Tuesday, October 27, 2020

If you want some really worrying reading today...

...try this abstract from a new paper in Nature Geoscience:

Equilibrium climate sensitivity above 5 °C plausible due to state-dependent cloud feedback

The equilibrium climate sensitivity of Earth is defined as the global mean surface air temperature increase that follows a doubling of atmospheric carbon dioxide. For decades, global climate models have predicted it as between approximately 2 and 4.5 °C. However, a large subset of models participating in the 6th Coupled Model Intercomparison Project predict values exceeding 5 °C. The difference has been attributed to the radiative effects of clouds, which are better captured in these models, but the underlying physical mechanism and thus how realistic such high climate sensitivities are remain unclear. Here we analyse Community Earth System Model simulations and find that, as the climate warms, the progressive reduction of ice content in clouds relative to liquid leads to increased reflectivity and a negative feedback that restrains climate warming, in particular over the Southern Ocean. However, once the clouds are predominantly liquid, this negative feedback vanishes. Thereafter, other positive cloud feedback mechanisms dominate, leading to a transition to a high-sensitivity climate state. Although the exact timing and magnitude of the transition may be model dependent, our findings suggest that the state dependence of the cloud-phase feedbacks is a crucial factor in the evolution of Earth’s climate sensitivity with warming.


No comments: