62 Years in the Making: Nyc's Newest Water Tunnel Nears the Finish Line
Key topics
As New York City's Water Tunnel #3 nears completion after 62 years, commenters are abuzz with fascinating facts and pop culture connections, like its feature in the movie Die Hard. The discussion takes a technical turn as users ponder the tunnel's 800-foot depth, with some speculating it's necessary to maintain a downhill gradient for water flow. One commenter cleverly compares this to the gentle gradient of rivers like the Mississippi, highlighting the tunnel's impressive engineering. The thread is a great blend of nostalgia, curiosity, and technical insight, making it a captivating read.
Snapshot generated from the HN discussion
Discussion Activity
Very active discussionFirst comment
25m
Peak period
59
0-6h
Avg / period
15.4
Based on 108 loaded comments
Key moments
- 01Story posted
Dec 28, 2025 at 6:05 PM EST
4d ago
Step 01 - 02First comment
Dec 28, 2025 at 6:30 PM EST
25m after posting
Step 02 - 03Peak activity
59 comments in 0-6h
Hottest window of the conversation
Step 03 - 04Latest activity
Dec 30, 2025 at 10:21 PM EST
2d ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.
> The Bronx and Manhattan already receive water from it, and the final phase — extending service to Brooklyn and Queens — is expected to be completed by 2032.
https://news.ycombinator.com/item?id=43816183
https://news.ycombinator.com/item?id=19785044
I wonder why 800 feet underground: Is that necessary to pass beneath all other infrastructure (to prevent flooding it?)? Remain beneath waterline to create negative pressure and reduce leaking? ?
Also, what is the general mathematical relationship between depth, rock pressure / weight, and energy required to drill? That is, what is the proportion of energy required to drill beneath 800 feet of material compared to drilling beneath 400 feet?
...
I'd guess the reason for the 800 ft is because the reservoir it'll draw from is near sea level.
1. https://en.wikipedia.org/wiki/Mississippi_River (search gradient)
(The Army Corps of Engineers has great detailed depth surveys for most of NY's waterways[1].)
Edit: There's also a higher-resolution render of the tunnel layout here[2].
[1]: https://www.nan.usace.army.mil/Missions/Navigation/Controlli...
[2]: https://extapps.dec.ny.gov/docs/water_pdf/nycsystem.pdf
I believe Tunnel #3 connects to the Catskill Aqueduct[1], which draws from the Schoharie and Ashokan reservoirs. Both are at least a few hundred feet above sea level (the Ashokan is about 600 feet above, since it was formed by flooding a valley in the Catskills).
But I have no idea why they dug it so deep, given that! Maybe to give themselves an (extremely) ample buffer for any future infrastructure in Manhattan.
One diagram I saw indicated 2 different layers of bedrock. I didn't find anything real clear, but it can be that the lower layer is a more suitable material for the tunnel.
Here you do deep tunnels to avoid the surface, in ways another poster said; everything is easier when nothing is in the way.
For the mathematical difference, 400 feet below sea level and 800 feet below are almost exactly the same: difficulties are water getting in to your pit, but the machines that work on rock, work on rock at the same speed regardless of depth, so the difference between 400 feet and 800 feet is best described as 400 feet difference. A big issue here is that they do not drill; they hammer. Pounding base pylons into bedrock causes dramatic rhythms in the surrounding 500m, but that's to deal with the bedrock, not depth.
What? I've never heard this. Everything I've read says pylons go into drilled holes.
There isn't any. It completely depends on the local geology.
Liquids are easy because there are no lateral load transfers, and the structures have to bear the weight of the entire water column above them. But with soil you get lateral load transfer, so the pressure on the tunnel is not easily relatable to its depth.
That's also why you can have mines that are kilometers deep, yet with tunnels held by wooden beams.
That depends on the rock type. In london, most things are clay, so not actually that solid (ie it needs shoring up immediately, and will collapse without supports, hence the travelling shield)
manhattan schist appears to be reasonably hard (not granite, but also no clay)
It feels like very soon, and coastal cities can stop relying on hinterland reservoirs for water.
There are Roman aqueducts in continuous operation for two millenia.
https://en.wikipedia.org/wiki/Acqua_Vergine
It's not entirely accurate to say that the West Coast doesn't have enough fresh water. Oregon and Washington have a lot of rain, and many groundwater resources.
California kneecaps itself with perpetual deeded water rights and mismanagement/closure/lack of improvement to reservoirs and related infrastructure. There's a long history of this kind of stuff in the state (see the watering LA desert, the Salton Sea experiment, and many others).
Certainly NY's government and budget are larger than other US cities, for obvious reasons.
https://www.nytimes.com/2017/12/28/nyregion/new-york-subway-...
> "the highest construction costs in the world"
> "The estimated cost of the Long Island Rail Road project, known as “East Side Access,” has ballooned to $12 billion, or nearly $3.5 billion for each new mile of track — seven times the average elsewhere in the world. The recently completed Second Avenue subway on Manhattan’s Upper East Side and the 2015 extension of the No. 7 line to Hudson Yards also cost far above average, at $2.5 billion and $1.5 billion per mile, respectively."
> For years, The Times found, public officials have stood by as a small group of politically connected labor unions, construction companies and consulting firms have amassed large profits.
> Trade unions, which have closely aligned themselves with Gov. Andrew M. Cuomo and other politicians, have secured deals requiring underground construction work to be staffed by as many as four times more laborers than elsewhere in the world, documents show.
> Construction companies, which have given millions of dollars in campaign donations in recent years, have increased their projected costs by up to 50 percent when bidding for work from the M.T.A., contractors say.
> Public officials, mired in bureaucracy, have not acted to curb the costs. The M.T.A. has not adopted best practices nor worked to increase competition in contracting, and it almost never punishes vendors for spending too much or taking too long, according to inspector general reports.
etc.
(The classic form of griping over NYC corruption is the MTA which is notable for not being administrated by the city.)
[1]: https://www.vitalcitynyc.org/articles/how-corrupt-is-new-yor...
> However, beginning in the 1990s, the number of corruption convictions there began to rapidly decline, so much so that by the beginning of the 2020s, Manhattan’s position relative to other areas had flipped. It now boasts the fewest corruption convictions of any major city area.
But as a few Germans have put it to me - sure, there's corruption here, but at least it still gets things built unlike _Italian_ corruption.
Which is an... ...interesting point of view.
[0]: https://en.wikipedia.org/wiki/Franz_Josef_Strauss
As for London, they built an entire industry around hiding money for oligarchs who stole it from their own countries. Maybe it's technically legal, but it's morally corrupt AF.
No studies, personal impressions, so I might well be wrong and maybe they all know but don't care. No majority that cares either way.
Low-level is when you bribe individual cops, city clerks, etc so they let you go instead of writing a speeding ticket or approving your house building plan.
High-level is when people like Merz receive a political donation from McDonalds, do some self-promotion in one, and then keep/lower the Mwst (VAT) for restaurants.
Germany unfortunately has high-level corruption but as far as I know, very little low-level. I think thats partially why people don't care to vote to differently. Yes, it happens, but there is a large disconnect between what Merz does and how it impacts an individuals bottom line.
If people would have to constantly hand out bribes to anyone then maybe its a different story.
It’s a big project, and it is tricky to patch it after release. The thing is supposed to last 300 years, and usually we use infrastructure well past it’s intended lifespan…
That one is ~2600 years old
Ask Europeans ? They're bangin'em out.
It's similar in scope to this recently-completed second phase of NYC Tunnel #3, albeit carrying sewage rather than fresh water: 25 km long, 7.2 m in diameter in London vs 29 km long, 4.9 m diameter in NYC. Flow volumes are likely similar (a sewage tunnel will rarely run full).
