Some Sketchy Climate Change Math
Which of your actions actually contribute the most to the coming catastrophe?
This post is the result of a project I've been thinking about for the past few months, due to a combination of influence from close friends and the recent national conversation sparked by Greta Thunberg and other climate activists.
There are lots of ways in which I personally contribute to the coming climate disaster: I use air-conditioning in the summer, I eat meat, I fly on airplanes. And I'm not going to stop doing all of those things, because I'm a human being and change is hard and I don't want to make the perfect the enemy of the good.
So instead, I want to do the effective-altruisty thing and figure out: assuming I'm only willing to make one life change, which one would most significantly affect my carbon footprint?
The problem is that climate change is something I'd consider a problem area for Effective Altruist (EA) methods: it's a problem that needs to be solved if we want human life to continue, so we should be devoting resources to solve it, but the sort of major, quick action we'll need to make progress are hard to quantify because we've never done something like this before. So I was unsurprised that I didn't find a lot of climate change info on the sorts of EA sites I usually trust on questions of "how do I best do (secular) good?"
This list is my answer to the question: how can I personally mitigate my contribution to climate change. It's not meant to be a 100% accurate guide, and it's also not meant to replace the sorts of large-scale national or international actions we'll need.
A caveat: my certainty in these numbers is extremely low. In part, I'm combining a lot of numbers from a lot of different sources and I'm not qualified to do a proper meta-analysis. More importantly, these effects are really really hard to measure. If I give up beef for a day, I don't directly affect the amount of carbon in the world, because whatever cow would have contributed to my dinner has already lived its life and been killed. But I do lower demand for beef by a slight amount, which might contribute to fewer cows being raised in the future! But if I measure that effect, should I measure the effect that my switching to (for example) Impossible Burgers will contributed to future meat-alternative research? How would I possibly quantify that?? So these are very back-of-the-envelope calculations, which given the small amount of money (relative to climate change) I donate seems like as much effort as is reasonable to put in.
The result is that I'm trusting reasonable-looking sites on these numbers and trying to get multiple verifications where possible, but I haven't checked each one individually, and I'm not correcting at all for comparing across studies. Based on my experience looking up numbers and seeing how the different sources compared, I'd say I trust each number as an order-of-magnitude estimate. So I'm not confident there's a real difference between chicken (5 kg CO2/ kg) and pork (7 kg CO2 / kg), but I'm fairly confident they're both significantly better than beef (26 kg CO2 / kg).
In what follows, "grams" always means "grams of CO2 equivalent". For comparison, the human body produces approximately 1 kg of CO2 per day.
If you don't care about how I came to my numbers, scroll down all the way to the bottom to Conclusions.
There seems to be a bit of uncertainty surrounding the carbon emissions of air travel, but somewhere in the range of 200 - 300 g / km per person, with longer flights being a bit smaller because takeoff/landing use an inordinate amount of fuel. I fly around 5-6ish times round trip per year, most often between Chicago and Northern California.
This is a distance of 3000 km or so, which means each flight counts for 600 - 900 kg. (I also travel to/from the airport, usually in a car, which by my estimates is around 4-10 kg of carbon dioxide, which is fairly negligible compared to the total.)
On the other hand, I don't think this is particularly fungible: I don't have another way to travel to California in a reasonable time-frame for the holidays, and I don't intend to stop seeing my family.
Living in Hyde Park, I walk most places, although most weekends I'll take a bus/train downtown and back. Carbon emissions depend significantly on the specific model of vehicle (think SmartCar vs. Hummer), but I found the following government statistics:
Typical Car: 440 grams per passenger mile
Typical Bus: 290 grams per passenger mile
Typical Light Rail Train: 163 grams per passenger mile
Realistically, however, my choices are between ubering somewhere and taking public transportation. My choice to use public transportation is unlikely to affect bus routes, which (I assume) are based mostly on things like politics, so me taking the bus should only very slightly increase carbon emissions (due to the small effect my ridership has on keeping routes in service, along with the slight increase in fuel use from my weight). My choice to uber, however, probably has a much larger effect on the amount of carbon, since the number of uber drivers isn't nearly as fixed. I don't know exactly how to quantify this, but I would guess that ubering is at least 200 g / mi worse than public transit.
While looking this up, I learned that Lyft has committed to carbon neutrality! This (in addition to the numerous other problems with Uber) seems like a very good reason to switch to Lyft, which I'm finally getting around to downloading
Here there's a lot of variance based on food! I looked at a couple of different sources, and eventually settled on a single metastudy that more or less matched all the other sources, and was presumably more systematic than my scattered googling.
There's quite a bit of variance, even among similar types of food! To produce 1 kg of poultry requires the equivalent of about 5 kg of CO2, while pork is about 7 kg and most fish are around 4 kg. Beef is a whopping twenty-six kilograms.
Fruits and vegetables tend to sit comfortably in the .3 - 1 kg range, although avocados get up to 1.5 and asparagus, for some unclear reason, surpasses most of the meats at nearly nine kilograms. My roommate and I get most of our vegetables from Imperfect Produce, which sells ugly or excess vegetables for lower prices - does this count as lowering our carbon footprint? I don't know enough about the economics to be sure.
Bread and breakfast cereals also hover around 1, and I imagine most processed food is a little but not significantly higher than the unprocessed version? The more processed it gets the harder it is to find good data so I'm not certain.
What about cooking?
My oven and stove run off of natural gas, which is measured in either BTUs or Therms. (1 Therm corresponds to 100,000 BTU). According to the EIA and EPA (probably working from the same sourceS), using one therm of natural gas emits about 5.3 kg of CO2.
A typical stove seems to use somewhere in the 10k - 20k BTU range per hour (on the highest setting), so let's say about 15k. Let's say (to make the numbers round) a typical meal takes about 12 minutes of stove time, so that's about 3,000 BTUs, or 150 grams. This is probably comparable to the emissions associated with veggies and breads, but much smaller than those of meats.
A typical oven seems similar, maybe on the higher end. The main difference is cook times are now longer - in addition to the 20 minutes I typically roast veggies for, we have about 15-20 minutes of preheating (our oven's not the fastest), so this gets us much closer to the .5 - 1 kg range. It seems like this is the main carbon component in anything except for meat (or is comparable to the eggs if you're baking.)
Which of course begs the question: how much is a kilowatt-hour?
This summer, I think largely because of our unit's central AC, my apartment used 25-30 kWh per day, or about 6-8 per person to day. This corresponds to roughly 2.5 - 4 kg per person per day.
Some example numbers:
Charging an iPhone overnight: 19.2 Wh, or roughly 8 grams of carbon.
Running a 100-Watt lightbulb for an hour: 100 Wh, or roughly 40 - 45 grams of carbon.
A load of laundry (+ dryer): 4 KWh, or roughly 1.6 - 1.8 kilograms of carbon.
Using my laptop for an hour: 60 - 100 Wh (? this seems low but all the websites I checked seemed to agree? Why is this similar to a lightbulb??), or roughly 25 - 45 grams of carbon.
A refrigerator/freezer running for a day: 1 - 2 KWh, or roughly 400 - 900 grams of carbon.
Heating and A/C:
We typically keep our apartment at 68 F at night and 72 F during the day. To get a rough estimate for how much natural gas we use per degree, I made a spreadsheet comparing our gas bill for each month to the temperature, and concluded that a degree of heat corresponds to approximately 2.42 therms per month (R^2 = .96), or about 13 kg of carbon dioxide per month.
I ran the same sort of analysis on our electric bill over the summer, and concluded that a degree of "cool" corresponds to approximately 70 kWh per month (R^2 = .98) , or about 28 - 32 kg of carbon dioxide per month. (Our utility seems to charge 10 to 15 cents per KWh, so it also costs us about seven dollars per month.)
The other kind of heating is hot water, which we mostly use for showering.
A typical showerhead uses about 2.1 gallons of water per minute, and a typical shower runs a bit over 100 degrees Fahrenheit. (This seemed high to me, but was on the low end of estimates?) It takes about 1 BTU to raise 1 gallon of water 1 degree, so a shower uses maybe 100 ( = 50 * 2) BTUs, or half a kg per minute. Looking at our natural gas bill this estimate seems a bit high, unless my roommates really like cold showers. My current guess would be closer to 100 g per minute.
I read a few articles on the effects of donating to various charities on climate change, and the one I found most convincing/well-researched was this Cause Area report by Founders Pledge. (This was my first interaction with Founders Pledge, so I don't have a strong prior on their work, but this report struck me as rigorous and carefully-researched.) There's a lot to be said, but here's the rundown:
CfRN is an intergovernmental group working to prevent deforestation, by helping to ratify, implement, and enforce forest-friendly treaty protocols. FP estimates that CfRN averts CO2e emissions at a rate of about $.12 / tonne. SoGive disagrees with this estimate, considering CfRN to be effective but not that effective. (Most of their concerns are qualitative, so they don't have an explicit replacement estimate.)
CATF is an American NGO working to reduce (both CO2 and other) air pollution through public lobbying and private partnerships. FP estimates that CATF has averted CO2e emissions at a rate of about $1.26 / tonne.
(Another climate change charity that's been fairly popular in effective altruist circles is Cool Earth, a group that helps communities fight deforestation. Early analyses of Cool Earth scored it at about $1.34 / tonne. However, more recent authors have disputed, these numbers, pointing out that some displacement effects have been ignored -- loggers stopped from logging in one part of a forest may simply move to an unprotected part. This does raise the associated costs, and so probably decreases logging to some extent, but not by 100%.)
If these seem too indirect, another option is Carbon Fund, who sell direct carbon offsets at a rate of $10/tonne. They claim some level of quality assurance, but their website didn't really have clear enough information for me to be sure that this is accurate and not some form of donor illusion.
A typical American carbon footprint is somewhere around 50,000 kg / year. So the changes below (other than donations) are fairly small, but not negligible.
If you recall, the main point of this brainstorming exercise was to think about ways I could personally mitigate my climate impact. Here are the ideas that came to mind that I'm willing to do, which seemed to have strong enough impacts to justify the effort:
Switching from Uber to Lyft: This is something I've been meaning to do for other reasons, but it has the bonus that Lyft is trying to be carbon neutral! Obviously this isn't an excuse to Lyft everywhere in place of e.g. public transit, but they seem to have similar prices in Chicago, so this is a benefit of 50 - 100 kg per year at essentially no cost.
Donating $25 per month to the Clean Air Task Force: I don't have a strong reason for the choice of $25 per month, or for the choice of CATF over CfRN, other than the fact that SoGive seconds FP's analysis of CATF, which makes me a little more confident since they have a history of spotting analysis errors. I would choose either organization over direct carbon offsets because they're working towards serious international climate action, which I think is very necessary and want to support. This will correspond to a bit over 200,000 kg per year, and is by far the single most important action on this list. That said, I think there's a problem with using donations to cover for bad actions, even if you could argue for it from a super strict utilitarian standpoint.
Shorter Showers: I have a really bad habit of taking long showers, with the water on on the order of 20 minutes. Maintaining a strict 5 minute shower rule (a reasonable length since I'm a guy with short hair) would save about 500 kg per year. (And give me 15 minutes more sleep!)
Less Beef, more Chicken and Pork: I'm not willing to go fully vegan, but I think I could cut my beef intake and increase my intake of chicken and pork. (In particular, notice that replacing beef with chicken or pork causes a much larger decrease than the corresponding change from chicken or pork to veggies.) If I replaced a quarter of a kg of beef per week, this would also save about 250 kg per year.
Apartment Temperature: I'll have to talk to my roommates about this one, but it seems we could adjust our summer temperature up and our winter temperature down by small amounts with minimal change in comfort. Assuming we run heat and AC for about five months per year each, each degree of change would correspond to about 200 kg per year.
This is all pretty sketchy math, so I'm definitely open to feedback or other ideas! We're all going to face the climate apocalypse together, so we should probably work together to figure out what to do.