Monday, March 12, 2007

So how much energy do we use to make ... energy?

I am working hard to nail down The Number that represents how many calories of fuel energy it takes to make one calorie of food. My previous post about dinosaurs in your food describes how this Number is on average 10 calories of fuel to make one calorie of food.

So where does this number come from? Below is a narrative followed by references to the researchers mentioned.

I accept the 10:1 ratio because I have read various data sets and combed numerous papers and studies related to this issue. Further research on this issue is needed, something I am pursuing every day, and I will revisit this ratio if needed.

The ratio is based off of a complex number of variables: amount of energy used in the food system, definition of food system used, amount of calories per person per day to be considered, the role of import and exports, and what percentage of total U.S. energy is used by the food system.

Pimentel and, separately, Hall have estimated the average to be 10:1, while Heller/Keoleian has estimated 7.3:1. Heller/Keoleian say that the food system consumes 10.2 quadrillion Btu's (quads) of energy and provides 1.4 quads back out, based off a diet of 3,800 calories per person per day (because we produce more food than we need, eat too much of it, and then throw some away). If we use a 2,500 calorie diet, we would get an 11:1 ratio; a 2,000 calorie diet means a 13.8:1 ratio.

On the side, grain-fed beef requires thirty-five calories for every calorie of beef produced (Horrigan), and a can of diet soda that provides maybe 1 calorie of energy needs 2,200 calories to produce (70% tied up in the aluminum can)(Heller/Keoleian).

I am still dissecting Heller/Keoleian's comprehensive paper. Heller/Keoleian estimates food energy use at 10% of total U.S. energy. Hendrickson studied 8 different studies from the 1970's and found an average for food system energy use to be 15.6%. This suggests we are already reducing energy use in ag. The Earth Policy Institute (EPI) created some nice graphs showing total U.S. and farm energy use base off the more current numbers. In one graph EPI show the whole U.S. food system uses 10.25 quads of energy (10,250 quadrillion Btu's) and ag production accounts for 21% of energy use, or 2.125 quads. What

What I take from this is this: if Heller/Keoleian are correct that ag uses 10% of energy, this number would roughly align with the 1.691 quad number, but if we use the 21% figure that comes from Heller/Keoleian, the amount of embedded energy (and therefore the calorie ratio) is much higher than Heller/Keoleian report. I am trying to get hold of the Heller/Keoleian team to ask them about these discrepancies.

In the meantime I accept 10:1 to be an average ratio I can support. I will keep resarching this issue and updating as needed. Thanks to Cookson Beecher, a reporter with the Capitol Press (Olympia, WA), for asking me to source this fact.


Heller and Keoleian's article "Life Cycle-Based Sustainability Indicators for Assessment of the U.S. Food System" is very comprehensive.

One article by Pimentel and Giampietro is "The Tightening Conflict: Population, Energy Use, and the Ecology of Agriculture".

Sustainable Table has a good article "Fossil Fuel and Energy Use" with strong references at the bottom.

Hall, C. A. S., C. J. Cleveland, and R. Kaufmann, "Energy and Resource Quality" Wiley Interscience, New York: 1986.

Heller, Martin C., and Gregory A. Keoleian, "Life Cycle-Based Sustainability Indicators for Assessment of the U.S. Food System", Ann Arbor, MI: Center for Sustainable Systems, University of Michigan, 2000.

John Hendrickson, “Energy Use in the U.S. Food System: A Summary of Existing Research and Analysis” Sustainable Farming, Vol. 7, No 4, 1997

Horrigan, Leo, Robert S. Lawrence, and Polly Walker. "How Sustainable Agriculture Can Address the Environmental and Human Health Harms of Industrial Agriculture." Environmental Health Perspectives 110, no. 5 (May 5, 2002)

David Pimentel and Mary Pimentel, "Energy Use in Fruit, Vegetable, and Forage Production", in "Food, Energy, and Society", ed. D. Pimentel, and M. Pimentel, revised edition. University Press of Colorado, Niwot, CO, 1996,
pp. 131-147.


Fat Knowledge said...

I have just found your blog and am enjoying reading through your posts.

I think the ratio of fossil fuel energy to calories is very interesting and important, and am glad to see someone is taking the time to go through the research reports to figure out what the best estimate is. I am curious to see where you finally arrive.

On this post, I got confused what you were doing in your last couple of paragraphs. You state: In one graph EPI show the whole U.S. uses 10.25 quads of energy (10,250 trillion Btu's) and ag production accounts for 21% of energy use, or 2.125 quads.

It seems like you are saying the 10.25 quads is for the entire economy, but doesn't this just refer to food production? I believe the US uses 100 quads of energy for the entire economy a year, so 10.25 quads would be around 10% of that. Earlier in the post you state that the Heller/Keoleian estimate for the food system is 10.2 quads, which I think is the same number.

