Management, detection and repair of small leaks in a distribution system are critical functions of system operation and maintenance, yet they are often neglected. Large water main breaks can cause sensational damage and draw media attention, but those catastrophic failures only account for about 1% of water lost to leaks (USEPA, 2009). Some small leaks are noticeable at the ground surface and are easily identified, but many leaks continue below ground for months or years. A minor leak of four liters per minute would likely continue for years before it was noticed, resulting in the loss of over two million liters per year. Leak management methods can prevent or reduce leakage volume and leak detection technology can improve the ability of water utilities to respond quickly and repair leaks (USEPA, 2009; Thomson and Wang, 2009).
Non-revenue water is the difference between the volume input to a municipal distribution system (e.g. from a treatment plant) and the billed authorized consumption (e.g. water for which the utility receives payment). The term “non-revenue water” (NRW) has largely replaced “unaccounted for water” (UFW) among water professionals and is used here. NRW is commonly expressed as a percentage of the total water volume input into a system. NRW typically accounts for a large percentage of the water input. Surveys of Latin American and Indian cities found that 40% or more of their input water was lost as NRW (Hinrichsen et al., 1997; McIntonsh, 2003; Farley, 2001).
NRW consists of three categories: unbilled authorized consumption; apparent losses; and real losses. Unbilled authorized consumption (e.g. water donated to a non-profit organization) usually makes up a small fraction. Apparent losses include unauthorized consumption (e.g. illegal connections) and meter inaccuracies; these often account for a considerable percentage of total NRW, especially in developing countries. Real losses consist of any water that is physically lost from the system before it reaches a consumer’s water meter. A small fraction of this may include overflow of storage tanks owned by the utility. However, the vast majority of real losses are due to leakage in distribution systems; this article focuses on detecting and addressing this leakage.
Leakage in distribution systems is a major problem for water utilities throughout the world, in both wealthy and developing countries. Water distribution pipes in many industrialized countries were installed decades ago and are approaching the end of their useful life. The US Environmental Protection Agency (USEPA) has declared that the replacement or rehabilitation of water distribution and transmission systems is one of the country’s biggest infrastructure needs. Leakage rates of 10-20% are considered normal and in some areas of the US, the aging infrastructure is losing up to 50% of water distributed (USEPA, 2009). In developing countries, common causes of leakage, in addition to aging pipes, include poor network design and construction, damage to exposed pipes, and leakage at poorly sealed connections (Farley, 2001).
Prior to implementing formal leak management, detection and repair programs, a water audit should be performed to quantify leakage and prioritize leak management activities. Water audits are typically conducted by monitoring water inputs, flow throughout the distribution system, and customer use during a low-flow period (at night); these are used to quantify losses and identify zones with high leakage. Further information and training materials for water audits can be found elsewhere (Farley, 2001; Maryland Department of the Environment, 2002).
Leak management: In some countries, utilities address leaks reactively, responding to identified leaks and water audits. However, the UK has developed pro-active approaches to prioritizing leak detection and controlling system pressure. In summary, isolating small areas of metered homes yields total leakage for a local area and allows more intensive detection methods to be targeted to key zones. Pressure management, lowering system pressures during times of low demand, can lead to a decrease in the long-term volume lost to leakage and extend the life of pipes (Thomson and Wang, 2009). However, 24-hour continuous pressure should be the first priority and pressure management should not be undertaken by utilities that struggle to maintain adequate 24-hr water pressure (Correlje et al., 2008). Extensive guidance on leak management is available (Fanner et al., 2008).
Leak detection: Many new technologies for leak detection have appeared in recent years. In the late-20th century, the primary methods used for leak detection included acoustic, infrared thermography, chemical tracer, and mechanical methods. Among the acoustic methods were ground microphones, acoustic loggers on pipe fittings, and tethered in-line leak detectors. New and emerging technologies include ground penetrating radar (GPR), combined acoustic logger and leak noise correlators, digital correlators, and radio-frequency interferometers (Pilcher, 2003). More advanced acoustic methods have also been developed recently, including un-tethered leak detection (e.g. the Sahara and SmartBall systems). Detailed discussion of some of these technologies can be found in the references (Thomson and Wang, 2009; Smith et al., 2000; Huniadi, 2000).
