US Carbon Dioxide Emissions – down by source

I’ve often said my email is like a fire hose. Here’s an example of something that got lost in the stream and I didn’t find it until last night while looking for another email.

Zeke Hausfather writes (In August 2012):

We have a new memo through Berkeley Earth that attempts to discern the causes of recent declines in U.S. CO2 emissions. The switch from coal to gas is the single largest factor, though others are important as well.

Some of the inspiration for the research came from a post you did last year: http://wattsupwiththat.com/2012/07/02/us-co2-emissions-may-drop-to-1990-levels-this-year/

The memo is available here: http://static.berkeleyearth.org/memos/explaining-declines-in-us-carbon.pdf

I’m also attaching the summary figure. The black portion represents actual emissions, with each “wedge” showing how much additional emissions would have occurred without the associated factor.

U.S. total co2 reductions

About these ads

42 thoughts on “US Carbon Dioxide Emissions – down by source

  1. Well, with the economy finally on the upswing after half a decade or so of doldrums the emissions will probably start to go up again.

  2. I’m going to venture a guess that the upswing after 2010 is partially in response to the drought. Energy use in various areas would have been ramped up to compensate?

  3. Seeing this chart though makes me wonder as to how much CO2 from Mother Nature is being generated from the 3.6 million square mile area of the USA.

  4. Something tells me that statistically, “Houston, we have a problem” with this research result. My guess is that the measurement error width is as deep as the actual y axis, and buries the derived line graphed data.

  5. Shouldn’t we all be starving and shivering in the dark if our carbon emissions are declining? I am pretty sure Marc Morano told me this.

  6. Rust, the drop in C02 emissions is because of the switch to natural gas from coal and because of the drop in US manufacturing. The drop in US manufacturing is a bad thing. You will be shivering in the dark when you are no longer able to afford electricity.

  7. Pamela Gray says:
    December 7, 2013 at 12:18 pm
    Something tells me that statistically, “Houston, we have a problem” with this research result. My guess is that the measurement error width is as deep as the actual y axis, and buries the derived line graphed data.
    >>>>>>>>>>>>>>>>>>>>>>

    Thanks for raising that. I’m so used to climate data graphs without error bars, and having any complaints about same fall on deaf ears, that I now find myself in need of reminding that NOT having them isn’t normal, isn’t right, and without them the graph is essentially meaningless. I suspect for a graph like this that adding the error bars would simply prove that the estimates are meaningless, which is most likely why they aren’t there.

  8. rustneversleeps: To get to Kyoto levels, you would be freezing in the dark. Take out the gas and wind from that chart, and its a little under 2/3 of the cuts are likley due to the effects of the worst recessions since the 1930s. If I am generous with the cuts due to economic reasons, that gives us 50% of the cuts are due to economic reasons.

    To get 6% under 1990 levels, the US would need to have the same cuts again, as seen to date from 2008. That would imply a recession that was 3 times worse than just suffered, just to meet 6% under 1990 levels. Now imagine the 80% cuts called for….

  9. GlynnMhor says:
    December 7, 2013 at 11:55 am

    “Well, with the economy finally on the upswing after half a decade or so of doldrums the emissions will probably start to go up again.”

    Yeah it is a good experiment to show what killing the economy does. Green jobs anyone (an oxymoron if I ever saw one).

  10. Looking at Berkely’s method in determining the effect of wind shows a basic misundestanding of how wind generation affects the grid and therefore emissions. Wind is uncontrollable and requires
    back up, which means spinning gas turbine power, which produces no power but emits carbon.
    The determination of how much carbon emissions are reduced thru the use of wind requires
    empirical data, not simply assuming a one to one reduction of wind for gas/coal, etc. which is totally wrong. They also, for some strange reason, used old data to determine the fuel mix displaced, with 50% coal which went away quite some time ago. It’s in the mid thirty percent range these days. This is a sloppy and lazy analysis that can be mostly ignored if one is looking for accuracy.

  11. As a chart-person, I object to the chart Zeke provided. What exactly is “CO2 Reductions by Month” Shown as a positive value. From what baseline, etc. Okay I can read it, but here is a more accurate chart from EIA (released Oct 21, 2013).

    US Energy CO2 emissions have fallen from 6.023 billion tons in 2007 to 5.290 billion tons in 2012 (or by 12.2 per cent). Who else had a 12% decline in emissions in the last 5 years.

    How about SO2 emissions. Down 72% between 1990 and 2011. The US air is cleaner than it has been in many decades.

