How much water covers the earth?
August 30, 2013 12:03 PM   Subscribe

If you looked at the planet Earth from space and were able to view the whole of the globe, on the average, what percentage would be water? (not as simple as it seems)

I thought of this question a time back when I read that you would be looking mostly at a wet planet. But then it went on to say that oceans cover 70.8% of the globe. That seems to vastly underestimate how much water you would be looking at. How much of the surface area of the planet is covered by lakes? They are usually figured into the land surface areas. I've found a number of sites that talk about cumulative lake volume. That's small, but a number of lakes are large and relatively shallow. I've not found a site that looks at total lake surface area for the world.

And then, according to this link, 60% of the globe at any given moment is covered by clouds, which are water. If this were divided equally between land and water areas (it's probably not) that alone would reduce the visible land from 30% down to 12%.

http://www.nytimes.com/interactive/2012/05/01/science/earth/0501-clouds.html?_r=0

Other vast areas, Antarctica, the Himalayas, many mountain ranges, Greenland, other northern territories are covered by snow or glaciers.

I'm imagining 90+% of the world is covered by water at any given moment. Has anyone figured this out? It's driving me crazy.
posted by dances_with_sneetches to Science & Nature (12 answers total) 1 user marked this as a favorite
 
I don't have the answer, but the season would also make a significant difference. Interesting question!
posted by theodolite at 12:13 PM on August 30, 2013


According to the US Geographical Survey, "About 70 percent of the Earth's surface is water-covered, and the oceans hold about 96.5 percent of all Earth's water." So the oceans, pardon the pun, pretty much have most of the amount covered. (The next highest category, ice caps and glaciers, make up 1.74%, but that amount represents about 68% of the planet's total fresh water.)

The USGS link does note that water exists in many different forms. in fact, if you were to count the water vapor in the air, nearly all the Earth's surface would be covered, though the amount makes up just 0.001% of the planet's total water. (And if it all fell as precipitation at once, the Earth would be covered with only about an inch of water.)
posted by Gelatin at 12:15 PM on August 30, 2013 [1 favorite]


(The USGS also points out that there is more fresh water underground than in liquid form on the surface.)
posted by Gelatin at 12:17 PM on August 30, 2013


(And as this image shows, the amount in freshwater lakes and rivers is tiny in proportion -- it's the little blue dot over Atlanta, Georgia, making up a little more than .007% of total water by volume.)
posted by Gelatin at 12:20 PM on August 30, 2013 [1 favorite]


With the proper imagery at any given point in time this would be a trivial spectral analysis problem from a remote sensing standpoint. The difficulty comes from an unexpected place:
If you looked at the planet Earth from space and were able to view the whole of the globe...
We're looking, in this problem, to resolve an image of the entire earth at a fixed moment in time. I'm not sure if the entirety of NOAA's remote sensing platforms could do this (even given license to interpolate data and a margin of temporal wiggle-room). I could be wrong, and I welcome correction if I am.

Other satellites could pick up some slack, of course; It'd be an interesting logistical puzzle to try to orchestrate such a thing. I'd be interested to learn more about this from someone more in the know (wouldn't be hard; my last involvement with remote sensing in a professional capacity was ten years ago).

There's also the problem of how you've defined "water". Water vapor makes up a non-trivial percentage of our entire atmosphere, clouds or no clouds. If we're arbitrarily limiting this to include only 'visible cloud cover', would we then define "covered by water" to be a patch of clouds that had a certain opacity? a certain albedo? would we define it by "amount of water per columnal meter"?

What about damp soil -- which would contain much more water per unit of volume than your clouds would?

Not at all trying to shit on your question, but for these and other reasons I think it needs to be cleaned up and more defined/focused before we could begin to attack it.
posted by jjjjjjjijjjjjjj at 12:25 PM on August 30, 2013


Response by poster: These are volumes, not surface areas. I realize lakes are shallow (especially when compared with the sea), but, for example, I read on one site Canada's lakes in surface area would combine to cover Manitoba. That's about 7 per cent of Canada's territory, not trivial.

We as creatures live in territory more than in volume. The depth of the Pacific affects us very little, but crossing the Pacific does effect us.

As for the soul of my question, I'm wondering what percent of the planet would look like water rather than earth. I realize you couldn't look at all sides at once, but if you did a few views you could cover the planet. Alternatively, you could ask what would happen with the average view (which would not necessarily be staring at the earth north pole on top, south pole on bottom as all of the spaceships in movies seem to do).

As for water vapor, short of clouds we wouldn't be seeing much of that. I'd classify mud as earth, we are not seeing the water, just its effect on soil.
posted by dances_with_sneetches at 12:35 PM on August 30, 2013


Best answer: Here's a link that covers everything except clouds: http://chartsbin.com/view/wwu

Looks reasonably referenced, here are the relevant percentages it gives for the surface area of the earth:
69.03% salt water
1.77% fresh water
5.84% snow covered land
total 76.64% water covered

Clouds would be tricky, but using your 60% of the earth is covered by uniformly distributed clouds assumption: 76.64+(.6)*(100-76.64)= 90.656%
posted by pseudonick at 12:36 PM on August 30, 2013


Response by poster: Final note: as the rubric above suggested a figure of 90% - to me that's a profound way of rethinking my planet.
posted by dances_with_sneetches at 1:04 PM on August 30, 2013


Response by poster: Okay, looking carefully at the site linked by pseudonick, the numbers seem a bit facile. The exact same figure is given for mountain land, land covered by snow and dry land (29,788, 205 square kilometers). And isn't much of mountain land covered by snow? The links on that page aren't much help.

Funny, I thought there'd be precise figures on all of this, somewhere.
posted by dances_with_sneetches at 1:21 PM on August 30, 2013


You mention clouds, which is not a standard way of discussing surface being "covered" by water. If you want to push this further, you'd have to include water vapour and if you were to observe the Earth using wavelengths slightly below that of visible (by humans) light, you'd get essentially 100% coverage by water i.e. you'd never see anything on the surface of the planet.
posted by aroberge at 3:46 PM on August 30, 2013 [1 favorite]


how would you take into account that clouds would be covering some surface area otherwise covered by a form of water?
posted by maximum sensing at 7:38 AM on August 31, 2013


Why don't you ask Randall Munroe?
posted by brianogilvie at 3:09 PM on August 31, 2013


« Older Mobile auto engine steam-clearing in the Portland...   |   Help me help my cat to want to pee in the right... Newer »
This thread is closed to new comments.