Abstract

Establishing the water balance developed by the International Water Association (IWA) is a worldwide applied approach to determine and analyse water losses in water distribution systems (WDS). The water balance covers those parts of a WDS within the responsibility of the water utility. Water losses occurring ‘before’ a customer meter are at the expense of the utility, while water lost or wasted ‘after’ the meter is paid for by the customer. This applies to systems where customer metering is in place and/or consumption is charged according to the consumed volumes. However, many WDS in the world lack customer meters, are operated intermittently and apply flat-rate tariffs. In intermittent supplies, a considerable amount of water is lost or wasted within the private properties. The flat-rate tariff might not cover this amount or part of the amount. Thus, actual consumption and wastage should be separately quantified with respect to the utility's water reduction measures. Accepting the described conditions, the authors have developed an adaption of the IWA water balance and the methods to establish the balance. In this paper the application of the developed approach in an initially unmetered WDS with intermittent water supply in the city of Tiruvannamalai, India, is presented.

INTRODUCTION

The water balance introduced by Lambert & Hirner (2000) of the Water Loss Task Force of the International Water Association (IWA) is a worldwide applied approach to assess water losses in a water distribution system (WDS) or district metered area (DMA); see, for example, US EPA (2010). The IWA balance covers those parts of a WDS within the responsibility of the water utility. The share of system input volume supplied to the billed customers, which is termed billed authorised consumption, is assumed to be paid for and thus to equal the volume of revenue water. The volume of non-revenue water comprises the volumes of unbilled authorised consumption and water losses, which are split into apparent and real losses along with their sub-components (Lambert & Hirner 2000). Thus, the standard IWA balance is based on the assumptions that customer metering is in place and/or water consumption is charged according to the consumed volumes and therefore, water lost or wasted ‘after’ the customer meter is at the expenses of the customers.

The conditions in a WDS might differ from the assumptions on which the standard IWA water balance is based. For the case of South Africa, Seago & McKenzie (2007) and later McKenzie et al. (2012) subdivided the revenue water component of the IWA water balance. Subsidised water volumes are defined as free basic water, billed and actually paid volumes are called recovered revenue water, and non-recovered revenue water is billed but not paid. Kanakoudis & Tsitsifli (2010) further modified that approach for the case of Greece by introducing the minimum charge difference. Customers pay a minimum charge equal to a certain minimum volume. The component comprises the proportion of the paid minimum volume actually not consumed.

However, in the developing world WDS often lack customer meters, operation is intermittent and flat-rate tariffs are charged, not reflecting the actual water consumption (Farley & Trow 2003; Kingdom et al. 2006; Fanner 2009; Klingel 2011). Furthermore, in intermittent supplies a considerable amount of water is lost or wasted within the customers' premises, for example by overflows of private tanks during supply or tank emptying prior to the supply period in order to get fresh water (Bradley et al. 2000; Totsuka et al. 2004; Butler & Memon 2006; Klingel 2011). As flat-rate tariffs are either based on fixed volumes or other characteristics like the diameter of the service connection or the size/value of the property, these tariffs might not cover the total amount or part of the wasted or lost water volumes (Raghupati & Foster 2002). Thus, the actual consumption and wastage should be separately quantified or estimated with respect to the utility's water reduction measures and the calculation of revenue water. The IWA water balance and the modifications published so far are not appropriate for the described case. Therefore, an adapted water balance for the application for WDS facing the above described constraints has been developed by the authors (Mastaller & Klingel 2017).

This paper presents the application of the adapted water balance and the methods to determine the balance components for an initially unmetered DMA with intermittent water supply in the city of Tiruvannamalai, Tamil Nadu, India. First, the case study is introduced, including the relevant boundary conditions regarding the establishment of a water balance. In the third section, the water balance adapted to these conditions is introduced and the methodology for determining the balance components is presented and applied for the case study. The fourth section presents and discusses the results of the adapted water balance. Finally, a conclusion is given in the last section.

