The city where I live estimates that we lose somewhere between 25% and 30% of drinkable water to leaky infrastructure.
We've had something close to a drought this summer -- unseasonally long periods without rain. You can see the young trees on the streets and trees in the middle of large parks suffer from it - wilted leaves and leaves dropping earlier than usual. BUT, large old trees seem to be thriving - full canopies, lush, firm leaves.
I've been suspecting the big street trees do so well because they benefit from the dilapidated state of our water delivery infra. It's nice to read of a study that confirms my amateur observations and musings.
Older, bigger plants have roots that go deeper and have access to more water. You can see the same effect in gardens, where new plants wilt sooner than established plants (and the care instructions advise frequent watering for the first couple of weeks).
Austin is close to 50% leakage. Now up to 110gal/day per connection in 2024.
https://www.kxan.com/investigations/city-of-austin-pipes-lea...
That’s really high. There’s either a big problem in your city, or they are making generous estimates to justify asking for more capital. 10-15% is more typical.
In my region, the street trees are usually getting sewer water. Residential service in older houses are usually clay pipes with lead solder that the tree infiltrates. It’s not a problem until the clay pops and roots clog it.
It varies a lot by region and jurisdiction. One of the cities near me made the mistake of using riveted pipe from rolled steel to save money 75 years ago, and regularly has catastrophic main breaks as the rivets aren’t as robust as a regular pipe.
OP numbers aren’t only a city problem, IIRC [0] the numbers are close here in France. There’s a startup that try to tackle it : www.leakmited.com/en I applied there 3 month ago and they never responded. Can’t blame them but I’m a bit sad: it’s the dream impact-job.
[0] 20% apparently https://www.eaufrance.fr/repere-rendement-des-reseaux-deau-p...
I wouldn’t be too sad. There are lots of companies offering ML leak detection. Most or many are using fibre optics and using them to detect changes in temperature or noise along the line. Sample the signal for period of time and run it through a model and you have leak detection.
My city loses about 30% of the water going through it's pipes, including leaking wastewater directly into rainwater drains ( left untreated).
I believe some of the plumbing was wood pipes in select very busy parts of the city until somewhat recently, as it was a nightmare to replace.
> as it was a nightmare to replace.
I suspect this is the problem with all aging infrastructure; 100 years ago or more, water, sewage and electricity was deployed everywhere, but then they just kept building and rerouting and now the systems have become unmaintainable.
I wonder if this has improved in any way since then in newly constructed areas / cities. I'm thinking a service tunnel underneath roads with all the pipes and lines clearly marked and installed in a highly accessible fashion. Said service tunnel can also be used for daring heists and escapes and the like.
Mexico City, by some estimates loses about 40% of its water that does enter its system, whether it's through leaky pipes or being stolen.
https://www.marketplace.org/story/2024/05/27/mexico-city-wat...
TIL clay pipes are a thing but it does make a lot of sense there would be.
These are probably the longest lasting option we have right now, and is far more eco-friendly than the closest alternatives, plastics or concrete.
Here in Germany, we estimate sewer infrastructure to last anywhere from 50-100 years, and water mains around 50-ish years. After that, it needs replacement or, that's the modern thing but it's a one-trick pony, re-lining.
The prudent thing would be to set aside and invest a tiny bit of money every year to fund a replacement, but unfortunately modern economic theory ("run lean") and manufactured income crises (aka, politicians going for lower taxes and utility rates) have led to a lot of infrastructure being utterly dilapidated and no savings left, and now we need to invest untold billions of euros raised from debt to keep it running.
Unfortunately, a lot of the deciders are already dead, and for those that still live, it's fallen out of favor to hold them accountable.
Smart is to have a crew that replaces a little every year. That way they build expertise in how to do it and there isn't a large expense all at once. You can likely get a discount with private plumbers because you want it done sometime and so they schedlue around other customers who want it now.
