Hacker News new | past | comments | ask | show | jobs | submit login
How Much Water Is in Earth's Crust? (worldatlas.com)
43 points by Brajeshwar on June 6, 2024 | hide | past | favorite | 44 comments



This is just water under the layers, right? Wasn't there an ocean of water trapped inside crystals underground like ringwoodite? I bet if all that water comes out, Earth will look like Waterworld.

Like this one: https://www.npr.org/sections/thetwo-way/2014/06/15/322246690...


The first question you should have asked is, does this include only chemically free water? It does not.

Geology 101 will tell you that a lot of water is chemically bound in rocks. The accessibility of said "water" is... questionable.

These aren't necessarily hydrates either; if it has oxygen and hydrogen, then you can extract water from it.


Earth surface area is 197 million square miles.

So the mean water thickness is 0.053 miles or 281 feet.


The droughts in recent years have had a big influence on the level of ground water. I heard on the radio a year ago that it will take for it to rain every day for the whole year, for it to recover to the usual level, at least in my area. And that it will only get worse as the climate is getting hotter, and the water evaporates at a faster rate.


Seeing how many here are having fun with the choice of units, here's a puzzle:

Assuming the Moore's law holds, how long will we have to wait before there are more transistors in a chip the size of a 8086 processor than moles of water in Earth's crust (the choice of time unit is left to the student) ?


10 million cubic miles of water is 4.2*10^22 ml or the same number in grams. The molecular weight of water is 18 g/mol, so dividing that out gets us 2.3*10^21 moles of water in the Earth's crust.

According to a quick 'n dirty Google search, in 2024 chips can have up to 208 billion transistors. This leaves us with a 1.1*10^10 improvement needed to hit our target, which works out to be about 33 doublings. Moore's law says one doubling happens every 1.5 years, so we're looking at a 50 year timeline.

Assuming Moore's law can hold for another 50 years is likely pushing it, but it's cool to see that if we manage to duplicate the last 50 years of progress we'll be approaching a mole of transistors per chip.


[stub for offtopicness]


Well, that's the first time I've seen cubic miles used as a unit. For reference it's about 4.4e19 litres or 4.4e16 cubic metres.


I don’t think it’s that uncommon. For one thing, it is listed in the NIST Guide to the SI (https://www.nist.gov/pml/special-publication-811/nist-guide-...)

It also, IMO, beats acre-foot as a generic way to describe volumes (who even knows an acre is a chain by a furlong?)


The length of a mile varies across jurisdictions though. You'd think the world atlas would be mindful of the gap.


It’s similarly difficult to have an intuitive grasp regardless. And I’m an SI-loving engineer.


e19 and e16 are both absurdly large, which is probably why they used miles. Maybe kilometers would be better for folks who need it in SI.

Edit: And I see someone quoted the value from the article apparently few of us read completely.


> The new figure stands at approximately 43.9 million cubic kilometers

quoted from the article


Cubic Miles? Who does scientific measurements in Miles? Some people apparently.


Also, who knows what a cubic mile of water is?

~10M cubic miles of water is ~10 Mediterranean seas.

That's a lot.

But the earth is enormous, so, is it really that surprising?


Any normal measurement unit here is going to be hard to grasp on it's own without translating it into an equivalency; XXXXXX olympic swimming pools or XX Y ocean.


My unit of choice here is teaspoons.


I prefer banana equivalent mass.


Framing the amount of water in the Earth's seas in terms of one of the seas feels a bit circular :D.


the article is talking about water within the crust, which is different from water in the seas

i.e. there's 10 mediterranean sea's worth of water underground that we can't sea from space


Some of us are measuring the space in seas.


The title was in suspended animation so I added the answer from the article. The article is trying to make its primary audience, the Americans, easier by converting it to miles.


I would imagine most Americans on this website are capable of comprehending liters.


I'm an American, and I like to think that I can comprehend liters, but I think for enormous volumes like this it's easier to visualize distances cubed.

Like for a million cubic miles, I know that's a cube with 1,000 mile sides. That's like New York to Miami (don't ask me about cities in Europe...okay I'll go out on a limb: Paris to Rome?). I can visualize that cube of water on a globe.

But how many liters or gallons is that? A lot! But billions? Trillions? I probably would instinctively say billions, but with a tiny bit of mental arithmetic I'm suspecting it's up in the trillions.

edit: I came back to confess that indeed I am an ignorant American who has no intuitive sense of exaliters. My SI volume comfort zone doesn't extend much past teraliters.


Are you suggesting that most americans have a good grasp of how much water a cubic mile is?

Surely the standard unit for this sort of thing is multiples of well known large lakes or seas?


I refuse to acknowledge anything but multiples of Libraries of Congress.


I think I'm going to need a conversion to bald eagle volumes, or square hamburger-feet.


This is liquid: Cans of Coke or Coors are standard units (CCC)

In this case, 1.48 * 10^20 CCC


A cubic mile isn't a widely-used measurement of volume. Everyone knows what a 2-liter of soda looks like.


But everyone knows how long a mile is, and can imagine a cube one mile long one mile wide and one mile high. Though we can't truly grasp the scale, we can at least understand the magnitude of a value of ~ten million.

But quoting a value of 4.4 x 10^19 liters is meaningless for most people.

"Of course, that's 22,000,000,000,000,000,000 two-liter soda bottles!"

So cubic miles seems like a reasonable unit for this pop-science article, despite the fact that you likely wouldn't use it in a published journal article.


> But everyone knows how long a mile is

Yeah. Long. I doubt most people can eyeball something in the distance and say with accuracy "yeah that is about a mile away" because a mile is really long and people are bad at estimating things. Now do it for 10 miles.

A relatable example like someone has mentioned "its about 10 dead seas worth" would have been a better play.


I’d guess a lot of Americans can look at a globe and eyeball 100 miles or 1,000 miles. At least American adults with a lot of driving experience. When you get into the millions of cubic miles of water, I think the best way to visualize it is a cube sitting next to the globe.


Just picture how long a mile is, and then imagine a cube where each side is that long.


In that case it should have used soda cans as the unit.


1 cubic mile is about 14 billion bathtubs.


finally a sensible metric!


well that's around 2 baths for the entire world


Cubic Miles? Who keeps the metric system down?

This is equivalent to: 1/30th the volume of Earths Oceans.

17x the volume of the Greenland icecap.

Or 3.72x10^20 US gills


Surprisingly little compared to the volume of the oceans. Makes me wonder how much is in Mars' crust


Or about 2e17 hogsheads


but how many cups is it?


About 1.8e20 US cups according to Google.


The fact that a site called "worldatlas" uses antiquated (and in this case, meaningless) U.S. units, tells you all you need to know about internet bias towards a US audience. Fun fact: there are ~1 billion English speakers outside of the U.S.


"Eschew flamebait. Avoid generic tangents."

"Please don't pick the most provocative thing in an article or post to complain about in the thread. Find something interesting to respond to instead."

https://news.ycombinator.com/newsguidelines.html




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

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

Search: