Temperature and rainfall thresholds corresponding to water consumption in Greater Melbourne, Australia

Abstract

Water demand modelling is a crucial issue in the current climate of water restrictions and water conservation in Melbourne, Australia. Average annual inflows into Melbourne's major reservoirs since 1913 dropped rapidly by almost 40% during the 1997-2009 (Millennium Drought). This led to the development of the Sustainable Water Strategy for Central Region by the Victorian Government. The Strategy identified wide range of actions to secure water supplies in Victoria s Central Region for the next 50 years. The Strategy also sets a per capita consumption reduction target of 25% from 1990 s average use by 2015, increasing to 30% by 2020. Water conservation efforts and initiatives were and are being implemented with interest in quantifying the impact of water conservation programs on reducing water demand. However, evaluation of the effectiveness of these options requires considerable effort since each of these options has a distinguishing set of environmental, social and economic outcomes. There has been significant modelling made to date to incorporate the effects of water conservation programs, although these are still in the early stage of development. Basic water demand models are based on the premise that total water use is made up of base use and seasonal use with base use characterised by the water use during winter. Previous models described base use to be mainly indoor use that is weather insensitive, however, a number of studies revealed that base use is weather dependent and water usage during winter months may include garden watering in some areas. In some cases, base use was modelled as representative of winter usage, based on the months of lowest usage in a year, but correlation of base use values with temperature and rainfall was not undertaken for residential water use. Even correlation of base use values with temperature and rainfall is still not undertaken for mixed water uses comprising of residential (indoor use), industrial, and commercial. In this context, there is still a need to investigate whether base use for water demand modeling is weather-insensitive or weather-sensitive. Hence, threshold of temperature and rainfall needs to be determined to investigate whether “base use” values which are mainly for indoor purposes are weather insensitive or whether “base use” values represent winter usage that may include gardening in other areas or cities i.e. weather sensitive and not mainly indoor use. This paper determines the thresholds at which water consumption is independent of temperature and rainfall. In general, when surface air temperature increases, consumption of water increases and when rainfall increases, water consumption decreases. However, a threshold point is reached beyond which increase of temperature and rainfall values no longer result to increase or decrease in water consumption. In one word, beyond the threshold line water use is independent of temperature and rainfall. To fulfil the aim of this research, daily water consumption of Greater Melbourne from Melbourne Water and daily temperature and rainfall data recorded by Bureau of Meteorology, Australia from January 01, 1980 to December 31, 2009 are analysed. From the analysis, temperature threshold is found as 15.53 C for Greater Melbourne while rainfall threshold as 4.08 mm. It could be noted that at temperature higher than 15.53 °C, daily water use increases as the temperature increases but below this threshold, daily water use seemed to be independent of temperature and increases in water use at this level could be attributed to population increase or other factors. The identified temperature threshold accounts for 23 percentile of the daily maximum temperature recorded from January 1980 to December 2009 and usually occurred during the months of May–September. Daily water use increases as the rainfall decreases but above this threshold of 4.08 mm, any more rainfall would no longer contribute to daily water use reduction. This could be due to the resulting saturated soil moisture content or water use has already been driven to its base use level. The identified rainfall threshold accounts for 70 percentile of the daily rainfall recorded from January 1980 to December 2009.