1 s2.0 S0921344915300756 Main PDF

G Model
ARTICLE IN PRESS
RECYCL-3096; No. of Pages 10
Resources, Conservation and Recycling xxx (2015) xxxxxx
Contents lists available at ScienceDirect
Resources, Conservation and Recycling

journal homepage: www.elsevier.com/locate/resconrec
Optimal sizing of storage tanks in domestic rainwater harvesting

systems: A linear programming approach
Chiemeka Onyeka Okoye a , Oguz Solyal b, , Bertug Akntug c
a
b
c
Sustainable Environment and Energy Systems, Middle East Technical University, Northern Cyprus Campus, Kalkanl, Mersin 10, Turkey
Business Administration Program, Middle East Technical University, Northern Cyprus Campus, Kalkanl, Mersin 10, Turkey
Civil Engineering Program, Middle East Technical University, Northern Cyprus Campus, Kalkanl, Mersin 10, Turkey
a r t i c l e
i n f o
Article history:
Received 6 July 2015
Received in revised form 25 August 2015
Accepted 27 August 2015
Available online xxx
Keywords:
Domestic rainwater harvesting
Sustainable water use
Tank storage
Linear programming
Cost optimization
a b s t r a c t
This paper proposes an optimization model to determine the optimal tank size of a single residential
housing unit for rainwater harvesting and storage. Taking into account the site specic data such as
the rainfall prole, the roof area of the building, the water consumption per capita and the number of
residents, an integrated optimization model based on linear programming is proposed to decide on the
size of rainwater storage tank to build such that the net present value of the total tank construction
costs and freshwater purchase costs is minimized. The proposed model was tested on a case study from
Northern Cyprus, the results of which emphasized the feasibility of rainwater harvesting as a sustainable
supplement to the depleting aquifers in the region. The study also offers managerial insights on the impact
of various parameters such as the number of residents, roof area, discount rate, water consumption per
capita, unit cost of building the rainwater tank, and rainfall characteristics on the optimal tank size and
on the net nancial benet gained from rainwater harvesting through detailed sensitivity analysis.
2015 Elsevier B.V. All rights reserved.
1. Introduction
The quest to curb the menace of water scarcity has motivated
considerable research interest in a wide range of applications aimed
at providing a sustainable solution to ensure water security in
both rural and urban areas. Desalination, greywater harvesting,
rainwater harvesting (RWH), and virtual water are some of these
notable applications with proven documented research results
(Bani-Melhem et al., 2015; Jiang et al., 2015; Morales-Pinzn et al.,
2015; Scarborough et al., 2015). Among these alternatives, RWH
systems have stood out and their application has gained wider
acceptance (Aladenola and Adeboye, 2010; Morales-Pinzn et al.,
2015; Silva et al., 2015; Unami et al., 2015) because these systems
are not only sustainable means of supplementing available water
resources to overcome the chronic water scarcity but also proactive ways of mitigating the menace of urban ood (Sample and Liu,
2014).
The domestic use of freshwater accounts for approximately 10%
of the total global freshwater consumption (Bocanegra-Martnez
et al., 2014). RWH has been widely applied for the domestic use
Corresponding author.
E-mail addresses: okoye.onyeka@metu.edu.tr (C.O. Okoye), solyali@metu.edu.tr
(O. Solyal), bertug@metu.edu.tr (B. Akntug).
under different climatic conditions (Domnech and Saur, 2011;

Hadadin et al., 2010; Silva et al., 2015; Ward et al., 2012). The
low-quality domestic use of rainwater includes but not limited
to toilet ushing, laundry, car washing, and irrigation (Villarreal
and Dixon, 2005), whereas the high-quality domestic use of harvested rainwater includes potable uses after some treatment.
Although the technology of RWH has been recommended for
areas with annual rainfall above 1000 mm (Aladenola and Adeboye,
2010), considerable research studies have been performed for the
areas characterized with low precipitation (Abdulla and Al-Shareef,
2009; Domnech and Saur, 2011; Hadadin et al., 2012).
Various models ranging from behavioral (Liaw and Tsai, 2005;
Palla et al., 2011) to probabilistic (Basinger et al., 2010; Kim et al.,
2012; Su et al., 2009) have been used in the literature for the rainwater harvesting practice. The assessment of suitability of some
models for domestic application was performed by Ward et al.
(2010). Campisano and Modica (2012) mentioned that the feasibility of RWH systems depends entirely on the characteristic of
the rainwater storage tank, water demand pattern of households,
rooftop effective area of the building, and rainfall prole of the site.
Similarly, Santos and Taveira-Pinto (2013) concluded that variation in rainfall prole has the most signicant effect on the optimal
tank size when they applied different criteria in the sizing of rainwater storage tanks. The mentioned characteristics not only affect
the water saving efciency but also the economy of the designed
http://dx.doi.org/10.1016/j.resconrec.2015.08.015
0921-3449/ 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Okoye, C.O., et al., Optimal sizing of storage tanks in domestic rainwater harvesting systems: A linear
programming approach. Resour Conserv Recy (2015), http://dx.doi.org/10.1016/j.resconrec.2015.08.015
G Model
ARTICLE IN PRESS
C.O. Okoye et al. / Resources, Conservation and Recycling xxx (2015) xxxxxx
Nomenclature
Acronyms
CV
coefcient of variation
IBR
increasing block rate
LP
linear programming
net nancial benet
NFB
RWH
rainwater harvesting
TDC
total discounted cost
Turkish lira
TL
Indices
j
t
price levels
periods of the year
Parameters
a
cost incurred per unit volume of rainwater tank built
Acol
area of the rooftop collector
cost per volume of purchasing water from the utility
btj
network in period t at the price level j
dimensionless runoff coefcient
cf
CostPFN total discounted cost of satisfying demand completely by purchasing water from the utility
network
CPt
cost of purchasing water from the utility network in
period t
dt
domestic household water demand in period t
fini
xed cost of installing the rainwater tank
discount rate
i
J
number of price levels
price level with the greatest unit price to be incurred
k
for a purchased volume of freshwater
number of residents
n
Nt
number of days in period t
rdt
measured rainfall depth in period t
amount of rainwater that can be harvested and
rt
stored in period t
smax
maximum size for the volume of rainwater tank that
can be built

