11 June 2012

Lowballing Carbon Dioxide Emissions Projections

The IEA has released a new analysis that helps to demonstrate the systemic failure of policy analyses focused on carbon dioxide emissions reductions.

In the new report the IEA projects that by 2030 the world will be emitting about 45 Gt (gigatonnes) of carbon dioxide. Yet, in 2008, just 4 years ago the IEA was projecting 40 Gt CO2 for 2030 (see it at p. 11 at this PDF).

Where did the extra projected 5 Gt CO2 for 2030 (just about equal to an extra United States) come from over the past 4 years?

It came from a systemic underestimate for future emissions that is built in to almost all such exercises. The IEA assumed in 2008 that future emissions would grow from 2005 to 2030 at 1.5% per year. Actually, from 2005-2010 emissions increased by 2.4% per year (data from PBL in this PDF). The 1990 to 2010 average was a 1.9% increase per year, and 2009 to 2010 was a whopping 5.8% increase.

Thus, in 2008 the IEA used a low-balled 1.5% annual rate of increase in projected emissions to 2030. In the years since, actual emissions have increased by much higher than this rate, which means that the new 2012 projection for 2030 needs to start at a higher starting point than was projected just several years ago. Hence, in the new 2012 report the IEA has quietly increased the 2030 level of emissions o 45 Gt from 40 Gt. I actually anticipated this revision almost exactly when the 2008 IEA report came out.

So, based on this experience, what rate is the IEA now using to project emissions from 2015 to 2030 under a business as usual scenario?

1.3% (45 Gt CO2 in 2030, 37 in 2015)

So much for learning from experience.

Such estimates matter because all of the estimates of cost of emissions reductions and level of policy effort required that derive from the IEA projections are contingent upon an assumption of a rate of emissions growth that is much lower than observed in recent years and decades.

This repeated failure to accurately describe the actual magnitude of the emissions reduction challenge occurs so widely that it is the systemic failure of policy analysis of CO2 emissions reductions. We repeatedly trick ourselves into thinking that the task is easier than reality shows.

Unless things change in how such analyses are done, we should expect that future IEA projections will be revised upwards and the charade to continue. A more technical explanation of this dynamic can be found here in PDF and an alternative approach to policy analysis of emission reductions can be found in TCF.

10 comments:

Bret said...

Are projected CO2 levels also being revised upwards an equivalent amount?

MattL said...

"It's worse than we thought!"

Paul said...

Roger, So with CO2 emissions higher than IEA projections & off the chart for Hansen's scenario C, and GMT flat for another half dozen years since your Prometheus comments on this subject I would really like dots to be connected by the 'consensus'.

Adrian said...

If there is one thing we learn from history, it is that we never learn from history.

Joel Upchurch said...

Since CO2 forecasts are economic forecasts rather scientific forecasts why would any sensible person assume that they are accurate?

Tom said...

The same is happening with energy consumption estimates. The DOE's EIA is lowballing energy consumption by about 21% in 2030, if my calculations are accurate.

Paul said...

"#5. Joel Upchurch said...Since CO2 forecasts are economic forecasts rather scientific forecasts why would any sensible person assume that they are accurate? Mon Jun 11, 03:38:00 PM MDT"

Joel, please elaborate.

It is my understanding that anthropogenic CO2 contributions are theoretic and based on MMORPGs and not direct scientific observation.

markbahner said...

"The IEA assumed in 2008 that future emissions would grow from 2005 to 2030 at 1.5% per year. Actually, from 2005-2010 emissions increased by 2.4% per year (data from PBL in this PDF). The 1990 to 2010 average was a 1.9% increase per year, and 2009 to 2010 was a whopping 5.8% increase."

"So, based on this experience, what rate is the IEA now using to project emissions from 2015 to 2030 under a business as usual scenario?

1.3% (45 Gt CO2 in 2030, 37 in 2015)

So much for learning from experience."

So...what is your estimate for the annual percentage emissions growth from 2015 to 2030?
The EIA says 1.3%. What do you say?

P.S. Just to keep things interesting, I'll predict 1.0% per year, from 2015 to 2030.

markbahner said...

Hi Roger,

I wasn't asking a rhetorical question. ;-) I really would like to get your prediction for the rate of growth in world CO2 emissions from 2015 to 2030. As I wrote previously, the EIA says 1.3%. I'll say 1.0%. What do you say?

As you have pointed out in other areas, "Prediction is very difficult, especially about the future."

Your statement that, "This repeated failure to accurately describe the actual magnitude of the emissions reduction challenge occurs so widely that it is *the* (your emphasis) systemic failure of policy analysis of CO2 emissions reductions" is pretty strong.

Personally, I've become much more humble in light of the failure (to this point, anyway) of my predictions, formulated circa 2004, that the world would most likely have emissions similar to the SRES B1 scenario over the course of the 21st century:

http://markbahner.typepad.com/random_thoughts/2006/04/complete_set_of.html

My only defense is that I didn't dream China would go the crazy coal route they've gone...I thought they'd sensibly go nuclear.

And if one looks at the graph that Hans Erren created circa 2003, my predictions actually looked like *over-estimates* of the likely CO2 emissions:

http://members.casema.nl/errenwijlens/co2/co2sres.gif

In fact, I still maintain that it is extremely unlikely that world CO2 emissions will continue to rise beyond ~2030. That's because:

1) Biofuels will start to replace oil (and by "biofuels", I mean more fuels from algae or genetically modified bacteria, not alcohol from biomass), and

2) I think world coal use will peak sometime between this year and 2030. (In fact, I think U.S. production is already at or beyond its peak.)

Roger Pielke, Jr. said...

-9-Mark bahner

Thanks ... I am not much for predictions (unless they are of soccer games;-), but in this context I think that an argument can be made that policies should be robust to such uncertainties.

The IEA policy recommendations are contingent upon the scenario it sets forth for future emissions rates (and they are of course not alone). That means that they policy will fail if their prediction is off. So why not think about policy options robust to predictive uncertainties?

But specifically to your question, I'll take the over on your 1%, happy to bet one beer;-)

Post a Comment

Note: Only a member of this blog may post a comment.