Warming hiatus sheds new light on climate sensitivity

Between 2000 and 2012, the Earth’s surface warmed less than we might expect from the highly complex models used to project climate change.

Models taking part in the fifth Climate Model Intercomparison Project (CMIP5) suggested that temperatures would rise by around 0.21˚C per decade during 2000-2012, but in reality they only rose between 0.08˚C and 0.16˚C (depending on the dataset used).

What caused the slowdown in global warming?

Climate scientists have spent much time and effort investigating what caused this slowdown in temperature rise, known as the “global warming hiatus”. Several theories have been put forward, including changes in the solar cycle, volcanic eruptions, and issues with the datasets, but the main cause is thought to be the changes in the heat structure of the ocean driven by internal variability within the climate system. Internal variability refers to variations in the climate that result from processes intrinsic to the climate system such as the El Niño/Southern oscillation – year-to-year tropical climate variability, characterized by changes in sea-surface temperatures in the tropical Pacific.

These natural cycles cause ups and downs in the rate of global warming, which can temporarily mask or add to the human influence on global temperatures.

What role does climate sensitivity play?

Climate sensitivity – a measure of how much the Earth will warm as CO2 concentrations increase – is the subject of much research within the climate science community, and the hiatus has revived interest in the potential link between internal variability and climate sensitivity.

One notable recent study found that models with higher climate sensitivity also have stronger decadal variability, making a hiatus which is internal variability-driven more likely to occur if climate sensitivity is relatively high.

We primarily focused on the observed 2000-2012 hiatus, looking at what the probability is of seeing a temperature rise of 0.1˚C or less over the hiatus period in different climate sensitivity worlds – in other words, what level of climate sensitivity would give us the best chance of seeing the slowdown in temperatures witnessed during 2000-2012. To do so, we used a simple theoretical climate model, whose results we supported with those from the recent state-of-art CMIP6 climate models.

We found that the higher the climate sensitivity, the more likely it was that human-caused warming would be strong enough to mask any temperature trends caused by internal variability. So, the hiatus would be less likely to have been detected if climate sensitivity were high.

Probability distributions of decadal temperature trends during 2000–2012 compared to ECS. Lower ECS values (yellow) have a higher probability of showing temperature trends below 0.1˚C/decade.

Why are the oceans important?

The oceans play a key role in the climate system: the upper ocean layer adjusts quite quickly to rising temperatures, taking up heat, but the deep oceans absorb heat more slowly. The delay in the ocean response means that at higher climate sensitivities, human activity could cause temperatures to rise faster than any change caused by internal variability. This in turn makes it less likely that we would see a slowdown in warming if climate sensitivity were high.

Overall, we found that as climate sensitivity increases, the change in surface temperature due to human emissions is likely to be stronger than any changes caused by the natural cycles, making it less likely we would have been able to detect the 2000-2012 hiatus.

Although we might not yet be able to use the hiatus to estimate climate sensitivity, our results can be combined with other evidence to improve understanding of how the climate changes with increasing CO2 concentrations.

Angshuman Modak, postdoctoral researcher at Stockholm University.
The full paper is The 2000–2012 Global Warming Hiatus More Likely With a Low Climate Sensitivity, published in Geophysical Research Letters (open access).