Dear Author

Here are the proofs of your article.

You can submit your corrections online, via e-mail or by fax.

For online submission please insert your corrections in the online correction form.
Always indicate the line number to which the correction refers.

You can also insert your corrections in the proof PDF and email the annotated PDF.

For fax submission, please ensure that your corrections are clearly legible. Use a fine
black pen and write the correction in the margin, not too close to the edge of the page.

Remember to note the journal title, article number, and your name when sending your
response via e-mail or fax.

Check the metadata sheet to make sure that the header information, especially author
names and the corresponding affiliations are correctly shown.

Check the questions that may have arisen during copy editing and insert your
answers/corrections.

Check that the text is complete and that all figures, tables and their legends are included.
Also check the accuracy of special characters, equations, and electronic supplementary
material if applicable. If necessary refer to the Edited manuscript.

The publication of inaccurate data such as dosages and units can have serious
consequences. Please take particular care that all such details are correct.

Please do not make changes that involve only matters of style. We have generally
introduced forms that follow the journals style.

Substantial changes in content, e.g., new results, corrected values, title and authorship are
not allowed without the approval of the responsible editor. In such a case, please contact
the Editorial Office and return his/her consent together with the proof.

If we do not receive your corrections within 48 hours, we will send you a reminder.

Your article will be published Online First approximately one week after receipt of your
corrected proofs. This is the official first publication citable with the DOI. Further
changes are, therefore, not possible.

The printed version will follow in a forthcoming issue.

Please note
After online publication, subscribers (personal/institutional) to this journal will have
access to the complete article via the DOI using the URL:
http://dx.doi.org/10.1007/s11269-014-0704-1

If you would like to know when your article has been published online, take advantage
of our free alert service. For registration and further information, go to:
http://www.link.springer.com.
Due to the electronic nature of the procedure, the manuscript and the original figures
will only be returned to you on special request. When you return your corrections,
please inform us, if you would like to have these documents returned.

AUTHOR'S PROOF

Metadata of the article that will be visualized in OnlineFirst

Article Title

Water Sav ing and Energy Reduction through Pressure

Management in Urban Water Distribution Netw orks

Article Sub- Title

Article Copyright Year

Springer Science+Business Media Dordrecht 2014

(This w ill be the copyright line in the final PDF)

Journal Name

Water Resources Management

Family Name

Particle

Given Name

Suffix

9
10
11

Corresponding
Author

Organization

Chen
Qiuw en
CEER, Nanjing Hydraulic Research Institute

Division
Address

Nanjing 210029, China

12

Organization

RCEES, Chinese Academy of Sciences

13

Division

14

Address

Beijing 100085, China

15

e-mail

qwchen@nhri.cn

16

Family Name

Xu

17

Particle

18

Given Name

19
20

Author

Suffix
Organization

21

Division

22

Address

23

e-mail

24

Family Name

25

Particle

26

Given Name

27

Author

Qiang
RCEES, Chinese Academy of Sciences
Beijing 100085, China
Ma
Jinfeng

Suffix

28

Organization

29

Division

30

Address

RCEES, Chinese Academy of Sciences

Beijing 100085, China

AUTHOR'S PROOF
31

e-mail

32

Family Name

33

Particle

34

Given Name

35

Suffix

36

Organization

37

Author

Blanckaert
Koen
RCEES, Chinese Academy of Sciences

Division

38

Address

Beijing 100085, China

39

Organization

Laboratory of Hydraulic Constructions, EPFL

40

Division

41

Address

42

e-mail

43

Family Name

44

Particle

45

Given Name

46
47

Organization
Division

49

Address

50

e-mail

51

Received
Schedule

53
54

Wan
Zhonghua

Suffix
Author

48

52

Lausanne 1015, Switzerland

Beijing 100025, China

8 July 2013

Revised
Accepted

Abstract

Anheng Science and Technology Co. Ltd. of

Beijing

28 May 2014

Water shortages and climate change are worldwide issues.

Reduction in water leakage in distribution networks as well as the
associated energy saving and environmental impacts have recently
received increased attention by scientists and water industries.
Pressure management has been proposed as a cost-effective
approach for reduction in water leakage. This study conducted a
real-world water pressure regulation experiment to establish the
pressure-leakage relationship in a district metering area (DMA) of
the water distribution network in Beijing, China. Results showed
that flow into the DMA was sensitive to inlet water pressure. A 5.6 m
reduction in inlet pressure (from 38.8 m to 33.2 m) led to an 83 %
reduction (12.1 l/s) in minimal night flow, which is a good
approximator of leakage. These reductions resulted in 62,633 m 3 of
water saved every year for every km pipe, as well as associated
savings of 1.110 6 MJ of energy and 68 t of CO 2 equivalent
greenhouse gas emissions. The results of this study provide
decision makers with advice for reducing leakage in water

AUTHOR'S PROOF
distribution networks with associated energy and environmental
benefits.
55