That doesn't track because the real cost to replacing underground infrastructure is not the digging, materials, or labor. We avoid such maintenance as long as possible because shutting down a road is usually very expensive in terms of second-order effects.
Digging up a pipe and replacing it is actually pretty cheap and easy. Disrupting a main thoroughfare is incredibly expensive in terms of lost productivity, transport, shipping.
This whole thread reminds me of the Edinburgh tramworks; despite there having been 19th-century trams on the same street, the modern trams are heavier. So the tram building project turned into a saga of underground "technical debt": sewers had to be rebuilt to take higher weight overhead, which meant digging up everything buried in the road, all the pipes and wires, plus extra bits of archaeology. Hence massive cost overruns.
It is not cheap nor easy. For many of those pipes, especially yhe waste water ones, there are no AS-BUILDs, and they are unmapped in a very real sense.
This just sounds like a different way of saying "one more lane bro, trust me". Usually one doesn't "shut down" a road, but divert lanes, perform construction outside of peak hours (e.g. at night) etc.
Finally, some people may even consider the construction too much of a hassle and use the car less. Empirical studies suggest that "removing" a road does not cause productivity loss overall, since having the road in the first place induces a lot of "non-essential" demand.
That arguement is stupid. the world is not only about productivity. If people want to do something just because they enjoy it then they should. If the lack of a way means they do something else that is a loss. The whole point of Cities is all the options, if you want lack of options there are a lot of small towns - and even they have more options than you want to allow.
now it need not be by roads - a great transit system should enable moreeobtions. However the point is all the things you can do if you choose not a train or road to nowhere.
there are enough roads in any city as to rebuild a few every year. Pipes don't last as long as pavement in general.
Pipes have lifetimes of 2-10x roads... where's your data from?
Roads[0]: Asphalt (18 years), Concrete (25 years) - requires good expansion gaps, good substrate, zero roadwork over its lifetime.
Pipes[1]: HPDE (50-100 years), PVC (50-70 years), Reinforced Concrete (75-100 years) Vitrified Clay Pipes (Several centuries), Galvanised Steel (40-70 years)
[0]: https://www.ayresassociates.com/the-long-and-short-of-it-lif...
[1]: https://trenchlesspedia.com/the-lifespan-of-steel-clay-plast...
A city could replace pipes preventatively as part of road resurfacing when the pipes are sufficiently old.
Not that easy. A road resurface, that can be done in a few days worth of work, less if you hire enough machines and staff.
A pipe replacement? Sewer mains at least here in Germany tend to be anywhere from 2 to 8 meters below ground. That's a lot of soil to move. Freshwater mains is below freezing depth, so usually around 1-2 meters below ground. And above that is a ton of other wiring... electricity, phone, fiber, cable tv, gas and district heat/cooling just to name a few, so when you want to replace the sewer mains, it involves a lot of companies, plus the city authorities for coordination, permits, traffic re-routing (a bus route is bad enough - a tram line or a legit full size train line is a nightmare).
Outside of immediate emergency work from a burst pipe, replacement works take years to plan.
In Los Angeles, they have basically given up on doing any preventive maintenance of the water/sewer lines. They just wait for something to break and replace it. We have areas with pipes over a 100 years old. Too expensive to dig up huge areas of the city to replace the lines all at once. This may also have something to do with the notoriously corrupt LA Dept. Of Water and power too. Other cities in LA County do periodically replace the lines, but it's still tremendously expensive.
Sidetracked by the nominative determinism in the article (researcher André Poirier's surname means "pear tree").
Or as Anthropic's Safety team would write: "Trees conspire to take down human race by sabotaging the underground water network."
> Maple trees need to consume around 50 litres of water per day. Since street trees can’t get much of this from rainwater, which falls on concrete and drains into the city’s sewers, Poirier says the most likely explanation is that it is coming from Montreal’s leaky pipes, which lose 500 million litres of water per day.
I feel like this is burying the lede.
What can be done to reduce leakage?