I am not sure how Heller/Keoleian handle the food that the US exports to the rest of the world. Do you? Might that explain part of the reason why the 3,800 calories a day is so high?

For a different way to look at this issue, you might want to read my Energy Opportunity Cost of Agriculture post.

Tim Crosby said...

Thanks for the question. I did transpose some numbers, using the food system energy use as the whole us energy use. I also mislabeled a quad as a trillion Btus.I have edited the post to show this correction.

It is good to see another blog out there investigating similar issues. Look forward to sharing info.


odograph said...

I looked into this briefly, in some discussion at env-econ.

In this one (down in the comments) a tried reality-checking the output of some scary "fossil fuel per calorie" numbers with actual US diesel consumption:

"Now, the problem with those numbers (and here I thought I could trust Pimentel because I like his ethanol skepticism) ... if "beef" is really 0.029, that's too much energy. We cannot possibly be putting that much (primarily diesel) energy into our beef production.

As we saw above, 0.03 works out to 35 billion gallons of diesel-equivalent energy, and we only consume 60 billion gallons nationally. We are not using half our diesel for beef production. No way."

And I found:

"Agriculture uses approximately 3 billion gallons of diesel fuel per year."

I'm not sure about those quads and trillions of BTUs, but how do they compare to the 3 billion gallons of diesel that we know are the actual upper limit?

odograph said...

I get a kick out of this kind of math, and I hope you do too. I hope also that you are ready to throw away some of the less supported numbers coming from activists.

I think I'm a sort of environmentalist myself, but unfortunately I've also got a chemistry degree and 20 years of engineering behind me.

Looking to play with the numbers again, I found this:

"Food delivered to grocery stores is measured in pounds of food. This does not include alcoholic beverages though. Unfortunately, this does include food that is stolen or spoils and is thrown away either by the grocer or consumer. Measured this way, the average American "eats" about 3,600 calories per person per day and has not changed much from 1909 when it was 3,500 calories per person per day. These calories come from milk (14%), meat (28%), eggs and legumes (5%), grain products (26%), fruits and vegetables (14%) and fats, sweets and beverages (13%)."

3600 Calories x 365 days x 300,000,000 people = 3.942e+14 Calories per year delivered in the US.

What to do with that number? We could apply the 10:1 ratio to it, and then convert it back into diesel gallons:

3.942e+14 x 10 / 32000 = 1.231e+11
gallons (or 123 billion gallons).

That's way more than we actually use for all purposes in the US.

If we work it the other way, and assume that Calorie-for-Calorie imports approximate exports, we can look at the ratio from the 3 billion gallons we actually use in US agriculture and see working ratio:

3.942e+14 food calories / (3e+9 gallons * 32000 Calories) = 4 food calories per diesel calorie.

Tim Crosby said...


Thanks for the input. I do get a kick out of the numbers with part of the kick being to figure out which ones hold true at the end of the day. That is one of the main reasons I started this blog; to filter through the fluff and find the facts that we can use to figure out proper solutions.

Your info on diesel usage is good, and thanks for link back to your blog. As the Earth Policy Institute graph shows in the post
And that graph only shows farm usage, which would exclude processing, transportation, storage refrigeration, and then waste disposal. I want to know the numbers that include the whole food chain.

Diesel use is part of the energy supply. We also use gas, natural gas, liquified gas, and electricity. The EPI graph shwos only a small amount of natural gas usage on farm, but I have been told there is a lot more in processing.

I bet that if we included these other energy sources the numbers would come close to alignment with the calorie input/output.



odograph said...

I think we don't know ;-)

I would guess that it will probably come down to how close we are staying to farm produce, and how far up the chain we are going in consumer delivery.

I do notice that bulk (dry) beans and flour are staying pretty cheap, while the cost at fast-food joints rise. That could be because the energy is coming at the "far end."

But, a lot of this stuff is not intuitive. I did a calc over at the Fat Knowldege page, based on the 75g CO2 number for a bag of potato chips:

For what it's worth, 75g is (75 * 12/44) = 20 grams carbon

Given Diesel carbon content per gallon (link below): 2,778 grams

One "standard bag" of chips contains the energy equivalent of 0.007 gallons of diesel fuel.

Geez that seems low. If we knew the size or calories in that bag of chips we could calculate the fossil fuel to calorie efficiency as well.

I don't know where that leaves us!

BTW, do you have an actual farm there? Maybe you could get a student to work totaling your yearly energy inputs and annual calorie outputs .. and we could use _that_ as a sanity check.



odograph said...

It suddenly occurred to me that we can translate that 0.007 gallons back into calories, using 32000 as the rough measure of calories in a gallon of diesel.

224 calories.

It's actually hard to think that their reference bag would have fewer than that!

So for something that I actually thought was one of the worst cases (fried food, trucked around in bags of mostly air), we may be 1:1 or better on fuel inputs.