Keywords
separated by ' - '

56

Foot note
information

Leakage reduction - Pressure management - Water distribution

network - Energy saving - Greenhouse gas emission reduction

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Water Resour Manage

DOI 10.1007/s11269-014-0704-1

1
3
2
4

Qiang Xu & Qiuwen Chen & Jinfeng Ma &

Koen Blanckaert & Zhonghua Wan

6
7

Received: 8 July 2013 / Accepted: 28 May 2014

# Springer Science+Business Media Dordrecht 2014

PR
O
O
F

Water Saving and Energy Reduction through Pressure

Management in Urban Water Distribution Networks

U
N
C
O
R
R
EC
TE
D

Abstract Water shortages and climate change are worldwide issues. Reduction in water
leakage in distribution networks as well as the associated energy saving and environmental
impacts have recently received increased attention by scientists and water industries. Pressure
management has been proposed as a cost-effective approach for reduction in water leakage.
This study conducted a real-world water pressure regulation experiment to establish the
pressure-leakage relationship in a district metering area (DMA) of the water distribution
network in Beijing, China. Results showed that flow into the DMA was sensitive to inlet
water pressure. A 5.6 m reduction in inlet pressure (from 38.8 m to 33.2 m) led to an 83 %
reduction (12.1 l/s) in minimal night flow, which is a good approximator of leakage. These
reductions resulted in 62,633 m3 of water saved every year for every km pipe, as well as
associated savings of 1.1106 MJ of energy and 68 t of CO2 equivalent greenhouse gas
emissions. The results of this study provide decision makers with advice for reducing leakage
in water distribution networks with associated energy and environmental benefits.

Q2

8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

Keywords Leakage reduction . Pressure management . Water distribution network .

Energy saving . Greenhouse gas emission reduction

24
25
26

1 Introduction

27

Along with rapid socio-economic development and urbanization, water shortages and climate
change are worldwide issues. According to the United Nations, around 1.2 billion people, or

28
29

Q. Xu : Q. Chen : J. Ma : K. Blanckaert
RCEES, Chinese Academy of Sciences, Beijing 100085, China
Q. Chen (*)
CEER, Nanjing Hydraulic Research Institute, Nanjing 210029, China
e-mail: qwchen@nhri.cn
K. Blanckaert
Laboratory of Hydraulic Constructions, EPFL, Lausanne 1015, Switzerland
Z. Wan
Anheng Science and Technology Co. Ltd. of Beijing, Beijing 100025, China

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q. Xu et al.

Q3

U
N
C
O
R
R
EC
TE
D

PR
O
O
F

almost one-fifth of the worlds population, live in areas of physical water scarcity (available
from: http://www.un.org/waterforlifedecade/scarcity.shtml). At the same time, water use has
been increasing at more than twice the rate of population growth in the last century. By 2025,
1.8 billion people will be living in countries or regions with absolute water scarcity, and twothirds of the worlds population could be under water stress conditions (WWAP 2006).
Water loss exists, however, in any water distribution system due to structural deterioration
of the pipe network. Each year 35 % of total water supplied is lost from water distribution
networks all over the world (Farley et al. 2008). Water supply is an energy-intensive industry,
which consumes 23 % of worldwide energy (James et al. 2002). The energy consumption rate
ranges from 242 MJ of energy per cubic meter of water depending on the water source
(Friedrich 2002; Racoviceanu et al. 2007; Lyons et al. 2009; Stokes and Horvath 2009; Mo
et al. 2011). Energy consumption and greenhouse gas emissions are closely related. According
to Stokes and Horvath (2009), 60.7 g of CO2 is emitted for 1 MJ of energy consumed in the
process of producing and distributing imported water. Therefore, water loss not only affects the
revenue of water utility providers, but also wastes a large amount of energy to treat source
water and distribute potable water, and increases the greenhouse gas emissions.
Theoretically, water leakage from the distribution network occurs when the residual
resistance of the pipe can no longer bear the impact of water pressure (Skipworth 2002).
Therefore, approaches for water leakage control can basically be classified into two categories:
improving pipe resistance and reducing water pressure.
The first category focuses on pipes. Breaks are detected and repaired, and deteriorated pipes
are repaired or replaced. Thus, the condition of the distribution network can be improved and
water leakage can be reduced. Researchers have investigated pipe break behavior to optimize
break detection and/or determine the economically optimal time for pipe replacement (Shamir
and Howard 1979; Walski and Pelliccia 1982; Park et al. 2008, 2010; Giustolisi and Berardi
2009; Carrin et al. 2010; Tsitsifli et al. 2011; Xu et al. 2011a, b, 2013; Tabesh and Saber
2012; Fontana and Morais 2013). These analyses mostly depended on collected pipe break
data. There are, however, many beaks that cannot be detected with available technologies. Pipe
replacement can further reduce water leakage, but replacement cannot be implemented on a
massive scale due to high costs and long implementation times.
The second category focuses on water pressure management (Araujo et al. 2006; Nazif
et al. 2010). Water leakage is positively related to water pressure, and reduction in water
pressure can be translated into reduction in water leakage. The total leakage in a pipe
distribution network is often estimated according to a pressure-leakage relationship in the
following form (Lambert 2001; Thronton 2003; UKWIR 2003; Thornton and Lambert 2005):
n
L kP

