Background
For a class at school, I performed a Life Cycle Assessment (LCA) of growing and transporting coffee in order to identify the entire environmental impact of growing and transporting coffee. The purpose of this study was to identify the areas of impact of producing coffee to see what drives the environmental impact.

What is an LCA
A Life Cycle Assessment (LCA) is a technique which calculates the total environmental impact of a system. In this case, the Eco-Indicator 99 tool was used. Eco-Indicator 99 (EI99) is a single factor LCA tool, which means it provides a result of a single number (measured in "EI99 Millipoints") which wraps up impact in a variety of categories. A higher millipoint value means more environmental impact. On their website, you can download the set of tables which were used in the study.

The most important part of performing an LCA is in choosing the system boundaries. A slightly different system boundary can completely change the impact and impact drivers of a system.

The Study Itself
Edwin Martinez of Finca Vista Hermosa(FVH) provided details about their farming techniques. It is important to note that FVH is a sustainably minded farm and that these results apply to this farm in specific. Portions of the study can be extrapolated and the practices at FVH provide a lower bound to most farms. Additionally, the data provided by EI99 for transportation and electricity are specific to Europe but were used anyway.

Finally, it should be realized that there might be inaccuracies in this paper. I am not very experienced with LCAs and have a great deal to learn about growing coffee. Keep this in mind when reading it, and if you have any questions, please contact me directly (benDOTsalinasATgmailDOTcom) or simply post here. I will be happy to go into more detail about any aspect of the study, the results, or the interpretation.

The paper is available for download here.

Discussion
I feel that this paper has the possibility to provide a good platform for discussion on environmental sustainability- an issue we like to talk a lot about, but have never really quantified. I look forward to your thoughts on the issue.

Summary of Results
The results of this paper are complex, and somewhat specific to the scenario provided. While this is a summary of the results, to adequately understand and discuss the results, it is recommended that you read the paper.

When considering the environmental impact of coffee on FVH, from when the trees are planted to when the green coffee is in the country of consumption, the vast majority of the environmental impact is in the transportation. In fact, based on the results shown here (and this set of system boundaries), everything that happens on FVH is negligible. The impact relating to transportation has 3 major parts

• Transportation In the Country of Origin
• Transportation to the Country of Consumption
• Transportation In the Country of Consumption

Within the country of origin, transportation to a nearby city occurs via inefficient and small trucks(34 EI99 millipoints/ton*km). Despite these trucks having 1.5-2 times the impact of trucks used at other stages in the process, this only contributed to approximately 15% of the overall transportation impact.

From the dry mill (in Huehuetenango) the coffee was transported to a port in a slightly more efficient truck (22mp/t km). This step provided 22% of the transportation impact.

Once at the port, it was loaded onto a freighter (1.1 mp/t km) and shipped to New York accounting to 34% of the transportation impact.

Finally, the coffee was transported to Boston, accounting for the last 30% of the environmental impact.

The take away message here is that more than 60% of the transportation impact occurred outside of the country of origin, and almost 1/3 of the environmental impact occurred shipping the coffee from New York to Boston. Had the coffee been shipped to Seattle instead of Boston, the overall impact would have been multiplied by 4.8 and the transport within the United States would have accounted for 85% of all the transportation impact.

It is also important to note that shipping via freighter is very efficient. If air shipment was used instead of boat, the overall environmental impact would have been multiplied by 17.

And so, it looks as though the impact of coffee is actually all in the transportation. For the sake of discussion, I added in brewing and roasting.

Roasting Coffee
Over the lifetime of a commercial roaster, it was assumed that the materials and manufacturing of the roaster would be negligible (because of the high use). Based on some data for the amount of gas used to roast coffee, it was found that roasting a kg of coffee used about 4.85 EI99 mp. This is approximately 1/4 of what was used during the transportation phase (but almost 5 times what was used on the farm). Very quickly the roasting of the coffee is matched by transporting the coffee to the consumer (to get the same 4.85 EI99 mp via transport of roasted coffee, that coffee would have to be transported to New Haven (150 miles away)). Now, imagine transporting that coffee to Georgia, or Texas, or Chicago, or San Francisco.

Brewing as Espresso
Finally, looking at the impact associated with brewing a kg of green coffee (after its been roasted, of course) was examined. In some work I did previously, I determined that the materials and manufacturing of an espresso machine is completely negligible and that some 98% of the environmental impact comes from electricity use. With this, it was determined that brewing the kg of green coffee as espresso in a coffee shop setting would account for 107.7 mp. This dwarfs the rest of the impact (assuming the coffee stays in Boston, this alone accounts for 82% of the overall impact).
If this was instead in a commercial espresso machine which sits on in a house, this impact would be even higher.

Conclusions
In other words, the burden of the environmental impact lies within the country of consumption and choices very much in the hands of consumers. However, the choices are not always clear. It is hard for a consumer to identify where within the country of origin the farm was located (and thus the impact associated with transporting it to the coast) and perhaps more importantly, where the coffee arrived in the country of consumption. But these are all important things to think about.

Properly farmed coffee is one of the ultimates of environmental sustainability as it sustains the wildlife, stewards,workers and a community too. A concept American have long ago forgot how to truly understand. Think how tiny the impact was back when we threw the tea in Boston harbor and went looking for coffee. It was sailed here and then transported by horses who could run on grass without having to turn it into ethanol at a refinery! The bummer about back then was they hadn't invented espresso yet Take some time to really understand "farming" before you try to key in too much on a form of farming.
farm

Hi Ben, nice work! Just one niggle -- once they stop grading you, you might try writing in English, rather than the American corporatese that provides a good living for so many comics in the UK.

There's a fun book, "The Carnivore's Dilemma," that underlines your main point, our (1st world) food and food delivery is about oil -- oil for fertilizers, for transport, for cooking. Apparently several 100s of calories for every calorie of food we actually eat.

But the book, and your report prompted a second thought: Coffee is a tropical crop, so transport costs will always be part of the equation. Should these should be allocated to the good itself, or to the transport infrastructure? As fuel costs rise, inland transport may go back to rail, or even to rivers and canals, while sea transport, already efficient, can be made more so by various means. Would such changes lower the environmental impact of coffee, or oranges, or teddy bears, etc or would it lower the envoironemntal impact of transport in general?

The type of non-financial accounting you are doing, e.g. single number for environmental impact, presupposes some sort of virtual-money that could be turned real by the appropriate legislation privatizing common property rights and creating markets. For instance, a mother earth corporation could be established that owns the world's water and air. In this case, Ma Earth would be charging the transport companies for polluting her air, she would not be charging the coffee growers or shippers. It is therefore the transport companies that would be responsible for the environemtal impact.

Please understand: whenever you say X has an environmental impact that can be boiled down to a single number, you are proposing that X should bear the financial and legal responsibility rather than discovering some natural fact about X.

But what does your number actually mean?
Does it mean that for every 50 millipoints God kills a little kitten and sets two trees on fire or raises the sea level with 2 nanometer or... or?

I'd like to know environmental impact and i appreciate your efforts but there just meaningless numbers as far as i'm concerned.

Think how tiny the impact was back when we threw the tea in Boston harbor and went looking for coffee. It was sailed here and then transported by horses who could run on grass without having to turn it into ethanol at a refinery!

Exactly. And compare that to the double shot of an air shipped, vac sealed at origin, roasted across the country and air shipped to one's doorstep on the2 group Linea that is kept on all the time at home.

There's a fun book, "The Carnivore's Dilemma," that underlines your main point, our (1st world) food and food delivery is about oil -- oil for fertilizers, for transport, for cooking. Apparently several 100s of calories for every calorie of food we actually eat.

