Showing papers by "Tim D. Fletcher published in 2013"

Adapting urban water systems to a changing climate: lessons from the millennium drought in southeast Australia.

Abstract: Feature pubs.acs.org/est Adapting Urban Water Systems to a Changing Climate: Lessons from the Millennium Drought in Southeast Australia Stanley B. Grant,* ,†,‡ Tim D. Fletcher, ⊥ David Feldman, § Jean-Daniel Saphores, †,§ Perran L. M. Cook, # Mike Stewardson, ‡ Kathleen Low, † Kristal Burry, ∇ and Andrew J. Hamilton ∥ Department of Civil and Environmental Engineering, E4130 Engineering Gateway, University of California, Irvine, Irvine, California 92697-2175, United States Department of Infrastructure Engineering, Melbourne School of Engineering, Engineering Block D, The University of Melbourne, Parkville 3010, Victoria, Australia Department of Planning, Policy, and Design, 300G Social Ecology I, University of California, Irvine, Irvine, California 92697-7075, United States Department of Agriculture and Food Systems, The University of Melbourne, 940 Dookie−Nalinga Road, Dookie College, Victoria 3647, Australia Melbourne School of Land and Environment, The University of Melbourne, Burnley Campus, 500 Yarra Boulevard, Richmond, Victoria 3121, Australia Water Studies Centre, School of Chemistry, Monash University, Victoria 3800, Australia Melbourne School of Land and Environment, The University of Melbourne, Parkville Campus, 207 Bouverie Street, Victoria 3052, Australia the way Melburnians source and use their water resources and discuss what these changes may portend for other large cities in water-scarce and climate-change-vulnerable regions of the world, in particular, the Southwest region of the United States. MELBOURNE’S WATER SUPPLY Melbourne sources most of its water from protected stream catchments located in uninhabited mountain ash (Eucalyptus regnans) forests to the north and northeast of the city (Figure 1). Runoff from these protected catchments flows by gravity into ten harvesting reservoirs and, from there, through a network of aqueducts and pipelines to storage reservoirs where it is distributed, after minimal treatment, to local service reservoirs. Since the first harvesting reservoir was built in the mid-1800s, Melbourne’s protected catchments have provided the city with a safe, low-energy, and mostly reliable source of high quality drinking water. However, they have also left the city vulnerable to water shortages during periods of very low precipitation. 5 To buffer against water shortages, Melbourne recently invested in various water supply augmentation schemes, including an interbasin transfer pipeline (the North−South or Sugarloaf Pipeline) and the largest desalination plant in the Southern Hemisphere (the Wonthaggi Desalination Plant) (Figure 1). These two projects were built at a capital cost of approximately AU$700 million 6 and AU$6 billion, 7 respec- tively, and can deliver annually up to 75 and 150 GL of water to Melbourne; combined, that equates to about 40% of the city’s present day municipal water demand. However, since their completion in 2010 (Sugarloaf Pipeline) and 2012 (Wonthaggi Desalination Plant), neither A LONG HISTORY OF DROUGHT IN MELBOURNE Australia is the world’s driest inhabited continent, and its population is one of the most urban. As of 2010, 89% of Australia’s 21 million inhabitants lived in urban areas. 1 Finding adequate water resources to sustain Australia’s cities is an ongoing challenge. 2 Nowhere is that more apparent than in Melbourne, a coastal city of approximately 4 million people located on the country’s southeastern coast. Over its 166-year history, Melbourne has experienced eight major droughts. The most recent one, known as the Millennium Drought, started in 1997 and lasted more than a decade. By 2009, below-average precipitation and above-average temperatures drained the city’s drinking-water reservoirs and stoked bush fires, including the “Black Saturday” fire that damaged 30% of the city’s water supply catchment and claimed 173 lives. 3 The Millennium Drought also altered public perceptions about global climate change, water conservation, and water-use behaviors, and energized city managers and politicians to adopt a wide range of approaches for augmenting water supplies and conserving water resources, although the contribution of climate change to the Millennium drought, while plausible, remains unproven. 4 In this paper, we explore how the Millennium Drought changed © 2013 American Chemical Society Special Issue: Design Options for More Sustainable Urban Water Environment Published: May 3, 2013 dx.doi.org/10.1021/es400618z | Environ. Sci. Technol. 2013, 47, 10727−10734

A two-dimensional model of hydraulic performance of stormwater infiltration systems