CASE STUDY TIRUVANNAMALAI

About 34,380 households with a population of about 145,000 people are registered in the city of Tiruvannamalai, located in the state of Tamil Nadu in South India (Government of India 2011). The city's WDS, operated by Tiruvannamalai Municipality, comprises four main pump stations from where the treated water is pumped to 11 elevated storage tanks (ESRs) which are used to supply water to the 11 distribution zones across the city. The supply of the approximately 16,941 connected households is intermittent with daily supply hours varying from 2 to 11 hours per day (Tiruvannamalai Municipality 2013a). Almost all of the households have installed private storage tanks to cope with the intermittent supply. In the entire WDS no metering devices are installed. Hence, system input and consumption is unknown and the customers are charged by a flat-rate tariff of 100 Indian rupees for the supply of a flat-rate volume of 7.5 m³ per month and service connection in addition to a constant water supply tax depending on the property size (Tiruvannamalai Municipality 2013b).

A DMA, supply zone 5 of the WDS, was chosen as a case study covering approximately 715 household service connections with a total population of 3,094 people (Figure 1). The water supply of the zone is mainly operated from the Thamarai Nagar elevated storage reservoir (Thamarai Nagar ESR) and partly from the Anna Nagar overflow tank (Anna Nagar OT).

Figure 1

WDS of the DMA with locations (in grey) of the sampled household connections (left), household characteristics of the WDS (right).

Figure 1

WDS of the DMA with locations (in grey) of the sampled household connections (left), household characteristics of the WDS (right).

Fifty four percent of the population of zone 5 only use water supplied by the public WDS. The other 46% use additional sources, such as private bore wells (25%), public taps (8%), the purchase of bottled water (9%) or all three sources (4%). Most of the households are equipped with a ground and a roof tank (70%). The roof tanks are filled by running a private pump. 15% only have a ground tank and 13% of the households directly pump water from the WDS into a roof tank. 2% neither have a ground tank nor a roof tank. They store water in containers, for example in buckets and barrels. 60% of the households have four or fewer members, while 32% have five to eight members. 8% of the households count nine or more people. Figure 1 summarizes the numbers.

APPROACH

Adapted water balance

Regarding the application of a water balance to assess the water losses in the WDS, the most relevant boundary conditions of the case study are intermittent water supply, absence of metering devices and the application of a flat-rate water tariff, which are typical for many WDS in developing countries. As these boundary conditions differ from the conditions the application of the standard IWA water balance is based on, a water balance adapted to the outlined boundary conditions has been developed by Mastaller & Klingel (2017), as shown in Figure 2. The component authorised consumption from the IWA water balance is replaced by the component Authorised Supply QS, which comprises the two components Flat-Rate Billed Authorised Supply QBS and Unbilled Authorised Supply QUS. The first represents the water volume actually supplied to the customers that are billed by the flat-rate tariff. The latter includes the volumes taken by unbilled customers, the utility and other authorised parties in accordance with the IWA definition of unbilled authorised consumption. The reason for using the term ‘supply’ instead of ‘consumption’ is that due to the flat-rate tariff not all of the water supplied to the customers' premises is actually consumed but partly lost or wasted. Hence, QBS is split into the two components Consumption QBSC and Wastage QBSW. The components System Input Volume QSIV, Water Losses QL and Real Losses QRL remain the same as in the IWA water balance. Apparent Losses QAL comprise only the component Unauthorised Consumption QALC, as water meters are not installed and thus billing is not based on meter readings. Known meter inaccuracies and systematic data handling errors of the monitoring system to quantify the volumes of System Input Volume QSIV as well as Flat-Rate Billed Authorised Supply QBS (outlined in the following sections) are already considered in these volumes and thus not depicted separately in the adapted water balance.

Figure 2

Adapted water balance for an intermittently operated WDS without customer metering and charging flat-rate tariffs in volume per balancing period (Mastaller & Klingel 2017).

Figure 2

Adapted water balance for an intermittently operated WDS without customer metering and charging flat-rate tariffs in volume per balancing period (Mastaller & Klingel 2017).

As the flat-rate tariff does not necessarily reflect the volume actually supplied to the customers, Revenue Water QRW and Non-Revenue Water QNRW cannot be calculated following the top down approach of the IWA water balance and have to be calculated separately. Three scenarios are possible. In scenario 1, customers are actually supplied with exactly the water volume that is billed (QRW = QBS). In scenario 2, customers pay for more water than they are supplied with (QRW > QBS). The amount of water consumers are not supplied with but pay for is defined as Excess Revenue Water QERW (QERW = QRWQBS). In scenario 3, consumers pay for less water than they are supplied with (QRW < QBS). The supplied water consumers did not pay for is called Billed Non-Revenue Water QBNRW (QBNRW = QBSQRW).

System Input Volume QSIV

The System Input Volume QSIV should be measured comprehensively and corrected for known meter inaccuracies. In order to determine QSIV of the DMA, a monitoring system was set up. At Thamarai Nagar ESR, a mechanical bulk meter connected to a data logger (Zenner WPH-I) was installed at the inlet pipe of the tank together with a water level recorder (OTT Orpheus Mini) and an ultrasonic flow measurement device (SebaKMT UDM 200-M) at the outlet pipe (Figure 3). This equipment allows for determining the system input and tank overflows at the reservoir. Another ultrasonic flow measurement device was installed at the outlet pipe of Anna Nagar OT, measuring the system input into the DMA.

Figure 3

Monitoring of the System Input Volume QSIV with a mechanical bulk meter (top left), ultrasonic flow measurement device (bottom left); household water metering and logging at a service connection (top middle) and downstream a roof tank (bottom middle) and consumption survey protocol (right).

Figure 3

Monitoring of the System Input Volume QSIV with a mechanical bulk meter (top left), ultrasonic flow measurement device (bottom left); household water metering and logging at a service connection (top middle) and downstream a roof tank (bottom middle) and consumption survey protocol (right).

Flat-Rate Billed Authorised Supply QBS

Full coverage with meters of the initially unmetered DMA was not possible for financial, social and political reasons. Hence, only a representative and random sample of households is equipped with meters (Zenner single-jet dry-running meters) at the service connections to estimate Flat-Rate Billed Authorised Supply QBS using statistical sampling methods (Figures 1 and 3). An accurate calculation of the adequate sample size is not possible due to the explorative nature of the method. However, a sample of 5% of the total number of households is assumed to ensure an adequate representativeness (W.-D. Heller, personal communication, January, 2014). Meters were installed at a sample of 45 service connections, which results in a coverage of approximately 6.3% of the total 715 service connections in the DMA. The households were selected based on a preliminary survey regarding household population and type of water supply as well as in consideration of the sample covering the whole DMA. The affected customers have been informed that they will still be charged as unmeasured by the utility to ensure no changes in their consumption patterns. The locations of the metered service connections are indicated in Figure 1.

Applying statistical methods, the mean volume supplied to the sample of metered service connections is calculated in order to estimate the total supply volume of QBS to a certain degree of uncertainty. Besides simple sampling, post-stratified sampling can be applied using auxiliary information about the households, e.g. occupancy or type of water supply, to separate the population into non-overlapping subpopulations (strata). By adequate stratum selection, the precision of the estimates can be increased. For simple sampling, the mean is calculated by dividing the sum of the sampled values y1,…,yn by the size of the sample n, as shown in Equation (1). The mean of the post-stratified sample is composed of the sum of the individual strata means . As shown in Equation (2), the means are weighted by the relative size of each stratum Nh/N, which automatically corrects the estimator for any bad balanced sample (Holt & Smith 1979).  
formula
(1)
 
formula
(2)
The total Flat-Rate Billed Authorised Supply QBS is estimated by multiplying the total population N, for example, the total number of service connections in the DMA, with either the mean value obtained by simple sampling or post-stratified sampling, see Equation (3). Afterwards, the estimated values can be evaluated using variance, standard deviation and confidence intervals. Mastaller & Klingel (2015) describe the application of the statistical sampling methods in more detail.  
formula
(3)
With the mean per household capita supply volume of the sample, the Flat-Rate Billed Authorised Supply QBS was estimated for the balancing period using the two statistical methods simple sampling and post-stratified sampling, see Equations (1) and (2). Comparing the results of the two methods, the post-stratification sampling method leads to a higher precision than the simple sampling method. The reason for that is the lower variance within the non-overlapping strata in which the metered values have been separated for the post-stratification method.