length of the planning horizon
purchased volume of freshwater
V
Vj
maximum cumulative volume of freshwater that
can be purchased at the jth price level
Wd
volume of water usage per day per capita
Variables
It
inventory level of the rainwater tank at the end of
period t
Ptj
amount of water purchased from the utility network
at the jth price level in period t
Rt
amount of rainfall stored by the rainwater tank in
period t
Tcap
volume of the rainwater tank to build
amount of water used from the rainwater tank to
Ut
satisfy demand in period t
objective function value
Z
storage tank. Often times, the economic potential of RWH exists due
to avoiding freshwater purchase but the overall feasibility of integrating a rainwater storage unit may still be infeasible due to initial
capital cost of installation (Kim et al., 2014). For this reason, most
governments are providing rebates in the form of exemption from
stormwater taxes or offset in the initial capital cost of installation
to encourage the deployment of the RWH systems (Domnech and
Saur, 2011; Imteaz et al., 2012; Rahman et al., 2012). Domnech

and Saur (2011) mentioned that subsidies up to 1200 D are granted
to a household installing a RWH system in Barcelona, Spain. Similarly, the Victoria Government and Sydney Water Corporation offer
up to Aus $500 and Aus $1400, respectively, as rebates to properties that have rainwater tanks installed in Australia (Imteaz et al.,
2012; Rahman et al., 2012).
Coombes and Barry (2008) compared the relative efciencies of
runoff into dams with rooftop RWH using duration curves developed for supplying water to the cities of Brisbane, Melbourne,
Perth, and Sydney. They concluded that RWH is more resilient
to the impacts of climate change. Ghisi (2010) considered the
parameters affecting the sizing of rainwater tanks for domestic use
and recommended that regional assessment of rainwater tank sizing be carried out by taking into account local rainfall data, roof
areas, number of residents, potable water demand, and rainwater demand. Tam et al. (2010) compared the cost of procurement,
installation and operation of rainwater tanks to the benets of
the use of a rainwater tank in an empirical study to aid residential decision-making. Domnech and Saur (2011) assessed the
social experience, freshwater savings, and economic costs associated with the use of RWH in single and multi-family buildings in
Spain. Imteaz et al. (2011) presented a daily water balance model
for domestic rainwater usage so as to provide decision support for
the performance analysis of rainwater tanks in commercial buildings with large roof area. The authors claimed that optimal tank
size was obtained by studying the effect of varying parameters of
tank size and roof areas on cumulative overow loss and cumulative water saved. Khastagir and Jayasuriya (2010) used multivariate
regression between domestic rainwater tank capacities and roof
catchment area to develop a dimensionless curve for assessing
water supply effectiveness. They considered the developed dimensionless curve as a step toward developing a web-based interactive
tool for optimum tank selection. Similarly, Campisano and Modica
(2012) developed a regression model which enables the evaluation
of water saving and overow discharge from domestic RWH systems. They evaluated the optimal tank size by applying a minimum
cost approach on the developed regression model and concluded
that the economic attractiveness of large tanks decreases as rainwater availability decreases. Morales-Pinzn et al. (2015) proposed
a predictive model for estimating the nancial and environmental
feasibility of RWH for different housing congurations in Spain.
Imteaz et al. (2012) assessed the rainwater harvesting potential for
southwest Nigeria using a daily water balance model. They found
that the analysis using monthly rainfall data tends to overestimate the required rainwater tank size and recommended the use
of daily data. L et al. (2013) presented a multi-criteria optimization approach for rainwater utilization, which was evaluated using
a case study in Shanghai, China. They concluded that the rainwater utilization could enhance the sustainability of cities with the
involvement of stakeholders preferences. Al-Ansari et al. (2013)
proposed a combination of linear programming model together
with a watershed modeling system to maximize the irrigation area,
which could be supplied from a selected reservoir. Huang et al.
(2013) proposed a stochastic optimization approach for the integrated urban water resource planning with the aim of optimizing
water ows in cities facing signicant water shortage. Sample and
Liu (2014) proposed a nonlinear metaheuristic search algorithm for
the identication of near-optimal least cost solutions for the dual
purpose of water supply and runoff capture across a wide range of
land uses and locations in Virginia, USA. They concluded that the
net benets are very sensitive to water and wastewater charges.
Gurung and Sharma (2014) presented the economies of scale on
communal rainwater tank system design. Bocanegra-Martnez et al.
(2014) proposed a nonlinear mixed integer programming model
to harvest, store, and distribute rainwater for multiple residential
G Model
ARTICLE IN PRESS
housing units. Their optimization model minimizes the total annual

cost and freshwater consumption. Their model was developed for
communal harvesting involving multiple housing units to decide
on the size of the storage tanks, the housing units to be used for
harvesting, the connections between housing units and the storage tanks, and the size of the elevated reservoir which stores both
freshwater and rainwater after treatment. The sizes of both the
storage tanks and the elevated reservoir are calculated using the
largest volume of rainwater stored over the planning horizon of 12
months.
In this study, we consider a single residential housing unit connected to the utility network and a decision is made on either
building a rainwater storage tank to satisfy some portion of the
water demand or meeting whole demand from the utility network. Building a rainwater tank incurs an initial capital installation
cost while purchasing water from the utility network incurs a cost
dependent on the amount purchased. The aim is to assess the economic feasibility of building a rainwater tank compared to the
alternative of meeting whole demand from the utility network by
taking into account site specic data including the associated costs.
Although existing studies in the literature offered some solutions to their respective case applications, there is still a need to
have an integrated approach to the domestic RWH problem considering tank costs and freshwater purchase costs. Therefore, we
propose for the rst time an integrated optimization model based
on linear programming (LP) to determine the optimal rainwater
tank size for the domestic rainwater harvesting and storage at a single residential housing unit by taking into account site specic data
such as the rainfall prole, the roof area of the building, the water
consumption per capita, the number of residents, the initial capital
cost of building a rainwater tank, and the cost of purchasing water
from the utility network. Our paper is closely related to BocanegraMartnez et al. (2014), but unlike Bocanegra-Martnez et al. (2014),
our model is for a single residential housing unit, involves decisions
on the amount of rainfall to store (or equivalently the overow
decisions), allows separate storage of rainwater and freshwater,
considers volume-dependent increasing unit prices for freshwater
purchased from the utility network, and is able to directly take into
account rainfall and demand data along the useful life of a rainwater
tank (i.e. over 20 years).
An important advantage of our LP model is that it can easily
be constructed and optimally solved within milliseconds because
of the availability of efcient commercial and non-commercial LP
solvers. In particular, our model has been coded in MS Excel and
solved using the open source solver, OpenSolver 2.6.1 (see Mason,
2012).
The proposed model was tested and validated on a case study
from Northern Cyprus. Finally, the study offers managerial insight
on the impact of various parameters such as the number of residents, roof area, discount rate, water consumption per capita, unit
cost of building the rainwater tank, and rainfall characteristics on
the optimal tank size and on the net nancial benet gained from
RWH through detailed sensitivity analysis.
The rest of the paper is as follows. The detailed description of the
problem addressed is provided in Section 2. Section 3 presents the
proposed LP model for domestic RWH and storage. The implementation of the proposed model to a case study is provided in Section
4. Finally, Section 5 concludes the paper.
Fig. 1. Increasing block rate tariff scheme.
available space for rainwater tank. The aim is to determine the

optimal size of rainwater tank to build at the minimum total cost
which is composed of the capital cost of installing the tank and the
freshwater purchase cost.
The problem considered can be described in different time scales
such as hourly, daily and monthly time periods depending on the
available input data (e.g., water demand and rainfall) resolution.
Although Imteaz et al. (2012) recommended the use of daily data
for more realistic results, due to the scarcity of daily rainfall data,
we consider monthly time periods and use month and period words
interchangeably. We dene as the length of the planning horizon
in terms of the number of months.
The capital cost of installing the tank is composed of a xed
cost and a variable cost. The xed cost of installing the tank fini
includes the costs of tank, pump, pipe, pressure control, lter, and
installation while the variable cost of installing the tank, a, is the
cost incurred per unit volume of tank built. The cost of rainwater
pumping is neglected as this cost is also incurred when purchasing
water from the utility network.
An increasing block rate (IBR) tariff scheme is used to calculate
the freshwater purchase cost. According to the IBR tariff scheme,
unit prices for water purchased from the utility network increases
as the purchased volume increases. The IBR tariff scheme is widely
applied in many countries such as Spain (Surez-Varela et al., 2015),
Portugal (Silva et al., 2015) and USA (Boyer et al., 2012), as an effective tool to prevent the wastage of the scarce water resources. In
IBR tariff scheme, btj is the cost per volume of purchasing water
from the utility network in period t at the price level j with J denoting the number of price levels, and Vj is the maximum cumulative
volume of freshwater that can be purchased at the jth price level.
Note that Vj Vj1 denotes the maximum volume of freshwater
that can be purchased at the jth price level. As unit prices increase
with the increasing purchase volume in IBR tariff scheme, we have
bt1 < bt2 < . . . < btJ for any period t. Dening V as the purchased volume of freshwater in any period t and k as the price level with the
greatest unit price to be incurred for V , the cost of purchasing water
from the utility network in period t, CPt , can be calculated as
CPt =
2. Problem formulation
In this section, the problem addressed in this study is presented.
A single residential housing unit is considered in a given location
with specied meteorological climatic variables, number of residents, daily water consumption per capita, available roof area, and
k1
j=1
btj (Vj Vj1 ) + btk (V Vk1 )
(1)
where V0 = 0 by denition and k = argmax0jJ {j|V > Vj } +1. An

example IBR diagram for freshwater consumption is presented in Fig. 1, where CP1 = bt1 V1 , CP2 = CP1 + bt2 (V2 V1 ),
CP3 = CP2 + bt3 (V3 V2 ), and CP4 = CP3 + bt4 (V4 V3 ). As an illustration of Fig. 1, we consider a household with a water purchase
G Model
ARTICLE IN PRESS
the tank to satisfy demand in the same period. At the beginning of

the planning horizon (i.e., t = 0), the initial water level is equal to
zero since the tank is newly installed and empty.
3. Optimization model
We present the proposed optimization model in this section. In
addition to the notation dened in Section 2, we dene the following decision variables:
Tcap : volume of the rainwater tank to build.
Ptj : amount of water purchased from the utility network at the
jth price level in period t.
It : inventory level of the rainwater tank at the end of period t.
Rt : amount of rainfall stored by the rainwater tank in period t.
Ut : amount of water used from the rainwater tank to satisfy
demand in period t.
We propose the following LP model to determine the optimal
size of the rainwater tank:
Find the optimal decision variables minimizing the objective
function
Fig. 2. Schematic of a typical domestic rainwater harvesting system.
V
< V
volume of
in period t. Assuming that V2
< V3 , the cost of
purchasing water from the utility network in period t is equal to
bt1 V1 + bt2 (V2 V1 ) + bt3 (V V2 ), as can be seen from Fig. 1.
Because some costs are incurred at the beginning of the planning
horizon while some others are incurred throughout the planning
horizon, they need to be in the same monetary value, and thus we
use the monthly discount rate i to calculate the present value of all
costs.
The amount of rainwater that can be harvested and stored rt
over a given roof area at any period t is a function of dimensionless
runoff coefcient and rainfall depth, hence can be estimated from
the expression presented in Aladenola and Adeboye (2010) as;
rt =
c f Acol rdt
1000
(2)
where cf is the dimensionless runoff coefcient, Acol the area of the

rooftop collector in m2 , and rdt is the measured rainfall depth in
mm at period t.
Monthly domestic household water demand dt is the total
amount of non-potable water needed, thus can be expressed as,
dt = Wd n Nt
(3)
where Wd is the water usage in m3 /day/capita, n is the number

of residents and Nt is the number of days in month t. Although the
harvested rainwater in this present analysis is considered for indoor
and outdoor non-potable uses like toilet ushing, laundry, and car
washing, it can also be considered for potable uses by including the
associated water treatment cost.
There is a maximum size smax for the volume of tank that can
be built for a residential housing unit. This maximum size limit
arises due to the available space for the rainwater tank, the desire
to preserve the esthetic view of the building, and the obligation for
the already designed structure to carry the weight of the storage
unit.
Ideally, a typical domestic RWH system consists of a rooftop, a
gutter system, and a storage unit as depicted in Fig. 2. In this system,
rooftop is used as a catchment area to harvest the rainwater which
is then collected using the gutter system and stored in a rainwater
tank. The water level of the rainwater tank at the end of a time
period is given as the sum of the water level in the tank at the end
of the preceding time period and the amount of rainwater stored
by the tank in that period minus the amount of water used from
Z = aTcap +
J
btj
j=1 (1 + i)t
t=1
Ptj
(4)
Subject to
It = It1 + Rt Ut
1 t
1 t
Rt rt
(5)
(6)
It Tcap
1 t
(7)
Rt Tcap
1 t
(8)
Tcap smax
Ut +
J
P
j=1 tj
(9)
= dt
Ptj Vj Vj1
1 t
1 t , 1 j J
I0 = 0
It , Rt , Ut , Ptj , Tcap 0
(10)
(11)
(12)
1 t , 1 j J.
(13)
The objective function (4) is the sum of variable cost of building a rainwater tank and the total discounted cost of purchasing
water from the utility network over the planning horizon. Eq. (5) is
the inventory balance equation for the rainwater tank stating that
the amount of rainwater kept in the rainwater tank at the end of a
period is equal to the amount kept from the previous period plus the
amount of rainfall allowed to ow into the tank minus the amount
used from tank to satisfy the demand in that period. Constraints (6)
ensure that the amount of rainfall allowed to ow into the rainwater tank in a period cannot be greater than the amount of rainfall in
that period. Constraints (7) stipulate that the amount of rainwater
kept in the rainwater tank at the end of a period cannot exceed the
size of the rainwater tank. Constraints (8) ensure that the amount
of rainfall allowed to ow into the rainwater tank in a period cannot
exceed the size of the rainwater tank. Constraints (8) are especially
needed when time periods are months because a heavy rainfall may
occur at the beginning of a month whereas water demand occurs
uniformly over the month. Thus, one needs to have a sufciently
large tank size that can store the rainfall. Constraint (9) states that
the size of the rainwater tank cannot exceed the maximum size that
can be built in a given site. Constraints (10) ensure that the demand
in a period can be satised from the rainwater tank and/or by purchasing water from the utility network. Constraints (11) stipulate
that the amount of water purchased from the utility network at the
jth price level in any period cannot be greater than the maximum
G Model
ARTICLE IN PRESS
CostPFN =
CPt /(1 + i) . If fini + Z* < CostPFN , then building a rain-
t
is economically feasible, otherwise not. Note