30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64

where L is the leakage, Pis the average pressure of the network, and k and n are parameters to
be calibrated. The exponent n ranges from 0.5 to 2.5 or even higher depending on the type of
leakage and pipe material (Lambert 2001; Thornton and Lambert 2005). For leaks from joints
and fittings, bigger values of n (>1) are usually obtained. For leaks from holes in pipe, n
usually has smaller values. Regarding to the pipe material, plastic pipes have bigger n values
than metal pipes (Lambert 2001). Obviously, water leakage will be sensitive to water pressure
if n>1.
Pressure management is an effective strategy to reduce water leakage in distribution
networks. Furthermore, it is the only strategy that allows for reductions in residual water
leakage due to undetectable pipe damage. In addition to reducing water leakage from existing
pipe breaks, pressure management also reduces the risk of new breaks and extends pipe

65
66
67
68
69
70
71
72
73
74
75
76

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q1

Water Saving and Energy Reduction through Pressure Management

77Q4=Q5
78
79
80
81
82
83
84
85
86
87
88
89

2 Material and Methods

90

U
N
C
O
R
R
EC
TE
D

2.1 Description of Experimental Pipe Network

PR
O
O
F

lifetime (Farley and Trow 2003; Thornton 2006; Lambert 2011). Energy consumption and
greenhouse gas emissions can accordingly be reduced.
In some countries, notably the UK and Japan, it has been recognized for over 30 years that
effective pressure management is the key to efficient leakage management (Thronton 2003).
Although more and more countries are realizing the importance of pressure management
(Marunga et al. 2006; Girard and Stewart 2007; Soriano et al. 2012), it is still not applied in
most developing countries for two reasons. The first is the lack of decision support tools that
accurately predict the benefits associated with pressure management and justify the investment. The second is that water distribution networks are usually not well configured for
effective pressure management (Mutikanga et al. 2013).
This study conducted a large-scale real-world experiment in a sub-network of Beijings
water distribution system to investigate the effect of pressure management on water leakage
and gain insight into associated energy and environmental benefits.

The water pressure regulation experiment was conducted in a district metering area (DMA) of
the water distribution network of Beijing, China. The network has one inlet where the water
pressure is measured and controlled, and where the flow is measured as well. Fig. 1 depicts the
layout of the DMA and Table 1 presents its general features.

#
I

#
I Pressure Monitoring
")

Inlet (flow and pressure monitoring)

Pipes

")
Fig. 1 Layout of experimental District Metering Area (DMA)

91
92
93
94
95

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q. Xu et al.

t1:1
t1:2
t1:3

Table 1 Basic information on the

experimental District Metering Area (DMA)

t1:4
t1:5

Item

Value

Pipe diameter (mm)

75300

Pipe age (years)

Total pipe length (km)

About 40
7.2

t1:6

Pipe material

Cast iron (97.6 %), Polyethylene (2.4 %)

t1:7

Water user types and

quantities

4,427 residential properties, 3 schools, 2

hospitals, and 1 police office

96

PR
O
O
F

2.2 Flow and Pressure Monitoring and Control

97
98
99
100
101
102
103
104
105
106
107
108
109
110
111

3 Results

112

The measured data of the inlet pressure and flow during the three investigated periods are
shown in Fig. 2 and summarized in Table 2. A strong relationship clearly existed between flow

113
114

U
N
C
O
R
R
EC
TE
D

Water pressure and flow at the inlet were automatically measured and recorded every 15 min.
To ensure that pressure was adequate for reliable water supply, the node (indicated by a
triangle in Fig. 1) with minimal pressure in the DMA was synchronously measured and
recorded. A minimal pressure of 18 m, which was a mandatory rule from the government of
Beijing, was requested at this point in all instances. In fact, considering the users satisfactory
and other possible demands for high water pressure, the minimal pressure was finally
maintained at 27 m. The experimental period was from September 1, 2012 to March 10, 2013.
Three strategies of pressure management were investigated and compared. In the first period of
45 days from September 1 to October 15, 2012, the inlet pressure was not manipulated and kept at
its original value of 38.8 m. This value was rather constant during a day. In the second period of
38 days from October 16 to November 22, 2012, the average pressure was reduced by 3.7 m, and
the pressure during peak hours was relatively low. In the third period of 98 days from November
23, 2012 to March 10, 2013, the average pressure was further reduced by 1.9 m. But the pressure
during peak hours was intentionally increased to compensate for the higher energy losses due to
larger water use. Details of the pressure patterns are presented in the results and discussion sections.

