Winter weather threatens London's leakage targets
The prospect of a colder than average winter has fuelled Thames Water's research into understanding the effects of cold water on its water distribution network in south-east England. Andrew Boyd of RWE Thames Water reports on the range of strategies under consideration.
In order to avoid a repeat of the last cold snap in February 2005, Thames Water is putting staff on standby to deal with the leaking pipes fractured by plunging temperatures. The company was criticised for missing leakage targets it was on course to meet when temperatures dropped harshly, leading to a sharp surge in burst pipes and a rise in leakage.
"Cold weather is a real headache for us, particularly in London," says David Dangerfield, Thames Water's operations manager. "Temperature fluctuations cause the capital's clay soils to shift, inevitably fracturing our water mains.
"We're determined to bring leakage rates down, so we will have extra gangs available during the cold spell to fix leaks as soon as possible, wherever they occur in our 32,000km network," he added.
A percentage of water is lost in transit from treatment pipes to consumers in most water distribution systems and leakage is the primary cause. However, for Thames Water leakage is exacerbated by an ageing pipe network - half the pipes are over 100 years old and a third are over 150 years old - buried deep below modern-day London.
The company's scientists are carrying out research to better understand the causes of such a temperature-inflicted rise in leakage and how it can be pre-empted, looking at the parallels between the cold weather and burst pipes. Recently, researchers have turned their attention to the issue of water temperature in order to understand the effect it has on leakage.
The amount of water needed to supply Thames Water customers follows a complex but predictable pattern throughout the year. The company has developed models that use a five-day weather forecast - the longest predictions for air temperature currently available - to predict water demand throughout summer and winter. However, during extreme weather conditions, other inter-relationships must be taken into account; as temperatures fall, there comes a point when the correlation between temperature and leakage breaks down.
Mark Simister, leakage strategy manager at Thames Water, says, "We obviously expect that winter weather will lead to an increase in leakage and can quantify the difference made by 'normal' winter temperatures. But we are working hard to better understand the changes brought about by more extreme conditions, such as those we saw in February 2005.
"Up to that point, the weather had been getting gradually colder. There was no rapid fall in temperature from one day to the next, but we suddenly reached a point where leakage rose dramatically.
"In the space of about two weeks, the leakage rate grew by over 13%. Until this point, we had been making good progress on tackling leakage and were on course to meet our target for the year. But the sudden increase meant we narrowly missed it," Simister explains.
Thames' overall leakage for the financial year 2004-05 was 915ML per day, across the whole region, against a target for the year of 905ML. In other words, Thames Water missed it by 10ML. However, Thames Water achieved the target set in the Thames Valley area outside London; here, the target was 115ML per day and 112ML was achieved. The overall target set for 2005-06 is 860ML per day.
The challenge for Thames Water scientists is to try to mitigate extreme weather conditions when they happen. Thames Water research and development director Tony Rachwal points out that, "The role played by water temperatures is very complicated and one that is little understood across the water industry.
"It might be tempting to think that, during the coldest weather, the metal pipes simply shrink and burst, but this is a over-simplistic approach to a much more complex system. At present, the five-day weather forecasts we monitor allow us to plan ahead to some extent and to increase the amount of water put into supply to match an expected rise in leakage."
Rachwal continues, "Ideally, however, we need to be able to predict exactly when the problem is likely to start. In other words, we need to know why the normal 'seasonal' leakage we expect to see in winter suddenly rises.
"One obvious solution would be to heat the water, but the amount we supply to London on a typical day is huge. Water has massive 'specific heat capacity' and that means that the amount of power needed would be enormous."
The Thames Water team has looked at whether a reservoir of warm summer water could be stored for use in the coldest weather. Evidence suggests there could be a beneficial effect during a short, extreme cold snap if a deep-water reservoir was used in this way.
However, current conditions of low rainfall compromise the benefits that could be outweighed by the effective loss of water storage as reservoirs are filled quickly to cater for 11 consecutive months of below-average rainfall. Further ideas concentrate on using warmer water from different depths in a reservoir employing stratification whereby water separates into layers at different temperatures.
While stratification occurs easily in the summer, during the winter, the wind circulates reservoirs enough so that that stratification only happens once the ice has formed on the surface. Under such conditions, water at the bottom of a reservoir, although very cold, is typically a degree or so warmer than that at the top.
Using this water for supply purposes could help reduce the maximum number of burst pipes that occur during the coldest weather. The team are also looking to take advantage of the fact that stratification may begin to take place when temperatures rise at the end of the winter, which could mean that the warmer water is instead found at the top of reservoirs.
A further approach being explored by the scientists looks at the possibility of using more groundwater, as its temperature is typically higher than the water taken from rivers. Tony Rachwal says, "A lot depends on the ambient temperature but surface water is typically from 4 to 7°C during a cold winter while water taken from wells is usually around 10°C.
"We have looked at taking winter supplies from our unique underground 'droughtbuster' in north London, where we can store surplus treated water in a natural aquifer and pump it out when we need it. The problem with this is that, while it is stored in the aquifer, the water loses some of the air dissolved in it.
"The water would be slightly warmer but it would cause problems when we feed it into our local works in Hornsey, as it would affect the bacteria used in the biological treatment process. A more viable concept would be to manage the balance of ground and surface water used in areas that are normally supplied by both.
"At present, we have looked at just one area in north Kent, where the design of our water supply network means we could potentially switch off a surface water supply and instead feed the area with groundwater. Unfortunately, the local geology which gives us access to the warmer groundwater also means that the local ground conditions are less 'aggressive' and therefore less likely to prompt leaks and bursts.
"We have been looking at the number of bursts and leaks that typically take place in that area during very cold weather to measure the difference that the warmer water might make. It is important to stress that we are only in the early days of researching this, but reducing leakage is a major challenge for the company and we are therefore keen to explore all avenues," Rachwal said.
Operations manager Dangerfield points out, "The last year has seen 11 consecutive months of below average rainfall, so it's important for us to save water where we can."
The prospect of long, dry summers, combined with freezing winters, are key drivers to stemming leakage. In addition to investigations by scientists, Thames Water's sustainable approach to combat leakage includes investing in a major pipe renewal programme for the capital, and, since work began three years ago, it has replaced more than 320km of Victorian water mains in 14 London boroughs at a cost of more than e170 million.
Plans are in place to invest over e740 million to upgrade a total of 1370km over the next 4 years. "In the meantime we must encourage consumers to use water efficiently and match their efforts by continuing to reduce leakage," says Rachwal.