    Coal CO2 emissions fell by 12% last year while Natural Gas emissions grew by 4.5%

    In terms of Electricity generation, the switch to Natural Gas is even more stark, Natural Gas rose 211 billion kWh’s, while coal fell 215 billion kWh (and wind provided a measly 20 billion kWh more).

  12. I would like to see a graph for a longer period, scaled 0-520. Why does the graph start in 1990, given that EIA shows US energy consumption data from 1776.

  13. GlynnMhor says:
    December 7, 2013 at 11:55 am
    “Well, with the economy finally on the upswing after half a decade or so of doldrums the emissions will probably start to go up again.”

    Finally on the upswing, that’s a good one. Like a junkie who manages to get rid of his withdrawal symptoms by shooting himself up 4 times a day saying, now that I feel better I think I’ll start confronting my addiction any moment now.

    The business cycle is nearly over, what Bernanke managed was to re-inflate the bubble but barely so, all your business activity is artificially pumped up by 85 bn USD a month money creation.

    You’re on the upswing when you manage to grow the economy without that artificial money creation.

  14. DirkH says (December 7, 2013 at 3:17 pm): “The business cycle is nearly over, what Bernanke managed was to re-inflate the bubble but barely so, all your business activity is artificially pumped up by 85 bn USD a month money creation.

    You’re on the upswing when you manage to grow the economy without that artificial money creation.”

    Current US fiscal policy explained (skip to 27:00):

  15. It is important to know these are not measurements, they are “guestimates”. They do not know how many miles I drove, nor how much CO2 my particular car emits.

  16. With temps of 10 degrees F, my CO2 emissions are definitely going up this week, as will the CO2 output of much of the U.S.

  17. data sources

    “Energy use and carbon emission data used in this article is available from the Energy Information Agency (EIA) Energy Monthly Review.2 Ground transportation data is taken from the Federal Highway Administration’s Traffic Volume Trends.3 Population data is taken from the U.S. Census.4 .

    ####################

    Let’s see if we can explain in simple terms what this analysis tries to do.

    1. We have data. It is taken as a given. If folks want to go argue with that data they can knock themselves out. We choose, as it our rational right, to accept the data as given and then
    try to explain the causes for the reduction in CO2, subject of course to the acceptance of that given.
    2. Errors. Error analysis is not a part of this exploration of the data. The first order of business is to see what if anything the data as given tells use. Can we begin to ferret out the causes
    of the decline in C02. Again, if folks want to take the next step they can knock themselves out.
    One of the points of a memo as opposed to a full blow paper is that its aims are somewhat limited.
    3. The biggest uncertainty is the counterfactual which everybody missed. Why? because they have canned responses for any claim.

    So, lets take Alvin as an example. These are not measurements THEY CANNOT BE MEASUREMENTS. that’s because the estimates are constructed as they can only be constructed, by a counterfactual. we cannot get in a time machine and go back to see what the world would have been like if we changed this or that. The best we can do is construct a good faith estimate. C02 declined. Why? what are the possible causes and how can we BEGIN to come to an understanding of it. And yes, Alvin of you drove a 100K miles it would not make a difference.

  18. Three thoughts:

    1. Steven, you write in a flippant style using phrases like “knock yourself out”. Bad form in my opinion.

    2. Data can only talk when you have first determined whether or not the data is significant (outside the standard error). It is your job to knock yourself out to assure us that your data has something to say beyond “look at my wriggles”.

    3. You must assure against significant minutia. This is when a string of data is taken from a longer string and given an enlarged view such that wriggles become obvious. It may look significant in this enlarged view, but it may still only be minutia when placed back in the longer string.

  19. Well when ‘you gotta put a price on carbon’ Ms Gillard and Co, the price will soon become apparent-

    http://www.dailytelegraph.com.au/news/labor-under-pressure-to-axe-the-carbon-tax/story-fni0cx4q-1226777879365

    ‘While the carbon tax is now $24 a tonne, the effective cost of the emissions reduction on the basis of revenue raised is $21,000 per tonne.’
    Unless of course Big Climate want to argue the finer points of methodology that we can all agree upon with such important calculations? That sort of dough would buy a heap of EU carbon credits to make us one of the cleanest, Greenest nations on the planet. Don’t you agree Mr Shorten and Co?