>Since street trees can’t get much of this from rainwater, which falls on concrete and drains into the city’s sewers, Poirier says the most likely explanation is ...
I feel like this is burying the lede. What can be done to reduce leakage?
This is hardly a pipe dream. Village Homes demonstrated the concept[0] over five decades ago.
While there are some interior pipe lining solutions, they are just kind of a short term bandaid that will cost more in the long run and has its own complications, the only real answer is to bury brand new pipe. And that can get expensive if a lot of stuff has been built on top and around it which is why most municipalities just live with even major leaks and kick the can down the road until it breaches the surface and starts flooding areas. It is probably cheaper to replace them sooner rather than later in the long run, but no politician cares about how well a city's finances will be decades after they are gone.
Hopefully the lede, lead would be an even bigger concern.
The lede rendering is optional:
>The spelling lede (/ˈliːd/, from Early Modern English) is also used in American English, originally to avoid confusion with the printing press type formerly made from the metal lead or the related typographical term "leading".
Hah, that was an unfortunate auto-correct. Fixed!
Montreal has plenty of lead pipes too. ;(
> What can be done to reduce leakage?
Do we really want that? Thousands of people are being killed each year by heat strokes. Keeping those trees alive by its environmental services is much more valuable in terms of lives and also energy saved. Maples have soft big leaves but also reduce the asphalt temperature by 5-10 degrees. If required just plant a tree species that can live with less water.
Yeah let's wilt all those pesky maple trees in Canada
Why should anything be done to reduce leakage? They take water out of the St Lawrence and, as much isn't diverted by trees, it goes back in (cleaner then when it came out.)
Because trees will clog those pipes eventualay and then we must dig them up. Pipes that don't leak don't attract roots and so last longer
Not only trees in cities do that. A lot of clogged home sewers are caused by trees that wanted more to drink. Once the sewer line is fully blocked, they've arrived in paradise. Now there is a constant supply in the permanently filled sewer line.
I use to live on a property in a development. On my property were a pair of big willows. The trees were large and healthy, well over 50' tall. That was strange, because the region is high desert with little water, and I made no effort to irrigate them.
One morning in spring, after I'd been living there about 15 years, the neighborhood streets flooded. There were geysers of water shooting up from manholes. Turns out, the willows had been planted over an irrigation ditch[1]. The willows had driven their tap roots into the pipeline and plugged it about 10' underground. When the water authority opened gates miles upstream, the water pressure blew water up the manholes into the streets and a few yards.
Farmers are very motivated to get their water. They, and the ditch company, rapidly cleared the plug and removed the trees.
Before this happened I had no idea the irrigation system ran through the property. I knew about an easement, but I thought it was for sewage, because the developer used an iron manhole cover from the local municipal waste management operation when they covered the irrigation ditch: it literally had "Sewer" cast into the iron.
[1] Formerly an actual ditch, later made into a pipeline and covered over, but still technically a "ditch" for purposes of water management.
Or at least, constant supply for the several hours it will take to call a rooter company and clear the drain line. Ironically, the tree would have better results if it only partly blocked the drain. I wonder if trees might ever evolve to strike that kind of balance, or if there's not enough selection pressure for that to happen.
Watch out, some tree roots may be stronger than the rooter used to clear them, in which case the rooter company may end up with a rooter trapped in your pipe, as happened to my parents. Ultimately they had to get the sewer line excavated and replaced where it met the city sewer, 15 feet down in the middle of the road. Cost like $30k.
Maybe ask the rooter company what happens if they end up with equipment trapped jammed down your pipe is all I'm saying.
That to me seems like something that should be covered by the company's insurance.
Any services company that comes out, ens up with breaking their tools because they used inadequate tooling, and causing more damage? I don't know how they managed to foist that on to your parents.
The issue wasn't the tree roots, it was the rooter company's poor investigation. Video scoping a sewer line is trivial these days.
Oh, they video scoped the sewer, and my parents (father is a lawyer) went after them. Ultimately, the rooter company decided they'd rather go for the full legal battle and my parents decided "eh, we know how awful a legal battle is, we'll back down and not sue."
Not just leaky fresh water pipes, but also sewer.
Or if you're a willow tree you make a leaky pipe.
We had a lemon tree that did this. The irrigation line connector was probably not 100% sealed and the roots grew to it slowly broke it. It enabled the lemon tree to gets lots of water and grow. Meanwhile the trees further down the irrigation line suffered.
Our pines had fun with our sewage back in the day...
Some genius planted a curly willow right in the middle of my house’s septic drain field. That tree cost me $20,000.
Reminds me when I installed a drain field for a guy that had to have trees in the woods cleared and dug up for space and everything pumped 100 feet up a hill to it due to the terrain. Just as I was finishing up he asked when he could park his 40 foot boat on top of it and was outraged when I said he can't/shouldn't do that, he then declined the offer to have us clear more area before we loaded up all the equipment. All I could imagine as I drove away was the river of shit flowing back down the hill to his house in a few years.
You’re not supposed to plane ANY trees in the middle of the septic system.
You're not. But people often don't even fully know exactly where the limits of their septic system (including leach fields) are. SUPPOSEDLY they know where their septic tank is but tree roots grow, records are lost, etc.
Same happened to me with a hackberry I didn't know was wild. Some genius added landscaping to make it look like a choice. You got off easy, I think after all was said and done I paid about the same out of pocket, but also was displaced for a long time due to the damages, and it cost insurance another $80+ grand that I'm sure I'll pay back over time in raised premiums.
Fun!
> While the park trees contained lead isotopes normally associated with air pollution, the street trees had isotopes found in lead water pipes, which were made with metal from geologically old deposits in nearby mines.
I don't understand this part. We didn't use different sources of lead to make leaded gas and lead pipes, no?
Tetraethyllead production was very centralized by Ethyl corp/DuPont and required a higher purity lead ore so their isotope ratios are very well known based on the deposits that they mined. More locally sourced lead used for construction will have different isotope ratios.
When you need a lot of lead (enough to build plumbing for a neighborhood), you probably want to source it locally. When "1 part TEL to 1300 parts gasoline by weight is sufficient to suppress detonation",[1] you can source the lead from just about anywhere and ship it with the fuel.
The radioactive lead isotopes come from decay of uranium and thorium, so lead from different mines will have different isotope ratios depending on how much U and Th happened to be in that ore.
Not all leaded gasoline was the same either:
> 206Pb/207Pb ratios commonly found in Pb ores throughout the world range between 16.0–18.5 and 1.19–1.25, respectively (Hansmann and Köppel, 2000). Exception to this rule is the commonly used Pb ore from the Broken Hill deposit, Australia, which is characterised by extremely low 206Pb/207Pb ratios (1.03–1.10). On the other hand, Pb originating from the Mississippi Valley ore deposit, USA, exhibits significantly more radiogenic Pb isotopic composition (206Pb/204Pb N20.0; 206Pb/207Pb= 1.31–1.35) (Doe and Delevaux, 1972). American leaded gasoline reflected therefore significantly higher 206Pb/207Pb ratios compared to European gasoline (Fig. 1). The introduction of the European leaded gasoline around 1945 resulted in a steep decrease of the 206Pb/207Pb ratio of atmospheric Pb (Weiss et al., 1999; data from peat deposits). The isotopic composition of leaded gasoline was to some extent dependent on economical factors, such as the availability and price of Pb ores and has evolved due to the different Pb ores used. For example, Pb used for French leaded gasoline originated from Australian, Moroccan and Swedish ores and the contribution of the separate ores changed during time (Véron et al., 1999). It is therefore indispensable to gather data concerning the origin of gasoline used in studied regions.
[from https://sci-hub.ru/https://doi.org/10.1016/j.envint.2007.10....]