Abstract: Stormwater infiltration systems are a popular method for urban stormwater control. They are often designed using an assumption of one-dimensional saturated outflow, although this is not very accurate for many typical designs where two-dimensional (2D) flows into unsaturated soils occur. Available 2D variably saturated flow models are not commonly used for design because of their complexity and difficulties with the required boundary conditions. A purpose-built stormwater infiltration system model was thus developed for the simulation of 2D flow from a porous storage. The model combines a soil moisture-based model for unsaturated soils with a ponded storage model and uses a wetting front-tracking approach for saturated flows. The model represents the main physical processes while minimizing input data requirements. The model was calibrated and validated using data from laboratory 2D stormwater infiltration trench experiments. Calibrations were undertaken using five different combinations of calibration data to examine calibration data requirements. It was found that storage water levels could be satisfactorily predicted using parameters calibrated with either data from laboratory soils tests or observed water level data, whereas the prediction of soil moistures was improved through the addition of observed soil moisture data to the calibration data set.

Setting objectives for hydrologic restoration: from site-scale to catchment-scale

Abstract: The protection and eventual restoration of natural ecological functions and values in urban streams requires approaches to stormwater management that restore natural hydrologic processes at small scales, with the ultimate goal of returning catchment-scale flow-regimes towards their predevelopment behaviour. Adoption of such approaches is however, currently limited by a lack of stormwater management design objectives applicable at the site-scale. In this paper, we outline a conceptual framework for setting such objectives based on the role that an individual site plays in delivering catchment-scale hydrologic outcomes. Objectives are provided for the proportion of rain falling on impervious areas that should be lost (evapotranspired and/or harvested) and infiltrated. We also propose an objective for equivalent initial loss, which characterizes the probability of surface runoff from a given rain-event, with the aim of restoring natural catchment-scale retention of storm events. It is apparent that the management of flow-regimes at small scales will require both retention (and loss through use or evapotranspiration) of stormwater and restoration of baseflow processes. Stormwater harvesting and infiltration-based techniques are thus required to manage stormwater at small scales.

Turning (storm)water into food; the benefits and risks of vegetable raingardens

Abstract: The adoption of stormwater source control techniques such as raingardens at the household scale depends on the perception of benefit to the householder. Vegetable raingardens integrate the ability to retain stormwater at the source, while supporting a growing interest in urban food production. We compared the performance of two designs of raingarden for vegetable growing with normal vegetable gardens fed either by potable water or rainwater collected in a tank. We evaluated their yield, stormwater retention and risk of contamination. The vegetable raingarden designed to promote stormwater retention and infiltration reduced the frequency of runoff by more than 90%, yet produced a vegetable yield comparable with traditional vegetable garden designs. The chemical and microbial contamination risk from raingardens irrigated with roof water is no higher than from a vegetable garden irrigated with potable water, but we note that results may be context-specific. The concept of a vegetable raingarden has promise for its ability to simultaneously reduce stormwater runoff and support urban food production; further studies are necessary to determine if it is suitable where the raingarden receives general stormwater runoff from urban impervious surfaces.

Long-term phosphorus accumulation in stormwater biofiltration systems at the field scale

Abstract: Previous research has demonstrated that biofilters are an effective technology for the removal of phosphorus (P) from stormwater. However, biofiltration is a relatively new technology and most field-scale systems are still fairly young, therefore little is understood about the long-term ability of biofilters to act as a sink for P. Studies from a board range of disciplines indicate that iron (Fe) and P interactions are an important mechanism for P sequestration in soils. To investigate long-term P retention dynamics and associations between Fe and P in biofilters we collected filter media cores from six biofilters in both Melbourne and Brisbane. The filter media was subjected to a four-step sequential extraction scheme designed to measure P associated with the following phases: i) Bioavailable P; ii) P-adsorbed to iron oxyhydroxides; iii) P associated with amorphous iron oxyhydroxides; and iv) Organic P. The results suggest that P accumulation varies spatially (areally and with depth) in biofilters. P concentrations were highest in the top 10 cm of the filter media and near stormwater inlets. In all biofilters tested, surface layer P was mostly associated with the amorphous Fe and organic phase, which is largely related to the build-up of trapped sediment. P concentrated in the Fe-adsorbed phase increased at lower depths suggesting that Fe-P sorption interactions may play an important role in long-term P retention. This result emphasises the importance of maintaining good hydraulic performance in biofilters, since Fe-adsorbed P may be sensitive to changes in redox potential, leading to release under reducing conditions. These findings may influence how we design biofilters and plan system maintenance to ensure effective long-term P removal.

