Modificación de la Norma Foral del Impuesto sobre Sucesiones y Donaciones

5.3.1.

Designing rain tank end use study experiment

As mentioned previously, the primary goal of the pilot study was to design an experiment that would enable the disaggregation of water consumption from both the potable and rain tank supplies into end-use events. Moreover, the study sought to design an experiment that could reveal the energy consumption of each and every end-use event supplied by the rain tank pump. The predominant issue resolved through the exploratory data analysis process in this study was the accurate disaggregation of events and water-balance calculations when the supply switched between the rain tank and potable supplies, even part way through high consumption events like irrigation.

Five GCC households were included in the pilot study. Stock inventory and socio- demographic surveys were completed for each of these households, providing an

understanding of the family makeup and its likely water-using behaviours (e.g. young children typically bathe at 5 pm), along with the model and star rating of every water-using fixture or appliance (e.g. 75 L/wash, 4-star rating washing machine). I also assessed the rain tank system in detail (e.g. pump size, switch system type, first flush size, etc.). This information aids the analyst completing the end use analysis process (see Willis et al. 2009 for overview of end use analysis process) and improves the accuracy.

Figure 13 depicts the experimental design, which includes three modified Actaris CT-5 high- resolution water meters (i.e. one at front of household; one each side of switch system), EDMI Mk7c high-resolution electricity meter (0.1 Wh/pulse), and two Aegis DataCell-R loggers (i.e. one at front of household; one at rain tank with two flow and one electricity meter inputs). Flow and energy data was collected at 5-second intervals and transferred via mobile phone network to the data server on a weekly basis. The pilot study period was only two months, which was sufficient to ensure that the experiment could handle end-use disaggregation of flow data when the rain tank supply switched to potable and vice versa.

As shown in Figure 13 and discussed in detail later, the three meters enabled the

determination of all end-use events in the household and their supply source, even when the event was split between supplies. For example, if the rain tank plumbed end use is supplied by potable water, the same flow trace will appear on the front of house meter and both meters located before and after the switch system. Otherwise, a flow-event pattern supplied by the rain tank only appears for the meter after the switch system. The experimental design proved to provide even higher accuracy of end-use categorisation than the ongoing potable-only end- use studies. Improved trace analysis came from the additional meters after the switch system, which improved the deductive process when multiple potable-supplied events are occurring simultaneously (e.g. shower and clothes washer running concurrently).

5.3.2.

Designing rain tank end use study experiment

The trace analysis and end use derivation processes for the pilot study site, as well as sample results for the experiment described above, is shown in Figure 13. The site had a 5000 L rain tank with a Davey Pump and associated RainBank switch system. The wireless loggers were set to record for a five-week period. The primary purpose of the logging period was to ensure that the supply sources switched over at least once in this period, capturing both the rain tank and potable supply of the three mandated end uses.

The flow trace process using Trace Wizard has an essential role in this particular study beyond disaggregating flow into end-use events. The process also has a critical role to reveal whether individual events from the rain tank plumbed end uses have been supplied by either the tank or potable water supply sources. For potable supplied events, the identical flow trace pattern will be clearly evident for the two smart meters each side of the switch system, and the flow pattern will also be displayed at the front meter along with all other events supplied by potable supply. As an example, Figure 14 shows a clothes washer event from the potable supply. For a rain tank supplied end use, a flow trace event will be evident only on the smart meter after the switch system (i.e. not the meter before the switch system). As an example, Figure 15 shows a toilet event that is supplied by the rain tank.

End-use analysis for the two smart meters each side of the switch enables the resolution of supply from the rain tank and potable supplies. Figure 16 illustrates the end-use summary for the smart meter before the switch (Figure 16a), after the switch (Figure 16b) and the resolved rain tank supplied water (Figure 16c), which represents the difference between (a) and (b). For example, the water consumed by the clothes washer from the rain tank is the difference between 1571 L and 922 L, equating to 649 L for the study period. The potable supply consumed 922 L, which will also be evident in the front potable water supplied end use breakdown, along with all other potable supplied end uses (Figure 17).

Figure 14: Trace Wizard software showing a clothes washer event supplied by potable water (a) potable supply smart meter (b) meter before switch (c) meter after switch

Front meter

Clothes washer

10.7.2010 8:00 – 12:16 AM

Meter before switch

Clothes washer

10.7.2010 8:00 – 12:16 AM

Meter after switch

Clothes washer

10.7.2010 8:00 – 12:16 AM

Figure 15: Trace Wizard software showing event supplied by rain tank

(a) potable supply meter (b) meter before switch (c) smart meter after switch

Figure 16: Resolving rain tank end uses (a) end-use summary of meter

after switch

(b) end-use summary of meter before switch

(c) rain tank end-use summary

Figure 17: Potable water supplied end-use summary

As noted previously, while the pilot study revealed the end uses of the potable and rain tank supplies, I used a modelling approach in this study for three reasons. Firstly, funds to

undertake a rain tank end use study were not forthcoming at time of fellowship project (it cost approximately $3000 per household for the experimental equipment. Secondly, Griffith University ethics approval and household consents were not available at the time of study. Finally, I would need to collect and analyse 12 months of end-use data for a sample of at least 50 households for this type of scheme to obtain a representative end-use summary for both the potable and rain tank supply sources. Due to these constraints, I adopted a

modelling approach to reveal the end uses of the potable and rain tank supplied end uses, as discussed below.

Front meter

No toilet event

21.7.2010 6:04 – 7:04 PM

Meter before switch

No toilet event

21.7.2010 6:04 – 7:04 PM

Meter after switch

Toilet event

21.7.2010 6:04 – 7:04 PM