Jim- if you're referring to "The Omnivore's Dilemma" then I have it and have read some of it (though I haven't yet finished it).
The author makes some great points about the luxuries we can afford of shipping food all over the world and not being "slave" to eating food that is "in season"

Please understand: whenever you say X has an environmental impact that can be boiled down to a single number, you are proposing that X should bear the financial and legal responsibility rather than discovering some natural fact about X.

I don't think I quite understand what you mean by this. Keep in mind that while the impact has been summed into a single number, it wasn't done by a committee saying "I think cars should be worth 29 mp/km" but rather by a committee of very experience environmental scientists/engineers who made (widely accepted) decisions on how impact areas relate to one another (in this case- "Human Health Impact should be weighted the same as Ecosystem Quality and twice as much as Resources"). The ratings of specific things within these categories are simpler decisions. In the case of human health, we have a pretty good idea of how 1 kg of a specific chemical affects human health. Then, by looking at the emissions (both in the traditional sense, and in terms of what products leak) and the inputs to making a truck as well as common use patterns, they generated a huge list of all the inputs and outputs to the system (for shipment via a truck, there is literally thousands of inputs and outputs).

My point is that while there were value judgements, they were on considerably smaller things. If I had access to the $5000 software, results could be considerably more specific and explain how much impact there is to ecosystems, human health, resource depletion, carcinogens, etc.

Just as with rolling a cup of coffee into a single number, a lot is lost in the move to single factor LCA, but at the same time a lot is gained in the simplification and ease of communication (just as with coffee).

But what does your number actually mean?

Very good question. Part of the purpose of LCAs is to compare the results to other things (i.e. "drinking coffee is worse than drinking tea" or "you are better off drinking coffee than going out to eat"... neither of which I am actually saying). But the other, more common, purpose is to use the results to identify where the large impact comes from. We can then say, "if you are looking to reduce your environmental impact of coffee production, look towards the transportation within the country of consumption and the brewing method" This prevents us from obsessing over something which might not make a huge difference. So, for example, we could spend a great deal of time focusing on reducing the amount of petroleum based fertilizer used, but ultimately this would have almost 0 reduction environmental impact since the bulk (more than 99%) happens after the coffee leaves the farm.

If you do want some comparisons to other things which might help, here are some numbers I ran from the EI99 tables. Assuming the scenario outlined in the paper of consuming coffee in Miami which was roasted in Boston as espresso in a coffee shop- a double shot of espresso uses about 3.83 mp. When transporting a person in a car is at 29 mp/km, we see that driving to the shop is clearly what matters. (This is not to say that we should stick a commercial grade espresso machine into every house, because that quickly turns the numbers on their head).
Likewise, that double shot of espresso is equivalent to leaving 2 75 Watt light bulbs on for an hour (assuming average European grid mix... which varies heavily)

I hope that helps some.

bsalinas wrote:Jim- if you're referring to "The Omnivore's Dilemma"

Yep.

bsalinas wrote:I don't think I quite understand what you mean by this. Keep in mind that while the impact has been summed into a single number, it wasn't done by a committee saying "I think cars should be worth 29 mp/km" but rather by a committee of very experience environmental scientists/engineers who made (widely accepted) decisions on how impact areas relate to one another (in this case- "Human Health Impact should be weighted the same as Ecosystem Quality and twice as much as Resources").

If you sum up, for instance, air pollution, resource expenditure, soil depletion, etc etc into a single number, you have to decide on how to balance the various players. How does earthworms' welfare trade off with those of guppies, those of pandas, and those of suburban coffee drinkers, etc etc? There is a lot of high grade engineering and science required to calculate the individual burderns. But once those burdens are calculated, you have to decide which burdens count for what. To claim it's all 'the environment and interconnected" is obfuscation. These connections determine how the impacts correlate, but they don't determine who gets screwed, who gets stuck with the costs, who gets to be extinct, and who doesn't. That part is always a monetary or pseudo-monetary allocation of burdens and benefits.