The impact of stormwater source-control strategies on the (low) flow regime of urban catchments L'impact des stratégies de contrôle à la source sur le débit d'étiage des bassins versants urbains

Abstract: Stormwater management strategies increasingly recognise the need to emulate the pre-development flow regime, in addition to reducing pollutant concentrations and loads. However, it is unclear whether current design approaches for stormwater source-control techniques are effective in restoring the whole flow regime, and in particular low flows, towards their pre-development levels. We therefore modelled and compared a range of source-control stormwater management strategies, including some specifically tailored towards enhancing baseflow processes. The strategies were assessed based on the total streamflow volume and three low flow metrics. Strategies based on harvesting tanks showed much greater volume reduction than those based on raingardens. Strategies based on a low flow rate release, aimed at mimicking natural baseflow, failed to completely restore the baseflow regime. We also found that the sensitivity of the low flow metrics to the proportion of catchment treated varied amongst metrics, illustrating the importance of metrics selection in the assessment of stormwater strategies. In practice, our results suggest that realistic scenarios using low flow release from sourcecontrol techniques may not be able to fully restore the low flow regime, at least for perennial streams. However, a combination of feasibly-sized tanks and raingardens is likely to restore the baseflow regime to a great extent, while also benefitting water quality through the retention and filtration processes.

Stormwater biofiltration-the challenges of inorganic and organic nitrogen removal

Abstract: Despite the demonstrated treatment efficiency of biofiltration, nitrogen remains a challenging pollutant to remove from urban stormwater runoff. This difficulty stems from the multiple forms in which nitrogen may be present, varying from bioavailable and inorganic to an array of largely uncharacterised organic compounds. Additionally, complex environmental influences on nitrogen transformation and removal processes affect nitrogen removal efficiency. An experiment with 245 biofilter columns of varied design indicated nitrate dictated performance variation during wet periods, suggesting differing rates of nitrification, denitrification or plant uptake. In dry months, organic nitrogen removal was additionally problematic, but displayed minimal variation between treatments. The results suggest that alternative biofilter design features are required to facilitate microbial processing of re-released organic nitrogen from plants.

Uniting Urban Agriculture and Stormwater Management: The example of the 'vegetable raingarden'