Acoustic methods are able to recognize leaks based on the characteristic patterns of sound that leaks create; they have been and continue to be the most common leak detection methods. The choice of an appropriate leak detection technology must consider the pipe material and pipe diameter of a system. Acoustic methods have been used successfully for leak detection in metallic pipes for many years. However, their application in non-metallic piping is more challenging; the sounds created in plastic and concrete pipes tend to be lower-frequency and attenuate more quickly. Despite these challenges, recent technological innovations have enabled the successful application of acoustic methods to these types of piping (USEPA, 2009; Farley, 2001; Hunaidi, 2000; Hunaidi and Chu, 1999).
Leak repair: New technologies enable rapid and accurate detection of leaks, but investing in rapid detection is futile unless repairs can be performed quickly. Repairs to pipes with holes generally involve either covering the hole from outside the pipe or inserting a smaller pipe inside the one that is leaking. The complexity and time for repairs varies widely, from one employee tightening a loose nut to large crews and excavators spending days repairing a deeply buried main. Of course, an analysis of repair costs versus replacement costs should be conducted for older pipes (Farley, 2001; Georgia Environmental Protection Division, 2007).
Water systems with intermittent supply cannot benefit from many of the methods for leak management and detection covered above. To compound this problem, leakage is even more serious in intermittent systems, where a residual chlorine concentration is not always maintained and infiltration is common (Hunaidi, 2000). High water pressure is required for leak detection equipment to be used effectively. Therefore, alternatives methods for leak detection in intermittent systems involve isolating a small zone of the network, closing the stop taps to customers, providing temporary water pressure to that zone, and then using conventional or modified leak detection methods. The basics of these methods can be found in the references (Farley, 2001).
Metered connections are essential for water conservation and they can also serve as a leak detection mechanism for pipe beyond customer meters. Customers will usually complain to the utility when they receive an unusually high water bill. Alternatively, an automated system to flag large increases in individual water bills or meter readings can alert utility personnel. A study conducted in the UK found that water consumption declined 10% following meter installation (Farley, 2001).
A warmer climate is highly likely to result in more frequent drought (IPCC, 2007). Additionally, growing population will push many countries into water stress and water scarcity in the coming decades. Detection and repair of leaks in water systems is an important part of comprehensive strategies to reduce pressure on existing water resources. Reducing water use in municipal systems also contributes to climate change mitigation by decreasing greenhouse gas emissions. Detecting and preventing leakage in piped water systems can lead to large savings in the energy used to transport, treat and distribute water (USEPA, 2010).
Increasing access to piped water at home leads to large gains in health and development. However, per capita demand for water increases rapidly during the development transition (Zhou and Tol, 2005). As population expands and water resources are stressed, economic development can be hindered (Gleick et al., 2002). Leakage prevention can slow the onset of water stress and preserve limited water resources. Additionally, these programs often pay for themselves through water conservation, reduced costs for treatment and distribution, and reduced maintenance and pipe replacement costs (see the section about financial requirements and costs below).
The costs of leak management, detection and repair include staff training, management, labor, and equipment. However, leak management, detection and repair programs generally pay for themselves by enabling early repair of leaks and reducing water waste (Georgia Environmental Protection Division, 2007). Leaks often damage pipes through erosion; therefore, additional benefits of early detection include reduced maintenance costs and lower probability of catastrophic failures. Monitoring systems remotely also enables confirmation that pipes are in good condition, preventing premature replacement (Thomson and Wang, 2009).
An extensive treatment of the costs and benefits of leak management, detection and repair programs is included in the WHO training manual (Farley, 2001).
The World Health Organization (WHO) training manual for leak management and detection is an excellent capacity building tool that can enable utilities to conduct their own high-level training at low cost (Farley, 2001). Leak detection technology is developing rapidly. Large utilities should have internal expertise in the technologies best suited for monitoring their system and emerging technologies from which they may benefit.
Institutional elements are largely governed by the perception of, and attitudes to, leakage and water waste within the water utility and political bodies (Farley, 2001). Within water utilities, an organizational climate of water conservation and financial sustainability can motivate employees to help reduce leakage. If water conservation is seen as a priority by the population, particularly under water stress or during drought, politicians who are made aware of the potential water savings may be more receptive to funding leakage management and detection programs.
Leak detection and repair can be undertaken in any piped water system. However, the technologies utilized for leak detection must be appropriate to the resources of the system. For community-managed and rural systems with above-ground pipes, detection and repair of these should be prioritized. Most small utilities should generally contract leak detection to a firm with appropriate expertise.
Motivation to prevent leakage may be low when water is inexpensive and abundant, and when water utilities are short-staffed or under-funded.
