Life-cycle perspectives for urban water systems planning

Abstract

Urban water systems around the world are going through a period of substantial change: they are evolving towards more complex water supply alternatives; are being placed under increasing pressure to achieve higher quality effluent and biosolids discharges; and are being confronted with a growing number of broader environmental management challenges. This thesis explored the use of the Life Cycle Assessment (LCA) methodology for assisting in that process of change, because of LCA’s strengths in combining practicality and flexibility with ‘big picture’ thinking. The overarching goals were to: (1) explore LCA principles that do or do not provide useful perspectives for urban water planners, and (2) identify situations where the benefits from life-cycle thinking will be impeded by gaps in data and modelling approaches. The analytical starting point was a comparison of two different configurations for a city-scale, integrated water supply and wastewater system: a ‘traditional’ approach dependent on low-energy dam-sourced mains water supply; and one with a more complex mix of contemporary water supply infrastructure intended to reduce the intensity of freshwater extraction and nutrient discharge. This change incurs a substantial increase in energy use, meaning the reduced pressure on local aquatic ecosystems may come at the expense of large increases in other life-cycle impacts. Notably however, the results generated in this thesis indicate that an exclusive focus on energy use is unlikely to be a robust approach to factoring these bigger picture environmental impacts into water industry decision making. Furthermore, it is the wastewater components of the system, rather than the water supply components, that make the largest contribution to most of the life-cycle impacts. An excessive focus on the energy or greenhouse gas (GHG) implications of growing urban water demands is, therefore, unlikely to chart the industry on an optimal course to a more environmentally benign system configuration. The estimates for direct (scope 1) greenhouse gas emissions in this thesis utilise a comprehensive set of locally relevant empirical data and expert knowledge. Based on that, direct emissions could comprise 20% or more of the overall GHG footprint for urban water infrastructure systems. The substantial spatial variability associated with all the largest direct emission sources should be an important consideration in the urban water decision making process. For assessing the option to dispose of sewage treatment plant (STP) biosolids onto farmlands, the uncertainty associated with estimating field fluxes of carbon and nitrogen is likely to be more important than the more traditional focus on biosolids transport energy. The second major case study considered in this thesis is focussed on the issue of biosolids reuse for agricultural purposes. When that practice is assessed against a broader set of impact categories than just energy use or GHG emissions, it becomes apparent that conventional life-cycle impact assessment (LCIA) models could bias against this as a preferred fertiliser source. With respect to nutrient discharge, metals toxicity, and phosphorus recovery, there is a disconnect between the results produced with these impact assessment models, and the scientific knowledge and industry priorities that currently guide the associated Australian policy debate. Growing use of LCA in the Australian agricultural sector will encourage the use of those very models that are least well placed to provide useful critique of biosolids applications to soils, hence could lead to a weakening of agricultural support for this practice. This could pose a risk for water utilities already dependent on farmers to absorb the majority of their STP biosolids. Phosphorus recovery and organics toxicity are both issues that could benefit from analysis incorporating the life-cycle perspective, since for both there is the prospect that water industry mitigation actions could shift the environmental burden to somewhere else in its supply chains. However, the analyses presented here suggest that the available LCIA models are not up to this task in either case. For the assessment of minerals resource depletion, the choice of impact assessment models could also have a substantial effect on the results that are obtained. A number of priority tasks are identified here, that would advance the LCA modelling framework so it can provide more meaningful contribution to urban water cycle planning. Ozone depletion assessment is another issue where the adoption of conventional LCIA approaches will fail to provide any useful insight to the urban water industry. There is a strong case for including N2O emissions in such assessments, and doing so clearly indicates this could have a material influence on the conclusions draw from analysis of water infrastructure systems. Quantifying the ozone enhancing effects of CO2 and CH4 emissions remains a bridge too far for the available LCIA models, but their increasing and complex influence suggests there may be a need for evolution in the metrics used to assess the ozone depletion issue. The urban water industry would likely be affected by any changes in international ozone-layer policy as a result of the increased scientific focus on these non-halocarbons, and should keep a watching brief on this issue. The work undertaken in this thesis clearly identifies the value that can be derived from the LCA approach to infrastructure and options analysis. Furthermore, the compilation of whole-of-system data provides benchmarks that offer valuable benefits for the task of considering environmental trade-offs – whether that be from comparing across different water system technology options, and/or comparing across impacts that occur at different localities or points in time, and/or comparing across different environmental issues. In all respects, the goal should be to strive for robust consideration across all important life-cycle contributions and impacts. The challenge is to appropriately direct effort into the issues that matter, rather than those that are easiest to deal with. Incidental benefits derived from detailed industry data collection can be substantial, however they do not necessarily ensure that LCA delivers its fullest value. Detailed consideration may well be required on uncertain issues that are beyond the traditional expertise of the analyst. Practitioners should also resist the temptation to uncritically adopt historical convention in the choice of LCA impact assessment models. Lack of rigour in the definition of inventory data and/or impact assessment models doesn’t prevent LCA studies from being completed, but it will greatly diminish the value of the LCA exercise.