  20. If accurate, which it probably is, this graph essentially encapsulates the actual picture of the U.S. economy in recent years, including the double-dip recession. It shows switching of some power from coal to natural gas actually had far less effect on CO2 emissions than the combined effects of less energy/material consumption (fewer miles people could afford to drive, drops in industrial production, etc). The latter, the enviro-activist dream, is also known as increased poverty. GDP reporting is biased towards max counting of surplus government spending growth and misleading tricks, but variation in the amount of carbon fertilizer emitted over the past several years is a near-proxy for actual economic consumption / production / prosperity (or lack of it) meanwhile.

  21. Henry, just in case, I wanted to clarify that GDP (gross domestic product) reflects goods made here, even if those goods are part of something that is put together in some other country. GNP (gross national product) reflects both GDP and overseas return on investments. Most countries report GNP. The US switched over to GNP to be comparable with the numbers being reported by other countries. It is always interesting to compare the two numbers. It can reflect the changing “mix” of just making things here compared to the combination of making stuff here AND investing in businesses in other countries to make stuff there. It is why we can be losing jobs here yet have a growing GNP.

  22. ” Errors. Error analysis is not a part of this exploration of the data. The first order of business is to see what if anything the data as given tells use. Can we begin to ferret out the causes
    of the decline in C02. Again, if folks want to take the next step they can knock themselves out.
    One of the points of a memo as opposed to a full blow paper is that its aims are somewhat limited.”

    Yes, and let’s put the cart before the horse. Without knowing the error limits of the data, how can you do any analysis? Did you ever receive any training in a science? In my day some of the first science classes in high school discussed this issue.

  23. DirkH says:
    December 7, 2013 at 3:17 pm

    The business cycle is nearly over, what Bernanke managed was to re-inflate the bubble but barely so, all your business activity is artificially pumped up by 85 bn USD a month money creation.

    You’re on the upswing when you manage to grow the economy without that artificial money creation.

    The 85 bn that the Fed spends monthly to buy Treasury Securities on the open market removes the interest income from the regular economy. Instead of that interest income going to the private sector, i.e: individuals, university funds, pension funds, businesses, banks, or trust funds–which would grow the economy–it is now going to the Fed, and that interest income after expenses must be returned to the US Treasury by law (since 1947). The 2012 Annual Report is here.

    The amount returned to the US Treasury, by law, in 2012 was $88 billion ($88,418 million)

    Comprehensive net income before interest on Federal Reserve notes expense remitted to Treasury totaled $90,516 million in 2012 (net income of $90,569 million, reduced by other comprehensive loss of $53 million). Distributions to Treasury in the form of interest on Federal Reserve notes totaled $88,418 million in 2012.

    NB: those are not ‘Federal Reserve Notes’ as in dollar bills (approx. $650 billion/yr), which the Fed buys, no matter what the face value, from the US Bureau of Printing and Engraving for about $0.07/ea ($0.08 or $0.09 for the new Oct/2013 $100 color bills, can’t remember). As explained in a New York Fed doc here, they are “purchases [that] kept the quantity of Treasury securities held by the private sector lower than it would have otherwise been in order to put downward pressure on longer-term interest rates.”

    So called Quantitative Easing, except that it is essentially an asset swap (from savings accts at the Fed to checking, or technically, from ‘securities’ accts to ‘reserves’) that winds up being a tax on the US economy to the tune last year of $88 billion!

  24. Bill,

    The black line is the EIA data, and shows a 12 percent decline. The wedges show what would have happened if each factor identified has remained at 2005 levels (either per-capita or constant, depending on the specific scenario).

    Col Mosby,

    We are using wind generation rather than wind capacity, with the assumption that if those wind turbines had not been built an equivalent amount of generation would have been required from other sources. I’m well aware that wind does necessarily offset baseload generation due to intermittency (though this is somewhat mitigated by backup gas turbines and grid storage), so in practice this might be somewhat overestimating wind-related carbon reductions as peaking/cycling plants tend to be a bit less carbon-intensive. The 50% coal number you mention is assuming a fixed 2005 grid mix as the counterfactual.

  25. Mike Jonas,

    Figure 24 in the memo shows a version with a 0 to 550 scale for reference.

    Its also worth noting that the Residential, Commercial, and Industrial categories represent non-electric CO2 emissions in each. The memo also breaks down electricity use reductions by end-use, but they are excluded from the summary graph to make it more readable. The interesting take-away is that almost all declines in electricity use and the majority of declines in non-electric CO2 emissions are focused in the Industrial sector, as heavy industries have outsourced production overseas over the last few decades.

    It is important to note that some of the emission reductions attributable to declining industrial energy use might not represent actual emission reductions from the perspective of a global carbon budget. As industrial production moves from the United States to countries like China, the energy and emission intensity may actually increase. However, as this analysis focuses solely on U.S. emissions, we do not attempt to account for these shifts in emission location.

  26. The US with no Carbon Tax but an effective fracking technology reduced its emissions by 208.0 million tons.

    Australia with its new Carbon Tax reduced its emissions by 0.3 million tons. Nice price per ton there.

    The UK with all its new wind power increased its emissions by 20.5 million. Germany with all its new wind and solar power increased its emissions by 14.0 million tons.

    The market and effective technology is the way to reduce emissions, not Carbon Taxes and pie-in-the-sky subsidized wind and solar power.

    Society needs to do things that simply work. Not things that make some people feel better but simply don’t work. Its a simple principle that took us out of the stone age.

  27. Bill Illis says:
    December 8, 2013 at 6:05 am

    The US with no Carbon Tax but an effective fracking technology reduced its emissions by 208.0 million tons. . . . Germany with all its new wind and solar power increased its emissions by 14.0 million tons.

    Bill: powerful argument. Would you have the links easily accessible for this? I could look them up on my own, but it would take me over 90 minutes because I don’t know where to go. Thx.

  28. “increasing gas prices in late 2012/early 2013 have led to somewhat increased coal use”
    This is an important point. Coal and NG are competitors. Without coal, NG would have a tendency to ramp up. But Obama’s EPA, in its infinite wisdom is trying to shut coal down.
    All of this talk about carbon accounting though, is pointless. In terms of climate, it means absolutely nothing. Its’ sole function lies in the squabbling among nations over who is to “blame”, who should pay, and how much.

  29. All this teeth knashing scrutiny and measuring of CO2 emission is like counting fleas and tics on deer. The purpose would be? Certainly the deer don’t care. Other animals don’t care. Hunters don’t care. But someone obviously cares and thinks that humans are to blame for fleas and tics on deer. So a tax on the fleas and tics on deer has been initated. Sounds silly doesn’t it. But apparently “silly” has managed to become the world’s new reality.

  30. @Bill Illis

    Powerful, just wow. And thanks for the follow-up with links. I can’t help but feel that including robust nuclear expansion would greatly enhance our energy portfolio, too. Like you, I just want energy technology that actually works.

  31. GlynnMhor says:
    December 7, 2013 at 11:55 am

    Well, with the economy finally on the upswing after half a decade or so of doldrums the emissions will probably start to go up again.

    Oh frack you. Jokes aside, I want co2 levels in our atmosphere to go up to 800ppm. I love co2.

  32. timetochooseagain says:
    December 8, 2013 at 12:34 pm

    I estimate that 92% of this reduction can be attributed to a poorly performing economy, not to any improvement in technology:

    http://devoidofnulls.wordpress.com/2013/12/08/back-of-the-envelope-estimate-of-emissions-reduction-due-to-the-great-recession/

    Hardly something to be too pleased about.

    In October of this year it was reported that US co2 emissions fell nearly 4% in 2012, reaching the lowest level since 1994, while the economy grew by 2.8%. Re-do your “back-fo-the-envelope-calculations”.

    http://www.usatoday.com/story/news/nation/2013/10/21/us-carbon-emissions-energy-lowest-1994/3146123/

  33. Jimbo- As it presently stands, my estimate is a “back of the envelope” calculation, and it is not as sophisticated as I would expect necessary for a formal paper. That being said, if you or anyone else has the skills and knowledge, I would be happy to collaborate on a formal paper on this subject.

    As for your belief that the economy growing and CO2 emissions shrinking in *any* way refutes my argument, no, it absolutely does not. In my counterfactual scenario, total private expenditures grew about 15% between 2007 and 2012. In *reality* what happened was that private expenditures were 3 percent *less* in 2012 than in 2007. In other words, the private economy is about 16% *smaller* than it it would have been if it had kept growing, rather than experiencing a recession. But my counterfactual scenario involves an economy which is 5% *more* efficient (metric tons/$) than the actual economy! I assume *faster* technological progress than actually occurred, and I *still* get the bulk of reductions coming from a poorly performing economy. Where did you think the other 8% of reductions came from? In my counterfactual scenario, emissions are still lower in 2012 than 2007-by by a very small amount, *rather than a very large amount.*

Comments are closed.