http://everylastdrop.co.uk/ says "When you consider the water required to produce all the stuff we consume, we actually guzzle a massive 4,645 litres every day". It then lists several specific cases, including:

  • 1 kg cotton = 20,000 litres, and
  • 1 cup of coffee. 130 litres

Are these two claims true?

  • 7
    I can't cite my own experience, so I'll just comment: I grow and process my own coffee. There's no inherent need for anything like that much water: water is used in the demucilaging bath, but that's low volume. So presumably they mean irrigation, but that varies by location (I use none, but nearby commercial farms have drip systems). Even if one does include irrigation, what's the definition of "taking" water? It's not removed from the water cycle. Many farms, at least here in Kona, Hawaii, are organic and don't use/minimize pesticides and runoff. Sep 7, 2013 at 21:34
  • 7
    define "use water" first Sep 7, 2013 at 22:42
  • 3
    @ratchetfreak The asker doesn't need to define the term. The source says that these figures represent "the water required to produce all the stuff we consume". If there's ambiguity in that, then so be it. A good answer will address that issue.
    – user5582
    Sep 7, 2013 at 22:45
  • 2
    Just for reference, here are the steps in producing coffee at the totally personal level: knowing.net/index.php/2009/06/09/coffee Sep 8, 2013 at 3:55
  • 2
    @Sancho not so. If that 2000L needs to be applied, it can be considered to have been used even if the net use after recovery and beancounting side benefits is far less. And that's the point. Without knowing what the actual claim is, there's no debunking it.
    – jwenting
    Sep 9, 2013 at 15:07

1 Answer 1


I get about:
coffee: 80 liters per cup
cotton: 10000 liters per kg
The differences could be due to the original calculation including run off pollutants. But the numbers are actually pretty close for a back of the envelope calculation. There are some details about this in the cotton water footprint report below.

looking at the averages of cotton requirements for the three time periods in the table I get:
(25 * .09) + (35 * .22) + (50 * .30) = 24.95 inches of water per acre
1 acre = 6272640 sq inches
6272640 sq inches * 24.95 inches = 156502368 cubic inches per acre
156502368 cubic inches = 2564610 liters per acre

Taking the average of "Cotton Yield: pounds per planted acre": 495.56, 667, 591
584.52 lbs per acre = 265.134 kg per acre
265.134 kg per acre / 2564610 liters per acre = 9672 liters per kg

This is a pretty interesting paper on the water footprint of cotton:
for virtual water average in the world total cost (pg. 192)
cotton lint: 8506 cubic meters per ton = 9376.2 liters per kg
end textile: 9359 cubic meters per ton = 10316.5 liters per kg
this seems to confirm my above calculations

claims 140 liters per cup

seems to agree with:
~2000 lbs of beans per acre
~17% weight loss (can apparently be higher or lower depending on the darkness of the blend) http://store.starbucks.com/Coffee-Preparation-FAQ/coffee-prep-faq,default,pg.html
A 1-lb bag of coffee yields approximately 64 5-oz cups* (320 oz) (53 6oz cups)
This is pretty close to most of what the individuals say that I could find ~50 cups so I'll use this number.
1 lb 11.8 oz (1.7375 lbs) = 240 6oz cups
But here is an up to amount from a folgers label: 240/1.7375 = 138 cups per lb

60 to 80 inches (I'll average the two and use 70)

70 inches * 1 acre = 7 195 310.72 liters per acre
2000 lbs of beans per acre
2000 lbs * .83 (less 17%) = 1660 lbs of roasted beans per acre
1660 * 53 = 87980 cups of coffee per acre
82 liters of water per cup of coffee

  • It important to say that this usage of water isn't "destroying" the water. It is cycled back to nature.
    – T. Sar
    May 26, 2017 at 10:48
  • No usage of water really "destroys" the water from an environmental standpoint other than cracking it open to get the hydrogen for use in nuclear fusion.
    – Ben Barden
    May 26, 2017 at 13:44
  • @BenBarden - That's different than what TSar is saying. If I remove water from the groundwater sources, use them for irrigation, for cranberry bogs, or other uses, then they are taken out of the normal water cycle. Aquifers, streams and water bodies lower or dry up completely. Trees and plants using normal rainwater is going to happen whether or not we plant, coffee trees, harvest the beans and drink the coffee is more the point. However, most coffee berries are "wet processed" to remove fruit pulp and get the "bean," so I'm not sure how much that plays into the original claim. May 26, 2017 at 15:01
  • @PoloHoleSet I'm reading the direct claim made by TSar, and things like irrigation and cranberry bogs absolutely have the water "cycled back to nature".
    – Ben Barden
    May 26, 2017 at 15:51
  • @BenBarden - not sure how dried up streams and lakes qualify as "cycled back to nature." By your standards, nothing, ever would not qualify, as you stated, which is obviously not what TSar is talking about. You're being excessively pedantic and literal. Try going with obvious intent, or ask for clarification if that intent isn't clear. That kind of strict literalism, if applied universally, would make communication impractically tedious. May 26, 2017 at 16:27

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .