Rb

Aa

b

a

ARR1A

KRLST

1

bAmisirinpaAt‘vfnot

0d

Resources, Conservation and Recycling 61 (2012) 16– 21

Contents lists available at SciVerse ScienceDirect

Resources, Conservation and Recycling

journa l h o me pag e: www.elsev ier .com/ locate / resconrec

ainwater harvesting in Greater Sydney: Water savings, reliability and economicenefits

taur Rahmana,∗, Joseph Keanea, Monzur Alam Imteazb

School of Engineering, University of Western Sydney, Locked Bag 1797, Penrith, NSW 2751, AustraliaSwinburne University of Technology, Melbourne, Australia

r t i c l e i n f o

rticle history:eceived 15 June 2011eceived in revised form7 November 2011ccepted 1 December 2011

a b s t r a c t

Due to greater environmental awareness and mandatory water restrictions in many Australian cities,rainwater tanks have become popular in recent years. This paper investigates the water savings potentialof rainwater tanks fitted in detached houses at 10 different locations in Greater Sydney, Australia. A waterbalance simulation model on daily time scale is developed and water savings, reliability and financialviability are examined for three different tank sizes, 2 kL, 3 kL and 5 kL. It is found that the average annual

eywords:ainwater tanksife cycle cost analysisustainable water useank rebate

water savings from rainwater tanks are strongly correlated with average annual rainfall. It is also foundthat the benefit cost ratios for the rainwater tanks are smaller than 1.00 without government rebate. Itis noted that a 5 kL tank is preferable to 2 kL and 3 kL tanks and rainwater tanks should be connectedto toilet, laundry and outdoor irrigation to achieve the best financial outcome for the home owners.The results from this study suggest that government authorities in Sydney should maintain or possiblyincrease the rebate for rainwater tanks to enhance its acceptance.

© 2012 Elsevier B.V. All rights reserved.

. Introduction

Australia is one of the driest continents in conjunction witheing one of the highest consumers of drinking water in the world.

growing urban population and frequent droughts due to cli-ate variability and change have made water supply to be a major

ssue in Australia (Ryan et al., 2009). A number of alternative waterources have received attention in recent years in Australia, whichncludes rainwater harvesting, grey water reuse and wastewaterecycling (Zhang et al., 2010). Among these, rainwater harvest-ng has received the greatest attention as rainwater is fresh inature and can be easily collected and used for non-potable pur-oses. However, many people in Australia still show reluctance indopting a rainwater harvesting system (RWHS). Statistics from theustralian Bureau of Statistics (ABS) show that about 47% say that

he main reason for not installing a rainwater tank is the perceivedhigher cost’ (ABS, 2009). Government authorities in Australia pro-ide financial incentives in the form of rebate to the home ownersor encouraging them to install rainwater tanks. For example, Syd-

ey Water Corporation in Australia offers a rainwater tank rebatef up to Aus$1,400 depending on the type of water use and size ofhe tank (Sydney Water, 2010).

∗ Corresponding author. Tel.: +61 247360145; fax: +61 247360833.E-mail address: a.rahman@uws.edu.au (A. Rahman).

921-3449/$ – see front matter © 2012 Elsevier B.V. All rights reserved.oi:10.1016/j.resconrec.2011.12.002

Muthukumaran et al. (2011) found that use of rainwater insidea purpose-built home in regional Victoria in Australia can save upto 40% of potable water use. Farreny et al. (2011) examined thequantity and quality of a RWHS in Spain and found that slopingsmooth roofs may harvest up to about 50% more rainwater thanflat rough roofs. Mun and Han (2012) developed a design and eval-uation method for a RWHS on the basis of water balance equationand found that a design based on sensitivity analysis can notablyimprove the operational efficiency of a RWHS.

Many house owners do not readily see the benefit of RWHSover longer term, which may be attributed to the limited under-standing of the life cycle costs of the system. A study by Rahmanet al. (2010) for multi-storey buildings in Sydney found that itcould be possible to achieve “pay back” for the RWHS under somefavourable scenarios and conditions. They found that a smallerdiscount rate is more favourable and the greater the number ofusers the higher the benefit–cost ratio for a RWHS. Domenechand Sauri (2010) investigated the financial viability of the RWHSin single and multi-family buildings in the metropolitan area ofBarcelona (Spain). In single-family households an expected pay-back period was found to be between 33 and 43 years dependingon the tank size, while in a multi-family building a payback

period was 61 years for a 20 m3 tank. Imteaz et al. (2011) foundthat for commercial tanks connected to large roofs in Melbourne,total construction costs can be recovered within 15–21 years timedepending on the tank size, climatic conditions and future water

servation and Recycling 61 (2012) 16– 21 17

ptttieiswttr

bJswbsttuWststietw6hboaK

itcst

iTdtuietpoleptban

ifiaR

Fig. 1. Selected study locations in Greater Sydney, Australia.

Table 1Summary of selected locations in Greater Sydney and rainfall statistics.

Location Rainfall station Rainfall records Average annualrainfall (mm)

Bankstown 066054 1986–2009 1009Campbelltown 068007 1900–2009 743Cronulla 066058 1910–2009 1085Hornsby 066158 1936–2009 1325Kellyville 067042 1978–2009 880Manly 066182 1957–2009 1376Parramatta 066124 1966–2009 963Penrith 063185 1970–2009 971

(average: 73 years) as shown in Table 1. Typical average monthlyrainfall distribution in the study region is shown in Fig. 2.

A. Rahman et al. / Resources, Con

rice increase rate. Tam et al. (2010) investigated the cost effec-iveness of RWHS in residential houses around Australia and foundhat this system can offer notable financial benefit for Brisbane,he Gold Coast and Sydney due to the relatively higher rainfalln those cities as compared to Melbourne. Zhang et al. (2009)xamined the financial viability of RWHS in high-rise buildingsn four capital cities in Australia and found that Sydney has thehortest payback period (about 10 years) followed by Perth, Dar-in and Melbourne. Khastagir and Jayasuriya (2011) conducted

he financial viability of RWHS in Melbourne, Australia and foundhat payback period vary considerably with the tank size and localainfall.

Notable researches have been conducted on the relationshipetween rainwater tank sizing and water savings. Khastagir andayasuriya (2009) used water demand and roof area to develop aet of dimensionless number curves to obtain the optimum rain-ater tank size for a group of suburbs in Melbourne. A paper

y Su et al. (2009) focused on the development of a relation-hip between storage and deficit rates for RWHS. Results showedhat as the deficit rate increased so too did the storage size ofhe tanks. Eroksuz and Rahman (2010) conducted research on these of RWHS for multi-unit blocks in three cities of New Southales, Australia. They found that in order to maximize the water

avings, a larger tank would be more appropriate and that theseanks could provide significant water savings, even in dry years. Atudy in Brazil by Ghisi et al. (2009) aimed to assess the poten-ial for potable water savings for car washing at petrol stationsn the City of Brasilia found that an increase in the tank sizenhanced the reliability of the rainwater tank notably in meetinghe demand. Kyoungjun and Chulsang (2009) showed that rain-ater collection would only be feasible in South Korea for during

months of the year. They also found that a benefit cost ratioigher than 20% could not be gained due to the cost of watereing so inexpensive in South Korea. They suggested that the costf water supply would need to be increased by a factor of fivepproximately for the RWHS to become financially viable in Southorea.

The research on financial viability of RWHS has not had signif-cant presence in the literature thus far and also the findings fromhese studies are often contradictory. The home owners do notlearly see the financial benefits of a RWHS. Also there is a lack oftudy on the adequacy of the current government rebate providedo the home owners for installing a RWHS.

