Andrew Jarvis

Lancaster Environment Centre

Lancaster University


plus44 1524 593280


The carbon pause?

The Guardian has twice drawn attention to the observed lack of growth in global anthropogenic carbon emissions over the last three years (1,2). Within these articles scientists were quoted as cautiously interpreting this as an indication that national efforts on emissions reduction may be starting to take effect. Although those offering such interpretations were necessarily reserved in their optimism, pointing out such effects fall far short of the requirements of the Paris Agreement, it is worth investigating the validity of these claims less we deceive ourselves.

The data on which these claims are based (3) are shown in Figure 1. I make the following four points in light of these observations and the associated claims being made:

i. Within the uncertainty specified for these data it appears statistically invalid to claim global fossil fuel and cement production emissions emissions have not grown for the past three years.

ii. Even if we ignore these uncertainties, as is being done in these reports, we would then have to also accept and report that similar ‘pauses’ occurred in the 1970’s, 80’s and 90’s.

iii. The climate system responds to total emissions and, therefore, it is important that we include land use change in our narrative. It is also important to appreciate that land use change emissions are as much a product of economic activity as those arising from fossil fuel use and cement production. It would appear that any recent stabilisation of fossil fuel and cement production emissions have been fully offset by an increase in land use change emissions.

iv. The media also reported that the recent pattern of fossil fuel and cement production was in the face of strong growth in the global economy. As the data in Figure 1 show, there appears to have been no significant departure from the long-run secular decrease in the carbon intensity of global Gross Domestic Product (GDP) during this period. This secular decrease simply arises from the fact that total emissions grow less quickly than GDP, but grow they appear to be still doing.

This debate parallels the one had over the hiatus in global mean surface temperature observed between ~1998 and 2013. That narrative developed because the scientific evidence strongly suggested any lack of growth in global mean surface temperature was transient, as indeed it turned out to be (4). Given the global economy is no less dynamic we should be equally cautious when attempting to interpret short-run variability in emissions. As with the global mean surface temperature data, our interpretations of such behaviour must be set within the appropriate long-run context.

Figure 1. Global anthropogenic carbon emissions from fossil fuel and cement production (), land-use change () and in total () (source – 3). Shaded areas represent one standard deviation uncertainty. Also shown is the ratio of total emissions to the World Bank global GDP (+) (source - 5).



3. Le Quéré C et al., Earth Syst. Sci. Data, 8, 605–649.

4. Marotzke J and Forster PM, Nature, 517, 565–570



Related papers

Energy returns on investment and the growth of global industrial society. Jarvis AJ. Ecological Economics, (in review)

Resource acquisition, distribution and end-use efficiencies and the growth of industrial society. Jarvis AJ, Jarvis SJ and Hewitt CN. Earth System Dynamics, 6, 1–14, (2015) (pdf)

Climate-society feedbacks and the avoidance of dangerous climate change. Jarvis, AJ, Leedal DT and Hewitt CN, Nature Climate Change, 2 (9), 668-671 (2012) (pdf)