As Flat-Rate Billed Authorised Supply QBS is the water volume actually supplied to the customers, possible metering inaccuracies need to be considered. Especially the meters installed at the service connection are affected by the intermittent supply, for example, by air intruded into the pipelines; see, for example Van Zyl (2011). The obtained meter values are therefore corrected in advance of the statistical extrapolation regarding over-registration due to air flow through the meters using the results of Staiger (2016) and Walter et al. (2016).

Consumption QBSC and Wastage QBSW

The household installations in the DMA consist of a private roof tank to cope with intermittent water supply. In the case of the system pressure being too low to directly fill the roof tank an additional ground tank is installed, which is supplied by the service connection (Figure 1). The roof tank in this case is filled by running a pump. Usually, most of the Consumption QBSC of the households is covered by the roof tank (Roof Tank Consumption QRTC). The Roof Tank Consumption QRTC,S at the sample households is measured with a second meter installed at the outlet of the roof tank in addition to the meter at the service connection (Figure 3).

However, part of the supplied water might be consumed directly from the ground tank or anywhere else upstream of the roof tank. This Ground Tank Consumption QGTC,S of the sample is estimated by a survey. Two surveys were conducted, one at 11 households for six days and one at eight households for five days. The customers were asked to note the approximate volume of ground tank consumption in a protocol over the survey period (Figure 3). Consumption QBSC,S is calculated by adding the Ground Tank Consumption QGTC,S and the Roof Tank Consumption QRTC,S, see Equation (4). Then the Wastage QBSW,S equals the Flat-Rate Billed Authorised Supply QBS,S minus Consumption QBSC,S of the sample household, as shown in Equation (5). The total volumes of the components Consumption QBSC and Wastage QBSW are then estimated, applying the statistical approach described in the previous section.  
formula
(4)
 
formula
(5)

Unbilled Authorised Supply QUS

Unbilled Authorised Supply QUS comprises volumes taken by unbilled authorised customers, the utility and other authorised parties within the DMA in the balancing period.

Water Losses QL

Adding Flat-Rate Billed Authorised Supply QBS and Unbilled Authorised Supply QUS gives Authorised Supply QS. Water Losses QL are determined by subtracting Authorised Supply QS from System Input QSIV. The difference between Water Losses QL and Apparent Losses QAL gives Real Losses QRL. As outlined before, QAL comprises only the volume of Unauthorised Consumption QALC. Leakage and Overflows at Storage Tanks QRLT, a subcomponent of Real Losses QRL, can be determined using the measuring data gained with the water level recorder installed inside the storage reservoir.

Revenue Water QRW and Non-Revenue Water QNRW

Revenue Water QRW is calculated by multiplying the fixed Flat-Rate Volume QFR, the number of customers NC billed with the flat-rate tariff and the duration of the balancing period t (in months), see Equation (6). In case the flat-rate tariff is not based on a certain volume, QRW can still be calculated using a volumetric water price (cost/volume) applied in a comparable city and the flat-rate tariff (cost/customer) billed in the case study.