This study focuses on the efficiency of RWHS in Sydney, whichs the largest city in Australia, with over 4.5 million populations.he city was under severe water restriction for about a decadeuring 1990s and 2000s, which prompted the search for alterna-ive water supplies in the Sydney region. To enhance the waterse efficiency and water conservation, the government authorities

n Sydney has introduced BASIX legislation, which requires thatvery new house in Sydney must have a rainwater tank. However,here has not been any in-depth study to determine an appro-riate tank size for a given house in a given location dependingn the roof area, family size and local rainfall. As Sydney is quitearge and has a high rainfall gradient, it is most likely that differ-nt parts of the City need different tank sizes for achieving the bestossible water savings and financial returns. This study examineshe water savings potential, reliability of water supply, financialenefits, and the adequacy of the current government rebate for

RWHS in detached house at different locations in Greater Syd-ey.

The research presented in this paper is undertaken

n the light of the current knowledge gaps to assess thenancial viability of a RWHS in Sydney to provide guid-nce to water authorities to enhance the acceptance of aWHS.

Richmond 067021 1902–2003 800Sydney City Centre 066062 1859–2009 1214

2. Description of data

The study considers 10 different locations across Greater Sydneyin Australia (Fig. 1). These locations are Bankstown, Campbell-town, Cronulla, Hornsby, Kellyville, Manly, Penrith, Parramatta,Richmond and Sydney City Centre. The daily rainfall data at each ofthese locations is obtained from the Australian Bureau of Metrol-ogy. The length of these rainfall data ranges from 31 to 150 years

Fig. 2. Average monthly rainfall in the study area.

18 A. Rahman et al. / Resources, Conservation and Recycling 61 (2012) 16– 21

Table 2Capital cost and government rebate (Sydney Water, 2010) for different sized rain-water tanks.

Tank size (kL) Capital cost (Aus$) Rebate from SydneyWater (Aus$)

2 $1,743 $1,150

tlamhe1aecuar

flwoicwwg1

abcficaS

3

ids

w(

R

R

watt

S

S

annual rainfall values at these two locations (1325 mm and 743 mm

3 $1,881 $1,1505 $2,113 $1,400

Three different combinations of water use are considered: (i)oilet and laundry (ii) irrigation and (iii) a combination of toilet,aundry and irrigation (combined use). Three different tank sizesre considered: 2 kL, 3 kL and 5 kL. A hypothesized new develop-ent is considered at each of the study locations with a single

ousehold having 4 occupants. A total site area of 450 m2 is consid-red with a roof, lawn and impervious areas of 200 m2, 150 m2 and00 m2, respectively. The rainwater tank is considered to be locatedbove ground. A mains top up system is considered to supplementach tank once the volume of water drops below 20% of the tankapacity. A pump is considered to serve indoor (toilet and laundry)se while a gravity system is assumed for irrigation use. First flushnd leaf-eater devices are considered to filter contaminants fromoof runoff before water enters into the tank.

Daily water demand data for toilet, laundry and irrigation useor residential properties are obtained from Sydney Water. The toi-et considered in the study is a 6-l AAA (the higher the A’s the more

ater efficient the device is) rated dual flush toilet, and a frequencyf toilet use of three times per person per day is assumed, whichs equivalent to 0.018 kL per person per day. The washing machineonsidered in this study is either rated 4A. It is assumed that theashing machine has a volume of 50 l and it is used 3 times pereek, which is equivalent to 0.0215 kL/day of water use. The irri-

ation demand per square metre of lawn area is assumed to be0 mm/day.

The life cycle cost analysis (LCCA) considers the capital and oper-ting costs of the RWHS plus the value of water savings deliveredy a RWHS. It is assumed that the RWHS has a life of 40 years. Theapital costs include the rainwater tank itself, concrete base, pump,rst flush, leaf-eater device, pump to tank connection kit, electri-al, plumbing supplies and necessary labour cost. Table 2 provides

summary of the capital cost and government rebate provided byydney Water for the installation of a domestic RWHS.

. Method

A water balance simulation model (WBSM) on daily time scalen excel is built which considers various factors such as tank size,aily rainfall, losses, daily water demand, mains top up and tankpillage.

In the WBSM, rainfall is regarded as inflow and the release asell as possible spill as outflow following the approach by Su et al.

