A few days ago the internet was abuzz with shocking headlines because the gentleman behind 'virtual water', professor John Anthony Allan of King’s College London, got an award from a water conservation group, the Stockholm International Water Institute (SIWI) for his work on advocating water conservation. Reading the news clippings about it, you'd have thought it was a Nobel prize for perpetual motion.

Perpetual motion is a good analogy. Generally, if you see something too ridiculous to be true, it's probably not true. A few weeks ago, for example, a VA Tech grad student got a prize for a 'gravity lamp' that was just the kind of alcohol-and-magic-fueled hocus-pocus that sets the internet on fire. It was green energy and cool tech all rolled into one. Except it didn't exist. Rather than being able to power a household bulb for hours, even an unsuitably-large one could only power a tiny 0.1-watt LED for 45 minutes. It's just physics.

So a few days ago people were aghast and outraged when they saw a number stating that 34 gallons of 'virtual water' went into a cup of coffee. I understand their panic. That means we only have about 9,588,235,294,117,647 cups of coffee left before all the water is gone.(1)

Except water doesn't actually disappear. There is the same amount of water on earth now as millions of years ago. That's right, you're drinking water a Neanderthal peed in. Water recycles. I understand that only 2.5% of all water is available 'fresh' water but that is why the outrage over coffee and hamburgers is unjustified.

Trees, for example, give off 70 gallons of water per day just in evaporation. That's two cups of coffee right there coming back to us. So water evaporating from trees provides enough water to result in 392 cups of coffee for every man, woman and child on the planet. Humans use, on average, 50 gallons per day directly, so we'd have to scale that coffee back to just over 390 cups per day.

My diatribe here is not about the water conservation industry. I am all for clean water. 2 billion people live without proper clean water and even in civilized areas there are 70,000 known chemicals in use that can contaminate water. I am very much against bad water. I am also against bad math.

Allan has been writing about the 'virtual water' concept for a while, notably in 1998 in the article "Virtual Water: A Strategic Resource Global Solutions to Regional Deficits"(2) published by Ground Water magazine. It had existed in concept before then, since 1993, and he even used it farther back himself under the term "embedded water" but he came up with "virtual water" because, as he said, the term embedded water "did not capture the attention of the water managing community."(3)

So at least we understand motivation.

He used it while doing examinations of the middle east and came up with a basic figure in how much water was contained in something like imported grain.(4) His feeling was that the 'virtual water' contained in grain imported into one region of the mid-east was equal to all the water in the Nile so there was trouble brewing, if you'll pardon the coffee pun. From there it took on a life of its own.

Since that time, the 'hydropolitics' of virtual water have been brought up on occasion yet the one place where it should be obvious now, the middle east, still has no signs of it. There hasn't been a war over water even though they fight over almost everything else.

There's a reason for that. Real water is what people care about, not pretend water. 'Virtual water' can be used for shock effect but it's basically an intangible and, as we shall see, the math is fuzzy. Even Allan never intended for it to be anything more than conceptual:

The recent publication by Hoekstra and Hung (2002), which will be discussed at the Third World Water Forum in Kyoto in March 2003, shows that some water engineers liked the idea so much that they have devoted months to adding quantitative substance
to the conceptual utility of the idea.

They basically wanted it to be true so they started matching data to the topology. Yet the concept has spread everywhere, sketchy math or not. As a comparison of how far it had spread, I did a search to find out how much water it takes to grow an egg. The most common result I found was 120 gallons of water. For one egg. Stephen Colbert's 'truthiness' says people are more inclined to believe something outrageous if it is vague and this is a perfect example, because I saw the same figure in dozens of places.

Dr. Jim Arthur of Hi-Line International, an egg producer in Iowa, doesn't know where that number comes from. He says 100 birds lays 80-90 eggs per day and use 4-5 gallons of water. He told Elizabeth Grossman when asked(5), "The 120-gallon figure would have to include water for something other than what the hen consumes directly" which is about 1 ounce of water per egg.

