Robbie Andrew

scholar orcid twitter
CICERO Center for International Climate ResearchHome   |   About me   |   Contact me

Drivers of atmospheric CO2

First published: 29 October 2016, with subsequent updates

Download video: MP4 video | animated GIF | high-resolution GIF


While we measure the amount of CO2 in the atmosphere in mere parts per million (ppm), this small amount has a substantial effect on the temperature at the surface of the Earth. The amount has now increased by 40% since the industrial revolution, leading directly to increased temperatures worldwide.

There are four main drivers of the changes in the level of CO2 in the atmosphere.

The seasonal cycle

The first is the seasonal cycle, dominated by the forests of the northern hemisphere, absorbing CO2 via photosynthesis in spring and summer, and then, in autumn and winter, the decay of their fallen leaves to release some of that CO2 back into the atmosphere. This cycle can be seen here as a pulse, sometimes imagined as the planet 'breathing'.


The second most important driver is our emissions from burning of fossil fuels and deforestation. This transfers carbon that has been stored for a very long time underground into the atmosphere in the form of CO2. Nature partially compensates for this increased level of CO2, with both forests and oceans absorbing more (the latter leading to acidification of the oceans), but this amounts to only a little over half of the new CO2, with 44% of each year's emissions remaining in the atmosphere.

The El Niño Southern Oscillation

The third driver is often referred to as El Niño, or more correctly as the El Niño Southern Oscillation (ENSO). ENSO describes a naturally occurring cycle of pressure and temperature differences across the width of the Pacific Ocean. In El Niño years there tend to be more droughts in important forested areas, and that reduces the productivity of forests, in turn reducing their absorption of CO2 from the atmosphere. The drier conditions can also lead to increased wildfires, sending even more CO2 into the atmosphere. Strong El Niño conditions can be seen here in 1983, 1988, 1998, and 2015.


The fourth major driver is the periodic eruption of volcanoes, the most significant of which in the last few decades was Mt Pinatubo in the Philippines in June of 1991. The enormous amount of fine debris thrown into the atmosphere - probably the largest since Krakatoa in 1883 - stayed there for many months, blocking sunlight, reducing global temperatures by about 0.6°C, but also increasing diffuse sunlight, which stimulates tree canopy growth, with the overall effect of increased natural carbon sinks. Despite a strong El Niño in 1992, atmospheric CO2 rose only very slowly, pulled back by the effects of the eruption. Note that it's not about the CO2 volcanoes emit when they erupt: Pinatubo emitted about 50 MtCO2 in 1991, compared with our own emissions of about 29000 MtCO2 in the same year.

Relative effects

The relative effects of ENSO and our emissions on the growth of atmospheric CO2 can be determined by analysis of these data. The bars at the bottom of the animation above show these effects, with the effect of emissions being always positive and growing, while that of ENSO flicks between positive and negative depending on whether the planet is currently in El Niño (to the right) or La Niña (to the left). The effect of the sea-surface temperatures used to measure ENSO on measured atmospheric CO2 is delayed by several months because of the natural processes involved. The year 2015 was a particularly strong El Niño year, and, when combined with our record-high emissions, atmospheric CO2 levels rose more sharply than usual.


NOAA releases weekly average concentrations of CO2 measured at Mauna Loa, Hawaii here. These data are similar to, but not the same as, estimates of global average concentration. Mauna Loa data are often used because they are the longest set of measured data.

Global emissions data are from the Global Carbon Project.

ENSO temperature data are from the Hadley Centre's Sea Surface Temperature dataset HadSST.3.1.

For the relative effects of ENSO and emissions I have relied on the relationship found by Richard Betts and colleagues.

Mauna Loa weekly PPM

blog post

In 2016 the atmospheric concentration of CO2 stayed above 400ppm all year.Learn more »

Reaching peak emissions

Global emissions growth appears to have slowed in the last two years. Rob Jackson and colleagues discuss the causes for this and the potential for emissions to peak in the near future.Learn more »

Fairness and Ambition

We are rapidly depleting the global budget for CO2 emissions determined by a 2C limit. How do major emitters’ pledges compare to the path we must take?Learn more »

Global Carbon Budget Figures

Every year the Global Carbon Project publishes a number of figures demonstrating the latest global carbon budget, and these are freely available for use in a number of formats.Learn more »

Web design by Robbie Andrew