Updated: 80% of world CO2 emissions are related to population levels.

[Iterate]

16. You were right to not use the GWP.

There are just too many variables embedded in that index, and too many non-consumptive ones to make it useful.

I think I'll have to disagree on most of the rest of it though. That seems a crazy tall order given how convincing your graph looks and how close the correlation is. It's almost too close. There's something wrong in this and I admit I don't have a full grasp of what it is. I suspect that the steel/cement variance substituted well enough for for both GWP and average CO2 per capita that it was a reasonable match to the observed CO2. I haven't plotted an alternative, but could if you want. Mainly I'd just point out that if you apply this to individual economic blocks it falls apart.

Here's a tidbit. We know about Bangladeshi population growth in the past decades.

Cement production in Bangladesh almost doubled in the past five years, but the country's per capita consumption is still one of the lowest in the world.

Even under-developed West Bengal of India consumes cement one and a half times higher than Bangladesh does.

“Per capita cement consumption in Bangladesh is less than 100 kilograms (kg), which is 150kg in West Bengal and more than 1,000kg in China,” said Mike Cowell, managing director of Lafarge Surma Cement, Bangladesh.

Cowell was talking to The Daily Star recently at his office in Gulshan. He has been heading the cement company in Bangladesh for the past five years.
www.thedailystar.net/newDesign/news-details.php?nid=234830

I didn't re-graph anything, but just quickly normalized on 1970 and checked the correlations of the variance pairs. They're all very high, higher than that of GWP and Steel+cement at .89. I didn't see how assigning a higher proportion of the value to population helped the overall accuracy.

Observations:

First of all, there's so much carbon embedded in the steel/cement proxy that it can stand in for the measured CO2 as well as the GWP. If consumption falls off the radar, either by substitution or efficiency, it doesn't get counted. Steel and cement are sensitive to development, not population. As demand gets largely satisfied in one developed market, production has moved elsewhere.

Second, the use of 'total population' might be good for giving an overall direction and fate, but it masks the bigger and more subtle story. In a sense there is no such 'thing' as population, it's just a construct that embeds a swarm of variables. It matters less to CO2 totals if you are born than it does where you are born and to which class you are born. Born in the USA means 20x the carbon of an average African, and born 1% might be 20x more.

Plus, the CO2 burden that comes with your birth doesn't get paid right away. First there are eighteen years or so of social cost before the CO2 payment of transportation, work, and setting up a household begins. Non-traveling, book-reading retirees are likely past their CO2 peak. So, in essence, a young country is one with a CO2 emitting future.

And that brings up another point. After looking through this for a day, I'm reminded just how deterministic the type of energy and consumption culture one is born to can be. I don't think that it's because of virtue that Europeans consume half as much as the US for the same life quality, but more a matter of the consumption culture, economic assumptions, infrastructure, and leadership. The bad news is that those things are nearly as resistant to change as is our DNA.

Off topic, but interesting. The European population levels stayed nearly constant from from about from the 4th to the 10th centuries, largely in part to slave raiding from the north and south. It was begun by the Romans, who didn't allow slaves to breed, worked them to death, then raided for more. That 'economic' pattern continued for six centuries after the Romans were gone. Those patterns die hard.

OK, now that I've worked myself into some dark thoughts, some tables I noticed. These slopes don't match population increase:

from: http://petrolog.typepad.com/climate_change/2010/01/cumulative-emissions-of-co2.html

Annual Emissions of Carbon Dioxide from the Combustion of
Fossil Fuels and Production of Cement by Region: 1850-2006
Source of data: Carbon Dioxide Information Analysis Center, sited above.


Cumulative Emissions of Carbon Dioxide and Percentage by Region: 1850-2006
Source of data: Carbon Dioxide Information Analysis Center, sited above.


And damn, somewhere here I had a tidy chart of Chinese steel production since 2000 that explained a post-2004 hook in your data. Now lost to the clutter, forever.
So I'll substitute with this. It's only for the EU and it's only one year, but it doesn't seem to correlate to population change, but it does to development.

http://setis.ec.europa.eu/newsroom-items-folder/per-capita-cement-consumption-in-selected-countries/image_large


Some sources of the week:
Analysis of the Relationship between Growth in Carbon Dioxide Emissions and. Growth in Income.
www.econ.cam.ac.uk/rstaff/grubb/publications/GA12.pdf

Economic Growth and CO2 Emissions: a Nonparametric Approach
webdoc.sub.gwdg.de/ebook/serien/e/CORE/dp2001-12.pdf

Population Is a Critical Factor for Global Carbon Dioxide Increase
jhs.pharm.or.jp/data/55(1)/55_125.pdf

Overpopulation and Climate Change
http://www.biologicaldiversity.org/campaigns/overpopulation/climate/index.html

Global and regional drivers of accelerating CO2 emissions
http://www.pnas.org/content/104/24/10288.abstract

One approach I'd like to look at is extrapolation of 1950-1960 data on consumption v. population to see where that leads. Another would be to use miles driven per capita as a consumption proxy v. CO2 v. population, since it might reflect changes in efficiency, or it might just be interesting for uncovering something. I'm not finding good data for those though.

Afterthought:
Thanks for doing this, though I have to admit that after having gone through it, between the recalcitrance of my fellow Americans and insight into development plans for some highly populated countries, I'm firmly more pessimistic than anyone else on the board.