So it must include the grain. Well, birds eat like ... birds. They don't eat a lot, about 1/2 cup per day for a hen laying those 8 eggs. A ton of grain is going to feed over 10,660 hens for a day and produce 85,300 eggs(6). That should take 10,236,000 gallons of water at 120 gallons each.

But the 1,000 cubic meters Allan says makes a ton of grain, even using his aggressive virtual water scenario, is only 1 million liters - 263,936 gallons. Somewhere these egg numbers include not only chickens drinking water, but the grain to feed them, the water the farmer drank, the well the water came from, etc., et al, until we end up with 40X the amount of water calculated using food for the hens.

In other words, it's just made up. So it goes with the odd '633 gallons of water to make a hamburger' claim. Vegetarians want to believe it, but they also wanted to believe walking to the store causes more global warming than driving your car there if you eat meat. As long as they didn't use a calculator, they were right and in that instance they were handed an easy-to-remember platitude ('a gallon of gas per pound of beef') and ran with it despite it being made up. Much like saying it takes 1,000 cubic meters of water to produce a ton of grain.

So it goes in this instance but there's more than math that's flawed in that reasoning. There's also ecology.

Let's suppose for argument's sake it does take 34 gallons of water to grow the beans for my coffee. The bulk of that will be rainwater which, as we we all know, happens in places like Colombia. The water is not trucked in from some distant land where it is hurting the ecosystem by being gone. If the water is not used in coffee plants, it will seep into the ground or flow into rivers.

If we're going to complain about coffee and hamburgers, the Amazon rainforest is one of the top water consumers in the world. If less coffee and hamburgers actually meant more water for African kids who have none, cutting down the rainforest would easily provide water for the 2 billion people without it. Yet no one thinks that's a good idea.

Clean water in places where people live that have no water is really the issue. We don't have a water shortage, a resource that needs to be rationed. There is a lot of water because, as I mentioned above, only 2.5% of what we have is fresh water. People who talk about a water crisis or wars over rivers are in the advocacy business. Like anything else, if a lobbyist, even one guised as a scientist, starts giving you numbers, you need to examine the data carefully. We have an energy issue when it comes to water, not a mitigation one, because we just need energy to make undrinkable water potable.

Renewable, clean energy will solve a lot of the water issues that remain, either by desalinatization where there is ocean water available or shipping where there is no water at all. Putting money toward that is a much better strategy than Bono birthday parties.

So drink all the coffee you want without guilt. Just make sure you pee once in a while and it will all come back around. Your Neanderthal ancestors did it for you, after all.

Back to this aggressive 'virtual water' math, we can actually have some fun with it. Just for kicks I intend to take on the FTC and see if they fold once I get out the calculator. If there are 34 gallons of water in my cup of coffee I am certain I can prove there are 'five ounces of milk' in a one ounce slice of Kraft cheese.(7)


(1) Lenntech Water Facts And Trivia. http://www.lenntech.com/water-trivia-facts.htm

(2) J.A. Allan (1998) Virtual Water: A Strategic Resource Global Solutions to Regional Deficits Ground Water 36 (4), 545–546 doi:10.1111/j.1745-6584.1998.tb02825.x

(3) S. Merrett, J.A. Allan, and C. Lant, Virtual Water - the Water, Food, and Trade Nexus: Useful Concept or Misleading Metaphor?

(4) 1,000 cubic meters of water to produce a ton of grain.

(5) High Tech Trash: Digital Devices, Hidden Toxics, and Human Health By Elizabeth Grossman Island Press (2006) p.63 ISBN 1559635541

(6) Hens don't lay eggs every day of their lives. They start at about 6 months of age and then taper off when they get older plus they rest in winter unless you leave a light on. Even using geriatric chickens laying 1 egg a week, the numbers cannot add up.

(7) August 10, 1992, Federal Appeals Court upholds FTC Ruling That Kraft Misrepresented Calcium Content Of Individual Cheese Slices, http://www.ftc.gov/opa/predawn/F93/kraft-app5.htm