Non-Revenue Water QNRW is determined by subtracting Revenue Water QRW from System Input Volume QSIV, as shown in Equation (7). In scenario 2, Excess Revenue Water QERW is calculated by subtracting Flat-Rate Billed Authorised Supply QBS from Revenue Water QRW, see Equation (8). Flat-Rate Billed Authorised Supply QBS minus Revenue Water QRW gives Billed Non-Revenue Water QBNRW in scenario 3, see Equation (9).  
formula
(6)
 
formula
(7)
 
formula
(8)
 
formula
(9)

RESULTS AND DISCUSSION

The adapted water balance of the DMA is established for a monitoring period of 82 days and the results are summarised in Figure 5. The System Input Volume QSIV is approximately 53,693 m³ (100%). The component Flat-Rate Billed Authorised Supply QBS is estimated based on the metering data of a sample of 44 households. As one household is supplied by two service connections, a total number of 45 service connections are metered. Figure 4 shows the daily supplied volumes per household (white columns) and, considering the known household occupancy, per household capita (black columns) within the water balancing period. The per household volumes vary from 0.1 to 2.6 m³ per day, whereas the variation of the per household capita volumes is smaller. This is due to the relativisation of high per household values by the high number of people living in the respective households. This effect is also shown by the application of the statistical sampling methods, where the post-stratified sampling method using per household capita supply values shows the highest precision (QBS = approximately 37,503 m³), using Equations (2) and (3). Subtracting inaccuracies of the customer meters due to air flow gives a corrected Flat-Rate Billed Authorised Supply QBS of approximately 31,978 m³ (60%).

Figure 4

Daily supply volumes per household and household capita of the metered sample within the water balancing period.

Figure 4

Daily supply volumes per household and household capita of the metered sample within the water balancing period.

Figure 5

Adapted water balance for the case study.

Figure 5

Adapted water balance for the case study.

Consumption QBSC of 26,723 m³ (50%) is calculated using the metered data of the roof tanks and the results of the household consumption survey (Equation (4)). Then, Wastage QBSW is estimated to be 5,255 m³ (10%), using Equation (5). As outlined in the previous section, no water is supplied to unbilled authorised customers, the utility and other authorised parties within the balancing period and thus, the component Unbilled Authorised Supply QUS comprises 0 m³. Following the top-down approach of the water balance, Water Losses QL are calculated to 21,715 m³ (40%) by subtracting the volume of Authorised Supply QS from the System Input Volume QSIV. No Unauthorised Consumption QALC is reported within the balancing period, and thus Apparent Losses QAL are 0 m³ (0%). Therefore, Real Losses QRL are 21,715 m³ (40%) and equal to the total Water Losses QL. Tank overflows could be determined by the water level indicator installed inside the reservoir to approximately 96 m³ (0.2%).

With the fixed Flat-Rate Volume QFR = 7.5 m³, the number of customers NC =715 and the duration of the balancing period in month t = 2.73, Revenue Water QRW can be calculated to 14,658 m³ (27%), using Equation (6). Thus, considerably more water is supplied to the customers than they are billed for by the flat-rate tariff (QBS > QRW, scenario 3). The supplied water consumers did not pay for, the component Billed Non-Revenue Water QBNRW, is 17,321 m³ (Equation (9)). The total volume of Non-Revenue Water QNRW adds up to 39,035 m³ (73%), of which 44% is due to the inadequate flat-rate tariff applied.

It has to be noted that the quantification of the water losses using the top-down calculation of the water balance is afflicted with uncertainties of the values of the individual components. Therefore, the measurement errors of the water meters and the confidence interval of the estimations applying statistical sampling methods have to be taken into account. With meter errors of 2% indicated by the manufacturers and the 95%-confidence interval of the estimated Flat-Rate Billed Authorised Supply QBS showing the upper and lower limit being ±10%, the combined uncertainty of the water losses in the balancing period is ±3,996 m³. Thus, the true value of the water losses can lie in the range from 33% to 47%, with the mean value of 40% indicated in Figure 5.

CONCLUSION

Many WDS, especially in the developing world, are operated intermittently and without metering. Flat-rate tariffs are applied due to the fact that the actual consumption is not metered. These boundary conditions lead to the necessity to adapt the IWA water balance in order to provide a suitable tool for managers and operators of water utilities to monitor and assess water losses and non-revenue water in WDS facing these constraints. The adapted water balance has been developed and introduced by Mastaller & Klingel (2017). In this paper, the application of the balance for a DMA within the WDS of Tiruvannamalai in Tamil Nadu, India, which is a typical example of the case, is presented.

