Johannesburg has a water leaks problem. Here’s what’s being done about it
Study found more leaked municipal water in the Jukskei River than naturally occurring water
The headwaters of the Jukskei River in Bertrams. Photo: Sean Christie
- Hydrological research found the Jukskei River contains more leaked water from pipes than natural water during dry seasons, with isotope analysis confirming the source as municipal water.
- About 35% of water entering Johannesburg’s reticulation system is lost through physical leakage, contributing to river pollution and ecosystem disruption.
- Johannesburg Water has acknowledged the problem, launching a 2024 Turnaround Strategy to reduce water losses through pressure management, leak detection and infrastructure repair.
In the dry season in Johannesburg, there is more leaked water running in the Jukskei River than naturally occurring river water, a study has found. This raises troubling questions about the state of Johannesburg’s water reticulation.
The river’s identity crisis was revealed by hydrologist Simon Lorentz, who started analysing water samples from the Jukskei’s daylight point (the point at which the water emerges above ground for the first time) in 2018.
Lorentz subjected the samples to isotope analysis, a process that investigates the ratio of oxygen and hydrogen isotopes in order to determine, among other things, where the water comes from.
Lorenz found that the “isotopic signature” of those water samples was consistent with water from Johannesburg’s water pipes.
Here’s the technical description in the report:
“The source of water sampled from the Jukskei main channel discharge … is either a reticulated water or sewage water leak. Characteristic of reticulated water isotope signals, those sampled from the taps in the upper Jukskei catchment reflect an evaporated state, being positioned to the right of the Local Meteoric Water Line (LMWL Pretoria). This verifies the evaporation process in the supply dams and reservoirs of the reticulated water. The water sampled from the Jukskei main channel have the same evaporated signature.”
The source of the Jukskei: the Jukskei spring can be seen here decanting into a stormwater tunnel, which already has a much higher volume of leaked water running in it. Photo: Stuart Dunsmore
Another way researchers can tell that a lot of the water flowing in the Jukskei comes from unnatural sources is to look at the amount of water entering the Hartebeespoort Dam downstream. Water management specialist Stuart Dunsmore, who has been working on a catchment management plan for the Upper Jukskei, has done just this.
According to Dunsmore, the Jukskei covers 19% of the Hartebeespoort dams’ supply catchment yet it contributes almost 40% of its annual inflow.
“Almost two thirds of this inflow is wastewater from consumption in the Jukskei catchment. Some of it is outflow from treatment works, and the rest is raw sewage that leaks from the city’s sewer networks,” he said.
The same is likely true of the water in the Klip River, says Andrew Barker, Chairperson of KlipSA (Klipriviersberg Sustainability Association).
“At the end of a rainless winter the Klip River is a torrent, thanks primarily to discharge from the three massive water treatment works we have in the catchment [Olifantsvlei, Bushkoppies and Goudkoppies],” he said.
Unsustainable
Professor Johann Tempelhoff explained that Johannesburg was built on a watershed, the Witwatersrand, extending from west to east. Although the streams that come off it eventually join the region’s biggest rivers – to the north the Limpopo, to the south the Orange – their ability to meet the city’s water needs had already been exceeded by the early 1900s.
“To solve the problem, the Rand Water Board [the utility responsible for supplying the industrialising South African highveld with water] looked to the waters of the distant Vaal river, completing an impoundment called the Vaal Barrage in 1923, and the much larger Vaal Dam in 1938,” said Tempelhoff.
Today, all of Johannesburg’s treated water comes from the 320km² Vaal Dam, which is fed by a network of rivers and dams called the Integrated Vaal River System, which includes two inter-basin transfer schemes that feed water into the system from the Drakensberg and Lesotho.
“It is abstracted through multilevel openings in a concrete intake tower, piped to a treatment works called Zuikerbosch, and then pumped as many as 390 metres uphill, via booster stations, into concrete reservoirs built atop Johannesburg’s ridges and koppies, so that it can be released under gravity to homes and businesses,” said Tempelhoff.