water tank of size Tcap
= 0, it implies Z* = Cost
that if Tcap
PFN .
We dene the net nancial benet (NFB) as the difference
between the discounted cost of fully satisfying the demand of
the residential housing unit from the utility network and the discounted cost of implementing a RWH system. In other words,
> 0 and = 0 otherwise.
NFB = CostPFN (fini + Z*) where = 1 if Tcap
Accordingly, positive NFB implies that the designed RWH system
has a marginal economic return in its lifecycle time. Otherwise, NFB
is equal to zero which implies that RWH is not feasible; thus, a rainwater tank should not be built and the residential water demand
should fully be satised from the utility network.
4. Case study
The proposed LP model has been tested on a case study in Northern Cyprus using 31 years monthly average rainfall data collected
from the local meteorological stations. Northern Cyprus, a typical
semi-arid country characterized by high seasonal and variable rainfall, suffers from severe water scarcity. The average annual rainfall
measured over 19 rain-gauge stations in the country from 1978 to
2009 is 361 mm. As expected of a typical island, the only source
of water is groundwater (aquifers) besides the short-lived surface water that remains after a rainstorm. The aquifers are being
depleted due to overdraw to meet the ever increasing demand. In
recent years, the country has witnessed several years of drought
and occasional heavy rainfall.
We have applied the proposed model to the city of Kyrenia in Northern Cyprus, which has an annual average rainfall of
464 mm. The average monthly rainfall prole of Kyrenia is shown
in Fig. 3. The mean and standard deviation of monthly rainfall in
Kyrenia are 38.7 mm and 38.8 mm, respectively. The coefcient of
variation of monthly rainfall for Kyrenia is 100%, depicting high
inter-variability.
As mentioned by Boyer et al. (2012), freshwater use is sensitive to the changes in prices and the price structures. For this
120
Average Monthly Rainfall (mm)
100
80
60
40
20
Aug.
Jul.
Jun.
May
Apr.
Mar.
Feb.
Jan.
Dec.
Nov.
Oct.
0
Sept.
amount allowed at that price level. Note that because of the convexity of the btj values (i.e., bt1 < bt2 < . . . < btJ for any period t), it is
always cheaper to purchase at the price level j 1 than the price
level j. (4) and (11) ensure that the amount purchased at the price
level j can only be positive if the total amount of freshwater purchased exceeds Vj1 , and thus the LP model correctly formulates
the total cost of purchasing water from the utility network according to the IBR tariff scheme. Because the LP model is proposed as
a tool to determine the rainwater tank size by considering its economic feasibility, there is no rainwater in the tank at time zero, as
stated by Eq. (12). Finally, constraints (13) ensure that all variables
are nonnegative.
Because the objective function and constraints are linear and
all decision variables are continuous, the proposed model is an LP
model, which can be solved quite efciently even for a large value.
In particular, we formulate the LP model in MS Excel and solve it
to optimality using the OpenSolver 2.6.1. The computation time
required to solve the LP model is very minimal showing the easiness
of solving the proposed model.
After solving the LP model to optimality, we nd the optimal
and the optimal objective function value
rainwater tank size Tcap
> 0, we add the xed installation cost f
Z*. If Tcap
ini to the optimal objective function value and determine the total discounted
cost of building a rainwater tank as fini + Z*. In order to check the
feasibility of building a rainwater tank, we need to calculate the
total discounted cost of satisfying demand completely by purchasing water
from the utility network, denoted by CostPFN . Note that
Fig. 3. Monthly average rainfall of Kyrenia (19782009).
reason, many residential households practice freshwater conservation in the presence of IBR tariff scheme. The IBR tariff scheme is
widely adopted and practiced by different municipalities in Northern Cyprus for freshwater prices in order to minimize water usage
from the aquifers. The IBR tariff scheme used in the Kyrenia region
is shown in Table 1. The cost is presented in Turkish Lira (TL) which
is the local currency used in the country. According to Table 1, J = 4,
V1 = 10, V2 = 30, V3 = 50, V4 =, bt1 = 2, bt2 = 2.5, bt3 = 6, and bt4 = 8.
Generally, the cost of freshwater in Northern Cyprus even with
the IBR tariff scheme is relatively low compared to other countries
and as such affects rainwater harvesting decision adversely. Several
municipalities in Northern Cyprus indirectly encourage building
RWH systems by charging a wastewater cost per volume of water
purchased from the utility network. Note that residents can avoid
this cost by building a RWH system. For the cost of wastewater, the
Kyrenia municipality charges 1.4 TL/m3 of water purchased from
the utility network.
Predominately, two different housing congurations namely
villas and apartments are found in the Kyrenia region, with each
having different roof areas and number of residents. We consider
a standard villa of 200 m2 roof area with 5 residents (i.e., Acol = 200
and n = 5). The xed capital cost and the unit cost of building a rainwater tank are estimated using the values obtained from a local
supplier as 3162.5 TL (i.e., fini = 3162.5) and 216 TL/m3 (i.e., a = 216),
respectively. The average daily water consumption per capita in the
country is 0.125 m3 (i.e., Wd = 0.125), and the corresponding runoff
coefcient is 90% (i.e., cf = 0.90) for pitched roof (Leggett et al., 2001).
The maximum size of rainwater tank is 20 m3 (i.e., smax = 20) so as
not to affect the esthetic view of the building. As in Okoye and
Atikol (2014), the annual discount rate is taken as 6% (i.e., i = 0.005
per month). The lifetime of the rainwater tank is taken as 25 years
(i.e., = 300).
When the LP model with the above mentioned data is solved
to optimality, the optimal objective function value of 8974.1 TL
Table 1
Monthly IBR tariff scheme in Kyrenia.
Water consumption (m3 /month)
Cost (TL/m3 )
010
1130
3150
>50
2.0
2.5
6.0
8.0
G Model
ARTICLE IN PRESS
6
3200
30000
11
27000
10
NFB
TDC
2800
9
24000
18000
1600
15000
1200
12000
800
2000
Tank Size (m )
8
21000
Total Discounted Cost (TL)
Net Financial Benefit (TL)
2400
6
5
4
3
9000
12
13
11
10
13
12
11
Number of Residents
10
3000
0
2
6000
2
400
Number of Residents
Fig. 4. Effect of varying the number of residents on the tank size and economic benet.
with an optimal rainwater tank size of 2.2 m3 is found. The net

