The hopeful prospect that doubling the atmospheric carbon dioxide levels might yield a relatively mild increase in global average temperatures is unlikely, according to a comprehensive new study.
Equilibrium climate sensitivity (ECS) is conventionally defined as the increase in Earth's average surface temperature that would occur if carbon dioxide concentrations in the atmosphere were doubled and the climate system was given enough time to reach an equilibrium state. Scientific American once called it "the most important number in climate change."
First, let's consider where the planet's climate stands right now. The level of carbon dioxide in the atmosphere has increased from the pre-industrial level about 280 parts per million (ppm) to 415 ppm today. The additional carbon dioxide has been added largely due to humanity's burning of coal, oil, and natural gas, along with plowing down forests. According to ice core data, today's carbon dioxide levels are higher than at any point in at least the past 800,000 years, and more recent research suggests that the current level is the highest it's been in the past 23 million years. Average global temperatures are currently about 1.2°C above pre-industrial levels.
In 1979, the U.S. National Academy of Sciences conjectured that ECS was probably somewhere between 1.5°C and 4.5°C per doubling of CO2. In 2013, the Intergovernmental Panel on Climate Change concluded that ECS is likely to be between 1.5°C and 4.5°C. In other words, the best estimate of sensitivity remained basically the same more than three decades later.
The new study narrows the range of the probable temperature increase. With doubled atmospheric carbon dioxide, the eventual warming would probably be between 2.6°C and 3.9°C. There would be less than 5 percent chance of staying below 2°C—and a 6 to 18 percent chance of exceeding 4.5°C.
Their evidence stems, in part, from a better understanding of the atmospheric feedback processes. The researchers evaluated the evidence for how changes in carbon dioxide, water vapor, surface reflectivity, and, most importantly, cloud cover affect global temperature trends. One of the greatest uncertainties with respect to future global temperature trends has been whether changes in clouds will tend to cool the planet. The researchers conclude that warming-induced changes in clouds will tend to boost rather than moderate future temperatures.
They also argue that the empirical data for historical warming of 1.2°C over the past century or so suggests that an ECS of 1.5°C is implausible. And as a third line of evidence, they looked deeper into prehistoric climates, including the succession of ice ages and the mid-Pliocene warm period, to see how changes in atmospheric carbon dioxide levels correlated with changes in global average temperatures. In the depths of the ice ages, the level of atmospheric carbon dioxide was just below 200 ppm; global temperatures were between 3°C to 7°C lower than the pre-industrial average.
During the mid-Pliocene warm period, carbon dioxide levels hovered around 400 ppm and average global temperature was 1°C to 5°C warmer than the pre-industrial average. Sea level was 20 to 30 meters higher than now, indicating significant reductions in Antarctic glacial ice. The paleoclimate data, they conclude, suggests the ECS is likely to fall within 1.5°C to 5°C, with highest likelihood around 2.5°C.
The researchers integrate the data from these three strands of evidence to find that the earlier lower-bound estimate of a 1.5°C ECS is improbable. Cloud feedbacks will not likely cool temperatures, historical temperature increases are already approaching the lower bound, and mid-Pliocene warming that was higher than now occurred with lower levels of atmospheric carbon dioxide. On the other hand, both historical and paleoclimate evidence suggest that an ECS greater than 4.5°C is also unlikely. The researchers conclude that there is a 66 percent chance that the ECS is 2.6°C to 3.9°C; they offer a broader 2.3°C to 4.5°C range to cautiously account for alternative views, assumptions, and unknown unknowns.
If the current annual increase in atmospheric carbon dioxide of 2.3 ppm is sustained, it will reach double the pre-industrial level before 2090. Keep in mind that the difference between an ice age when continental glaciers buried about one-third of the global land area and today, when they cover around 10 percent, is a temperature increase of as little as 3°C.
For more background, see "What Climate Science Tells Us About Temperature Trends" and "Climate Change: How Lucky Do You Feel?"