Hacker News new | past | comments | ask | show | jobs | submit login
Oldest water on earth, in a Canadian mine (2021) (macleans.ca)
110 points by walterbell on March 2, 2023 | hide | past | favorite | 29 comments

This would be a good place to discuss (1) what 'old' water means; isn't Earth's H2O mainly circulating and, hence, 'old' in the sense that all molecules are? But of course water can react with other elements and likewise result from chemical reactions, so is it that what the mean? (2) How can you just fill a vial with some water and then go and analyze the sample in a mass spectrometer to see what other stuff is in there and then go and claim attested age for that sample without loosing sleep over the question that we can not possibly know all of which might have been going on with that water over the centuries, millennia, millions and, in this case, billions of years? Could I, if you gave me a vial of ultrapure water, fake that result in a well-equipped laboratory or at least write a peer-reviewed paper that details alternative mechanisms plausibly possibly present in geological conditions that would cause the water to attain said properties in a fraction of a billion years?

Sadly none of these questions are addressed in the article.

> isn't Earth's H2O mainly circulating and, hence, 'old' in the sense that all molecules are?

A lot of it may be, but some of it isn't.

You can buy quartz, for example that contain water or petroleum sloshing around inside [1] although some think the water could seep in and out over time.

But quartz with fluid inclusions (small bubbles containing water) are completely sealed and have the same composition as they were up to 3 billion years ago. "the fluid in the inclusion is an actual sample of fluid in which the mineral was growing" [2]

[1] https://en.m.wikipedia.org/wiki/Enhydro_agate

[2] https://www.gamineral.org/writings/enhydros-gray.html

To expand on this

Groundwater "age" is typically considered in terms of residence time, that is the period of time between entering the subsurface and exiting at some discharge location. These analyses are often based upon the hydrogeochemistry and the decay rate of known isotopes, hence the mass spectrometer.

Shallow, unconfined aquifers often contain "young" groundwater, on the order of decades or centuries. For example snow melts on a mountain, downhill into a flatter coarse grained area where it enters the groundwater system. The groundwater then moves downdip along the bedrock contour through and into the deeper alluvial basin, eventually naturally discharging into a river or is pumped out at a well.

It would be interesting to read their paper to understand how they (i.e. what isotope) can date back to 1.6 billion years.


Yeah, the article is really frustrating because it only addresses that question in a roundabout way. It says "The Precambrian Canadian Shield that stores vast mineral deposits is 2.7 billion years old, and once formed an ocean floor." - so I guess this water got trapped while the ocean floor was subducted or overlaid with sediments or whatever happened that brought it to its current depth?! So it's not the H2O molecules, but the whole liquid which has stayed there for that amount of time.

Referring back to your original question, it would be fascinating to find out what percentage of the water molecules in circulation on Earth today date back to the time when water first condensed out of the atmosphere (IIRC ~ 4 billion years ago) - but I'm not sure if that's possible.

One might also consider the question "If there is old water, does that mean there is also new water?" - additionally, how does one actually date a molecule? :-/

They are addressed reasonably well in Lollar's published work on the Kidd Creek water: https://www.tandfonline.com/doi/abs/10.1080/01490451.2019.16...

ETA: (Unfortunately, it's paywalled at a high price, which I didn't realized when I originally posted this. Apologies).

From the article:

> We might know water as H2O, forgetting everything that’s dissolved in it. For example, a small vial of the Kidd mine water transferred in 2020 to Ingenium—a Crown corporation that runs the Canada Science and Technology Museum in Ottawa—features sediment at the bottom. That’s iron oxide, which precipitated from the water after exposure to oxygen from the air.

I did read that sentence and its among the few instances where the article goes a little into the details of the evidence rather than the circumstances of their acquisition, like that journey down the mine shafts. The introductory sentence is fine and a good reminder given the likely audience. It does a good job for setting the stage. The next two sentences then thrash that hope by telling lots of irrelevant detail in the hope to tell an interesting story, and we're left with, ok, there's a certain amount of iron oxide in the water, which fell out because of contact with oxygen. This is literally what can happen to you when you bore a meter-deep well in your garden; in regions with iron in the ground, the water that comes up will have lots of iron oxide in it that will color everything in a reddish hue when drying. How does that address any one of my questions?

> 'old' in the sense that all molecules are

Trying to remember school chemistry lessons, when they explained pH, ... I think H2O is in equilibrium with H+ and OH-, perhaps with 1 in 10^7 of the molecules being disassociated at any time (hence pH 7), so a particular H2O molecule doesn't survive for very long. Minutes? Hours? Not years, I think.

Just waiting for a company to bottle and sell it based on its “ancient powers”.

I wonder what its expiration date will be.

It’s 10 times saltier than seawater.

People used to drink radium water.

That just proves how effective and unique it is. And it’s “natural”!

As a bonus it will leave you thirsty, which will encourage you to open another bottle...

Aside from water created through burning Hydrogen and Oxygen I just assumed all the water I have ever consumed was the same water consumed by billions of animals, humans and plants and then excreted out back into the environment to be filtered by the earth or atmosphere and then I drink it again and again. Is that a gross way to think about it?

Somewhere around one ten millionth of the water you drink is currently ionized. You better hope it is, anyway. Go ask a chemist what pH actually means. Non chemists seem surprised to discover that essentially all water is partially ionized all the time.

Anyway my point is if 1e-7 of the water you drink technically is not a water molecule at that instant, you can draw graphs of how much of the water you drink has been ionized at some point in the past, which requires a lot more motivation to determine the kinetics than I'm capable of.

So, no, essentially none of the water you drink was recently "pee water" or whatever. Yes technically the individual hydrogen and oxygen atoms have quite an extensive ownership history, but the average "age" of an individual water molecule as three atoms with polar covalent non-ionized bonds is surprisingly low in your average glass of tap water. A decent shoulder shrugging "meh" estimate is figure the water molecules in your average pH 7 tap water are about half a day old. Its a lot more than a couple hours and a lot less than a couple months, that's for sure.

The "age" of water in the linked article is how long that bucket was isolated and sealed off from the general ecosystem which conceptually is a different kettle of fish. Or kettle of dissolved isotopes in this case.

I'm no physicist but I'm pretty sure oxygen is the most prevalent element on Earth mass-wise, and hydrogen is the most abundant element in the universe, so it seems likely that there are a good number of water molecules that have never existed before, and that most (all?) water molecules will not exist indefinitely.

A lake can be a lake but that doesn't mean the water is the same water.

In any solution, the bonds of water molecules are being continuously rearranged. The odds that any of the water molecules leaving such a solution have any of the atoms they started with are infinitesimal.

For the most part, we live in a giant soup of chemistry on this planet. Everything changes all the time, nothing is forever.

There's a song in Spanish called "Mi Agüita Amarilla" by "Los Toreros Muertos"[1] who basically tells the story of a guy who drank 40 beers and then proceeds to describe how his pee is transformed into water again to be drank by cows, people and the fish that you eat, to then ascend to the atmosphere to rain down on schools and your house.

This song haunted me back when I was a kid and learned about the water cycle.

1. https://www.youtube.com/watch?v=Jaf1yvWb1jw

I know this is the prominent question here, but isn't virtually all water on earth more or less the same age?

I think the title here is failing to be an accurate description, unless I missed something in science class.

The problem is defining "water".

The 7 in pH 7 meaning neutral comes from 1 in 1e7 molecules of water having broken apart and ionized at any given instant. Like the Taylor Swift song says, they're never ever getting back together. Ever. So the average lifespan of a rando H-O-H thrupple in plain old tap water is about half a day. So by the definition of "hydrogen atom with UUID X" in a monogamous bond with "oxy atom with UUID Y" in a monogamous bond with "another hydrogen atom with UUID Z" that relationship on average lasts about half a day before getting stirred up into a new relationship. In that sense, most water is less than a day old.

The article is defining age of water as in how long it was sealed off from the rest of the ecosystem. In that sense I have a bottle of Crown Royale containing about 60% water in my basement that's been sealed off from the environment, theoretically, for 18 years before it hit the liquor store shelf, or so the marketing claims. Actually a little longer now that it's chilling in my basement. These folks found some water that's been sealed off for a billion years.

So there's decent arguments for both that "a bucket has been sealed for a billion years" vs "this particular organization of atoms has existed substantially less than one day"

I suppose all protons are the same age and how they are configured is irrelevant?

My atoms are 13 billion years old or whatever it is. I guess you, me, and water are all the same age. After we die and desiccate our water will be joined to other fremen. Wait..

For some reason I can’t edit my comment so I’ll reply to it. Protons aren’t all the same age. One prominent example being radioactive decay (beta negative). I suppose most protons should be the same age but not all.

I remember visiting the Kidd Creek Mine, in the 1970s (my mother was a geologist, and used to work for TexsaGulf).

That is like a scifi movie. Weird place.

The zinc electrolyzing pool is ... invigorating to walk near.

Maybe this is a silly question, but how can a mass spectrometer determine the age of water?

They are not really determining the age of the water, but rather how long the water has been isolated from other water, and thus changing entirely as a result of local interactions with the containing substrate, and the entrained microbial life in the water. That's based on the isotope ratios of the oxygen and hydrogen in the water.

by measuring the Deuterium content

proud moment for Canada

and therefore the world

Never has comedy described a people more accurately than this SP exchange.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact