Water Footprint of Real Food

May 2nd, 2015

California’s drought has spawned a flurry of social media articles with titles like “It takes how much water to grow an Almond?” and “Don’t eat meat! Vegan diets use less water”, and graphs of how much water it takes to make a pound of different kinds of food. But for practical purposes, these articles are useless because A) nutrition is not measured in pounds, and B) they are picking on specific foods instead of considering a complete balanced diet.

It’s time to run the numbers!

I researched water footprints of 150 different ingredients from waterfootprint.org, and nutrition information from wolframalpha.com. I also filled in 12 additional essential nutrients from whfoods.com. I even put in the 9 essential amino acids. “Essential” means your body cannot make them, so you have to eat them. All in all I analyzed 43 essential nutrients, plus cholesterol. I put in an option to take a multivitamin pill, with or without vitamin A, since too much vitamin A is associated with reduced bone mineral density. For fun I generated nutrition labels using caloriecount.com.

I ask, what is the minimum total water usage that can produce a healthy diet, according to the FDA’s recommended daily intake of each of these 43 essential nutrients? Well that’s just a linear complementarity math problem, and luckily Excel can solve those! Into the spreadsheet it goes.

Without further ado, here is the minimum water usage diet, assuming vitamin pills and fish use very little water:

28g carrotsminwaterpill
223g cranberries
276g molasses
24g palm oil
259g salmon
21g soybean oil
340g tomato paste

Boom, only 845 Litres of water per day! You’d take a multivitamin to make up for the missing nutrients, especially folate, making sure not to add any more vitamin A. Unfortunately I don’t know how much water it takes to make one multivitamin pill, so this might not be fair. More importantly, this diet is totally gross. So let’s remove the vitamin pill:

4g beef liverminwater_nopill
42g broccoli
25g brussels sprouts
2g butter
450g cauliflower
37g coconut
5g ginger
394g green peas
137g molasses
75g peanuts
12g poppy seed
69g salmon
317g tomato paste
1g turmeric
429g turnips

This diet requires 1560 Litres of water per day. I capped the quantities for spices, otherwise you end up eating lots of ginger and turmeric. It’s slightly less gross than the first diet, but it’s still a lot of food and it’s heavy on starches.

Let’s try making diets that are less gross. I added some minimum requirements for the 4 basic food groups from Harvard’s Healthy Eating Plate. First, allowing the vitamin again:

50g carrotsharvard_pill
201g cassava
54g cranberries
50g molasses
299g onion
24g palm oil
130g salmon
21g soybean oil
0.3g spinach
119g string beans
96g tomato paste
250g bread

This balanced diet requires 1389 Litres of water, plus whatever water a vitamin pill uses. It actually sounds edible, thanks to the balancing of the food groups! Now without the vitamin pill:

4g beef liverharvard_nopill
51g broccoli
283g brussels sprouts
8g butter
6g carrots
172g cauliflower
25g cheese
18g molasses
12g palm oil
45g peanuts
15g poppy seed
69g salmon
92g string beans
272g tomato paste
250g turnips
211g wheat
39g bread

Well now we’re up to 1918 Litres of water per day. Such is the price of a balanced diet with food-sourced vitamins. At least it sounds delicious!

Okay now how about a vegetarian diet? Let’s remove all the ingredients that involve eating dead animals that are large enough to see with the naked eye, sticking with the no-vitamin rule:

51g broccoliveggie
360g brussels sprouts
16g carrots
99g cauliflower
62g cereal
1g ginger
102g milk powder
133g molasses
22g palm oil
63g peanuts
37g tofu
159g tomato paste
1g turmeric
350g turnips

The vegetarian diet uses 2057 Litres of water per day. Clearly, adding more constraints on your options can only worsen the objective (minimizing water). How about vegan? We’ll remove all ingredients that are stolen from helpless animals, again with no vitamin pills:

51g broccolivegan
283g brussels sprouts
4g carrots
408g cauliflower
176g cereal
77g molasses
25g palm oil
62g peanuts
163g tofu
276g tomato paste

That’s 2873 Litres of water per day for a healthy vegan diet. I hope you like cauliflower. Anyway, lowering your water footprint by going vegan is clearly a myth. That’s just simple math: adding more constraints can never improve the objective. To be fair, let’s find the optimal omnivore diet if we require one 4 oz steak per day:

113g beefsteak
51g broccoli
432g brussels sprouts
17g cassava
105g cauliflower
372g cranberries
82g eggplant
50g lemon / lime
4g maple syrup
133g molasses
17g onion
19g palm oil
31g peanuts
11g poppy seed
69g salmon
6g soybeans
2g soybean oil
359g tomato paste
126g turnips

This steak diet requires 3271 Litres of water per day. It’s more than the vegan diet, but not that much more. I’m going to say, myth busted. Vegetarians win here, actually. I had to cap the molasses to 133g (the same as vegetarians to be fair), otherwise it was over 200g and, well, yuck. How about requiring 4 oz chicken?