We didn't put elemental lead in gasoline, it was a very different molecule with a single lead atom. Given how dense lead is, you want to source it from as physically close as you can. A foundry making pipes in a city with a lead mine nearby will obviously use the local lead.
For gasoline, all production had to be centralized in a few refineries. The lead would have been shipped in, and would have been largely the same quality and age, likely coming from the same mine, or geographically close mines. Plus the absolute quantity of lead added to gasoline is relatively small. In the 60 years the US used TEL, we processed about 8 million tons of lead. Averaged out, it's 133 thousand tons a year. It would only take a few mines to provide that much. Probably not more than five or ten, but I can't immediately find good data on this.
One would expect that the lead used in gasoline is pretty homogeneous across time, and that intensive lead use (as in casting into solid metal object like pipes) would use the nearest available source, and use that source for as long as possible.
lead from pipes was mined localy, but the lead in parks soil is from airborn pollution and so the isotope signature will be quite different
I wonder how many of the pipes are made of wood. I forget the source, but I heard a decent number of pipes in Montreal are very old and made of wood (which is better than the proliferation of lead pipes that are still being removed)
One of the houses across the street from me had its driveway dug up, all the way from the street to inside the garage, for what I first thought was for putting a bathroom or something in the downstairs level. When I walked past the old pipes they took out they were full of roots.
Not really mentioned, but I suspect that at least part of the time, the leaky pipes are caused by the trees. I used to own a house in southern California with a sprinkler system and a couple Chinese elms. Trying to track down why some sprinkler heads weren’t sprinkling, I discovered that the trees had sent roots into some of the (PVC) piping, going as far as six feet of the pipe. I ended up digging up and replacing a lot of piping on that project, although I wouldn’t be surprised if the Chinese elms (which are nasty nasty trees that I will hate until I die) didn’t pull the same stunt on subsequent owners of the house.
This is not news. Having to rebore pipes due to tree roots has been around for decades
Heh. Yesterday, we had a plumber over who told us we have to rebore our sewage pipes because roots got in. It is an old house with cast iron pipes and they still got in.
Cast Iron’s worse than plastic because they always leak a tiny bit and that means that the roots can “smell” the water and go for it.
Plastic either is impervious or completely fucked.
I get tree root intrusion where my newish plastic sewer main joins the street sewer.
Yeah joints are always the problem.
Which is why you want the joints to be someone else’s problem.
I've never heard it called "reboring" - wonder if there is a different procedure for when it gets really bad, but I'd have thought problems (backup) would happen pretty quickly, so wouldn't be too bad as long as you take care of it.
They basically use something like a weed whacker fed down the pipe, except it uses a short bit of chain instead of trimmer line, and will pulverise any intruding roots.
There are at least two techniques: sewer rooting (low velocity and high torque) and sewer de-scaling (opposite).
Decades? Try millennia.
there is a new to me datum in that trees along residential streets are experiencing less water stress than trees in parks, due to city water leakage that was demonstrated by doing core samples on the trees to show how lead isotopes differed in the two populations of trees. it highlights a growing concern with water in general and how carefull water monitering and management is becoming, and how what was primarily interesting to civil engineering types, has a wider audience
I wonder how much human health is impacted by these leaky pipes.
I would like to see a city where pipes are guaranteed leak free, for example by making them double walled with high pressure air in the outer layer, and then seeing if disease levels in the city are lower.
That double wall thing is a red herring. The water system already protects against intrusion because the water itself exerts pressure on the pipes. Thus leaks are typically of the water getting out and not contamination of the water (most of the time, anyway.)
Contamination rarely happens outside of the source of supply, and not somewhere along the pipeline.
And the sewage pipes...?
Water pipes are under pressure. So outside water and pollutants do not infiltrate into leaky pipes. Unless you have a water shutoff. But those situations are minimal.
City department of water and power was doing some unrelated sewer work but ran into a problem, and shut off water in my neighborhood. No alert to warn of the shutoff, or an alert when it came back on. I found out because I worked from home and noticed the tap wasn’t working and went out to ask the workers on the road if they shut the water off. Chances are virtually the entire neighborhood besides myself and another curious local pedestrian did not flush their taps after that loss of pressure no doubt induced contamination (most of the housing stock is 100 years old so runs into homes are probably in terrible shape).
Globally they're definitely not minimal, nor are they unlikely even in developed countries.
Alameda county does something similar for health reasons. All home sales require pressure testing the sewer lateral. With replacement required if it fails before the deed can transfer or a loan is funded.
Does the buyer or seller pay for the repairs?
In practice the seller. Because they can’t sell to anyone without fixing it. But it’s negotiable.
> I wonder how much human health is impacted by these leaky pipes.
Not much, because the water mains pressure keeps nasty things from entering the pipe.
However, when the system is depressurized due to a power outage or due to running out of water, nasty things can happen (stuff entering the water pipes, oxygen from air bubbles causing rust), and that's why after such events boil-off orders are issued for a few weeks afterwards until it can be reasonably assumed that all pipes have been flushed and all air bubbles have gone.
No warning went out when they shut my water off most recently. Certainly no boil off advisory.
Those leaks are 99.9% one way.
That way being 'sewage leaking into the ecosystem '?
Followed by ecosystem being collected and put back into drinking water, most of which only has pretty lightweight treatment which doesn't even involve testing for any viruses which have snuck through.
The pipes are pressurized, so I would expect there to be limited avenue for infiltration. (Also, for sewage exposure, you’d need two leaks close together. Not impossible or anything, but much less likely.
Most of the world relies on drinking water collected from rivers downstream of other inhabited areas, so the impact of pollutants like sewage entering the watercourse does not have to enter a leaky water system to have a disastrous impact.
Most of the time the local soil is well able to purify the small amount that leaks. Most is key, see and expert on your local conditions for details.
For what it's worth, utilities do care quite a bit about leaks (it's service they are providing which they can't bill for!) and use various testing apparatus to locate leaky parts of underground systems for repair and replacement.
Considering the difficulty and cost of repairing underground anything, most of which will be there for many many decades, it's never going to be perfect, but there's a lot of resources that do go into improving this.
Why can't they bill for it? It's not like they are losing money on it, it's simply getting priced into the billable services they provide. Utilities are usually monopolistic, so there is little incentive for them to fix this.
They do, but rates are regulated, they can’t just tack on this month’s pipe replacement expenses. They have to make a guess, propose a rate increase, and get it approved by the relevant regulator or government authority.
So they cut corners in other ways?
> so there is little incentive for them to fix this.
This kind of stuff is typically death by a thousand cuts.
Add on that a lot of the places it leads are under roads that will have to be shut down for weeks/months and you start to realize the costs and impact of fixing these leaks are enormous.
If it is a local leak they can fix it in a few hours but most often the whole pipe leaks (or maybe every joint) and so the whole road needs be redone - thus it is worth waiting for the road to wear out.
Fundamentally it is getting priced into services provided to residents as a whole. But you're also forgetting water is generally not provided by a business: Local government is not a profit enterprise, and generally has a lot of pressure to reduce (or limit the rise of, anyways) the bill.
I can tell you factually a lot of work goes into measuring leakage, narrowing down what part of the water system it is coming from (most active components are metered in some way, and you can use math to determine where all of the water is not making it through a segment), and correcting those issues where it is cost-effective to do so.
For profit water companies are common. Municipalities do not have the capital to replace infrastructure, so private companies like American Water buy the pipes and plants, make the minimum fixes, and jack up rates to pay back investors.
We already have this data in a way, from cities where there is no running water and people rely on bottled water for drinking and washing.
That’s not biased at all.
500 million liters of water a day lost to leaks!
For those wondering how big this is, that's about (80 meters)^3.
Cope? Won't they be drinking from leaky pipes regardless of a drought?
I DONT CARE