Water Journal : Water Journal April 2011
refereed paper technical features 138 APRIL 2011 water water supply options is shown in Figure 3 to range from 2kL (the minimum size required to receive the rebate) to 85kL, but with 75% of tanks less than or equal to 10kL. Household connectivity information was collected during the rebate approval process. Connectivity is defined as connection to the bathroom, kitchen, laundry, toilet and external use. Properties may have any combination of these connections. All houses with internal connections to the rainwater tank are required to be plumbed so that metered water replaces tank water when the tank becomes empty. The distribution of household connection combinations in the rainwater tank group is shown in Figure 4. Approximately 60% of households have opted for external use only (1-e) connections. Water demand is affected by a range of factors including household occupancy, weather, climate, water outlets, garden water use, personal habits and a range of other variables, so it varies widely between dwellings. Except for weather/ climate, none of the variables affecting water demand are available for analysis. Prior to tank installation, total household water demand can be measured exactly as the metered water consumption of the household. After tank installation, because the proportion of the household's water use that comes from the rainwater tank is unmetered, total household water demand is unknown. This study uses the average daily water consumption of all single residential dwellings within an LGA during a quarter meter period as the measure of underlying water demand during that quarter. This is termed single residential water consumption and is abbreviated to srwc hereafter. srwc is the product of all factors driving household water demand and so provides a quantified rate of water demand against which the metered water use of individual households or groups of households can be compared. srwc acts as a control group in this analysis. There are a number of factors limiting the accuracy of srwc to deliver a true measure of total (metered plus unmetered) water demand. Among these are that it was not possible to remove water use by the rainwater tank rebated dwellings analysed by this study, or by dwellings constructed under BASIX building requirements, from the calculation of srwc. Both of these factors should result in progressive underestimation of srwc over time, resulting in an underestimate of water savings from tank installation. These groups are small in number relative to the total number of single residential dwellings in the Rous Water supply area (at most 5% of all single residential dwellings), so their affect on srwc is likely to be small. Figure 5 shows a time-series of srwc for each LGA across the period of this study. The timing and severity of water restrictions is indicated. srwc was very low during the summer of 2002--03, a period of hot, dry (drought) conditions during which water restrictions were imposed between January 9, 2002 and May 10, 2003. The very low rates through the summer of 2002--03 coincide with category 3, 4, 5 water restrictions. There have been no periods of water restriction in the Rous Water supply area since May 2003, and all but two of the 3,309 post-tank meter records are from quarters without restrictions. Linear correlation Figure 3: Histogram of tank volumes. Figure 4: House-tank connection types. Number-letter codes beneath each bar indicate the number and type of connections: e-external; k-kitchen; l-laundry; t-toilet; b-bathroom; h-whole house. Figure 5: Time-series of single residential water consumption (srwc) and water restrictions for the four constituent council areas supplied by Rous Water, 2002-2010.
Water Journal March 2011
Water Journal May 2011