Much of Johannesburg’s water infrastructure is old and due for replacement and a large volume of water leaks out of the system into city rivers.
“What you have is a situation where clean, potable water is imported from as far as Lesotho via the Vaal Dam, then mixed with effluent in Johannesburg and exported to another part of the country. It’s an unsustainable and deeply unjust situation,” said Dunsmore.
Given that more leaked water flows in the Jukskei River than naturally occurring water, it could be argued that the true source of the Jukskei is Katse Dam in Lesotho. Johannesburg receives all of its drinking water from the Vaal Dam, and the major source of the water in the Vaal Dam is Katse Dam. Photo: Sean Christie
The amount of piped water running in Johannesburg’s streams is a problem. University of Johannesburg-based water scientist Kyle van Heyde has been working on Johannesburg’s rivers for much of his professional life. He explained to GroundUp why the volume of leaked water running in Johannesburg’s streams is a problem.
“First, this is water that has been treated and pumped a long way at significant expense, only to be lost from the system. In a water scarce country, and in a city that struggles to consistently supply water to its residents, this is problematic,” he said.
“Second, it is well established that a lot of the leaked water in rivers is from the city’s sanitation system, which means the leakage is polluting the rivers and negatively impacting biodiversity and the ecosystem. Instead of coming from groundwater, the base flow in city rivers is coming from pipes, and this means there is a direct tie between the city’s water infrastructure and the rivers – when the infrastructure fails, the rivers fail.”
The health of Johannesburg’s rivers is tied to the city’s water infrastructure. When the infrastructure fails, the rivers fail. Photo: Sean Christie
A third problem is erosion and sedimentation. “One of the big things that we see in the Jukskei and other city rivers is a lot of sediment deposition, and a portion of this results from the fact that there is now significantly more water running in these systems as a result of leakage.
“Change in rivers typically occurs slowly, over hundreds of years, but in Johannesburg’s rivers we are now observing strange behaviour from a geomorphology and a hydrology point of view, linked to higher flow rates in the last several decades.”
Sedimentation is causing the city’s wetlands to rise, Van Heyde said.
“There is a bridge in Lenasia that crosses a wetland on the headwaters of the Klip River south of Johannesburg, and when it was originally constructed the water was several metres below the bridge but today as a result of sedimentation the bridge actually looks like it’s sitting at water level,” said Van Heyde.
The south-flowing Klip River, which flows through several important wetlands, is also likely to be carrying more treated water than naturally-occurring water. Photo: Sean Christie
Water losses
How much piped water is Johannesburg losing? At a roundtable discussion on 10 December, Johannesburg Water’s managing director Ntshaveni Mukwevho said the City has 46% non-revenue water, which is water that is put into the reticulation system, for which no money is recovered.
However, this does not mean that Johannesburg Water is losing nearly half of the water delivered to it.
“The two biggest components of non-revenue water are physical leakage, which is almost always the bigger share, and commercial losses, which can be anything from meter errors, to billing system issues, to reservoirs that overflow. The portion of non-revenue water that is physical water losses is around 35% [of the total amount of water that enters the system],” he said.
Water demand management specialist Ronnie Mckenzie, who brought the methodology used for calculating non-revenue water to South Africa in the 1990s, said Johannesburg’s water losses are not remarkable.
“Johannesburg water is not in a good position in terms of non-revenue water, but there are many municipalities doing a lot worse. A good target value for water losses would be 10–15%, however that can only be achieved through some serious interventions, which most towns and cities in South Africa just can’t afford,” Mckenzie said.
However, there are those that question the integrity of the data.
“There isn’t a huge amount of accurate data to work with, and so the specificity you see in reports around water losses is problematic. Instead of Johannesburg pegging water losses at 35%, it would be more honest to give a range, say 35-45%,” said Van Heyde.
Daryl Spencer, an instrumentation engineer and water metering specialist who has worked in the South African water sector for several decades, explained that most of the data utilised by water authorities and engineers to calculate system efficiency “is acquired by using water meters installed throughout the entire water reticulation system”. However, for the readings to be accurate, the entire reticulation system has to be fully charged with water all of the time.
