves but, in keeping with the laboratory results described above, was followed by the emigration of juvenile bivalves from the site. Future work will investigate rates and patterns of larval recruitment into contaminated and uncontaminated sediments.

The long-term outcome of this programme of research will be the ability to predict environmental impacts of proposed urban development in terms of rates of accumulation of contaminants in receiving environments, and the likely ecological consequences of this. This information will aid environmental managers in developing strategies for urban expansion and in prioritising resources for mitigating environmental effects of existing and future discharges.

Don Morrisey works at the National Institute of Water and Atmospheric Research Ltd in Hamilton.