Integrated modelling of residential water-related energy

Abstract

Around the world, urban demand for resources is increasing over time. In Australia, 90% of the population resides in cities, which are growing in both population and housing density. These factors place greater demands on water and energy, and associated greenhouse gas (GHG) emissions. Water use, energy use and GHGs are strongly interconnected, and thus actions to reduce water or energy consumption can have unintended consequences. Consequently, reducing water use without increasing energy use and GHGs as well as reducing energy use without increasing water use is important. One key area where this can be achieved is in the reduction of water-related energy (WRE) use.WRE consumption occurs in two distinct sectors: the water sector through water supply and sewage collection services, and the residential sector through water end uses. WRE use of both utilities and end users are interconnected through infrastructure, environment, technology, behaviour, and policies. WRE use of water supply (10%) and sewage collections services (10%) are both within the control of utilities, however, the most WRE intensive component of the residential urban water cycle is residential end use (80%), which is largely outside the control of water utilities. This thesis used a systems approach to WRE modelling, across the water utility and residential sector interface, to investigate opportunities for whole-of-system reduction in resource consumption.Firstly, this study investigated an interaction between the infrastructure and the environment through the cold water temperature (CWT) variability impact on household WRE. The spatiotemporal variability in CWT was determined using 5760 measurements from 1255 sampling locations across Yarra Valley Water, Melbourne, Australia. The monthly CWT varied across the 4000 km2 study site from 12-28°C during summer and 9-15°C during winter. Spatial clusters of hot spots and cold spots were observed. Variation in CWT was calculated to affect annual household WRE demand by -17 to +19%. Variability in results demonstrated the difference in household WRE demand in hot spots, cold spots, and neutral zones.Monthly mean CWTs for the study site diverged from hot water system (HWS) energy consumption guidelines by -21 to +47%. The CWT variability impact on household water heating varied up to three times the energy used by the water utility for water supply and sewage disposal services in this region. Results demonstrated the importance of modelling interactions between infrastructure and the environment. Quantifying the variability of CWT increased the accuracy of predicting regional WRE demand and HWS energy consumption.Secondly, this study investigated how technology and behaviour influenced resource consumption at regional scales. This objective explored how the interactions between household composition, HWS type, shower use, and clothes washer use affected regional resource consumption in Reservoir, Melbourne, Australia. The regional ResWE model of water, WRE and GHGs, was upscaled from a previously established household model using local water authority information and census data to capture end use variability between individual households. In total, 320 household types were used to characterise end use variability.Shower systems were found to be the largest lever for reducing resources. Changes in shower technology and behaviour together were predicted to generate annual water reduction of 27% WREelectricity reduction of 15%, WRE-gas reduction of 48%, and WRE-GHG reduction of 28%. Clothes washing highlighted the importance of accounting for interactions between behaviour and technology to reduce regional resources i.e. 100% penetration of front loaders reduced regional water use but increased regional WRE and GHGs because front loaders used more energy than top loaders for a cold wash cycle and 70% of households used a cold wash cycle. Economies of scale was a factor in the household composition effect on resource consumption where larger occupancy households were the lowest consumers per capita. In contrast, smaller occupancy households were the highest consumers: 53% of the population lived in 73% of the household stock and consumed 59% of resources. Thus indicating, prediction models of either water or energy use need to consider projected changes in demographics to effectively capture changes in resource use. Overall, results demonstrated how household interactions between technology and behaviour significantly determined regional resource consumption.Lastly, this study proposed a conceptual model which coupled the regional ResWE model of residential water-energy interactions to a geographical information system platform. The conceptual model proposes integration of information from top-down data (utility decisions) and bottom-up data (end user decisions) over a range of spatial and temporal scales. This enables the evaluation of the cumulative impacts of changes in factors that influence WRE use across infrastructure, environment, behaviour, technology and policy decisions.Integrated WRE modelling demonstrated the significant difference between household and regional water-energy interactions. This result would not have been found through a linear scale-up of household results. In conclusion, implications of both household and regional water-energy interactions need to be accounted for in the formation of policies related to water, energy or GHGs.