nancial benet (NFB) is equal to 675 TL depicting a marginal
economic return over the planning horizon for building the RWH
system. Considering the occasional ooding situation occurring in
the country in recent times, there can be an important environmental benet associated with RWH implementation besides its
nancial benet. Considering this environmental benet, even if
there is no NFB (i.e., NFB = 0) of building of a RWH in a region, it
can be appropriate for the governments to offset some portion of
the cost of the system as widely observed in other regions such as
Barcelona in Spain and Victoria and Sydney in Australia (Domnech
and Saur, 2011; Imteaz et al., 2012; Rahman et al., 2012).
In order to observe the impact of varying the parameters, we
identied and evaluated different scenarios which are marked as
Case A, B, C, D, E and F in the following. In all cases, the effect of varying one parameter at a time on the NFB, the total discounted cost
(TDC) of satisfying water demand (i.e., TDC = min {CostPFN , fini + Z*})
and the rainwater tank size is evaluated while all other parameters remains the same as previously dened. The effect of varying
the number of residents from 2 to 13 in Case A, the roof area from
780
80 to 300 m2 in Case B, the discount rate from 3% to 12% in Case

C, the average daily water consumption per capita from 80 to 195
Liters in Case D, the unit cost of building rainwater tank from 156
to 336 TL/m3 in Case E and nally, several rainfall prole characteristics in Case F were examined and critically analyzed.
4.1. Case A
The impact of varying the number of residents from 2 to 13 is
examined and the results are presented in Fig. 4, which indicates
the optimal tank size, the resulting NFB and TDC. The results show
that when the number of residents is less than 5 or greater than 12,
building a rainwater tank is not nancially feasible. In other words,
the RWH system would not be able to recover the capital investment over its useful lifetime. Thus, in the presence of less than 5 or
greater than 12 residents, it is cheaper to use water from the utility network in catering for the demand. The NFB is positive for ve
persons, increases steadily to its maximum at nine residents, and
then decreases gradually as the number of residents increases as
shown in Fig. 4. The corresponding optimal tank size is 4.9 m3 for
3.5
9080
NFB
TDC
760
9060
740
9040
720
9020
700
9000
680
8980
660
8960
640
8940
620
8920
600
8900
580
8880
3.0
Tank Size (m )
2.5
2.0
1.5
1.0
Roof Area (m )
300
280
260
240
220
200
180
160
140
120
100
0.0
80
300
280
260
240
220
200
180
160
140
120
100
80
0.5
Roof Area (m )
Fig. 5. Effect of varying the roof area on the tank size and economic benet.
G Model
ARTICLE IN PRESS
2800
13000
2.5
NFB
TDC
12000
2400
2.0
11000
9000
1200
8000
800
7000
Tank Size (m )
10000
1600
2000
1.5
1.0
0.5
400
6000
Consumption per cubic meter per capita
0.20
0.19
0.18
0.17
0.16
0.15
0.14
0.13
0.12
0.11
0.10
0.08
0.20
0.19
0.18
0.17
0.16
0.15
0.14
0.13
0.12
0.11
0.10
0.09
0.09
0.0
5000
0.08
Consumption per cubic meter per capita
Fig. 6. Effect of varying daily water consumption on the tank size and economic benet.
the predicted maximum NFB of 2983 TL. Note that more rainwater and/or freshwater is needed to satisfy increasing water demand
due to a larger number of residents and the way to increase the supply of rainwater is to build a larger tank size. We indeed observe
a larger rainwater tank size in Fig. 4 as the number of residents
increases (i.e. 2.2 m3 for 58 residents, 4.9 m3 for 9 residents and
10.7 m3 for 1012 residents) until the latter becomes 13. However,
the rainfall may not ll up the larger tank sizes sufciently and one
may still need to purchase freshwater to satisfy the water demand.
This is what happens when the number of residents exceeds 12
and it becomes nancially better to satisfy all water demand by
purchasing freshwater than by building a large tank size incurring
a high capital installation cost besides the cost of purchasing freshwater. We also observe that the TDC increases exponentially as the
number of residents increases as depicted in Fig. 4.
other words, increasing the roof area leads to increase in both the
tank size and the NFB, and decrease in the TDC.
4.3. Case C
The effect of changing the daily water consumption level
between 80 and 195 l per capita is presented in Fig. 6, which indicates the optimal tank size, the NFB, and the TDC. As can be observed
from Fig. 6, the optimal tank size predicted by the model is not
sensitive to the daily consumption level. When the daily water consumption per capita is greater than 0.10 m3 , the optimal tank size
to build stays constant at the size of 2.2 m3 whereas no rainwater
tank is recommended to build when the daily water consumption
per capita is less than 0.10 m3 . On the other hand, the NFB of implementing a RWH system enhances with an increase in the daily
water consumption per capita as expected.
4.2. Case B
4.4. Case D
For the considered roof areas from 80 to 300 m2 , the results are
presented in Fig. 5, which shows that RWH roof area has a linear
relationship with the optimal tank size, the NFB, and the TDC. In
2800
The results due to varying the discount rate between 3% and 13%
are presented in Fig. 7. The results in Fig. 7 reveal that the optimal
11500
4.5
NFB
TDC
2400
11000
4.0
10500
3.5
10000
3.0
8500
1200
8000
7500
800
7000
9000
1600
Tank Size (m )
9500
2000
2.5
2.0
1.5
1.0
6500
400
0.5
6000
Discount Rate (%)
13
12
11
10
0.0
2
13
12
11
10
5500
Discount Rate (%)
Fig. 7. Effect of varying daily water consumption on the tank size and economic benet.
G Model
ARTICLE IN PRESS
8
900
9150
4.5
9100
4.0
NFB
TDC
850
3.5
9050
800
3.0
3
340
320
0.0
300
8750
280
0.5
340
320
300
280
260
240
220
200
180
160
550
8800
160
600
1.0
260
8850
1.5
240
650
220
8900
2.0
200
700
2.5
180
8950
Tank Size (m )
750
9000
Unit Cost of Tank (TL/m )
Unit Cost of Tank (TL/m )
Fig. 8. Effect of varying the unit cost of building the tank on the tank size and economic benet.
4.5. Case E
140
Kantara
Karpaz
Kyrenia
Gzelyurt
Dortyol
Nicosia
Iskele
120
Average Monthly Rainfall (mm)
tank size to build, the NFB, and the TDC decrease as the discount
rate increases. In particular, when the discount rate is greater than
8%, the proposed model does not recommend building the RWH
system (i.e. tank size is equal to zero). This result is due to the fact
that the net present value of the cost of purchasing water from the
utility network gets smaller as the discount rate increases whereas
the capital cost of building the RWH system stays constant. Thus,
high discount rates make the option of purchasing freshwater to
satisfy all water demand nancially more attractive than building
a RWH system.
100
80
60
40
20
4.6. Case F
In this case, rainfall data of seven rainfall stations in Northern
Cyprus are utilized in the sensitivity analysis. The stations were
selected in a way not only to represent all the regions in the country but also to reect the differences in the statistical coefcient
of variation (CV) values. The rainfall station location and regions
are as follows: Karpaz for the Karpaz Peninsula with a CV of 1.01,
Kyrenia for the North Coast with a CV of 1.00, Iskele for the East
coast with a CV of 0.90, Guzelyurt for the West Mesaoria Plain with
a CV of 0.89, Kantara for the Besparmak Mountain with a CV of
0.86, Dortyol for the East Mesaoria Plain with a CV of 0.82, and
Nicosia for the Middle Mesaoria Plain with a CV of 0.74, respectively.
Aug.
Jul.
Jun.
May
Apr.
Mar.
Feb.
Jan.
Dec.
Nov.
Oct.
0
Sept.
Changing the unit cost of building a rainwater tank has a

signicant impact not only on the economic benet associated
with implementing the RWH system but also on the optimal
tank size as presented in Fig. 8. As expected, the NFB and the
tank size decrease with the corresponding increase in the unit
cost of building the tank. In Fig. 8, it is observed that there is a
threshold value of this cost after which it is not recommended to
build a rainwater tank. Specically, when the unit cost of building the tank exceeds 336 TL/m3 , the model recommended not
building a tank. On the other hand, the TDC increases with a
gradually decreasing rate as the unit cost of building the tank
increases.
Fig. 9. Average monthly rainfall distribution of the selected stations.
The monthly distribution of the selected rainfall stations is depicted

in Fig. 9, which indicates that the maximum rainfall occurs in
December while the minimum occurs in July and August. In Fig. 10,
the optimal tank size and the NFB associated with seven locations
are presented. It is observed that contrary to the assumption that
increase in rainfall will lead to an increase in the optimal tank size
and the NFB, the observed result shows that optimal tank size and
the NFB is actually more sensitive to the distribution of rainfall
over the months than the annual average rainfall. For example, the
optimal tank size for Dortyol, which has an annual average rainfall depth of 268 mm, is 2.8 m3 relative to the sizes of 2.2 m3 and
1.5 m3 predicted for Kyrenia and Guzelyurt with average annual
rainfall of 464 mm and 281 mm, respectively. Although Dortyol has
the lowest annual average rainfall, its NFB is more than that of Guzelyurt and Karpaz which have better rainfall amounts on average.
The results in Fig. 10 reveal that it is wrong to make assumptions
on the nancial benet of RWH systems based on average rainfall data of an area without considering variability in the rainfall
amounts.
G Model
ARTICLE IN PRESS
1.2
9040
4.5
CV
NFB
9000
720
8980
700
8960
680
8940
660
Coefficient of Variation
Tank Size
9020
TDC
740
1.1
4.0
1.0
3.5
0.9
3.0
0.8
2.5
0.7
2.0
0.6
1.5
0.5
1.0
Tank Size (m3 )
760
Dortyol
Guzelyurt
Nicosia
Iskele
Dortyol
Guzelyurt
Nicosia
Iskele
Kyrenia
Karpaz
8880
Kantara
620
Kyrenia
8900
Karpaz
640
Kantara
8920
Fig. 10. Effect of varying rainfall proles on the optimal tank size and economic benet.
5. Conclusion
In this paper, a mathematical model based on linear programming has been proposed and used in the optimal sizing of rainwater
storage tank for domestic rainwater harvesting and storage. The
proposed optimization model determines the optimal size of the
rainwater tank to build at minimum total discounted cost. The
model was applied to a case study from Northern Cyprus, which
showed through sensitivity analysis how some parameters affect
both the net nancial benet and the optimal rainwater tank size.
The sensitivity analysis reveals that the optimal tank size increases
with the roof area, but decreases with an increase in the discount
rate and the unit cost of building the tank. On the other hand, the
net nancial benet associated with implementing rainwater harvesting increases with an increase in the roof area and the daily
water consumption per capita but decreases with an increase in
the discount rate and the unit cost of building the tank. Until (resp.
after) a threshold value, an increase in the number of residents
leads to an increase (resp. decrease) in the optimal tank size and
the net nancial benet. We have also found that the daily consumption level per capita has no effect on the optimal tank size
whereas the monthly distribution of rainfall signicantly affects
both the optimal tank size and the net nancial benet.
As discussed in Section 2, we considered monthly time periods
in our LP model due to the scarcity of daily rainfall data. However,
the use of daily rainfall data leads to more realistic results than the
use of monthly data because consecutive heavy rainfalls in a month
may cause overowing of the rainwater tank several times and this
can only be captured by a daily analysis. Therefore, if daily rainfall
data is present, it is better to perform an analysis with daily data
which can be done by adapting the proposed LP model to the use
of daily rainfall and demand data.
Last but not least, as shown by the case study application, the
proposed LP model is an effective tool that can be used by public authorities or individuals to make feasibility analysis of RWH
systems at residential housing units.
References
Abdulla, F.A., Al-Shareef, A., 2009. Roof rainwater harvesting systems for
household water supply in Jordan. Desalination 243, 195207, http://dx.doi.
org/10.1016/j.desal.2008.05.013.
Aladenola, O.O., Adeboye, O.B., 2010. Assessing the potential for rainwater
harvesting. Water Resour. Manag. 24, 21292137, http://dx.doi.org/10.1007/
s11269-009-9542-y.
Al-Ansari, N., Ezz-Aldeen, M., Knutsson, S., Zakaria, S., 2013. Water harvesting and
reservoir optimization in selected areas of south sinjar mountain, Iraq. J.
Hydrol. Eng., http://dx.doi.org/10.1061/(ASCE)HE.1943-5584.0000712
(120903004647009).
Bani-Melhem, K., Al-Qodah, Z., Al-Shannag, M., Qasaimeh, A., Rasool Qtaishat, M.,
Alkasrawi, M., 2015. On the performance of real grey water treatment using a
submerged membrane bioreactor system. J. Memb. Sci. 476, 4049, http://dx.
doi.org/10.1016/j.memsci.2014.11.010.
Basinger, M., Montalto, F., Lall, U., 2010. A rainwater harvesting system reliability
model based on nonparametric stochastic rainfall generator. J. Hydrol. 392,
105118, http://dx.doi.org/10.1016/j.jhydrol.2010.07.039.
Bocanegra-Martnez, A., Ponce-Ortega, J.M., Npoles-Rivera, F., Serna-Gonzlez, M.,
Castro-Montoya, A.J., El-Halwagi, M.M., 2014. Optimal design of rainwater
collecting systems for domestic use into a residential development. Resour.
Conserv. Recycl. 84, 4456, http://dx.doi.org/10.1016/j.resconrec.2014.01.
001.
Boyer, C.N., Adams, D.C., Borisova, T., Clark, C.D., 2012. Factors driving water utility
rate structure choice: evidence from four southern US States. Water Resour.
Manag. 26, 27472760, http://dx.doi.org/10.1007/s11269-012-0043-z.
Campisano, A., Modica, C., 2012. Optimal sizing of storage tanks for domestic
rainwater harvesting in Sicily. Resour. Conserv. Recycl. 63, 916, http://dx.doi.
org/10.1016/j.resconrec.2012.03.007.
Coombes, P.J., Barry, M.E., 2008. The relative efciency of water supply catchment
and rainwater tanks in cities subject to variable climate potential for climate
change. Aust. J. Water Resour. 12, 85100.
Domnech, L., Saur, D., 2011. A comparative appraisal of the use of rainwater
harvesting in single and multi-family buildings of the Metropolitan Area of
Barcelona (Spain): social experience, drinking water savings and economic
costs. J. Clean. Prod. 19, 598608, http://dx.doi.org/10.1016/j.jclepro.2010.11.
010.
Ghisi, E., 2010. Parameters inuencing the sizing of rainwater tanks for use in
houses. Water Resour. Manag. 24, 23812403, http://dx.doi.org/10.1007/
s11269-009-9557-4.
Gurung, T.R., Sharma, A., 2014. Communal rainwater tank systems design and
economies of scale. J. Clean. Prod. 67, 2636, http://dx.doi.org/10.1016/j.
jclepro.2013.12.020.
Hadadin, N., Qaqish, M., Akawwi, E., Bdour, A., 2010. Water shortage in Jordan
sustainable solutions. Desalination 250, 197202, http://dx.doi.org/10.1016/j.
desal.2009.01.026.
Hadadin, N., Shawash, S., Tarawneh, Z., Banihani, Q., Hamdi, M.R., 2012. Spatial
hydrological analysis for water harvesting potential using ArcGIS model: the
case of the north-eastern desert, Jordan. Water Policy 14, 524, http://dx.doi.
org/10.2166/wp.2011.058.
Huang, Y., Cheng, J., Zeng, S., Sun, F., Dong, X., 2013. A stochastic optimization
approach for integrated urban water resource planning. Water Sci. Technol. 67,
16341641, http://dx.doi.org/10.2166/wst.2013.036.
Imteaz, M.A., Adeboye, O.B., Rayburg, S., Shanableh, A., 2012. Rainwater harvesting
potential for southwest Nigeria using daily water balance model. Resour.
Conserv. Recycl. 62, 5155, http://dx.doi.org/10.1016/j.resconrec.2012.02.007.
Imteaz, M.A., Shanableh, A., Rahman, A., Ahsan, A., 2011. Optimisation of rainwater
tank design from large roofs: a case study in Melbourne, Australia. Resour.
G Model
10
ARTICLE IN PRESS

