Climate models help researchers predict how much Earth’s temperature will rise because of greenhouse gas emissions. However, even with the same starting parameters, different models can predict a wide range of potential warming. Becker and Wing provide new insights into why global warming predictions can vary so widely.
The new findings stem from the Radiative-Convective Equilibrium Model Intercomparison Project (RCEMIP). RCEMIP aims to enable comparisons among different climate and weather models by configuring them according to a simplified yet fairly accurate representation of tropical weather patterns. This mathematical simplification treats the tropics as a closed system in which radiative cooling and convective heating balance each other out.
The researchers analyzed climate sensitivity, or the amount of climate warming expected from an increase in greenhouse gas emissions, and found very different results for the 31 models included in RCEMIP. The results revealed that more than 70%–80% of the variation in climate sensitivity across models can be explained by differences in how the models simulate the influence of rising temperatures on shallow cloud cover and on how convection—which involves warm, rising air that forms thunderstorms—clumps together.
The team also found that global climate models may underestimate climate sensitivity, whereas cloud-resolving models tend to predict a greater degree of warming.
More work is needed to tease out exactly why the models produce the different warming-dependent changes in cloud cover and convection seen in the study, the researchers suggest. But they note that these findings could help researchers use radiative-convective equilibrium and other mathematical tools to reduce uncertainties in predictions of emissions-driven warming. (Journal of Advances in Modeling Earth Systems (JAMES), https://doi.org/10.1029/2020MS002165, 2020)
—Sarah Stanley, Science Writer