At first, an adapted monitoring system is implemented comprising full metering of the system input volume as well as a comprehensive metering of the supplied, consumed and wasted volumes at a sample of households. In case a full coverage of household metering is not possible due to financial, social and/or political reasons, the monitoring of a sample of households (about 6%) and the application of statistical sampling methods in the case study prove to provide proper results regarding the assessment of water losses using the adapted water balance. The volume supplied to the customers, depicted as component Flat-Rate Billed Authorised Supply QBS in the water balance, is estimated at 60% of the System Input Volume QSIV by applying the statistical method post-stratified sampling for extrapolation. The application shows that additional information on household occupancy increases the precision of the estimate and should therefore be acquired by a household survey or census data. Further, a distinction between the actual Consumption QBSC (50%) and the Wastage QBSW (10%) within the customers' premises can be made by installing water meters at the roof tank outlets and executing a survey regarding the direct withdrawals at the house tanks. The volume of wastage, which is at the expenses of the water utility due to the flat-rate tariff, could be tackled, for example, by supporting the installation of floating valves at house tanks to prevent tank overflows and by awareness campaigns. The total volume of Water Losses QL within the DMA is calculated top-down to 40%, with a combined uncertainty of ±7%.

A further advantage of the adapted water balance is that it directly addresses possible discrepancies between water volumes actually supplied to the customers and the fixed (theoretical) water volumes on which the flat-rate tariffs are often based. In this case study, more water is being supplied to the customers than is paid for. As a result, almost half of the considerably high volume of Non-Revenue Water QNRW of 73% is due to the inadequate flat-rate tariff applied by the water utility. This volume of supplied water consumers did not pay for is directly depicted in the balance by the component of Billed Non-Revenue Water QBNR.

The application of the adapted water balance in the case study shows that this approach can provide the basis for an adapted and sustainable water loss monitoring and assessment program in WDS with intermittent water supply, no full metering coverage and a flat-rate tariff being billed to the customers. The required monitoring system to be implemented for establishing the water balance is simple, robust and financially affordable to initiate a continuous data recording system in a WDS. The adapted water balance explicitly addresses the most relevant components of water losses and non-revenue water in WDS facing the outlined constraints, like wastage within households' premises or the inadequacy of the flat-rate tariff structure. Therefore, managers and operators of water utilities will directly benefit from frequently establishing the adapted water balance in order to initiate targeted measures to reduce the volumes of water losses and non-revenue water in their WDS.

ACKNOWLEDGEMENTS

The authors are grateful to the German Federal Ministry of Education and Research for funding the work (funding code: 02WCL1300B).

