Reconciled: Climate sensitivity estimates between models and historical data

Such agreement is not trivial; for example, the Intergovernmental Panel on Climate Change's most recent Summary for Policy Makers notes that "No best estimate for equilibrium climate sensitivity can now be given because of a lack of agreement on values across assessed lines of evidence and studies."

While climate science has firmly established the role of carbon dioxide emissions in global warming, because climate models (which are based on paleoclimate reconstructions) suggest a bigger range temperature response to this process than do estimates from recent history, some have interpreted this to mean these models are overly sensitive -- that the ultimate amount of warming Earth will experience at a given concentration of greenhouse gases is less than some claim.

Here, Christian Proistosescu and Peter Huybers developed a method to parse different response timescales, or modes, within climate models. Using this procedure, they show that an important mode of warming associated with the Southern Ocean and the Eastern Equatorial Pacific takes a century or more to fully manifest.

Critically, such slowly evolving changes to climate, or "slow modes," those that ultimately account for the greater warming found in climate models, are missed when looking at the relatively brief span afforded by historical data, the authors report.

When they accounted for slow mode feedback in their longer-term analyses, the temperature range of climate sensitivity from historical data was consistent with the range reported by climate models today -- a range implying a higher sensitivity of climate to carbon dioxide emissions.

The authors note that the slow mode warming phenomenon implies that future changes may be distinct from past patterns, making it difficult to predict future warming by relying on past observations.