Planning started in 2001, with construction beginning in 2016. It opened in May 2025, at a cost of around £5bn: https://en.wikipedia.org/wiki/Thames_Tideway_Tunnel
Construction on New York's Tunnel #3 began in 1970. It was 28 years before any part of it was operational. A second section came online 15 years later (2013). The final stage isn't expected to be completed until 2032, a full 62 years after construction began. I'm unaware of any comparable tunnel project which has progressed at this slow of a pace.
However, this is only the second phase of the plan, with two more phases broken out into separate projects. I've no idea if those were supposed to be a part of the original 50 year timeline or not.
These two questions are casually put next to each other in the same sentence but they're incredibly different. Personally, I don't think that corruption is a significant factor in how long it took. The second question is way too leading/framed - "necessary" doesn't exist past the physical limits.
For example, would the same project have taken the same time in China? No. Does that mean it should've taken as long as it would've in China, as clearly it took longer "than necessary"? Not by definition.
There typically are no technical solutions to rhose.
It's not needed now, but we think that it will be needed in the future
It's needed now, but we don't know if we will use it in the future
How MUCH will it be needed in the future
Will there be a future technology that makes this investment unnecessary, or even obselete before the project ever completes
For the latter, a big argument of "No need to invest in commuter trains" argument was "self driving cars are 'just around the corner' and they will make mass transit a quaint thing of the past" was used to deny investment in trains.
They use them heavily when they're available. The NYC subway is very popular and successful, and many see it as a selling point of the city.
> Most Americans don’t want to commute sitting next to strangers.
I never hear city residents talk about 'strangers'. Interacting with others is a pleasure of cities, in fact - it's energizing, it builds social trust. We're social animals. I've never gotten on public transit, or walked down a busy sidewalk, and thought about 'strangers'. Most of those people are pretty sociable.
I think your numbers are wrong: the city's foreign born population has been stable for at least 15 years[1]. We're not even at historic highs; those were before WWI.
[1]: https://cmsny.org/publications/data-briefing-on-new-york-cit...
This statement doesn't contradict the one about international immigrants keeping the city from shrinking. It is easy to imagine how immigrants come to NY, give birth to natural born Americans, who then move out of the city. This process can come to some kind of a dynamic equilibrium with a stable population of foreign born people.
People use the trains in places like Chicago and NYC not simply because they are available but because owning and driving a car in the city center is very expensive and impractical for most people.
Anywhere less dense, people prefer to drive their own cars.
Did you hear about the other other lawsuit about people burning to death in their cybertruck? Should we compare horrific deaths per passenger? Per mile traveled?
Does everything work perfectly all the time? No, not in anything. If you cross the street, maybe someone will drive right into you. But I cross streets without a problem.
Maybe if they make you uncomfortable, the lack of trust is in the mirror.
Not that they can't, but they won't.
This is where we’re at.
Building a train requires coordination. Building an autonomous vehicle requires technological innovation and convincing a few people at the top levels of government. The specifics matter (and the Abundance guys have done a great job summarizing them) but it's due to an entrenchment of certain styles of laws.
So the answer to "why do Americans build self-driving cars to ease transport when Europeans just built subway systems?" is "we do these things not because they are hard, but because they're actually much easier than the other thing you find easy".
I went looking for an article I read a decade ago about the challenges of supplying water to NYC and maintaining the aging infrastructure. Part of the "race" to build new capacity is so they can actually turn off some of this supply for extended periods to repair it. Millions of gallons of water leaks or is just unaccounted for every day.
I didn't find it but this [1] kind of goes into it.
And since you can't turn the water off (generally), you need to do repairs in fairly extreme environments and use materials that don't corrode over very long periods of time. IIRC some pump or valve infrastructure was made out of manganese bronze for this purpose.
[1]: https://nysfocus.com/2024/11/27/new-york-water-leaks-drought
I'll never be a billionaire, but I'll also never spend $77 billion with so little to show for it.