013.
Jiang, Y., Cai, W., Du, P., Pan, W., Wang, C., 2015. Virtual water in interprovincial
trade with implications for Chinas water policy. J. Clean. Prod. 87, 655665,
http://dx.doi.org/10.1016/j.jclepro.2014.10.074.
Khastagir, A., Jayasuriya, N., 2010. Optimal sizing of rain water tanks for domestic
water conservation. J. Hydrol. 381, 181188, http://dx.doi.org/10.1016/j.
jhydrol.2009.11.040.
Kim, B., Anderson, K., Lee, S., Kim, H., 2014. A real option perspective to value the
multi-stage construction of rainwater harvesting systems reusing septic tank.
Water Resour. Manag. 28, 22792291, http://dx.doi.org/10.1007/s11269-0140613-3.
Kim, H., Han, M., Lee, J.Y., 2012. The application of an analytical probabilistic model
for estimating the rainfall-runoff reductions achieved using a rainwater
harvesting system. Sci. Total Environ. 424, 213218, http://dx.doi.org/10.1016/
j.scitotenv.2012.02.021.
Leggett, D., Brown, R., Staneld, G., Brewer, D., Holliday, E., 2001. Rainwater and
Greywater Use in Buildings: Decision-making for Water Conservation. CIRIA,
London.
Liaw, C., Tsai, Y., 2005. Optimum storage volume of rooftop rain water harvesting
systems for domestic use. J. Am. Water Resour. Assoc., 901912.
L, Y.P., Yang, K., Che, Y., Shang, Z.Y., Zhu, H.F., Jian, Y., 2013.
Cost-effectiveness-based multi-criteria optimization for sustainable rainwater
utilization: a case study in Shanghai. Urban Water J. 10, 127143, http://dx.doi.
org/10.1080/1573062X.2012.682591.
Mason, A., 2012. OpenSolver an open source add-in to solve linear and integer
programmes in Excel. In: Klatte, Diethard, Luthi, Hans-Jakob, Schmedders, K.
(Eds.), Operations Research Proceedings 2011. Springer, Berlin Heidelberg, pp.
401406, http://dx.doi.org/10.1007/978-3-642-29210-1 64
(http://opensolver.org).
Morales-Pinzn, T., Rieradevall, J., Gasol, C.M., Gabarrell, X., 2015. Modelling for
economic cost and environmental analysis of rainwater harvesting systems. J.
Clean. Prod. 87, 613626, http://dx.doi.org/10.1016/j.jclepro.2014.10.021.
Okoye, C.O., Atikol, U., 2014. A parametric study on the feasibility of solar chimney
power plants in North Cyprus conditions. Energy Convers. Manag. 80,
178187, http://dx.doi.org/10.1016/j.enconman.2014.01.009.
Palla, A., Gnecco, I., Lanza, L.G., 2011. Non-dimensional design parameters and
performance assessment of rainwater harvesting systems. J. Hydrol. 401,
6576, http://dx.doi.org/10.1016/j.jhydrol.2011.02.009.
Rahman, A., Keane, J., Imteaz, M.A., 2012. Rainwater harvesting in Greater Sydney:
water savings, reliability and economic benets. Resour. Conserv. Recycl. 61,
1621, http://dx.doi.org/10.1016/j.resconrec.2011.12.002.
Sample, D.J., Liu, J., 2014. Optimizing rainwater harvesting systems for the dual
purposes of water supply and runoff capture. J. Clean. Prod. 75, 174194,
http://dx.doi.org/10.1016/j.jclepro.2014.03.075.
Santos, C., Taveira-Pinto, F., 2013. Analysis of different criteria to size rainwater
storage tanks using detailed methods. Resour. Conserv. Recycl. 71, 16, http://
dx.doi.org/10.1016/j.resconrec.2012.11.004.
Scarborough, H., Sahin, O., Porter, M., Stewart, R., 2015. Long-term water supply
planning in an Australian coastal city: dams or desalination? Desalination 358,
6168, http://dx.doi.org/10.1016/j.desal.2014.12.013.
Silva, C.M., Sousa, V., Carvalho, N.V., 2015. Evaluation of rainwater harvesting in
Portugal: application to single-family residences. Resour. Conserv. Recycl. 94,
2134, http://dx.doi.org/10.1016/j.resconrec.2014.11.004.
Su, M.-D., Lin, C.-H., Chang, L.-F., Kang, J.-L., Lin, M.-C., 2009. A probabilistic
approach to rainwater harvesting systems design and evaluation. Resour.
005.
Surez-Varela, M., Martnez-Espineira,