51g broccolichicken
275g brussels sprouts
2g butter
6g carrots
242g cauliflower
91g cereal
113g chicken
128g cranberries
50g lemon / lime
133g molasses
21g palm oil
29g peanuts
8g poppy seeds
57g salmon
7g soybeans
281g tomato paste
1g turmeric
191g turnips

That requires 2134 Litres of water, which is less than vegan and very close to vegetarian. Wow. So chicken actually does save water compared to steak. Okay, all these have broccoli, brussels sprouts, and cauliflower. Can we remove those to see what an alternative omnivore diet might look like?

152g avocadoalternate
49g barley
2g liver
216g cabbage
24g cheese
27g coconut
65g eggs
60g green peas
2g mushrooms
28g oats
24g poppy seed
173g rye
91g salmon
44g string beans
298g tomato paste
32g turnip
262g watermelon
250g zucchini

This diet uses 2514 Litres water per day. Yum, avocado! The green veggies got replaced with cabbage and zucchini. Eggs show up. But apparently leaving out the three magic green veggies has boosted the water usage quite a bit. Continuing this trend, if we omit cabbage and zucchini, they are replaced with celery and cucumbers, increasing the usage to 2786 Litres per day. I won’t bother showing it. If we remove celery and cucumbers, they are replaced with eggplant, okra, and soybeans, increasing the water to 2823 Litres per day. I think you get the idea.

Now let’s add some context. What’s the worst water usage I can come up with for a reasonable diet? Here we go:

15g almonds
454g apples
75g beef
9g broccoli
217g cauliflower
33g coconut
84g figs
50g flaxseed
50g lemons / limes
110g milk
0.5g nutmeg
64g oats
10g olives
454g peppers
454g plums
327g quinoa
61g salmon
13g sesame seed
18g sunflower seed
443g whey

That diet uses 20050 Litres of water per day. I’m not going to show the label because caloriecount.com is way off for the peppers, quinoa, and whey. Strange. Anyway, the peppers, plums, apples, and quinoa contribute the bulk of the water here because you can eat a lot of them. And this diet has a lot of them, which I capped at 1 lb each. Almonds show up, and they do use more water per gram, but you just can’t eat that many almonds without blowing your calorie and fat budget. Beef contributes more than the almonds here, as do the figs, flaxseeds, sesame seed, and whey. The other stuff is mostly just contributing nutrients and not so much water.

So far we’ve seen that purposefully mismanaging your diet can raise your water usage to more than 20 thousand Litres per day for food. If you’re paying attention, you can get it down to 2 or 3 thousand Litres. Now the real context: how much water do other daily activities consume? Let’s use the Personal Water Footprint calculator from waterfootprint.org. I entered zero for all the food and typical values for everything else, and I got roughly 2000 Litres per day. Wow, actually the food choices do have an impact.

Some Surprises

I was surprised how often “bad foods” showed up in these optimized diets, like butter and palm oil. I think these foods are under-appreciated because people are making blanket statements about them, instead of considering the diet as a whole. Smarten up, people! Also, what’s the deal with molasses? I guess it’s the cheapest calories per water footprint, so it showed up a lot.

Some Not-So-Surprises

Some foods that we expected to show up did show up a lot, like broccoli, brussels sprouts, and cauliflower. Even in a drought, you can count on the classics.

[Slashdot] [Digg] [Reddit] [del.icio.us] [Facebook] [Technorati] [Google] [StumbleUpon]


May 17th, 2012

The Character of Physical Law (7 lectures, 1964)

1) The Law of Gravitation:

2) The Relation of Mathematics to Physics:

3) The Great Conservation Principles:

4) Symmetry in Physical Law:

5) The Distinction of Past and Future:

6) Probability and Uncertainty:

7) Seeking New Laws:

Other Lectures

Quantum Mechanics:

The Pleasure of Finding Things Out:

Dirac Memorial Lecture:


[Slashdot] [Digg] [Reddit] [del.icio.us] [Facebook] [Technorati] [Google] [StumbleUpon]

What could $700,000,000,000 buy?

September 29th, 2008

Dr. Albert Evil

What would you do with $700,000,000,000?  Well, you could stave off the recession for a few months, but it seems congress didn’t buy into that.  But let’s put this into perspective.  Do you have any idea what you could do with seven hundred billion dollars?  Here’s an idea: solve green energy.