Opportunities for leakage management, detection and repair programs should abound when decisionmakers are made aware that the economic benefits often outweigh the costs. The economic benefits of these programs are especially great when: (1) energy costs for transport, treatment and distribution are expensive; (2) infrastructure is aging and leakage is high; (3) high-profile water main breaks lead to media attention and political pressure; (4) under water stress or water scarcity conditions; and (5) water conservation is valued.
Correlje, A.F., de Graaf, R.E., Ryu, M., Schuetze, T., Tjallingii, S.P., Van de Ven, F.H.M. (2008) Every Drop Counts: Environmentally Sound Technologies for Urban and Domestic Water Use Efficiency. Deflt University of Technology and UN Environmental Programme. Nairobi. http://www.unep.or.jp/Ietc/Publications/Water_Sanitation/EveryDropCounts... 
Fanner, P., Sturm, R., Thornton, J., Liemberger, R., Davis, S., and Hoogerwerf, T. (2008) Leakage Management Technologies. AwwaRF Report 91180. IWA Publishing. London. http://www.iwapublishing.com/template.cfm?name=isbn9781843398134 
Farley, M. (2001) Leakage management and control: A Best Practice Training Manual. World Health Organization. Geneva. http://whqlibdoc.who.int/hq/2001/WHO_SDE_WSH_01.1_eng.pdf 
Georgia Environmental Protection Division (2007) Water Leak Detection and Repair Program: EPD Guidance
Document. Accessed 01 December 2010 http://www1.gadnr.org/cws/Documents/Leak_Detection_and_ 
Gleick, P.H., Chalecki, E.L., and Wong, A. (2002) Measuring Water Well-Being: Water Indicators and Indices. In “The world’s water, 2002-2003: the biennial report on freshwater resources” Ed. Gleick, P.H. Island Press. Washington, DC.
Hinrichsen, D., Robey, B. & Upadhyay, U. (1997) Solutions for a Water-Short World. Population Reports, Series M. No. 14. Johns Hopkins Bloomberg School of Public Health, Population Information Program. Baltimore. http://info.k4health.org/pr/m14/m14print.shtml 
Hunaidi, O. (2000) Leak Detection Methods for Plastic Water Distribution Pipes. AWWA Research Foundation and NRC Canada. Accessed 01 December 2010. http://www.nrc-cnrc.gc.ca/eng/projects/irc/leak-detection.html 
Hunaidi, O. Chu, W.T. (1999) Acoustical characteristics of leak signals in plastic water distribution pipes. Applied Acoustics. Vol. 58:235-254. http://www.nrc-cnrc.gc.ca/obj/irc/doc/pubs/nrcc42673.pdf 
IPCC (2007). Climate Change 2007: Synthesis Report.
Maryland Department of the Environment (2002) Water Supply Program: Water Audit Guidance. Accessed 01 December 2010. http://www.mde.state.md.us/programs/Water/Water_Supply/Documents/www.mde... 
McIntonsh, A.C. (2003) Asian Water Supplies: Reaching the Urban Poor. Asian Development Bank. Chapter 9: Non-revenue Water. http://www.adb.org/documents/books/asian_water_supplies/ 
Pilcher, R. (2003) Leak Detection Practices and Techniques: A Practical Approach. Water21. IWA Water Loss Task Force, December 2003. pp. 44-45. www.iwapublishing.com/pdf/ACF1B18.pdf 
Smith, L.A., K.A. Fields, A.S.C. Chen, and A.N. Tafuri (2000) Options for Leak and Break Detection and Repair of Drinking Water Systems. Battelle Press. Columbus, Ohio.
Thomson, J. and Wang, L. (2009) Condition Assessment of Ferrous Water Transmission and Distribution Systems: State Of Technology Review Report. US Environmental Protection Agency Document EPA/600/R-09/055. http://www.epa.gov/nrmrl/pubs/600r09055/600r09055.pdf 
USEPA (2009) Pipe Leak Detection Technology Development. EPA/600/F-09/019 http://www.epa.gov/awi/pdf/600f09019.pdf 
USEPA (2010) Saving Water Saves Energy. Document EPA-832-K-08-001. http://www.epa.gov/WaterSense/docs/drops-to-watts508.pdf 
Zhou, Y. And Tol, R.S.J. (2005) Water Use in China’s Domestic, Industrial and Agricultural Sectors: An Empirical Analysis. Working paper FNU-67. University of Hamburg: Research Unit Sustainability and Global Change. https://www.fnu.zmaw.de/fileadmin/fnu-files/publication/working-papers/W...