R., Gonzlez-Gmez, F., 2015. An analysis of
the price escalation of non-linear water tariffs for domestic uses in Spain. Util.
Policy 34, 8293, http://dx.doi.org/10.1016/j.jup.2015.01.005.
Tam, V.W.Y., Tam, L., Zeng, S.X., 2010. Cost effectiveness and tradeoff on the use of
rainwater tank: an empirical study in Australian residential decision-making.
Resour. Conserv. Recycl. 54, 178186, http://dx.doi.org/10.1016/j.resconrec.
2009.07.014.
Unami, K., Mohawesh, O., Shari, E., Takeuchi, J., Fujihara, M., 2015. Stochastic
modelling and control of rainwater harvesting systems for irrigation during
dry spells. J. Clean. Prod. 88, 185195, http://dx.doi.org/10.1016/j.jclepro.2014.
03.100.
Villarreal, E.L., Dixon, A., 2005. Analysis of a rainwater collection system for
domestic water supply in Ringdansen, Norrkping, Sweden. Build. Environ. 40,
11741184, http://dx.doi.org/10.1016/j.buildenv.2004.10.018.
Ward, S., Memon, F.A., Butler, D., 2010. Rainwater harvesting: model-based design
evaluation. Water Sci. Technol. 61, 8596, http://dx.doi.org/10.2166/wst.2010.
783.
Ward, S., Memon, F.A., Butler, D., 2012. Performance of a large building rainwater
harvesting system. Water Res. 46, 51275134, http://dx.doi.org/10.1016/j.
watres.2012.06.043.

1 s2.0 S0921344915300756 Main PDF

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

1 s2.0 S0921344915300756 Main PDF

Hochgeladen von

Copyright:

Verfügbare Formate

G Model

RECYCL-3096; No. of Pages 10

Resources, Conservation and Recycling xxx (2015) xxxxxx

Contents lists available at ScienceDirect

Resources, Conservation and Recycling

Optimal sizing of storage tanks in domestic rainwater harvesting

under different climatic conditions (Domnech and Saur, 2011;

Saur, 2011; Imteaz et al., 2012; Rahman et al., 2012). Domnech

housing units. Their optimization model minimizes the total annual

Fig. 1. Increasing block rate tariff scheme.

available space for rainwater tank. The aim is to determine the

btj (Vj Vj1 ) + btk (V Vk1 )

where V0 = 0 by denition and k = argmax0jJ {j|V > Vj } +1. An

the tank to satisfy demand in the same period. At the beginning of

Fig. 2. Schematic of a typical domestic rainwater harvesting system.

where cf is the dimensionless runoff coefcient, Acol the area of the

where Wd is the water usage in m3 /day/capita, n is the number

CPt /(1 + i) . If fini + Z* < CostPFN , then building a rain-

is economically feasible, otherwise not. Note

Average Monthly Rainfall (mm)

Fig. 3. Monthly average rainfall of Kyrenia (19782009).

RECYCL-3096; No. of Pages 10

Total Discounted Cost (TL)

Net Financial Benefit (TL)

with an optimal rainwater tank size of 2.2 m3 is found. The net

80 to 300 m2 in Case B, the discount rate from 3% to 12% in Case

Total Discounted Cost (TL)

Net Financial Benefit (TL)

RECYCL-3096; No. of Pages 10

Total Discounted Cost (TL)

Net Financial Benefit (TL)

Consumption per cubic meter per capita

Consumption per cubic meter per capita

Total Discounted Cost (TL)

Net Financial Benefit (TL)

Discount Rate (%)

Discount Rate (%)

RECYCL-3096; No. of Pages 10

Total Discounted Cost (TL)

Net Financial Benefit (TL)

Unit Cost of Tank (TL/m )

Unit Cost of Tank (TL/m )

Average Monthly Rainfall (mm)

Changing the unit cost of building a rainwater tank has a

Fig. 9. Average monthly rainfall distribution of the selected stations.

The monthly distribution of the selected rainfall stations is depicted

RECYCL-3096; No. of Pages 10

Net Financial Benefit (TL)

Total Discounted Cost (TL)

Tank Size (m3 )

Conserv. Recycl. 55, 10221029, http://dx.doi.org/10.1016/j.resconrec.2011.05.

Surez-Varela, M., Martnez-Espineira,

Das könnte Ihnen auch gefallen