REFERENCES

REFERENCES
Bradley
,
R. M.
,
Weeratane
,
S.
&
Mediwake
,
T. M. M.
2000
Water use projections in developing countries
.
Journal AWWA
94
(
4
),
52
63
.
Butler
,
D.
&
Memon
,
F.
2006
Water Demand Management
.
IWA Publishing
,
London
,
UK
.
Fanner
,
P. V.
2009
Determining a validated water balance in an unmeasured system and understanding unmeasured residential customer consumption
. In:
Proceedings of the Water Loss Specialist Conference 2009
,
Cape Town, South Africa
,
26–30 April 2009
,
International Water Association
, pp.
184
188
.
Farley
,
M.
&
Trow
,
S.
2003
Losses in Water Distribution Networks – A Practitioner's Guide to Assessment, Monitoring and Control
.
IWA Publishing
,
London
,
UK
.
Government of India
2011
Census of India
.
Ministry of Home Affairs, Office of the Registrar General and Census Commissioner
,
New Delhi
,
India
.
Holt
,
D.
&
Smith
,
T. M. F.
1979
Poststratification
.
Journal of the Royal Statistical Society, Series A
142
(
1
),
33
46
.
Kanakoudis
,
V.
&
Tsitsifli
,
S.
2010
Water volume vs. revenues oriented water balance calculation for urban water networks: the ‘Minimum Charge Difference’ component makes a difference
. In:
Proceedings of the IWA International Water Loss Management 2010 Conference
,
Sao Paulo, Brazil
,
6–9 June 2010
,
International Water Association
.
Kingdom
,
B.
,
Liemberger
,
R.
&
Marin
,
P.
2006
The Challenge of Reducing Non-Revenue Water (NRW) in Developing Countries – How the Private Sector Can Help: a Look at Performance-Based Service Contracting
,
Water Supply and Sanitation Sector Board discussion paper series, Paper No. 8
,
World Bank
,
Washington, DC
,
USA
.
Klingel
,
P.
2011
Technical causes and impacts of intermittent water distribution
.
Water Science and Technology: Water Supply
12
(
4
),
504
512
.
Lambert
,
A.
&
Hirner
,
W.
2000
Losses from Water Supply Systems: Standard Terminology and Recommended Performance Measures
,
The Blue Pages, International Water Association (IWA), London, UK, October 2000
.
Mastaller
,
M.
&
Klingel
,
P.
2015
Estimating domestic water consumption for water balance calculation using statistical sampling methods
. In:
Proceedings of the Water Loss Management 2015
,
Bucharest, Romania
,
15–17 June 2015
,
International Water Association, ARA Publishing
, pp.
176
187
.
Mastaller
,
M.
&
Klingel
,
P.
2017
Adapting the IWA water balance to intermittent water supply and flat-rate tariffs without customer metering
.
Journal of Water, Sanitation and Hygiene for Development
7
(
3
),
396
406
.
McKenzie
,
R.
,
Siqalaba
,
Z.
&
Wegelin
,
W.
2012
The State of Non-Revenue Water in South Africa (2012)
.
Report TT 522/12
,
South African Water Research Commission
,
Pretoria
,
South Africa
.
Raghupati
,
U. P.
&
Foster
,
V.
2002
Water Tariffs and Subsidies in South Asia – A Scorecard for India
.
Paper 2
,
Water and Sanitation Program – South Asia
,
Washington, DC
,
USA
.
Seago
,
C. J.
&
Mckenzie
,
R.
2007
An Assessment of Non-Revenue Water in South Africa
,
Report TT 300/07
,
South African Water Research Commission
,
Pretoria
,
South Africa
.
Staiger
,
D.
2016
Approach for Determining Meter Inaccuracies in Intermittent Water Supply for Water Balancing
.
Bachelor Thesis (unpublished)
,
Karlsruhe Institute of Technology (KIT), Institute for Water and River Basin Management
,
Karlsruhe
,
Germany
(in German)
.
Tiruvannamalai Municipality
2013a
House Service Connections, Tariffs & Revenues
.
Computer Section Tiruvannamalai Municipality
(accessed December 2013)
.
Tiruvannamalai Municipality
2013b
By-Law, Sub-Rules for Distribution of Water Supply
.
Tiruvannamalai Municipality
.
(accessed September 2013) (in Tamil)
.
Totsuka
,
N.
,
Trifunovic
,
N.
&
Vairavamoorthy
,
K.
2004
Intermittent urban water supply under water starving situations
. In:
Proceedings of the 30th WEDC International Conference
,
Vientiane
, pp.
505
512
.
US EPA
2010
Control and Mitigation of Drinking Water Losses in Distribution Systems
,
Report EPA 816-R-10-019
,
United States Environmental Protection Agency (US EPA)
,
Washington, DC
,
USA
.
Van Zyl
,
J. E.
2011
Introduction to Integrated Water Meter Management
,
Report TT 490/11
,
Water Research Commission (WRC)
,
Gezina
,
South Africa
.
Walter
,
D.
,
Mastaller
,
M.
&
Klingel
,
P.
2016
Accuracy of single-jet and multi-jet water meters under the influence of the filling process in intermittently operated pipe networks
. In:
Proceedings of the IWA Water IDEAS 2016
,
Bologna, Italy
,
20–21 October 2016
,
International Water Association
.