40

period 1

period 2

period 3

pressure

60

flow

40
30

20

20
10
10
0
2012-9-1

0
2012-10-1

2012-11-1

2012-12-1

2013-1-1

date
Fig. 2 Inlet pressure and flow during the three investigated periods

2013-2-1

2013-3-1

flow (l/s)

pressure(m)

50
30

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q1

Water Saving and Energy Reduction through Pressure Management

Table 2 Inlet pressure and flow statistics during the three investigated periods

t2:2

Period

Period 1

Period 2

Period 3

t2:3

Date

01.09.201215.10.2012

16.10.201222.11.2012

23.11.201210.03.2013

t2:4

Duration (days)

45

38

98

t2:5

Pressure (m), meanS.D.*

38.80.41

35.11.75

33.21.58

t2:6
t2:7

Flow (l/s), meanS.D.

MNF* (l/s), meanS.D.

31.29.37
14.51.37

23.17.97
7.21.05

16.98.32
2.41.67

t2:8

Proportion of MNF to flow (%)

46.5

31.2

14.2

t2:9

Water saved compared to period

1 (m3/yr/km)

35,476

62,633

*S.D. is Standard Deviation and MNF is Minimal Night Flow

PR
O
O
F

t2:1

115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130

4 Discussion

131

4.1 Inlet Pressure and Corresponding Flow and Leakage

132

The data shown in Fig. 2 reveals that a reduction in pressure led to a considerable reduction in
flow. In the first several days of period 2, the flow remained at a relatively high level compared
to the following days. This was attributed to water uptake by boilers for urban heating, which
started in Beijing around November 1, 2012. The flow subsequently decreased until it reached
a lower and stable level. In period 3, a sharp decrease in flow occurred from February 10 to 17,
2013. This period coincided with the Chinese Spring Festival holiday, when approximately
two thirds of Beijing residents were out of town.
To further investigate the effects of pressure regulation, Fig. 3 shows hourly variations in
pressure and flow during a representative 24-h period for each period. The representative 24-h
period was obtained by averaging the measured pressure and flow at the same hour for seven
consecutive days. Any seven consecutive days consist of five workdays and two weekend
days. Hence, averaging seven consecutive days considers differences in water use between
workdays and weekend days. The seven consecutive days for periods 2 and 3 were November

133
134
135
136
137
138
139
140
141
142
143
144
145

U
N
C
O
R
R
EC
TE
D

and pressure. In period 1, the average pressure was 38.8 m and flow was relatively high, with
an average value of 31.2 l/s. The average pressure in period 2 was reduced by 3.7 m, resulting
in a decrease in average flow of 8.1 l/s (or 26 %). The average pressure was further reduced by
1.9 m in period 3, resulting in a further reduction in average flow of 6.2 l/s (or 20 %). It should
be noted that average pressure was restored to the original level of 38.8 m for 8 days from
December 15 to 17 and 20 to 24, 2012, to verify the flow response to changes in inlet pressure.
The data from these days was excluded when calculating pressure and flow statistics. Table 2
lists the statistical characteristics of the three periods. Compared with period 1, the reduction in
the inlet pressure for the considered DMA in periods 2 and 3 allowed yearly savings of
35,476 m3 and 62,633 m3 of water per km pipe, respectively.
Although minimal night flow (MNF) consists of normal night use, e.g. toilet flushing, it is
often adopted as an estimation of leakage in cases where no data on night use is available. In
this research, the night water use was not surveyed, thus MNF was directly adopted as an
estimation of leakage. The average MNF in period 1 was highest (14.5 l/s), and a reduction of
7.3 l/s (or 50 %) and 12.1 l/s (or 83 %) occurred in periods 2 and 3, respectively. The leakage
rate dropped from 46.5 % in period 1 to 31.2 % in period 2, and 14.2 % in period 3.

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q. Xu et al.

60

40

pressure (m)

30

P1
P2
P3

F1
F2
F3

L1
L2
L3

40
30

20

PR
O
O
F

20

flow & leakage (l/s)

50

10

10

0
00:00

06:00

12:00
time

18:00

0
24:00

U
N
C
O
R
R
EC
TE
D

Fig. 3 Hourly inlet pressure (P1, P2 and P3), flow (F1, F2 and F3) and leakage estimated according to calibrated
equation (5) (L1, L2 and L3) during a representative 24-h period for each period

15 to 21, 2012, and November 23 to 29, 2012, respectively. December 15 to 17 and 20 to 24,
2012 were used to represent period 1 when the pressure was restored to the original level. This
period was chosen because the pipe conditions could be assumed to be the same during the
period from November 15 to December 24, and also the temperature was similar, which might
influence water consumption. Fig. 3 shows that the flow in period 1 was globally higher than
in the other two periods due to higher pressure. Similarly, the flow in period 2 was globally
higher than that in period 3. Daily averaged flow discharges were 23.6 l/s, 18.6 l/s (21 %
reduction compared to period 1) and 16.4 l/s (31 % reduction compared to period 1) in periods
1, 2, and 3, respectively. Comparison of periods 2 and 3 clearly revealed that higher differences
in pressure resulted in higher differences in flow. Specifically, differences in pressure and flow
were high during the night (0:00 am to 5:00 am), but low during peak hours (8:00 am to
12:00 am and 6:00 pm to 8:00 pm) when the pressure was almost identical. These observations
complied with the hypothesis that differences in flow were mainly incurred by differences in
inlet pressure.
Figure 4 shows measured minimal and average (average from 2:00 am to 5:00 am) night flow
and corresponding inlet pressure during the three periods. The MNF was more stable than the
average night flow, which lends further credit to the hypothesis that MNF is a good estimator of
leakage. Therefore, MNF was used hereafter to establish the pressure-leakage relationship.

146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163

4.2 Pressure-Leakage Relationship

164

To quantify the relationship between inlet pressure and leakage, the following equation, similar
to Eq. 1, was fitted to the measured data:

165
166

L e
kPn

where P is the inlet pressure (m), L is the leakage (l/s), e

k and e
n are the parameters to be calibrated.
In this study, e
k and e
n were calibrated using the MNF and the corresponding inlet pressure.

168
167
169

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q1

Water Saving and Energy Reduction through Pressure Management

period 1

period 2

30

period 3

pressure when minimal flow happens (m)

minimal night flow (l/s)
average night pressure (m)
average night flow (l/s)

20
10
0
2012-9-1

2012-11-1

PR
O
O
F

pressure and flow

40

2013-1-1
date

2013-3-1

Fig. 4 Average (from 2:00 am to 5:00 am) and minimal night flow and corresponding inlet pressure during the
three investigated periods

U
N
C
O
R
R
EC
TE
D

A relatively short period of data (November 1, 2012 to January 31, 2013) was used for the
calibration instead of the whole series because it can be assumed that the pipe conditions, e.g.
amount of breaks, did not change over a short period. In the dataset, the pressure and flow
characteristics of all three periods were included. Data from December 15 to 17 and from 20 to
24 represented period 1, data from November 1 to 22, 2012 represented period 2, and the
remaining data represented period 3. The calibration was based on the measured MNF and
corresponding inlet pressure, as shown in Fig. 4. This calibration resulted in parameter values
of e
k = 1.511012 and e
n = 8.02 (Eq. 2). Fig. 5 illustrates the pressure-leakage curve.
The calibrated leakage relationship, Eq. 2, estimated leakage for every inlet pressure. The
estimated leakage for a representative 24-h period in all three periods is included in Fig. 3.
Daily averaged leakages were 8.5 l/s, 4.0 l/s (53 % reduction compared to period 1) and 2.2 l/s
(74 % reduction compared to period 1) in periods 1, 2, and 3, respectively.
In the absence of experimental data, Farley and Trow (2003) proposed adopting an
exponent n = 1 in Eq. 1, meaning that a 10% reduction in the average pressure of the network
will lead to a 10% reduction in leakage. The exponent e
n = 8.02 in the pressure-leakage
equation (Eq. 2) was much larger than the value of n=1 and also much larger that the
exponents n=0.5 to 2.5 reported for Eq. 1 (Lambert 2001; Thornton and Lambert 2005),
12

leakage (l/s)

10

observed data
fitted curve

R2=0.92

6
4
2
0
30

32

Fig. 5 Calibrated pressure-leakage equation (2)

34
36
pressure (m)

38

40

170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q. Xu et al.

187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204

where L is the leakage (l/s), Pi is the pressure at the leakage location i (m), and ki and ni are the
parameters associated with leakage location i. Pi could be much smaller than the inlet pressure
of the pipe network if the leakage occurs in service pipes with high level. Therefore, if a
pressure reduction is implemented at the inlet, the change rate of Pi can be much larger than
that of the inlet pressure P. To a given change rate of leakage, a more sensitive leakagepressure relationship will be obtained when using the inlet pressure. In this study, Eq. 2 was
used to approximate Eq. 3, which explained why a big value of parameter n was obtained.
Other possible reasons included the type of leakage and pipe material, as introduced in the
introduction section. However, it is hard to determine the exact reasons because these leaks are
usually undetectable.

205
206
207
208
209
210
211
212
213
214
215

U
N
C
O
R
R
EC
TE
D

PR
O
O
F

which suggests that leakage was sensitive to pressure management. Direct comparison of e
n
and n was, however, not meaningful, because fundamentally different pressure variables were
used in Eqs.1 and 2: the networks average pressure was used in the former and the inlet
pressure in the latter.
In practical applications, water utilities directly regulate inlet pressure to a DMA. Therefore
it is important to determine the relationship between inlet pressure and leakage according to
Eq. 2. This relationship is, however, case dependent and needs to be calibrated for each DMA.
In fact the pressure at the leakage location should be used to calibrate the leakage-pressure
relationship function. The pressure inside the water mains at the leakage location varies
between a maximal value given by inlet pressure P, and a minimal Pmin, which was measured
in the present real-world application (Fig. 1). Fig. 6 shows estimates of n according to Eq. 1 for
values of P = (1-a) P+aPmin where 0a 1, which corresponds to P in the range between the
inlet pressure and minimal pressure . From Fig. 6, a range between 7.41 and 8.02 was obtained
for n, which decreased linearly as a increased. A reasonable estimation of n was obtained by
adopting the average value 0.5 (P+Pmin) in Eq. 1, which yielded k=5.91 1012 and n=7.73.
It was concluded that leakage was sensitive to pressure management, and that pressure
reduction provided an efficient method to reduce leakage.
For a pipe network with multiple leakage points, the following equation could be used.
X
L
k i Pni i
3
i

8.2
8.0

7.8
7.6
7.4
7.2
0.0

0.2

0.4

0.6

0.8

1.0

a
Fig. 6 Estimates of n depending on adopted average pressure P = (1-a) P+aPmin (P=inlet pressure, Pmin =
minimal pressure, 0a 1) according to equation (4)

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q1

Water Saving and Energy Reduction through Pressure Management

216

According to Table 2, total flow reduction by decreasing pressure from period 1 to period 3 was
14.3 l/s, of which leakage reduction (indicated by MNF) was 12.1 l/s. The reduction in leakage
accounted for 85 %, while the reduction in water use by residents accounted for only 15 %, which
was comparatively negligible. It also indicated that pressure management would not significantly
affect normal water use of the residents. The 2.2 l/s water use reduction was not included in the
further analyses, thus the estimations of energy saving hereafter were in fact conservative.
The reduction in leakage and water use was accompanied by savings in energy consumption and greenhouse gas emissions. Stokes and Horvath (2009) reported that 18 MJ of energy
was needed to treat and distribute 1 m3 of imported water, and associated CO2 emissions were
1.093 kg. Because Beijing will use imported water as a source of drinking water from 2014,
the two above figures were adopted in the present study for estimating energy savings and CO2
equivalent greenhouse gas emission reduction associated with the proposed pressure management. In this study, 62,633 m3 of water was saved every year for every km pipe by adopting
the pressure reduction of period 3, which corresponded to 1.1106 MJ of energy and 68 t of
CO2 equivalent greenhouse gas emission.

217
218
219
220
221
222
223
224
225
226
227
228
229
230
231

4.4 Potential of Further Reductions in Leakage, Energy Consumption and Greenhouse gas
Emissions

232
233

As introduced in section 2.2, a minimal pressure of 18 m has to be guaranteed at all instances.

According to pressure monitoring, however, the actual average pressure at the minimal
pressure point (triangle in Fig. 1) was 35.2 m, 31.2 m, and 29.5 m in the three periods
respectively. This implied there was still large potential to further reduce the inlet pressure.
Supposing pressure was reduced by another 10 m, according to Eq. 2 the minimal night flow
would have dropped to 0.1 l/s and approximately 9,900 m3 of water per year per km pipe
would have been saved, corresponding to 1.8105 MJ of energy and 10.8 t of CO2 equivalent
greenhouse gas emissions. Unfortunately, the water supply company declined to further reduce
the pressure after considering the customer satisfactory and firefighting requirement.
In many cases, water pressure is regulated by pressure reduction valves (PRVs). The system
pressure and consequently leakage can be well controlled through this method. However, the
mechanism of this method is that local energy loss is provoked at the PRV, implying that
energy for pumping water is not really saved. Therefore, there is a potential to reduce energy
used in water distribution networks by lowering redundant water pressure at the pumping
station. By combining PRVs and pumping regulation, a two-level water pressure management
scheme can be outlined, especially for large water distribution network with multiple water
plants. The first level is large-zone pressure management by regulating the pumps, and the
second level is pressure reduction at the inlet of DMAs using PRVs. Figure 7 illustrates this
two-level pressure management strategy. If the water pressure of the whole pipe network is too
high, the pressure should be first reduced at the pumping station. Then if there are still some
areas having high water pressure because of other reasons, e.g. low ground elevation, further
pressure reduction at DMA level should be considered. It must be realized that water pressure
management for water distribution pipe network is very complex, thus the present sptiotemporal distribution of water pressure, which is influenced by water plants locations, urban
topography, characteristics of water use and etc., should firstly be investigated. Based on the
results, careful calculations should be made to compare different scenarios so as to obtain an
optimized cost-effective management. The optimized pressure management strategy is case
dependent, and there does not exist a strategy that suits for all pipe networks.

234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261

U
N
C
O
R
R
EC
TE
D

PR
O
O
F

4.3 Associated Benefits from Pressure Management

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q. Xu et al.

ENERGY SAVING
1st level:
pump regulation

2nd level:
PRVs control

pump station

pipe

DMA boundary
Fig. 7 Two-level pressure management strategy

PR
O
O
F

PRV

262
263
264
265
266
267
268

5 Conclusions

269

This study investigated the possibility and efficiency of regulating water pressure in a water
distribution network in order to reduce leakage, as well as associated energy consumption and
greenhouse gas emissions. A large-scale real-world experiment was conducted in a DMA of a
water distribution network of Beijing, China. We found that flow was considerably more
sensitive to pressure than expected. By analyzing minimal night flow, a pressure-leakage
relationship was established, which improved knowledge on water leakage behavior. In the
experimental DMA, average flow decreased from 31.2 l/s to 16.9 l/s (or 46 %) after pressure
management. Reduction in leakage and water use contributed 12.1 l/s and 2.2 l/s, respectively,
to this total flow reduction of 14.3 l/s. The corresponding savings per year and per km pipe
were 62,633 m3 of water, and 1.1106 MJ of energy and 68 t of CO2 equivalent greenhouse
gas emissions. These numbers could be even larger if a more complicated pressure management method, e.g. combining pump regulation and PRV control, was applied. These results
should be persuasive for decision makers to recognize the importance and benefits of
managing water pressure in water pipe networks.
The paper investigated the response of water leakage as well the associated energy
consumption and greenhouse gas emissions to the water pressure reduction of distribution
system. However, more studies are requested when optimize the water pressure management
for practical implementation. First, the energy consumption and greenhouse gas emissions
should be estimated for the pilot pipe network based on the lifetime cycle analysis of the water
supplied. Second, other factors should be explicitly considered in the formulation of a water
pressure reduction scheme, such as the potential of water discoloration and increased mineral
build-up in pipelines due to the reduced water velocity, the possibility of failures of the citys

270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291

U
N
C
O
R
R
EC
TE
D

Although the results obtained for this DMA cannot be directly extrapolated to the entire city
of Beijing, the benefits of the pressure management would likely be large if the proposed
strategy was implemented. For a city like Beijing, whose total water supply is about 2.5
million m3 per day, even a 1 % reduction in leakage would result in 9.1 million m3 of water
saving per year and 1.6108 MJ of energy and 9.9103 t of CO2 equivalent greenhouse gas
emission. These values would make a relevant global contribution to the sustainable management of water resources.

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q1

Water Saving and Energy Reduction through Pressure Management

heating system due to reduced water pressure, etc. All these aspects have not yet included in
the current work, and will be the focus of our future studies.

292
293

Acknowledgments The authors are grateful for funding from the Ministry of Sciences and Technology of the
Peoples Republic of China (No. 2012ZX07408-002) and National Natural Science Foundation of China
(No.51309216). Blanckaert was partially funded by the Chinese Academy of Sciences Visiting Professorship
for Senior International Scientists (No. 2011T2Z24). Appreciations are extended to the anonymous reviewers
whose comments are of great value to improve the quality of this manuscript.

294
295
296
299
297
298

References

PR
O
O
F

300

U
N
C
O
R
R
EC
TE
D

Araujo L, Ramos H, Coelho S (2006) Pressure control for leakage minimisation in water distribution systems
management. Water Resour Manag 20(1):133149
Carrin A, Solano H, Gamiz M, Debn A (2010) Evaluation of the reliability of a water supply network from
right-censored and left-truncated break data. Water Resour Manag 24(12):29172935
Farley M, Trow S (2003) Losses in water distribution networks: a practitioners guide to assessment, monitoring
and control, IWA, London
Farley M, Wyeth G, Ghazali ZBM, Istandar A, Singh S (2008) The managers non-revenue water handbook: a
guide to understanding water losses. Ranhill Utilities Berhad and the United States Agency for International
Developement, Bangkok
Fontana M, Morais D (2013) Using Promethee V to select alternatives so as to rehabilitate water supply network
with detected leaks. Water Resour Manag 27(11):40214037
Friedrich E (2002) Life-cycle assessment as an environmental management tool in the production of potable
water. Water Sci Technol 46(9):2936
Girard M, Stewart RA (2007) Implementation of pressure and leakage management strategies on the gold coast,
Australia: case study. J Water Res Pl-ASCE 133(3):210217
Giustolisi O, Berardi L (2009) Prioritizing pipe replacement: from multiobjective genetic algorithms to operational decision support. J Water Res Pl-ASCE 135(6):484492
James K, Campbell SL, Godlove CE (2002) Watergy: Taking advantage of untapped energy and water efficiency
opportunities in municipal water systems. Alliance to Save Energy, Washington DC, USA
Lambert A (2001) What do we know about pressure/leakage relationships in distribution systems. In:
Proceedings of IWA Specialised Conference: System Approach to Leakage Control and Water
Distribution Systems Management, Brno, Czech Republic.
Lambert A (2011) Thornton J (2011) the relationships between pressure and burstsa state-of-the-art update.
Water 21:3738
Lyons E, Zhang P, Benn T, Sharif F, Li K, Crittenden J, Costanza M, Chen Y (2009) Life cycle assessment of
three water supply systems: importation, reclamation and desalination. Water Sci Technol 9(4):439448
Marunga A, Hoko Z, Kaseke E (2006) Pressure management as a leakage reduction and water demand
management tool: the case of the city of mutare, Zimbabwe. Phys Chem Earth 31(1516):763770
Mo W, Zhang Q, Mihelcic JR, Hokanson DR (2011) Embodied energy comparison of surface water and
groundwater supply options. Water Res 45(17):55775586
Mutikanga H, Sharma S, Vairavamoorthy K (2013) Methods and tools for managing losses in water distribution
systems. J Water Res Pl-ASCE 139(2):166174
Nazif S, Karamouz M, Tabesh M, Moridi A (2010) Pressure management model for urban water distribution
networks. Water Resour Manag 24(3):437458
Park S, Jun H, Kim B, Im G (2008) Modeling of water main failure rates using the Log-linear ROCOF and the
Power Law process. Water Resour Manag 22(9):13111324
Park S, Choi C, Kim J, Bae C (2010) Evaluating the economic residual life of water pipes using the proportional
hazards model. Water Resour Manag 24(12):31953217
Racoviceanu AI, Karney BW, Kennedy CA, Colombo AF (2007) Life-cycle energy use and greenhouse gas
emissions inventory for water treatment systems. J Infrastruct Syst 13(4):261270
Shamir U, Howard C (1979) An analytical approach to scheduling pipe replacement. J Am Water Works Assoc
71(5):248258
Skipworth PJ (2002) Whole life costing for water distribution network management, Thomas Telford, London
Soriano RN, Ramos RU, Padua RC (2012) Introduction to Maynilad pressure management program. In:
Proceedings of waterloss 2012, Manila, Philippines

301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346

AUTHOR'S PROOF

JrnlID 11269_ArtID 704_Proof# 1 - 03/06/2014

Q. Xu et al.

U
N
C
O
R
R
EC
TE
D

PR
O
O
F

Stokes JR, Horvath A (2009) Energy and air emission effects of water supply. Environ Sci Technol 43(8):2680
2687
Tabesh M, Saber H (2012) A prioritization model for rehabilitation of water distribution networks using GIS.
Water Resour Manag 26(1):225241
Thornton J, Lambert A (2006) Managing pressure to reduce new breaks. Water 21 December 2006: 2426
Thornton J, Lambert A (2005) Progress in practical prediction of pressure: leakage, pressure: burst frequency and
pressure: consumption relationships. In: Proceedins of IWA Specialised Conference Leakage 2005,
Halifax, Nova Scotia, Canada
Thronton J (2003) Managing leakage by managing pressure: a practical approach. Water 21 October 2003: 4344
Tsitsifli S, Kanakoudis V, Bakouros I (2011) Pipe networks risk assessment based on survival analysis. Water
Resour Manag 25(14):37293746
UKWIR (2003) Leakage index curve and the longer term effects of pressure management. UKWIR Report (Ref
03/WM/08/29), ISBN 1840572809
Walski TM, Pelliccia A (1982) Economic analysis of water main breaks. J Am Water Works Assoc 74(3):140
147
WWAP (World Water Assessment Programme) (2006) World water development report 2: water: a shared
responsibility. UNESCO/Berghahn Books, Paris
Xu Q, Chen Q, Li W, Ma J (2011a) Pipe break prediction based on evolutionary data-driven methods with brief
recorded data. Reliab Eng Syst Saf 96(8):942948
Xu Q, Chen Q, Li W (2011b) Application of genetic programming to modeling pipe failures in water distribution
systems. J Hydroinformatics 13(3):419428
Xu Q, Chen Q, Blanckaert K, Ma J (2013) Optimal pipe replacement strategy based on break rate prediction
through genetic programming for water distribution network. J Hydro-Env Res 7:134140

347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370

AUTHOR'S PROOF
AUTHOR QUERIES
AUTHOR PLEASE ANSWER ALL QUERIES.

U
N
C
O
R
R
EC
TE
D

PR
O
O
F

Q1. Please check the suggested running page title if appropriate.

Otherwise, please provide short running title with maximum of
65 characters including spaces.
Q2. Please check if the affiliations are presented correctly.
Q3. The citation Tsitsifli 2011 (original) has been changed to
Tsitsifli et al. 2011. Please check if appropriate.
Q4. The citation Thornton and Lambert 2006 (original) has been
changed to Thornton 2006. Please check if appropriate.
Q5. The citation Lambert and Thornton 2011 (original) has been
changed to Lambert 2011. Please check if appropriate.