2009). The release is estimated based on the following equations:

t = Dt if It + St−1 ≥ Dt (1)

t = It + St−1 if It + St−1 < Dt (2)

here Dt is the daily demand (m3) on day t, St−1 is the tank storaget the end of the previous day (m3), Rt is release from rainwaterank (m3) and It is inflow (m3). Spill (SPt) (m3) is calculated fromhe following equations:

Pt = It + St−1Dt − SMAX if It + St−1 − Dt > SMAX (3)

Pt = 0 if It + St−1 − Dt ≤ SMAX (4)

Fig. 3. Reliability of rainwater tanks at 10 selected locations for toilet and laundryuse.

where SMAX is the design storage capacity (m3). The tank storage St

at the end of day t can be calculated using the following equations:

St = SMAX if SPt > 0 (5)

St = St−1 + It − Rt if SPt ≤ 0 (6)

The reliability of a rainwater tank is calculated as the ratio of thenumber of days when intended demand is met fully by the availablerainwater and the total number of simulated days.

To carry out the LCCA, all the present and future values are con-verted to present day dollar value using a discount rate of 3%. Weadopt the concept of nominal cost (the expected price that will bepaid when a cost is due to be paid, including estimated changesin price due to changes in efficiency, inflation/deflation, technol-ogy and the like) and nominal discount rate (the rate to use whenconverting nominal costs to discounted costs). To convert a nomi-nal cost (CN) to a discounted cost (CD), following equation is used(AS/NZS, 1999):

CD = CN ×(

1(1 + dn)y

)(7)

where dn is the nominal discount rate per annum and y is the appro-priate number of years. The benefit cost ratio is estimated as theratio of the sum of all the discounted benefits and discounted costs.The benefit comes from the water savings achieved from a RWHS,which is estimated as the product of mains water savings and waterprice. The value of possible environmental benefits from a RWHS(e.g. flood control and removal of pollutants from roof runoff) isnot considered in the LCCA due to limited data availability, whichwould have disfavoured the RWHS.

4. Results

4.1. Reliability of rainwater tanks in meeting the demand

Fig. 3 provides a summary of the reliability of the three differ-ent sized rainwater tanks for toilet and laundry use across all the10 locations. The average reliability is also included in Fig. 3, whichshows that on average a 2 kL rainwater tank can meet the demandfor toilet and laundry use for 86% of the days in a year, whichincreases to 97.5% for a 5 kL tank size. Among the 10 locations,Hornsby has the highest reliability (99% for a 5 kL tank) and Camp-belltown has the lowest one, which is well related to the average

for Hornsby and Campbelltown, respectively). As can be seen fromFig. 3, the differences in reliability across different locations reduceas the tank size increases.

A. Rahman et al. / Resources, Conservation and Recycling 61 (2012) 16– 21 19

Fu

fttbvHHt

i8i(cMIltito

4

swst5

F

the benefit cost ratios for these three locations without consider-

ig. 4. Reliability of rainwater ranks for toilet, laundry and irrigation (combinedse).

For irrigation, the highest and lowest reliabilities are found to beor Hornsby (72% for a 5 kL tank) and Campbelltown (39%), respec-ively, which is well related to the average annual rainfall values ofhese two locations (1325 mm and 743 mm for Hornsby and Camp-elltown, respectively). Fig. 4 provides a summary of the reliabilityalues for the toilet, laundry and irrigation use (combined use).ere again, the highest and lowest reliabilities are found to be forornsby (69% for a 5 kL tank) and Campbelltown (38%), respec-

ively.On average a 5 kL rainwater tank for combined use in Hornsby

s found to be empty, more than half-full, and full for 114, 211 and8 days/year, respectively. Fig. 5 indicates that a tank in Hornsby

s most likely to be empty during the months of July to Septemberwhen the average monthly rainfall is the lowest in the region asan be seen in Fig. 2) and most likely to be full during January toarch (when the average monthly rainfall values are the highest).

t may be noted from Fig. 5 that a rainwater tank in Hornsby isikely to be at least half full for more than half of the month for allhe months except for July, August and September. Campbelltowns found to be the worst performing location with a 5 kL rainwaterank being empty for 227 days/year and full for only 38 days/yearn average (Fig. 6).

