Go back to main page


Posted on 28 September 2016 by George Kunkel, P.E
Posted in FluksAqua Community, FluksAqua insights, Water and wastewater community,

North American water utilities have been successful in delivering safe, reliable water service that has allowed communities to grow and prosper.  Treated drinking water is delivered to customers through a vast network of underground piping systems.  Unfortunately a significant portion of this water never reaches customers since it escapes the system in the form of leakage.  Water utilities have been increasingly employing traditional leak detection technologies to detect and pinpoint individual hidden or unreported leaks and target them for repair.  This work is very necessary and all water utilities should have such a program in place.  However, most water utilities conduct leak detection intermittently, patrolling portions of the system once every 1-5 years.  This method misses newly emerging leakage that occurs between the leak detection intervals. Additionally, detection of individual leaks alone is not a fully sufficient leakage management approach since a portion of leakage – tiny weeps and seeps known as background leakage – cannot be detected via traditional means.


The most strategic approach to managing leakage relies on optimizing the three time periods that define the life of a leak: the Awareness time (time to become aware of the leak), the Location time (time to deploy personnel to pinpoint the leak), and the Repair time (time when water supply to leak is shut-off in advance of the repair). Traditional leak detection has shortcomings in that it cannot find all leaks in the distribution system, cannot provide a volume quantity to the leakage, and cannot optimize the awareness time since patrols occur intermittently, and new leaks are always emerging.

The best practice approach to optimizing leakage awareness time involves tracking flows in District Metered Areas (DMA) in the water distribution system.  A DMA is a small zone created by closing distribution system valves to form a boundary, while leaving at least one measured (metered) supply pipeline open to send water into the DMA. The normal flowrates into the DMA should be sufficiently small that the emergence of a new leak produces a discernible increase in flow that alerts the operator to newly forming leakage.  This allows the municipality to then strategically send leak detection personnel to the DMA to pinpoint and repair the leak very soon after it emerges.  Conversely, when input flows are in a normal low range, no leak detection crews need be deployed to the DMA, and wasted crew time is avoided.

DMAs don’t pinpoint specific leak sources; but instead serve as any early alert of leakage occurring in a specific area of the water network.  DMAs give utilities the ability to:

  1. Become aware of leaks shortly after they emerge, thereby minimizing the awareness time;
  2. Deploy leak detection personnel strategically in areas of known emerging leakage; and not deploy them where routine flowrates are indicated;
  3. Place a volume quantity on the bulk leakage in the DMA, and the ability to conduct testing to measure the amount of background leakage vs. unreported leakage;
  4. Drastically improve the efficiency of narrowing the location of leaks within the DMA, by successively closing portions of the DMA at night;
  5. Function dually as a Pressure Managed Area (PMA) where advanced pressure management methods can be used to better control areas of excessive and/or erratic pressures, and control high background leakage losses. A PMA boundary may coincide with a DMA boundary, or it may be a separate design.


The best example of effective leakage management using DMAs is the experience of Halifax Water in Halifax, Nova Scotia, Canada.  For over 16 years Halifax Water has optimized leakage management by having a sectorized distribution system of 75 DMAs, a number of which also function as PMAs.  Over this period, leakage has been reduced by 35 mega-liters per day (10 million gallons per day) at a savings of $CAN 650,000 annually.  Halifax Water has also realized enhanced benefits of more and better data, insightful information on the system, enhanced emergency capabilities; and has used their leakage management program as the focus of a utility-wide change management approach.


DMAs must be carefully designed in order to meet the leakage management objectives and not create unintended consequences such as insufficient pressure or compromised water quality.  The first step is to identify areas of high leakage, high pressure, and frequent water main breaks (bursts); these conditions will benefit from enhanced leakage management.  A typical DMA should generally encompass from 1,000 to 3,000 customer service connections; sometimes with a higher number in dense urban settings, and fewer in rural settings.  Look for advantageous locations for DMA boundaries to be created and/or work with existing “natural” boundaries where the distribution systems is interrupted by a major highway, railroad, park, or other notable feature.

District Metered Areas (Farley, 2001)

The design should ensure the ability of flows to meet normal and peak customer demands, as well as emergency demands from fire-fighting flows or high flows incurred during water main breaks.  Good water circulation should be provided within the DMA in order to prevent water quality degradation.  Flow, pressure, and water quality data – along with leak and main break data from recent years – should be gathered from within the target DMA region before the DMA configuration is put into place.  This data serves as the baseline measure of the “before” conditions that should be compared with similar “after” conditions data gathered once the DMA is in place and operating.

A minimum of one flowmeter must be installed and other equipment (pressure sensors, communication equipment) will likely be desired depending upon the specifics of the design.  If a wide variation in the normal vs. emergency demands is expected, a second supply feed can be designed for the system in order to supplement supplies to meet the occasional emergency flow.  This feed can be left unmetered and outfitted with a pressure reducing valve (PRV) that remains closed under normal conditions.  The PRV should be set to open only when a low “trigger” pressure level is detected; a pressure that would stem from a high emergency water demand on the system.  The PRV will close once flows return to a normal condition.

Once the DMA is established, testing should be conducted to ensure that all boundary valves are closed and “holding” water tightly.  Valves that don’t hold tightly or “pass” water compromise flow measurements as an unknown volume of water will flow out of the DMA.  Flow and pressure data should be continuously monitored by a SCADA or similar system and the data analyzed continuously to observe trends.  Under normal demands with minimal leakage conditions a low flow profile of water demand should be observed; but, as new leakage emerges, flows gradually rise.  It is important that the water operator have a reliable means to identify this subtle change in water demand as newly emerging leakage.  Utilities should employ an appropriate analytic software to continuously monitor and assess flows relative to the baseline normal flows as a means to automate the leak alert process.  Reliable analytic software – such as the FluksAqua Leak Detection App – becomes essential when a water system is sectorized into many DMA’s and multiple flow patterns must be analyzed on a continuous basis.


A vast amount of precious drinking water is wasted through utility leakage every year. Utilities should use traditional leak detection technology to pinpoint leaks and target them for repair.  However, leak detection alone is not a comprehensive leakage management strategy since it has a number of limitations.  Utilities can strategically manage leakage by continuously monitoring flows into individual District Metered Areas and employing analytic software such as the FluksAqua Leak Detection App in order to identify areas of emerging leakage, and deploy leak detection staff in an optimized manner.  This saves crews and equipment time and reduces the utility’s treatment and pumping costs by supplying only the water that the customer population needs.


About the author

Mr. Kunkel is Principal of Kunkel Water Efficiency Consulting, a firm that

specializes in water loss control in drinking water utilities. He has over

35 years of water utility and consulting experience and led the

successful water loss control program in Philadelphia for over 20 years.

He is active in the American Water Works Association, having served in

multiple roles, including chair of the Water Loss Control Committee.

Mr. Kunkel is currently the chair of the sub-committee that publishes

AWWA’s M36 guidance manual on water loss control and is a co-author

of the AWWA Free Water Audit Software. He has been involved in a

number of water loss projects in AWWA and the Water Research

Foundation, and was the recipient of the 2010 Water Star Award from

the Alliance for Water Efficiency.