According to Spencer, these fundamental metering conditions are not being achieved in the Highveld, where ongoing water shortages have resulted in continuous emptying and refilling of pipelines, resulting in air entering the system.
“Air in the system causes massive metering inaccuracies to both consumer meters and the bulk pipeline water meters, because most water meters have either a volumetric piston, or turbine, that rotates to measure the volumetric water consumption passing through the water meter. Air causes this device to spin uncontrollably, which massively affects the accuracy of water meter readings. Consumers who have air passing through their sales meters are unfortunately being billed for air,” he said.
Inconsistent meter readings is not the only consequence of air in the system, nor the gravest.
“When air enters the system you get water hammer, which is basically a spike in pressure in the system. If that air doesn’t find a way out it will blow a hole in the pipe,” Mckenzie said.
“As soon as you start putting air into your network, you’re on a downward spiral of creating leaks in your system, and if you move on to intermittent supply, your leaks will increase. If your leakage is at 20% at the start of the intermittent supply, it will be up at 30 or 40% after six months. It takes a day to introduce intermittent supply, but it can take years to get over the damage it causes,” he said.
The daylight point of the Jukskei. At this point, the river is mostly leaked sewage. Photo: Sean Christie
Reducing losses
What is being done to reduce water losses in Johannesburg?
According to Mukwevho, Johannesburg Water’s efforts to reduce physical water losses is guided by the entity’s 2024 Turnaround Strategy. Immediate interventions include reducing the pressure in the system overnight, improving the entity’s capacity to both identify and fix leaks, taking action to shut off illegal connections, and encouraging private-public partnerships to increase private investment and reduce non-revenue water.
“This strategy directly responds to years of under-investment, deteriorating infrastructure, and weak financial performance,” he said.
In response to GroundUp questions, Johannesburg Water spokesperson Nombuso Shabalala indicated that dealing with air in the system was a priority, both from a data integrity perspective, and to reduce physical water loss from damaged infrastructure.
Key to limiting damage to infrastructure is the repair and replacement of air relief valves that, when functioning properly, allow air to escape from pipes.
Some 600 air relief valves need replacing in the northern areas of the city alone, and Johannesburg Water is seeking external funding for this. The entity is also planning to reduce pressure in the system, using pressure reducing valves (PRV).
“The City of Johannesburg currently operates a total of 730 PRVs. However, some of these valves are bypassed due to vandalism or require some maintenance. To address this, the company has established a framework contract to procure spare parts, enabling its teams to repair and reinstate PRVs more efficiently,” said Shabalala.
She said Johannesburg Water is implementing a project to install smart controllers in areas where water consumption is high.
“Fifteen units were installed in the last financial year, with plans underway to install an additional 12 units in the current financial year,” she said.
To improve the integrity of data from metering, Shabalala said Johannesburg Water has tested “technologies that can negate the reading of air in conventional meters”, and technologists are “awaiting direction from the Operations and Metering and Revenue teams”.
Shabalala said the entity also plans to incorporate technologies that are not affected by air in the system, such as electromagnetic and ultrasonic metering devices.
There are spruits in Johannesburg like this one near Tembisa that reportedly stop flowing when the municipal water supply is shut off. Photo: Sean Christie
In this quote: “Change in rivers typically occurs slowly, over hundreds of years, but in Johannesburg’s rivers we are now observing strange behaviour from a geomorphology and a hydrology point of view, linked to higher flow rates in the last several decades.”
"several decades" originally read as "30 years".
© 2024 GroundUp. This article is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.
You may republish this article, so long as you credit the authors and GroundUp, and do not change the text. Please include a link back to the original article.
We put an invisible pixel in the article so that we can count traffic to republishers. All analytics tools are solely on our servers. We do not give our logs to any third party. Logs are deleted after two weeks. We do not use any IP address identifying information except to count regional traffic. We are solely interested in counting hits, not tracking users. If you republish, please do not delete the invisible pixel.