.2. Water savings

For toilet and laundry use, for a 5 kL tank average annual wateravings range from 32.8 kL (Campbelltown) to 34.7 kL (Bankstown),

ith a mean value of 34.1 kL, which show a relatively uniform water

avings across the 10 study locations. For irrigation use, for a 5 kLank average annual water savings range from 43.1 kL (Sydney) to7.5 kL (Penrith) with a mean value of 48.5 kL. Fig. 7 shows water

ig. 5. Frequency of tank being full, half-full and empty at Hornsby (combined use).

Fig. 6. Frequency of tank being full, half-full and empty at Campbelltown (combineduse).

savings for combined use. For a 5 kL tank, Manly has the highestaverage annual water savings (69.5 kL), Richmond has the lowestannual water savings (54 kL) and Parramatta corresponds to theaverage annual water savings over the 10 locations (59.5 kL). Inter-estingly, Manly and Richmond have the highest and lowest averageannual rainfall values of 1376 mm and 800 mm, respectively, andParramatta has the average annual rainfall value (963 mm) veryclose to the mean value (1048 mm) over the 10 locations. Indeed,it is found that the average annual water savings for combined usein Greater Sydney is strongly correlated with the average annualrainfall as expressed by the following equations:

AWS2 = 33.56 + 0.0067 × AAR, r = 0.49 (8)

AWS3 = 40.41 + 0.0085 × AAR, r = 0.50 (9)

AWS5 = 44.95 + 0.0162 × AAR, r = 0.63 (10)

where AWS2 is average annual water savings (kL) for toilet, laundryand irrigation (combined use) for a 2 kL rainwater tank and AAR isaverage annual rainfall value (mm). AWS3 and AWS5 represent 3 kLand 5 kL rainwater tank, respectively, and r is Pearson correlationcoefficient.

4.3. Life cycle cost analysis (LCCA)

In the LCCA, three locations are selected, which are Manly, Parra-matta and Richmond as they correspond to the highest, average andlowest average annual water savings, respectively. Table 3 provides

ing any government rebate. For toilet and laundry use, the benefitcost ratio is the highest for Manly and the lowest for Richmond. Forirrigation use, the highest benefit cost ratio is achieved with a 5 kL

Fig. 7. Water savings for combined use.

20 A. Rahman et al. / Resources, Conservation and Recycling 61 (2012) 16– 21

Table 3Benefit cost ratios at Manly, Parramatta and Richmond without government rebate.

Tank size (kL) Toilet and laundry use Irrigation use Toilet, laundry and irrigation use (combined use)

Manly Parramatta Richmond Manly Parramatta Richmond Manly Parramatta Richmond

2 0.48 0.45 0.42 0.52 0.46 0.49 0.66 0.58 0.523 0.47 0.46 0.43 0.65 0.56 0.64 0.76 0.67 0.605 0.45 0.44 0.43 0.72 0.62 0.74 0.90 0.77 0.70

Table 4Benefit cost ratios at Manly, Parramatta and Richmond with government rebate.

Tank size (kL) Toilet and laundry use Irrigation use Toilet, laundry and irrigation use (combined use)

Manly Parramatta Richmond Manly Parramatta Richmond Manly Parramatta Richmond

2 0.91 0.87 0.81 0.56 0.49 0.45 1.27 1.11 1.003 0.86 0.84 0.80 0.69 0.60 0.55 1.40 1.22 1.105 0.90 0.89 0.86 0.84 0.72 0.66 1.82 1.55 1.40

Table 5Rebate required to achieve benefit cost ratio of 1.00 at Richmond.

Tank size Toilet and laundry use Irrigation use

Current rebate (Aus$) Rebate required (Aus$) % Increase Current rebate (Aus$) Rebate required (Aus$) % Increase

2 $1,150 $1,535 33 $150 $850 4663 $1,150 $1,585 38 $150 $710 3735 $1,400 $1,740 24 $400 $800 100

Table 6Benefit cost ratio vs. water price in Richmond (without increasing current government rebate).

Tank size (kL) Benefit cost ratio (toilet and laundry use) Water price ($/kL) Benefit cost ratio (irrigation use) Water price ($/kL)

4

4

4

t(hfTt

aatb0a

lhrbbtrwwrttt

4

n

2 1.012 2.513 1.001 2.515 1.082 2.51

ank. The highest benefit cost ratio for toilet, laundry and irrigationcombined use) is 0.90 at Manly with a 5 kL tank. In summary, theighest benefit cost ratios (0.70–0.90) are achieved for a 5 kL tank

or the combined use, but this is below the desired value of 1.00.hese results confirm that a government rebate needs to be in placeo make the rainwater tanks financially viable to home owners.

Table 4 provides the benefit cost ratios for Manly, Parramattand Richmond when the government rebate is included. For toiletnd laundry use benefit cost ratios are still below 1.00 in all thehree locations for all the three rainwater tank sizes. The highestenefit cost ratio for toilet and laundry use is found at Manly with.91 for a 2 kL rainwater tank. A 5 kL rainwater tank at Parramattand Richmond has a benefit cost ratio of 0.89 and 0.86, respectively.

Benefit cost ratios for irrigation use are relatively smaller at allocations due to the low government rebate for irrigation use. Theighest benefit cost ratio is 0.84 at Manly for a 5 kL tank. The highestatios at Parramatta and Richmond are 0.72 and 0.66, respectively,oth with a 5 kL rainwater tank. The benefit cost ratios for com-ined use are above 1 in all the locations for all the tank sizes withhe highest being 1.82 at Manly with a 5 kL tank. The benefit costatios in Parramatta and Richmond are 1.55 and 1.40, respectively,hen a 5 kL tank is used. In summary, the highest benefit cost ratiohen the current government rebate is considered is 1.82 for a 5 kL

ainwater tank at Manly for combined use. These results suggesthat a 5 kL tank is preferable to 2 kL and 3 kL tanks and rainwateranks should be connected to toilet, laundry and outdoor irrigationo achieve the best financial outcome for the home owners.

.4. Assessment of rainwater tank rebate

Results from Section 4.3 show that a government rebate iseeded to make a rainwater tank financially viable to home owners.

1.002 $3.9791.217 $3.9791.451 $3.979

Even when the current government rebate is included in the LCCA, abenefit cost ratio of 1 is unachievable for toilet and laundry use. Ben-efit cost ratios for irrigation is below 1 due to the low rebate offeredfor irrigation use. To determine the required rebate for achievinga benefit cost ratio of 1 for toilet and laundry, and irrigation usea sensitivity analysis is carried out. Richmond is selected for thispurpose as it has the lowest water savings. Table 5 shows that fortoilet and laundry use, the current rebate needs to be raised by24–38% and for irrigation use by 100–466% to make the rainwatertank financially viable to home owners.

4.5. Assessment of water price

The water price considered in this study is Aus$1.811/kL (2010Sydney Water price). A sensitivity analysis is conducted to findthe required increase in water price without increasing the cur-rent government rebate to achieve a benefit cost ratio closer to1.00. Once again, Richmond is selected for this purpose as it has thelowest water savings. Table 6 shows that an increase in the cur-rent water price from $1.811 to $2.514/kL (which represents a 39%increase) would give a benefit cost ratio of 1.001 for a 3 kL rainwa-ter tank for toilet and laundry use. An increase in the current waterprice from $1.811 to $3.979/kL (which represents a 120% increase)would give a benefit cost ratio of 1.002 for a 2 kL tank for irrigationuse.

5. Conclusions

This paper examines the performance of a rainwater har-vesting system (RWHS) in 10 different locations in GreaterSydney, Australia. Three different combinations of water use are

servat

cttacndflmtaitbttma

R

A

AD

E

A. Rahman et al. / Resources, Con

onsidered: (i) toilet and laundry (ii) irrigation and (iii) a combina-ion of toilet, laundry and irrigation (combined use). Three differentank sizes are considered: 2 kL, 3 kL and 5 kL. It is found that theverage annual water savings from rainwater tanks are stronglyorrelated with average annual rainfall. A rainwater tank in Syd-ey is expected to be empty and full with the highest frequencyuring July to September and January to March, respectively. It isound that the benefit cost ratios for a RWHS for all the 10 studyocations and three tank sizes are smaller than 1.00 without govern-

ent rebate. A benefit cost ratio greater than 1 is achievable withhe current government rebate for combined use only. To achieve

benefit cost ratio near 1.00 for toilet and laundry use, an increasen the current rebate by 24–38% is needed. It is noted that a 5 kLank is preferable to 2 kL and 3 kL tanks and rainwater tanks shoulde connected to toilet, laundry and outdoor irrigation to achievehe best financial outcome for the home owners. The results fromhis study suggest that government authorities in Sydney should

aintain or possibly increase the rebate for RWHS to enhance itscceptance.

eferences

ustralian/New Zealand Standard (AS/NZS). Life cycle costing – an application guide.AS/NZS 4536; 1999.

ustralian Bureau of Statistics (ABS). Population by age and sex. ABS; 2009.omenech L, Sauri DA. Comparative appraisal of the use of rainwater harvesting in

single and multi-family buildings of the metropolitan area of Barcalona (Spain):social experience, drinking water savings and economic costs. J Cleaner Produc-tion 2010;11:1–11.

roksuz E, Rahman A. Rainwater tanks in multi-unit buildings: a case study for threeAustralian cities. Resour Conserv Recycl 2010;54:1449–52.

ion and Recycling 61 (2012) 16– 21 21

Farreny R, Morales-Pinzon T, Guisasola A, Taya C, Rieradevall J, Gabarrell X. Roofselection for rainwater harvesting: quantity and quality assessments in Spain.Water Res 2011;45:3245–54.

Ghisi E, da Fonseca T, Rocha VL. Rainwater harvesting in petrol stations in Brasilia:potential for potable water savings and investment feasibility analysis. ResourConserv Recycl 2009;54:79–85.

Imteaz MA, Shanableh A, Rahman A, Ahsan A. Optimisation of rainwater tank designfrom large roofs: a case study in Melbourne, Australia. Resour Conserv Recycl2011;55:1022–9.

Khastagir A, Jayasuriya N. Optimal sizing of rain water tanks for domestic waterconservation. J Hydrol 2009;381:181–8.

Khastagir A, Jayasuriya N. Investment evaluation of rainwater tanks. Water ResourManage 2011;25:3769–84.

Kyoungjun K, Chulsang Y. Hydrological modeling and evaluation of rainwater har-vesting facilities: case study on several rainwater harvesting facilities in Korea.J Hydrol Eng 2009;14(6):545–61.

Su M, Lin C, Chang L, Kang J, Lin M. A probabilistic approach to rainwater harvestingsystems design and evaluation. Resour Conserv Recycl 2009;53:393–9.

Mun JS, Han MY. Design and operational parameters of a rooftop rainwaterharvesting system: definition, sensitivity and verification. J Environ Manage2012;93:147–53.

Muthukumaran S, Baskaran K, Sexton N. Quantification of potable water savings byresidential water conservation and reuse – a case study. Resour Conserv Recycl2011;55:945–52.

Rahman A, Dbais J, Imteaz MA. Sustainability of RWHSs in multistorey residentialbuildings. Am J Eng Appl Sci 2010;1(3):889–98.

Ryan A, Spash C, Measham TG. Socio-economic and psychological predictors ofdomestic greywater and rainwater collection: evidence from Australia. J Hydrol2009;379:164–71.

Sydney Water. Rainwater tank rebates; 2010, http://www.sydneywater.com.au/.Tam VWY, Tam L, Zeng SX. Cost effectiveness and trade off on the use of rainwa-

ter tank: an empirical study in Australian residential decision-making. ResourConserv Recycl 2010;54:178–86.

Zhang Y, Chen D, Chen L, Ashbolt S. Potential for rainwater use in high-rise buildingsin Australian cities. J Environ Manage 2009;91:222–6.

Zhang Y, Grant A, Sharma A, Chen D, Chen L. Alternative water resources forrural residential development in Western Australia. Water Resour Manage2010;24:25–36.