I know, you say, you can’t solve a difficult problem just by throwing money at it.  But my point is, you can solve a difficult problem by throwing money and a million scientists at it.  Seriously.  Back of the envelope:

1 – Hire one million scientists and engineers full time for a few years, and ask them to solve green energy.  Tell them to form groups and share ideas however they like.  Plan on spending about $390 billion here.

2 – To keep them from slacking off, tell them you’ll pick the top 1000 most promising projects after a few years and give them an additional one hundred million dollars funding each for a second phase.

3 – To keep the people in phase 2 from slacking off, tell them you’ll pick the top 100 most promising projects after another few years for an additional one billion dollars funding each for a third phase.

4 – To keep the people in phase 3 from slacking off, tell them the top 10 out of those projects will be selected after ten years or so for a further ten billion dollars funding each for a final phase.  This should be enough for the actual build-out of the technology.  The LHC cost less than that.

You’d even have enough money left over to give the most successful project at the end a ten billion dollar bonus.

The money you’d make back with the technology would be many times the original investment, and the trade it would generate would likely rescue the economy big time.  Speaking of the economy, the companies you’d steal the scientists from are probably considering layoffs anyway, so it won’t have much short-term negative impact.  Also, you don’t need to borrow the full amount at the beginning, since the projects are funded in phases.  This is a doable idea if you’re actually planning to invest $700,000,000,000 into something useful, as opposed to just handing it over to your buddies.

Come on, people, this is just the first thing off the top of my head.  Can’t we get anyone to think more than a few months ahead around here?

[Slashdot] [Digg] [Reddit] [del.icio.us] [Facebook] [Technorati] [Google] [StumbleUpon]

Pick of the Litter

September 26th, 2007

Coke on a beach
Maybe it’s just me, but I can’t understand why any well-minded human being would ever litter. There is an increasing amount of litter in even the most beautiful areas, such as the beach in my photo. Even with posted fines in the thousands of dollars, there are people littering daily. I did pick up that can and throw it away, but all too often the litter around me is too nasty to even go near.

I have only witnessed a few people in the act of littering, and those I have witnessed fall into two categories. The first was a man who was crazy. He was walking down the street talking to himself, and simply dropped the (empty?) grocery bag he was carrying. I can understand this, because this man clearly had no idea where he was, let alone where the nearest recycling bin was located. The other people I have seen litter were all punk kids, trying to impress their friends. I can understand the urge and so-called coolness of defying authority for the sake of defying authority, but littering is one of those things that is not about authority. It’s about the fricking environment. Under no circumstances is this kind of littering not completely stupid. There must also be innumerable litterers out there who do not fall into any category.

It could be helpful to adopt better recycling programs, like some other countries have, that make it easier for people to recycle. But littering is not caused by lack of recycling. Littering is, in my judgment, an unsolvable problem.

Am I wrong? How to solve it?

[Slashdot] [Digg] [Reddit] [del.icio.us] [Facebook] [Technorati] [Google] [StumbleUpon]

Renewable Fuel Straight Out of Sci-Fi

September 16th, 2007

glowing green tubes of bubbling liquid
Professor Pengchen (Patrick) Fu is using cyanobacteria to produce ethanol from sunlight, carbon dioxide, and water. I got to see this first-hand at Wired NextFest. This is exceedingly cool since ethanol, a useful fuel, burns cleanly to produce energy, carbon dioxide, and water. See the pattern? It’s a sustainable cycle, essentially storing solar energy in a fuel that we already know and love. There is very little waste in this process, and it’s quite harmless, in the form of oxygen and some dead bacteria at the end of their natural life cycles. This is already cool enough that I think I need to sit down for a while and take it all in, but there is more! In case the photo didn’t make things amply evident, let me state clearly: Pengchen (Patrick) Fu has come up with a possible solution for renewable energy using large, glowing cylinders full of bubbling green liquid. These things look like they came straight off the set of a science fiction movie. You know, the good kind, with lots of explosions.


Some more details: The tubes shown in the pic each yield about 5 grams of ethanol per day. Fu expects improvements to this yield after more research into the process. He also talked about setting up kiloliter-sized tanks of this stuff for larger output.


5 grams of ethanol contains about 134 kilojoules of energy. If we had solar panels instead, assuming about 8 kWh/m^2/day, and 40% efficient solar panels, we would need about 116 square centimeters of solar panel to get the same yield (an 11 cm x 11 cm patch would do). I would estimate that Fu’s process is therefore less than one order of magnitude less efficient than the best solar cells at the moment. That’s not shabby, considering that the energy is stored in a fuel instead of a (costly) battery. Deeper vats of the stuff would mean more sunlight gets absorbed by the cyanobacteria, so the kiloliter tanks would probably have much better efficiency.

[Slashdot] [Digg] [Reddit] [del.icio.us] [Facebook] [Technorati] [Google] [StumbleUpon]