Simon Donner makes a disturbing point:
If you look at the global fossil fuel emissions data, all of the major disruptions to energy and oil use in the past 60 years caused carbon emissions to drop or level off. Annual emissions would later continue to rise at a rate similar to that before the disruption, but the total annual emissions would not “catch up” to where it “would have been” without the disruption.
The recent world financial crisis appears, on the surface at least, to be an exception. Carbon emissions stopped rising in 2008 and 2009, but rebounded so strongly in the past couple years, that emissions have reached the level to which they appeared to be headed, presuming linear extrapolation, before the crisis.
As an aside, this behavior of the timeseries not returning to its previously expected value is a sign of a non-stationary process (economic output, to which CO2 emissions are linked, is often assumed to be non-stationary, which leads to a unit root in the timeseries (http://en.wikipedia.org/wiki/Unit_root).
In contrast, global avg temperature data do show that after a dip (e.g. due to a strong volcanic eruption), the data go back to the level that they would have had in the absence of the dip. I.e. global avg temperature data are trend-stationary.
This relates to the “debate” whether global warming could be just a random process (implying non-stationary behavior), e.g http://ourchangingclimate.wordpress.com/2010/03/18/the-relevance-of-rooting-for-a-unit-root/ and refs therein.
I wonder what the graph would look like if you separated into two graphs, one for OECD countries, another for the rest of the world? Donner mentions China – but yes, his basic point stands: globally, there are only signs of increased carbon fuel consumption. We’re going the wrong way with no real strategy for changing direction. Global institutions (e.g. the World Bank’s tilt to green growth) face the realpolitik of a largely still-developing world who have no intention of sacrificing basic economic advancement.
The material provided in this post is the real reason your other post regarding why climate sensitivity is not as relevant as previously thought.
As Dan Nocera, Roger Pielke Jr. and my humble self have been noting for different lengths of time, energy consumption is growing more quickly than most have predicted, especially the DOE’s EIA and the IEA. At some point, even at a very low value for sensitivity, our cumulative emissions will prove significant enough to drive climate change.
The IPCC previously estimated this would occur in the 2030s or 2040s. My own back-of-the-envelope calculations put it at between mid-century and 2075.
But this is a real issue, unlike weird things like new hockey sticks or Xtreme weather. To the extent that you explore this issue at the expense of phony ones, you will be doing the climate discussion a real favor.