Abstract: Stormwater runs off impervious urban surfaces at artificially high rates, and erodes and pollutes local waterways. Raingardens, as biofiltration systems, are garden beds that are designed to capture and filter runoff using sandy soils and resilient plants. For healthier waterways, the construction of raingardens is being actively promoted in many cities, including Melbourne. However, raingardens might have another significant benefit; as sites of food production, using captured stormwater (runoff) for irrigation. The use of stormwater is an increasingly popular practice for overcoming water scarcity issues, which can constrain home vegetable gardening and urban agriculture. Nonetheless, the use of raingardens for food production has not been explored and vegetables represent a significant departure from the types of plants that are conventionally used in these systems. We investigated the potential to produce vegetables in raingardens through a 5-month greenhouse (pot) experiment and a 1.5-year field trial. The results indicate that it is possible to produce adequate yield in raingardens and the function of raingardens in reducing urban runoff (in terms of discharge to waterways) can be retained. An infiltration-type raingarden, sized 7.5% of its catchment area, reduced both the volume and frequency of runoff by > 90%. However, “vegetable raingardens” must be designed and managed effectively. Design issues include the use of sub-irrigation to ensure food safety and limit plant stress, and choosing filter/growing media that sustains vegetable growth while meeting runoff management objectives. Introduction: Water in Australian cities Stormwater runs off the impervious surfaces of Australian cities at artificially high rates, which poses a significant threat to the health of local waterways. For example, large volumes of urban runoff can cause severe channel erosion and lead to a loss of habitat, as well as transporting elevated concentrations of pollutants (Paul & Meyer 2001). Water Sensitive Urban Design (WSUD) is a way of incorporating treatment of this runoff into urban landscapes (Denman, May & Breen 2006; Lloyd, Wong & Porter 2002). It has involved the design and installation of a wide range of technologies, including “biofiltration” or “bioretention” systems (Davis et al. 2009). These technologies improve runoff management by intercepting stormwater flows and restoring the flow regime closer to the pre-developed, natural level (Bratieres et al. 2008; Williams & Wise 2006). In contrast to the large quantities of runoff in the urban landscape, the amount of water available from the traditional supply networks can be limited, particularly in times of drought. The water scarcity issues that affect many Australian cities are primarily caused by below-average runoff into urban water catchments (Edwards 2011). Melbourne, for example, has been affected by substantial decreases in rainfall since 1960 and some exceptionally severe droughts, the most recent lasting from 1997 to 2009 (Barker-Reid, Harper & Hamilton 2010). Water scarcity will become an even greater concern for Melbourne if long-term predictions of drier, hotter weather, as well as increasing consumption, are accurate (Edwards 2011; Howe et al. 2005). Many Australian cities have responded to water scarcity by implementing water restrictions, which require users (particularly households) to avoid or ration some uses of water. Such restrictions were in place in Melbourne for over ten years (Edwards 2011), but were lifted in December 2012 with the possibility of reinstatement during future drought periods. The most severe stage (Stage 4) included a ban on all outside watering, which significantly constrains urban food production. Urban food production is an important practice; traditional home vegetable gardening is often driven by economic motivations and cultural influences (Gaynor 2006), while a recent resurgence of interest in urban agriculture is a response to issues of environmental sustainability and food security (Barker-Reid, Harper & Hamilton 2010; Dixon et al. 2009). The use of stormwater and wastewater, which is not limited by water restrictions, has become an important component of Melbourne’s response to water scarcity (Barker-Reid, Harper & Hamilton 2010; Hatt, Deletic & Fletcher 2007; Misra, Patel & Baxi 2010). There is considerable opportunity for further expansion of stormwater reuse practices in urban food production, and for incorporation of urban food production into WSUD; at least on a small, non-commercial scale, as an extension of traditional home vegetable gardening. For example, green roofs are primarily used for stormwater management and other environmental benefits, but exploratory research has indicated that they could also be used for vegetable production (Whittinghill, Rowe & Cregg 2013). When it rains, such technology can both protect local waterways from the negative effects of stormwater and use this stormwater as a sustainable resource. The “vegetable raingarden”: Opportunities and challenges Raingardens are another prime example of a WSUD technology that has the potential to be used for vegetable production, but this has not yet been tested. Raingardens, as biofiltration systems, are garden beds which are engineered, typically using resilient plants and sandy soils with low-organic content, to capture and treat stormwater that runs off roofs and other impermeable surfaces. The use of vegetables represents a significant departure from the plant species conventionally used in biofiltration systems, which tend to be perennial, native species selected for their capacity to survive the extreme wetting-drying regime in a raingarden, and their ability to remove pollutants from runoff (Read et al. 2008). In comparison to conventional raingarden plants, vegetables are generally much more sensitive to drought and overwatering, both of which can lead to relatively poor growth and yield, and ultimately plant death in severe conditions. Vegetables typically require significant irrigation to supplement rainfall, even in traditional growing systems. Water availability is therefore a critical issue in a vegetable raingarden. In order to adapt raingardens to function effectively as sites of vegetable production, there are several knowledge gaps and design issues that need to be considered. Foremost among them is the method for delivering runoff water to the raingarden. Water usually enters a raingarden at the surface, but it might be preferable to “invert” a vegetable raingarden so that it is subirrigated. The use of sub-irrigation can offer higher water use efficiency than spray and drip irrigation, as demonstrated for tomato production (Ahmed, Cresswell & Haigh 2000; Goodwin et al. 2003; Incrocci et al. 2006; Santamaria et al. 2003). Sub-irrigation might also be beneficial for food safety, whereby pollutants can be filtered out of the runoff water as it moves upwards through the raingarden, before coming into contact with the plants. The choice of soil type is another important issue. Typically, loamy sand or similar is used in a raingarden, primarily to improve the quality of urban runoff (Bratieres et al. 2008; FAWB 2009; Henderson, Greenway & Phillips 2007). However, the relatively low water holding capacity of sandy soil makes it generally not well suited to vegetable production. Finally, there is a risk that modifying a raingarden for vegetable production will compromise its runoff management function. For example, a vegetable raingarden might be less able to capture urban runoff it is watered excessively, or if waterproof lining is required to maintain adequate soil moisture for vegetable production. Lining is also required if the raingarden is constructed close to a permanent structure, or if it is necessary to reduce the ingress of salt into the treated water, which is important in western Sydney and some other areas of Australia. Such lined systems are regarded as having poor hydrologic performance because they inhibit infiltration into underlying soils (Li et al. 2009), and fail to restore the baseflows that are lost following the creation of impervious areas. Research project summary